View Full Version : PowerPC
Well I've been silently following this tread since it was back around page 2. I do have a few questions for the great and honorable Morpheus (and anybody else:) ).
1. Are the 970MP chips still on track or have they been hit by delays in the last few months? If I remember right you projected sometime Q1 2005 they would be in production.
2. What chips do you expect to be available to Apple for an update to the PowerMac around WWDC 2005, still 970MP or something else.
Thanks
gosh dave, i do miss the playhouse. the day i tried to go there and the link disappeared on me, a dark day indeed. made me sad, it did.
gosh dave, i do miss the playhouse. the day i tried to go there and the link disappeared on me, a dark day indeed. made me sad, it did.
Diablo Numero Uno, I presume? ;)
Diablo Numero Uno, I presume? ;)
Nope... that would be an invalid presumption... :lol:
and since when were maxwell's first molecule shuffler and i the same person? now, it'd much rather be associated with my dear irresponsible amigo than with l'il sheba (that one baffles me, of course it was otto), but still!
johnwillo
2004-11-01, 15:11
Monks, and the shades of monks, everywhere. The One Demon has not seen fit to (dis)grace his older haunts for some time. Not that I was ever inside the sacred cloisters.
Nope... that would be an invalid presumption... :lol:
It was a joke. I never expected that the guy was actually Steve Jobs. (Also joking) All this monk stuff is starting to piss me off, though. :p
SonOfSylvanus
2004-11-01, 17:05
</incoherent crap, cos I'm in the cool gang>
:devil:
:p
So DaveGee, how about a new home for us Monks? The locals here seem to be bothered and spork's such a pain in the ass that everybody's leaving there in droves. it's been what, 10 months since I was active there? and a good 9 since the monastery went bellyup. not like i got too much to say, but brother morpheus sure seems to!
Morpheus
2004-11-02, 01:04
Baha... Good to see you around... know what you mean re the playhouse... the Monks need a new monastery... any suggestions? Maybe here would be fine... should we ask for a room with a lockable door? Or maybe in Nick's diggs?
Morpheus
Morpheus
2004-11-02, 01:10
So DaveGee, how about a new home for us Monks? The locals here seem to be bothered and spork's such a pain in the ass that everybody's leaving there in droves. it's been what, 10 months since I was active there? and a good 9 since the monastery went bellyup. not like i got too much to say, but brother morpheus sure seems to!
Crossed posts Baha... :D
I guess the locals are still a bit shocked with my arrival, (not to speak of others) but it can also be a good think... I mean... raw and maybe suitable material if you catch my drift...
Morpheus
Look at this at IBM (http://www-132.ibm.com/webapp/wcs/stores/servlet/ProductDisplay?productId=8774209&storeId=1&langId=-1&categoryId=2586156&dualCurrId=73&catalogId=-840). A JS20 blade with 2x 2.2 GHz 970 processors.. with 1 MB L2 cache each. AFAIK, this must be the 970GX.
Please don't go away. As for me, I was just trying to have a little fun taunting Dave. He has always hung around the same haunts as I, but I'm a little jealous that he was allowed in the Monastery and I was not. Come to think of it, the last I remember, he was spreading the prophecies of Codename. Were the monks aware of this heresy? There I go again! Maybe I should just go back to my poor little orphan act from a few pages back...
Please forgive me, Brother Morpheus. Thanks ever so much for the kind serving of eClipz. It certainly brought a smile to the other orphan's faces as well. I haven't seen little Eugene so excited since the orphanage accidentally assigned a girl as his bunk mate.
I'm coming! I'm coming!
Are there any threads on the IBM Cell processor on Applenova? I had read a disturbing article accusing IBM of plotting against Apple by holding back the G5 while releasing the Cell processor to there own benifit.
http://linuxinsider.com/story/34994.html%5D
This chip sounds amazing, I'd love to learn more about it from the insiders here. tia
frontline lamb
2004-11-03, 20:04
I'll second that request. The takeover of the old place by partisan hacks got to me....
Quagmire
2004-11-03, 22:04
Two questions. Will the 970MP be trouble free on yields? Are yield problems will only be with the 970fx? Also, can the 970fx(GX?) fit in the 12" Powerbook footprint safely? I would take it as a no since there are only 2 Pbook numbers and not three.
FearlessLeader
2004-11-03, 23:32
I suspect that the existence of only two PowerTune references means that, as of a few months ago, Apple could only fit the G5 into two PB form factors.
So it seems likely that Apple has had an enclosure ready for a while, but has no G5s to put in it. That doesn't mean that, in the interim, that enclosure hasn't been updated, refined, or even shrunk a little while the chips have caught up to the design - or that other projects to fit the G5 into a smaller enclosure haven't also matured.
I would be surprised to see the G5 in anything smaller than the 15", but I am very curious about the changes we will see beyond just the G5 chip (and if it is just a 1.6 - 1.8 GHz G5 we are talking about, it will be those other changes that make a huge difference). Enough time has now elapsed so that what we see in January (if we see it then) may not be the G5 PB that was originally planned.
IBM and Apple are showing that they can make real progress on these chips _very_ quietly. But IBM's recent statements seem to be very confident, and I'm guessing they'll be under rather than over-promising at this point. I'm guessing that come January we will see more than we expect, but perhaps less than we would hope. And what could be more Apple than that? :)
wizard69
2004-11-05, 14:00
IBM and Apple are showing that they can make real progress on these chips _very_ quietly.
OK I had to ask; how can we know that they are making progress if they are being very quiet about it? As far as I know nothing has been publically said about the progress they have made.
Please don't misunderstand me though as I SUSPECT that progress is being made. That suspicion has a lot to do with the fact that the starting point (970FX) allows them alot of potential to improve things. But my suspicions are just that and I'm not convinced of anything beyond the fact that the industry moves forward or dies.
FearlessLeader
2004-11-05, 15:12
Hi Airsluf, and thank you!
What I meant by the first para was just that those two entries really only point to where Apple was at a few months ago. It could point to really good progress having been made on all the other elements of a G5 PB, but hopefully even more progress woud have been made since then.
Wizard69, I was making a general statement about IBM's quiet progress. Cool announcements, such as cached chips and big production hikes, that were unheralded in advance. From a big company like IBM, that seems to me to express a quiet confidence that they can get on with things and let the results speak for themselves.
I doesn't definitively point to any particular improvements, but hopefully it is suggestive of movement in the right direction.
Cheers!
Question on OSX. Isnt OSX only optimzed for 2 processors? If so, how dificult would it be to bring quad processor support to the OS? Would other programs that suport mulitprocessing like PS have to be recompiled?
Question on OSX. Isnt OSX only optimzed for 2 processors? If so, how dificult would it be to bring quad processor support to the OS? Would other programs that suport mulitprocessing like PS have to be recompiled?I don't know to what extent OSX's kernel is specially optimized for 2 processors but it is relative to just one processor.
OSX will probably out of the box run just fine on a machine with >2 processors but Apple can tweak the kernel to perform even better. Usually there's no need to recompile any apps to especially take use of 4 or more processors, the big step is for apps to make use of more than one processor and that usually is a larger job than just a recompile.
The problem is really the kernel, ie the lowest level of the operating system. It has to manage the traffic between the procesosrs, keep track of what everyone is doing, and balance the load.
wizard69
2004-11-12, 20:28
Question on OSX. Isnt OSX only optimzed for 2 processors? If so, how dificult would it be to bring quad processor support to the OS? Would other programs that suport mulitprocessing like PS have to be recompiled?
AS far as OS/X goes most of my information is second hand so take this with a grain of salt. Apparently OS/X has a kernel with a few large global locks, these can cause a process to wait a long time. Linux was in a similar situation but much effort was put into improving the lock situation in the latest kernel. In any evetn it does appear that Apple has considerable room to grow with respect to pefromance increases on SMP hardware. No specifics in the way of response but, would have to say working or not with 4 processors isn't a concern, what is is the potential to go father with what they have.
No evidence mind you but I suspect that OS/X could handle 4 processors now.
As to the programs run on the SMP MAC's that really depends.
Thanks
Dave
No evidence mind you but I suspect that OS/X could handle 4 processors now.There are strings in Darwin that indicate that OSX can handle 3, 4 or more processors. I have no idea why these strings are present if Darwin isn't cabable in some respect to make use of these comfigurations.
Programmer
2004-11-13, 00:19
This come up due to rumors of MS using 6 cores on their new xBox contained in two chips. I can't see the wisdom in IBM building both two and three core variants of the the 970.
What MS is doing has nothing to do with anything Apple is using. And the rumors you are reading have the details wrong.
Quagmire
2004-11-13, 10:26
Here are some notes about the PowerPC 970MP, 970GX, and the low power 970.
http://www.thinksecret.com/news/0411ppc.html
wizard69
2004-11-13, 20:38
There are strings in Darwin that indicate that OSX can handle 3, 4 or more processors. I have no idea why these strings are present if Darwin isn't cabable in some respect to make use of these comfigurations.
Well that phrase 'capable in some respect' pretty much sums it up. OS/X might be able to handle 4, or more for that matter, processors now, but that does not imply that the system will be well optimized to use them.
Is OS/X ready to scale to large multiprocessors like a SGI operating system? My geuss would be no. It is even less of a guess to think that Tiger is addressing its behaviour on multiprocessors greater than 2X.
Now people may think that I'm a bit crazy to even suggest that large multiprocessor systems are coming any time soon. Apples specific plans are anybodies geuss but I think the industry trend is clear. We are very likely to see systems at reasonable prices with 8 hard threads per machine and SMT raising the number of logical threads to 16 by 2006 if not sooner. For Apple getting to 4 processors should be a snap at the begining of next year. The i86 world should be able to deliver 8 processors early next year also, these on affordable systems.
The biggest problem with Apple is that they seem to be back pedalling with respect to SMP. Why they bothered to introduce the single processor PowerMac is a mystery to me. It would have made more sense to just wait until the dual core chip was at hand.
Thanks
Dave
....The biggest problem with Apple is that they seem to be back pedalling with respect to SMP. Why they bothered to introduce the single processor PowerMac is a mystery to me. It would have made more sense to just wait until the dual core chip was at hand.
Thanks
DaveWait... how does the one processor system show they are backpedalling on SMP? SMP has nothing to do with the systems available (well, ok, some systems need to have more than one processor, but not all of them) - it's simply programming support.
The one processor system was introduced as a low-end option. It has nothing to say about future support of multiprocessors, but rather that processors are expensive, and that there's a market for a lesser-cost machine.
I wouldn't be surprised if a single processor option remained even when dual core chips are available simply from a price standpoint.
Programmer
2004-11-14, 12:21
The biggest problem with Apple is that they seem to be back pedalling with respect to SMP. Why they bothered to introduce the single processor PowerMac is a mystery to me. It would have made more sense to just wait until the dual core chip was at hand.
I have to agree with Majost on this one -- Apple built the cheapest tower they could with virtually zero development effort by reusing the iMac chipset and only a single processor. Everybody is always demanding lower cost machines, and here they do a minimal effort move in that direction and you find it mysterious? Heh. Hopefully it sells well and that leads them to realize there is a market there that is worth a bit of development effort.
Quagmire
2004-11-14, 12:32
Yeah, it was a no brainer to re-intro the single 1.8 Ghz Pmac. People wanted an expandable mac at low cost. This computer must be the easiest developed machine apple ever did. Now apple has 4 choices when dual core comes out.
1. Kill(again) the single 1.8 Ghz Pmac and intro a single dual core monsters at the lowered price of $1799.
2. Kill(again) the single 1.8 Ghz Pmac and intro dual dual Chips at the price of $1799.
3. Keep the single Pmac and upgrade it to dual core 2.5 Ghz(whatever the lowend is) and have the dual dual Pmacs at $1999.
4. Keep the single Pmac and just keep it single core at the lowend speed and have the rest of the Pmacs at a single dual core CPU at $1799.
The price points are all speculation.
wizard69
2004-11-15, 14:34
Wait... how does the one processor system show they are backpedalling on SMP? SMP has nothing to do with the systems available (well, ok, some systems need to have more than one processor, but not all of them) - it's simply programming support.
"simply programming support"? That is a very simplistic way to describe SMP and frankly not all that accurate. First off SMP is a hardware approach to system design, it does not imply that all software running on the system supports greater than one processor. Software that does can of course benefit but the system as a whole benefits even if the majority of the software running on it is not multithreaded.
In any event Apple needs to push technology forward and continue to innovate. Multiple CPU's are one avenue which can provide significant performance enhancements. The quick adoption of this sort of technology does have everything to do with the systems available. Introducing a single processor tower at this point in the technology development cycle seems at the very least to be very misguided. Sort of like introducing a steam powered care just as the internal combustion engine is taking off.
The one processor system was introduced as a low-end option. It has nothing to say about future support of multiprocessors, but rather that processors are expensive, and that there's a market for a lesser-cost machine.
Since I'm one of those people not at all happy about Apples pricing I have nothing agianst a low end machine. The problem is that the 1.8GHz single tower is not a low end machine.
I wouldn't be surprised if a single processor option remained even when dual core chips are available simply from a price standpoint.
Well maybe for awhile but I can't see this being a long term trend. In fact I'd give the market about 6 months after the first dual core chips are introduced before we start to see a rather fast migration to 2X SMP systems across the board. For the average user such systems will provide a bigger payoff than buying the next generation single core processor.
Apple can either lead here or follow the rest of the market. We all know that Apple does not do well when in following mode. After all why by OS/X if it is not leading edge both platform wise and software wise. The platform of tommorrow is a multiprocessor, so yeah I see the 1.8Ghz machine as backpedaling. Especially in light of the other single processor machines Apple has delivered.
Thanks
Dave
Ok Morpheus, I take the red pill.
Show me how deep the rabbit hole goes! ;)
MWSF is approaching, any sweet news from the real world...?
turncoat
2004-11-23, 23:46
Nick convinced me to register here, and what do I find but old (and maybe current?) friends. I'll have to scour the postings by Web Ex-er ex-runcibles, though I regret the bad feelings about that old place. Can't we all get along?
Would that fallen first drop me a line.
Pesticidally and initially,
turncoat
Morpheus
2004-11-29, 03:43
Keep your eyes open November 30... There will be an announcement...
Morpheus
November 30? That's a Tuesday! :) Of course, the announcement is most likely iTMS in Canada :D Care to be more specific?
Morpheus
2004-11-29, 04:27
November 30? That's a Tuesday! :) Of course, the announcement is most likely iTMS in Canada :D Care to be more specific?
iTSM? wrong thread for that... :no: :no:
it will be internal to begin with, but I'm sure someone will find something...
Morpheus
Knowing Morpheus, it won't be an announcement from Apple but rather IBM, even if it is indeed a tuesday. Let's guess that it has something to do about Antares and/or Cell.
I was hoping to answer my own questions with this pressrelease from IBM, Sony and Toshiba (http://money.cnn.com/2004/11/29/technology/cell.reut/?cnn=yes) but it said nothing new. Odd time to release.. nothing!
10.3.7?
some type of listing inside?
I just saw a link from cnn.com to an article talking about the Cell processor from IBM/Sony. They are claiming a 10X speedup for content creation such as video. Does anyone have any info on Apple's ability to use this technology? Will Sony try and lock out Apple? Looks like the speed breakthough everyone's been waiting for. I'd assume a 10x speedup would basicaly be realtime everything for any PS/2D work no mater how big the file. Will these systems need to run on Lynix that Sony is producing???
http://money.cnn.com/2004/11/29/technology/cell.reut/index.htm?cnn=yes
Does anyone have any info on Apple's ability to use this technology?I think IBM, Sony and Toshiba will sell this technology to whomever wan't to buy it. It is intended to be THE new platform for evenry home appliance imaginable from toasters to super computers through PDAs, TVs, DVD players, game consoles and workstations. Sony, IBM and Toshiba can't possibly handle such an enourmous market on their own.
IBM is trying to expand their PowerPC eco system throug their Power Everywhere initiative and Linux. I think Cell will fit nicely into this vision. My most inner hope is that IBM, Sony and Toshiba will do their best to crack the Leviathan that is Microsoft and Intel. But the Cell architecture they are making is on an collision course with the wills of MS and Intel.
cell -
would apple buy it?
would they need it?
rumors of a home entertainment tivo type/music device early next year - dated out of the box?
http://www.macminute.com/2004/11/29/entertainmentserver/
10x! speed on a cell - throw a cell in a quad 970mp power mac - pretty hot machine.. :) (battery cells may not be included) :lol:
Quagmire
2004-11-29, 16:37
I think it might be the G5 mobile( aka low power 970) or the 970GX. The dual core version of Antares I think will ship in January having Pmac updates at WWDC. The 970GX might go in iMacs when updated. But, what do I know.
Not that I believe them but MOSR says:
"...but more recent reports from IBM sources have cast doubts on previous predictions which relied on the 970MP for future PowerMac updates. And the 970MP is not believed capable of four-way operation in its initial silicon."
Please, Morpheus, say it ain't true.
Morpheus
2004-11-29, 22:40
Not that I believe them but MOSR says:
.. And the 970MP is not believed capable of four-way operation in its initial silicon."
Please, Morpheus, say it ain't true.
Does anyone still listen to MOSR?
Morpheus
Quagmire
2004-11-29, 22:49
Does anyone still listen to MOSR?
Morpheus
I don't. There news is something that the guy gets from Macrumors, ts, ai, or us. The others are crap. Can you please tell me if I was close on my guess with the low power G5 or the 970GX tomorrow news or is it something that will be available immediately? I would suspect the X800 to come but, that doesn't have to do with the PowerPC.
Morpheus
2004-11-29, 23:09
Undertaker, you are off the mark.
Power.org (http://www.power.org/) (public on the second or third...)
Morpheus
Quagmire
2004-11-29, 23:23
Undertaker, you are off the mark.
Power.org (http://www.power.org/) (public on the second or third...)
Morpheus
Darn, the link says go to cadville.com, will it still be the same clue/place?
Morpheus
2004-11-29, 23:36
Darn, the link says go to cadville.com, will it still be the same clue/place?
they have vacated the place...
Morpheus
Here's the IBM release for the cell stuff: http://www-03.ibm.com/chips/news/2004/1129_cell1.html
IBM, Sony Corporation, Sony Computer Entertainment Inc. (Sony Corporation and Sony Computer Entertainment Inc. subsequently referred to as Sony Group) and Toshiba Corporation today unveiled, for the first time, some of the key concepts of the highly-anticipated advanced microprocessor, code-named Cell, they are jointly developing for next-generation computing applications and digital consumer electronics.
The four companies also announced that they would reveal technical details of Cell at the International Solid-State Circuits Conference (ISSCC) to be held from February 6-10, 2005 in San Francisco.
Specifically, the companies confirmed that Cell is a multicore chip comprising a 64-bit Power processor core and multiple synergistic processor cores capable of massive floating point processing. Cell is optimized for compute-intensive workloads and broadband rich media applications, including computer entertainment, movies and other forms of digital content.
Other highlights of the Cell processor design include:
Multi-thread, multicore architecture.
Supports multiple operating systems.
Substantial bus bandwidth to/from main memory, as well as companion chips.
Flexible on-chip I/O (input/output) interface.
Real-time resource management system for real-time applications.
On-chip hardware in support of security system for intellectual property protection.
Implemented in 90 nanometer (nm) silicon-on-insulator (SOI) technology.
Additionally, Cell uses custom circuit design to increase overall performance, while supporting precise processor clock control to enable power savings.
...
Cell provides a breakthrough solution by adopting a flexible parallel and distributed computing architecture consisting of independent floating point processors for rich media processing. Cell supports multiple operating systems, including PC/WS operating systems, as well as real-time CE/Game operating systems. In addition, the Cell processor is scalable and can be utilized in a variety of applications — from small digital CE systems within the home to entertainment applications for rendering movies, to scientific applications, such as supercomputers.
A team of engineers from IBM, Sony Group and Toshiba are collaborating on the design and implementation of Cell which is expected to deliver vast floating point capabilities, massive data bandwidth and scalable, supercomputer-like performance. The design work is taking place at a joint development lab the three companies established in Austin, Texas, after the project was announced in 2001.
IBM plans to begin pilot production of Cell microprocessors at its 300mm wafer fabrication facility in East Fishkill, NY during the first half of 2005. The first computing application IBM plans for Cell is the Cell processor-based workstation it is developing with SCEI.
Not only that, but IBM, Sony and SCEI power-on Cell processor-based workstation prototype (http://www-03.ibm.com/chips/news/2004/1129_cell2.html), too. Here's a particularly interesting quote about that:
IBM, Sony Corporation (Sony) and Sony Computer Entertainment Inc. (SCEI) announced today that they have powered-on the first Cell* processor-based workstation.
The prototype workstation is the first computing application planned for the highly-anticipated Cell processor.
The companies expect that a one rack Cell processor-based workstation will reach a performance of 16 teraflops or trillions of floating point calculations per second.
16 TF in one Rack!!! :eek:
But this all happened yesterday... and people have hinted at it before. So, what's new today?
Is there any technical reason (or business reasons) that would prevent Apple from using this technology? Windows cannot run on PPC, so would that leave Linux for the Sony workstation? It seems if Apple can use this tech, OSX would be the ultimate OS to run the cell!
Because Cell is heavily distributed, software will have to be modified to run well on the platform. It will require some effort, but hey, the 68k to PPC transition worked out fine, as did the Classic to X transition.
Anyone else up for another "short term pain, long term gain" move by Apple?
Cell supports multiple operating systems, including PC/WS operating systems, as well as real-time CE/Game operating systems. In addition, the Cell processor is scalable and can be utilized in a variety of applications — from small digital CE systems within the home to entertainment applications for rendering movies, to scientific applications, such as supercomputers.
Didn't directly see apple listed... would apple WANT to use this?
Why? (with power5's around the corner and of course 970mp)
-but perhaps for media creation and testing on a g5/g6?
Would this technology make its way into the next gen apple chips?
morpheus?
Didn't directly see apple listed... would apple WANT to use this?Apple usually don't show up on ANY lists hinting to new and exotic hardware. It's not their way to pre announce stuff like this. And.. Mac OS X _is_ a PC/WS operating system if I'm reading the acronyms correctly.
16TF in one rack -
I wonder how many chips?
$$$$$
20 chips X let's say $400 per chip (very high) = $8000 + $4000 to $8000 for subsystems and box?(i have no idea) = $16000+- for a 16TF computer!
WOW
Talk about a shift in power, the virginia tech computer with all of the xserves does a 12 to 14TF or in that range...
this is killing me not knowing if Apple will take advantage of this. Anyone have any info? I'll take wild speculation at this point...
Quagmire
2004-11-30, 14:47
That cell thing was released yesterday so it wouldn't be what morpheus is talking about. Also, I wonder if the supply problems will only be with the 970fx and not the new 970GX,970MP, and G5 mobile. Morpheus?
Programmer
2004-11-30, 17:13
I wonder if the supple (http://dictionary.reference.com/search?q=supple) problems will only be with the 970fx
You think the 970fx was too easily bent?
thegelding
2004-11-30, 17:29
could somebody explain the cell processor for us tiny brains?
thanks
g
Quagmire
2004-11-30, 18:26
You think the 970fx was too easily bent?
fixed. I meant supply.
cell - it's going to be 4 processors in one and cost under $100 according to an article i just read in USA today
FallenFromTheTree
2004-12-01, 00:04
More than one UNIX based OS like LINUX & LONGHORN?
And this too!
The processor, code-named Cell, will handle vastly more memory than today's consumer chips as well as
<<enable hardware-based copyright protection>> and allow multiple operating systems to run at the same time. It also will feature multiple cores, or logic engines, on a single die.
:no:
Morpheus
2004-12-01, 20:50
Ok, try now:http://power.org/
"Coming together to form the Power.org community are AMCC, Bull, Cadence Design Systems, Chartered Semiconductor Manufacturing, Culturecom, IBM, Jabil Circuit, Novell, Red Hat, Sony Corporation, Shanghai Belling, Synopsys, Thales, Tundra Semiconductor and Wistron."
mmm... where is Apple? Shouldn't they be in?
Morpheus :D
cell - it's going to be 4 processors in one and cost under $100 according to an article i just read in USA today
IBM is going to do all this for $100 but they can't make enough 970fx chips for Apple's needs. I find this cell chip a little hard to believe. In fact with the performance of IBM in producing 90 nm chips for Apple, I don't see how they can do half the things they are promising. Cell processors, Xbox 2, and I thought Nintendo too. In some ways it doesn't seem possible.
Quagmire
2004-12-01, 21:08
Ok, try now:http://power.org/
"Coming together to form the Power.org community are AMCC, Bull, Cadence Design Systems, Chartered Semiconductor Manufacturing, Culturecom, IBM, Jabil Circuit, Novell, Red Hat, Sony Corporation, Shanghai Belling, Synopsys, Thales, Tundra Semiconductor and Wistron."
mmm... where is Apple? Shouldn't they be in?
Morpheus :D
Apple will begin to use Cell processors? Or apple might go back to Moto?(shivers)
Apple will begin to use Cell processors? Or apple might go back to Moto?(shivers)
A third possibility is that Apple and IBM are working on something different than Cell. :confused:
mmm... where is Apple? Shouldn't they be in?What about Freescale, Nintendo, Microsoft and Toshiba? Don't they want to play with the other Power people?
What does this mean for us? Will we get more information from IBM regarding the future of PowerPCs?
from Power.org, december 2 news
"In a related product development, TimeSys, headquartered in Pittsburgh, PA, announced the availability of its TimeStorm Linux Development Kits for IBM PowerPC 750 FX and GX. Thousands of embedded Linux developers worldwide from a range of industries have chosen TimeSys solutions to streamline the development, customization and validation of homegrown or commercial Linux-based embedded systems."
750 GX PM in january? Morpheus....
Huh? Isn't a 750GX something of a G3 type variant?
Huh? Isn't a 750GX something of a G3 type variant?
That's right... I find it strange too... the news is dated dec 2 2004.
In fact I wrote "750 GX PM" but it would have been better "750 GX Mac".
I've always tought the GX variant of the 750 was nothing but vaporware!
I've always tought the GX variant of the 750 was nothing but vaporware!The 750GX has been shipping for some time AFAIK.
You might think of the 750VX that's pretty vaporous.
Apple skipped the 750GX and wen't directly to the G4, and that's probably why we havn't seen much of it in the Mac related press. It has larger cache than the 750FX and is clocked a bit higher and supports a 200 MHz bus. It's still a 130 nm part though and I'm amazed that we havn't seen a 90 nm 750 class processor yet. It'd be a killer in the embedded market.
The 750GX has been shipping for some time AFAIK.
You might think of the 750VX that's pretty vaporous.
Apple skipped the 750GX and wen't directly to the G4, and that's probably why we havn't seen much of it in the Mac related press. It has larger cache than the 750FX and is clocked a bit higher and supports a 200 MHz bus. It's still a 130 nm part though and I'm amazed that we havn't seen a 90 nm 750 class processor yet. It'd be a killer in the embedded market.
yep... my fault. with all this fx, gx , vx stuff I messed up the processor family referring the GX step of the 750 to the GX step of the 970...
Not bad... I've made 4 posts and 3 of them are total bullshit... :-))) nice record!
Not bad... I've made 4 posts and 3 of them are total bullshit... :-))) nice record!Haha :) No worries. For what it's worth the "GX" moniker seems to indicate some fabrication enhancements and increased cache and frequency . That's pretty much what we are hoping the 970GX will bring us.
thegelding
2004-12-02, 10:44
nice site...lots of hype...kinda hard for me to find the info i want...too dim i guess
here (http://www-03.ibm.com/technology/power/)
g
There is some speculation at a certain place (sorry name escapes me ;)) that Cell tech could work with the G5 as a sort of next generation Altivec. Possible? :confused:
Programmer
2004-12-03, 11:07
Considering we know practically nothing about the Cell except that at least some part of it uses the POWER ISA, it is a possibility. The Cell's vector processors aren't really like AltiVec -- they are seperate cores with a (presumably) different ISA. AltiVec is an extension to the PowerPC ISA and therefore runs as part of a PowerPC core.
If that isn't clear, try this: a processor core (or SMT thread) follows its own stream of instructions that controls what it does. The instructions that it understands are its ISA (instruction set architecture), i.e. a list of specific instructions that tell the core what to do. Adding AltiVec created the ability to insert a new set of instructions into the stream for the G4/G5 PowerPC core. Adding Cell vector processors adds additional cores that each have their own stream of instructions which they execute at the same time as the POWER core. The instructions found in the streams for the vector processors are different than those found in the POWER core's stream, just like a POWER core and an x86 core have different instructions in their streams. It is possible that IBM/Sony/Toshiba used some portion of the existing POWER instructions, but I doubt it as that would serve little purpose -- either a core understands the whole stream, or it will crash while trying to execute it (or emulate it really slowly).
What Cell vector processors and PowerPC AltiVec (and Intel's SSE/SSE2/..., and most modern GPUs) do have in common is 128-bit registers which can hold, among other things, 4 32-bit floating point numbers at a time. These registers are used with SIMD instructions, which stands for "Single Instruction Multiple Data" -- i.e. a single instruction does the same thing to all the data elements found in the register (i.e. the 4 32-bit floating point numbers). This is as opposed to conventional instructions which typically do one thing to one piece of data, i.e. a register that holds 1 value.
I ask for the pardon of technical readers for glossing over details and over-generalizing things.
...SMT thread... ISA (instruction set architecture)... instructions found in the streams for the vector processors...Intel's SSE/SSE2.... 128-bit registers which can hold, among other things, 4 32-bit floating point numbers...SIMD instructions, which stands for "Single Instruction Multiple Data"...the 4 32-bit floating point numbers....
I ask for the pardon of technical readers for glossing over details and over-generalizing things.
Do you also ask for the pardon of the non-technical readers, who don't know what the hey you are talking about? ;) I rememer early on in AN history, Mr. Gelding saying how much he missed Programmer: although g never really understood what he was saying, he felt smarter after reading his posts. I second that.
In any case, Programmer's post here is certainly more comprehensible (to me) than some others he has posted and, for that matter, some of unixguru's posts. I note that one of the 'guru's recent messages concluded as follows
ohci_hcd 0000:21:00.1: OHCI Host Controller
ohci_hcd 0000:21:00.1: irq 35, pci mem e000000080021000
ohci_hcd 0000:21:00.1: new USB bus registered, assigned bus number 2
usb usb2: Product: OHCI Host Controller
usb usb2: Manufacturer: Linux 2.6.5-7.97-pseries64 ohci_hcd
usb usb2: SerialNumber: 0000:XX:00.X
hub 2-0:1.0: USB hub found
hub 2-0:1.0: 3 ports detected
NET: Registered protocol family 10
IPv6 over IPv4 tunneling driver
st: Version 20040318, fixed bufsize 32768, s/g segs 256
eth0: no IPv6 routers present
eth1: no IPv6 routers present
And this is just a small sample of that post.
I suppose that we have people of all ranges of technical knowledge here on AN. I've been meaning to post a poll to try to get a better idea of "who we are", at AN, in that regard. I'll get around to it one of these days.
Ehh, the guru post was just a cut-n-past of the output from the dmesg command or boot.log file (some Linux distros have it). Nothing special except that this is some interesting hardware that just about nobody else has their hands on. Most of it isn't terribly relevant to anything we'll see in a Mac.
IBM and AMD have found a way to improve transistor performance by up to 24 per cent - without increasing the power draw - using a tweaked implementation of Big Blue's 'strained silicon' process.
The process is not only going to be applied to upcoming AMD64 and PowerPC chips, but is likely to underpin the production of Sony's 'Cell' CPU and future Macintosh computers.
They're calling the technique 'DSL' - you can read the full Register article here (http://www.theregister.co.uk/2004/12/13/ibm_strained_silicon).
They're calling the technique 'DSL'.OMG, couldn't they have come up with an acronym that wasn't allready well established in the computing industry? Are they running out of three letter abbreviations?
unixguru
2004-12-13, 19:08
Good news for a Powerbook G5 and faster desktops! I can see a 2 Ghz powerbook in Jan.
I can see a 2 Ghz powerbook in Jan.
Didn't the article say first half of 2005 that they would be starting. I'm guessing we won't see it until the WWDC 2005. But then again I would love to be proven wrong. Hopefully we'll see over 3 Ghz by then and dual core chips.
Morpheus
2004-12-15, 00:42
Ryan (http://www.macosrumors.com/121304.php) has been catching up with old newspapers (http://www.poughkeepsiejournal.com/projects/ibm/bu111404s2.shtml) again?
Morpheus
:no:
By the way, any new ppc rumor before christmas?Morpheus?
I've always wondered where IBM would manufacture the Xbox 2-processors? If East Fishkill have problems saturating the Mac market then they are certainly in trouble since Microsoft probably will need at least five times as many processors. The Cell processors that Sony and Toshiba will use won't be manufactured by IBM if i recall correctly. IBM might want to build their own Cells though.
Power5 Power Mac? (http://www.macworld.co.uk/news/index.cfm?RSS&NewsID=10439)
:lol: Either it's the best kept secret or I'll eat my hat if they show one in jan. for sale...
I, too thought what ever happened to the power5/980 chip and have been asking on this thread in an endless fashion...
ain't gonna happen till oct. or later.... my complete and total guess...
to hell with the 970mp :smokey: give me a quad 980! :lol: :)
*************************************
980 q3 of 2004 was the prediction before....
*************************************
now that i said oct. of 2005, it probably will come out in jan. :)
ablative
2004-12-17, 07:10
Power5 doesn't have VMX/Altivec. So I doubt Apple will want it, unless
they only want it for servers.
What I'm really waiting to see is if Apple uses Cell. They would be crazy not to. Power6/Eclipz is still a ways off I think, I don't know what else they could use in between the dual core 970 and the p6. Maybe Morpheus can drop some more hints?
oldmacfan
2004-12-17, 09:39
As of yet not much information is available about Cell. But from what I have read, it seems to me that this chip will serve many functions. Multi-media is the biggest one I see. What if Apple where to some how incorporate this chip into the Mac as a co-processor of sorts. Only to be used for certain operations and at the beckon call of the main processor/s.
The IBM tech interviewed by Ars hinted to a dual core derivative of Power5. H certainly didn't rule it out, and he pointed out that all high performance processors would be dual core from now on. I think that if IBM would strip a Power5 the same way that they stripepd Power4 to make 970, then they would get a pretty decent successor to 970MP, and that before we see something Cell/Power6 based.
Reading Morpheus hints of Power6 and what we know of Cell, it seems to me that Power6 is Cell based. It has been stated all along that Power6 would concentrate on extreme frequenzies and massively multi cored. To me, that equals Cell. They will arrive in about the same time frame too.
As of yet not much information is available about Cell. What if Apple where to some how incorporate this chip into the Mac as a co-processor of sorts. Only to be used for certain operations and at the beckon call of the main processor/s.
Could Cell be popped onto a PCI-X card? If so, dig audio/video could be handled by the card, no? Might work real well with DigiDesign, Avid, FC Pro.
dave
You can slap a CPU on any PCI-family card, you probably wouldn't be happy with the memory bandwidth related performance though.
oldmacfan
2004-12-18, 10:36
I wouldn't see a need to put it on a PCI card, unless it is the video card PCI-E x16 lanes ought to give it some bandwidth. Or you could put in on the motherboard with a HyperTransport 2 connection. That ought to keep it happy.
ryguy777@frontiernet.net
2004-12-18, 15:50
I've always wondered where IBM would manufacture the Xbox 2-processors? If East Fishkill have problems saturating the Mac market then they are certainly in trouble since Microsoft probably will need at least five times as many processors. The Cell processors that Sony and Toshiba will use won't be manufactured by IBM if i recall correctly. IBM might want to build their own Cells though.
IBM is building an addition onto their current 300mm fab... It's looking like they'll be doubling their fab and the project will probably be finished within the year (2005, I mean...)
Yeh, they might still have production problems in terms of quantities, but it's better than nothing...
IBM is building an addition onto their current 300mm fab... It's looking like they'll be doubling their fab and the project will probably be finished within the year (2005, I mean...)To me, it doesn't seem nearly enough.
The 970 isn't the only thing currently fabbed at EF. Doubling overall capacity more than doubles available 970 family capacity. And with the 90nm yield increases they have reported and stated they will expect in the future, a mere doubling in fab capacity should do them quite nicely.
wizard69
2004-12-20, 10:38
Evidence seems to be mounting that SMT in PPC will happen sooner or later. Just found this snippet in the kernel change log:
[PATCH] Create cpu_sibling_map for PPC64
<<<<<<<<<<<<<snipped>>>>>>>>>>>
SMT is becoming more popular and is
turning up in more and more architectures. I don't think it will be too
long until this feature is supported by most arches...
Now I don't knwo if that means anything is specifically happening in the near future with respect to PPC. It is interesting though that people working directly on the PPC kernel suspect that SMT will be wide spread soon.
Dave
Evidence seems to be mounting that SMT in PPC will happen sooner or later.There's SMT in POWER5 today and it's fully PowerPC compliant. As far as I know, there's no special Linux-version for POWER-processors, all are considered PowerPCs and rightfully so.
Evidence seems to be mounting that SMT in PPC will happen sooner or later. Just found this snippet in the kernel change log:
Now I don't knwo if that means anything is specifically happening in the near future with respect to PPC. It is interesting though that people working directly on the PPC kernel suspect that SMT will be wide spread soon.
Dave
That's about as newly interesting as reporting the sun will rise tomorrow. With Intel already having HT and POWER5 having SMT it doesn't take a rocket scientist to figure as long as you are adding SMT for one platform--get it ready in the most basic way for another as well. 'Cause it's probably coming somewhere down the line and it will be easier to add if it was considered up front than tacked in as an afterthought. it's coming. Inevitably. Just a matter of exactly when and which chip lineage.
There's SMT in POWER5 today and it's fully PowerPC compliant. As far as I know, there's no special Linux-version for POWER-processors, all are considered PowerPCs and rightfully so.
POWER is not 100% compatible with PowerPC. All PPC are based on POWER but POWER is not a PPC. Endian-ness is the first categorical deal-breaker there.
Born of an alliance between Apple, IBM, and Motorola (also known as the AIM alliance), the PowerPC was based on POWER, but with a number of differences. For instance, PowerPC is open-endian, supporting both big-endian and little-endian memory models, where POWER had been big-endian. The original PowerPC design also focused on floating-point performance and multiprocessing capabilities. Still, it did and still does include most of the POWER instructions. Many applications work on both, perhaps with a recompile to make the transition.
POWER is not 100% compatible with PowerPC. All PPC are based on POWER but POWER is not a PPC. Endian-ness is the first categorical deal-breaker there.
Well, the 970 does not run the earlier versions of Virtual PC. It does not have the hardware little-endian converter found in the G4. So the G5 is big-endian only from what I understand, just like the Power series. I don't think anyone else ever used the little-endian feature of the G4 PPC. I suspect Apple's future PPC processors will be big-endian only too. Anyway, this is what I've read. I'm no expert.
EDIT: Added a little more explanation.
PowerPC is a subset of POWER (with the exception of the PowerPC 601 which included the extra POWER instruction sets - trivia), you are right, but anyone who uses the non-PPC instruction sets available in POWER is one trippin' foo'.
POWER is not 100% compatible with PowerPC. All PPC are based on POWER but POWER is not a PPC. Endian-ness is the first categorical deal-breaker there.The POWER5 is fully PowerPC compliant. The POWER processors have all had the full PowerPC ISA (32 and 64 bit) since POWER3 (in the beginning called PowerPC 630) in 1999. That's more than we can say for regular PowerPC processors since they all lack the 64 bit instruction set (all but 970). Before POWER3 IBM used the *Star processors (RS64-I to -IV) in its RS/6000 line and they were too all fully 32/64 bit PowerPCs. POWER (1990), POWER2 (1993) and P2SC (1995) are more of a curiosity now days than anything else, but it's true, they are not PowerPCs.
The 970 lack little endian mode, but this bi endian stuff is no part of the PowerPC specification if my sources are correct even if some G3 and G4 type PPCs do have this functionality.
This is so much hair-splitting and semantics. There is more to a PPC chip than the PPC ISA, or you couldn't call an implementation with VMX/Altivec a PPC, but we do. In the narrowest and most correct sense the POWER ISA is a true superset of the PPC ISA, but a POWER processor is not a true superset of a PPC processor as there are many things that are architected in and onto PPC silicon which are not architected or designed into POWERx silicon. We have met the land of once pure technical terminology being co-opted by the markeeting department as "close enough", which it is until it leads to confusion back at the technical discussion.
wizard69
2004-12-21, 19:04
In any event it look more and more like we will be getting Power5 derived technology soon in the "PPC" line. SMT and a Hypervisor look like a go. The question is when.
Dave
Here's another doom and gloom article about IBM sticking it to Apple. Not saying I believe it, but I just wanted to help cool down the Macworld frenzy. :D The end of the world as we know it. (http://www.macnewsworld.com/story/38633.html)
Here's another doom and gloom article about IBM sticking it to Apple. Not saying I believe it, but I just wanted to help cool down the Macworld frenzy. :D The end of the world as we know it. (http://www.macnewsworld.com/story/38633.html)
Very old news. Note the tagline at the bottom of the story: "This story was originally published on July 8, 2004, and is brought to you today as part of our Best of ECT News series". This guy didn't know squat before and time hasn't been any kinder to his technical obliviousness.
not sure if this is a re-hash or any new info... came out today: Thursday, January 06, 2005 3:37 a.m. ET
NANOTECH RESEARCH
Nanotechnology refers to the study of atom-sized materials no bigger than a few thousandths of a human hair across -- the scale at which modern computer chip production increasingly operates.
http://wireservice.wired.com/wired/story.asp?section=Breaking&storyId=971300&tw=wn_wire_story
IBM, infineon, Sony, Toshiba, Chartered, AMD and Samsung are announcing (http://www.state.ny.us/governor/press/year05/jan5_5_05.htm) that they are investing 1.9 billion in new facilities in East Fishkill. IBM, Infineon, Chartered and Samsung have already announced that they are collaborating on 65 mn technology and beyond. Really nice to see AMD joining in.. Are we seeing a future PowerPC/Cell licencee here?
Thread is spot on - Go Morpheus - dual doors of the drama unfold.
oldmacfan
2005-01-16, 12:09
Morpheus, you seem very quiet as of late.
colonelforbin
2005-01-16, 15:25
Morpheus, you seem very quiet as of late.
yes, i'd be interested to hear your thoughts on the feasibility of the digitimes report.
yes, i'd be interested to hear your thoughts on the feasibility of the digitimes report.
As would I. Also is there anything new in the area of processors for the PowerMacs, advancements or setbacks I'll take anything.
Morpheus
2005-01-18, 03:08
Morpheus, you seem very quiet as of late.
??? (http://us.movies1.yimg.com/movies.yahoo.com/images/hv/photo/movie_pix/walt_disney/monsters__inc_/waternoose.jpg)
Morpheus
??? (http://us.movies1.yimg.com/movies.yahoo.com/images/hv/photo/movie_pix/walt_disney/monsters__inc_/waternoose.jpg)Let me be the first to say: Huh?
Bucolic Old Sir Henry
2005-01-18, 04:16
Henry J Waternoose (Monsters Inc):
"today's children aren't easily scared, and profits in the scare industry are shrinking. Waternoose feels the pressure mounting, and he is desperate to ensure that his family's company survives". (Source) (http://www.tiscali.co.uk/events/2002/monsters/cast_waternoose.html)
I think the God Of Sleep and Dreams may be a little busy.
Dream a little dream for me...
Pip pip!
Ghstmars
2005-01-18, 04:36
morpheus?
Six legs! G6 this summer!!! W00T!
:D
Screed
propropro
2005-01-18, 09:15
Morpheus predicted the Powerbook G5 1.6 and 1.8 GHZ (and possibly 2.0 GHZ) for January, maybe he would like to tell us if the keeps his predictions ;)
propropro
2005-01-18, 09:17
6 legs... 6th month?? June?
oldmacfan
2005-01-18, 10:34
??? (http://us.movies1.yimg.com/movies.yahoo.com/images/hv/photo/movie_pix/walt_disney/monsters__inc_/waternoose.jpg)
Morpheus
When your ready.
Those bastards at Pixar are hogging all the G6's? ;)
Quagmire
2005-01-18, 14:41
It is too early for the G6 IMO. So I think the 6 means June or means a day like Feburary 6th. I will be pushing for june meaning the G5 pbook at 2 Ghz.
naw - g6 is here! powerpc 980 - quad with spoilers :) It was supposed to be here last year ..... :lol:
From AppleInsider:
Nov 20 2003
02:10 PM EST
PowerPC 980 based Power Macs not too distant
Apple and IBM have a contract for at least 4 generations of the PowerPC processor over the next 5 years, sources tell AppleInsider. The first of these microprocessors is the current PowerPC 970, which will be followed by the PowerPC 980 in Quarter 3 of next year.
The PowerPC 970 is said to max out at approximately 2.6GHz, shortly after the release of the second generation Power Mac G5s, reported on yesterday. The Power Mac product line will reportedly continue with the 'Power Mac 8,2' in late 2004. These units will sport the 90nm PowerPC 980, which Apple may brand the 'G6' according to a single source.
Ranging between 2.8 and 3.2GHz out the start gate, the PowerPC 980 will deliver the 3GHz speeds promised by Steve Jobs and ultimately top-out at over 5GHz on a 65nm process. In addition, the PowerPC 980 is said to implement hypertransport directly.
The 980 PowerPC enabled 'Power Mac 8,2' is projected to provide speed gains of over 40 percent when compared to a current Power Mac G5 of equivalent clock speed, and will use a crossbar switch to speed communication between processors. Performance gains will be even greater for applications written to take advantage of hyperthreading, sources said. The units will utilize the 'U4' system controller, which will allow front side buses to reach speeds of up to 2Ghz in its first iteration.
The PowerPC 980 will reportedly go through 3 revisions during its 18 month life-cycle, as opposed to the PowerPC 970 which will see only 2 revisions during a 12 month life-cycle in pro desktop
It is too early for the G6 IMO. So I think the 6 means June or means a day like Feburary 6th. I will be pushing for june meaning the G5 pbook at 2 Ghz.
Isn't Feb 6 Superbowl Sunday? So I doubt there is anything going to be released that day. June is more likely. Maybe the Power 6 is going to be ready in June?
" Waternoose (Intel) feels the pressure mounting, and he is desperate to ensure that his family's company survives". (Source)
This sounds like something Intel is feeling and is probably more logical than the 6 legs thing IMHO. After all the 5 eyes could be Power 5 based G5's.
Another idea.
3 + 3, dual core 3 Ghz G5.
This would probably make Intel crap it's pants if IBM + Apple started producing PowerMacs with these in them in the next month or two. I doubt AMD would like it either.
It might also explain the PowerMac 9,1 in the plist files.
Isn't Feb 6 Superbowl Sunday? So I doubt there is anything going to be released that day. June is more likely. Maybe the Power 6 is going to be ready in June?
Sounds like the perfect time to me to launch a new product with a hardcore ad campaign.
Make the most out of whatever the next version of the switch campaign turns out to be.
6 cores? at 6 AM? on the 6th day? of the 6th month? of 2006? :smokey:
It might also explain the PowerMac 9,1 in the plist files.Mac Mini maybe?
Morpheus: I believe it is our fate to be here. It is our destiny. I believe this night holds, for each and every one of us, the very meaning of our lives. This is a war and we are soldiers. What if tomorrow the war could be over? Isn't that worth fighting for? Isn't that worth dying for?
Give us another hint....
Quagmire
2005-01-18, 16:23
Morpheus gave us enough info from that pic. We can pretty much guess accuratly what it means. So we have the 6 legs probably meaning the 6th month(June), and 5 eyes meaning probably Power5 based G5's. Now, what would be interesting is that the G5 in the pbooks will be Power5 also. Or could this mean that the Power5 PPC 9?? could start production and means that it will have to be in the rev d Powermacs? Also, will it be dual core? That we would have to wait and see.
When we've colonized our sixth planetary system, obviously.
propropro
2005-01-18, 18:00
Powerbook G5 (eyes) over June (6 legs) ? :confused:
I wouldn't be susprised if it was released in June as many people (included our friend Morpheus) thought it was plausible a PBG5 in January...
But when June arrives, I'll buy the Powerbook that's selling at that moment, I'm very tired of this waiting... :grumble:
Mac Mini maybe?
I doubt it because it is bundled with the G5 based Macs like the PowerMac, xServe and iMac. It could be a G5 eMac but it's a bit early for that. I'm more thinking the PowerMac goes dual core with PCI-e which would require a motherboard revision big enough to warrant the new listing.
checksum
2005-01-18, 19:37
3rd generation. Befitting vest.
5 eyes, mouth sewn shut?, 6 fingers on each hand (+ thumb) with what would be the 3rd finger on the left hand extended, 6 legs, overall color BLUE.
Is this IBM giving Apple the finger?
Or Revelation's Beast.
Dave
oldmacfan
2005-01-18, 22:11
People, you are not thinking like Morpheus. There was a question asked and that question was answered.
next question(s) asked:
Does the monster figure relate to release date? or chip type? or other?
Do you really believe it is our fate to be here, that it is our destiny?
People, you are not thinking like Morpheus. There was a question asked and that question was answered.
Ok lets try it your way. I'll be the first since I was the last question before he replied so it might be a reply to my question about how the PPC's were advancing in regards to the PowerMacs. He then gave us the image.
What do we know about this image and where it was from:
He is from Monster's Inc which was produced by a company owned by Steve
- a connection to Apple (I know it's obvious but might be important)
He was in charge of a company that was having a difficult time but later dominated the market (if I remember the movie right)
- sounds like IBM in that the 90 nm process wasn't yielding well but now they are like in the movie
has 5 eyes
- Could be a symbol that the next PPC is based on Power5
has 6 legs, three on each side
- Dual core at 3 Ghz 3+3 but this is open to interpretation
- could also be a time frame, 6 days, 6 weeks, 6 months, 6th month etc
has 7 fingers counting thumbs
- could also reference to time frame 14 days (7 + 7)
This is what I'm seeing anyway, not so much based on the physical character but in the situation the character was in. The physical parts might just happen to fit in. Looks like it's time for a trip to Blockbuster
Let the flaming begin :lol:
Uh-oh, I don't like this at all...
Henry J. has been around for 50 years but now feels threatened - I don't think we can ignore that bit of symbolism. (If what the image represents is as important as the physical characteristics.)
Henry is now training a protege; why do I fear that this protege isn't Apple? Because IBM spun off their PC division? Because Apple isn't one of the 16 signatories to Power.org?
Or ... ??
Dave
Programmer
2005-01-19, 01:30
Are you guys all numerologists, or what?
Henry J Waternoose represents the boss of a scary company. Perhaps Morpheus is the one who is scared and this is why he has been quiet lately?
Indeed. That the boss is also a Pixar character wasn't lost on me either.
Are you guys all numerologists, or what?
Henry J Waternoose represents the boss of a scary company. Perhaps Morpheus is the one who is scared and this is why he has been quiet lately?
Nah, that makes too much sense. ;)
oldmacfan
2005-01-19, 10:56
Ok lets try it your way. I'll be the first since I was the last question before he replied so it might be a reply to my question about how the PPC's were advancing in regards to the PowerMacs. He then gave us the image. Let the flaming begin :lol:
Hello, go back and read his post, he quoted a specific question. :confused:
:lol:
??? (http://us.movies1.yimg.com/movies.yahoo.com/images/hv/photo/movie_pix/walt_disney/monsters__inc_/waternoose.jpg)
Morpheus
so so Morpheus, you seem frightened by the boss of Pixar.
You don't want to get the same action that Nick de Plume ?
:o
satansam
2005-01-19, 11:58
if you really want to look deep into this;
he has 5 eyes and 6 legs. symmetry. so halved would make 2.5 and 3.
throw in speculation of dual core which gives the PB range a clock speed of 2.5ghz and 3ghz, halved that would give the single CPU speeds (1.25ghz and 1.5ghz).
juxtapositioning; his head is at the top (duh) and would be lighter and smaller. 12"? and his legs are big, strong looking and heavy. 15+17"?
could we see dual core? hmmm ;) nah
of course i doubt that mr. Jobs intentionally created this guy as a rumor for new processors a few years later... but who knows?
Morpheus' response was in answer to a question about PowerMacs, not PowerBooks.
Dave
Programmer
2005-01-20, 01:16
Actually it was more likely to be in response to what he quoted:
Morpheus, you seem very quiet as of late.
Ghstmars
2005-01-20, 02:53
so then everything is still on schedule?970mp and 970gx coming out second half
of the year?
An interesting read:Cell Architecture Explained (http://www.blachford.info/computer/Cells/Cell0.html)
Just found this via The Inquirer (http://theinquirer.net/?article=20847): Slide (http://pcweb.mycom.co.jp/photo/news/2004/11/29/011bl.jpg) from ISSCC (edit: this is concerning Cell).
"Technology features:
4.6GHz clock speed
1.3V operation
85 degree C operation using a heat sink
6.4 Gbp/s off-chip communication"
Screed
oldmacfan
2005-01-21, 11:20
Just saved a hard copy of the report. Now I'll have some light reading this weekend.
The beloved and trustworthy ;) MacOsRumors site stated on saturday january 16:
"Is Apple developing a counterpoint to the Mac Mini? A new high-end workstation Mac with a "full-height" tower case, more expansion room than a PowerMac, and early-adopter features like DDR2-533 memory, PCI Express, as well as IBM POWER5-class processors has been spied in prototype form by reliable sources."
"...reliable sources"...
When I smoke too much I tend to trust my magic 8 ball too... :smokey:
oldmacfan
2005-01-24, 11:18
When I smoke too much I tend to trust my magic 8 ball too... :smokey:
They don't have a magic 8-ball, they have a mood ring. :lol:
"Is Apple developing a counterpoint to the Mac Mini? A new high-end workstation Mac with a "full-height" tower case, more expansion room than a PowerMac, and early-adopter features like DDR2-533 memory, PCI Express, as well as IBM POWER5-class processors has been spied in prototype form by reliable sources."
THIS WOULD BE GREAT!
A real response to the users - two optical drives (or more) and space for many more hds and how about 5(?) PCI slots (for many uad-1 and powercores) - Very Nice if True....
Morpheus
2005-01-24, 16:08
Uh-oh, I don't like this at all...
Henry J. has been around for 50 years but now feels threatened - I don't think we can ignore that bit of symbolism. ...
Dave
30
:smokey:
30
:smokey:
30 Cores?!?!?!!!! ;) :p :lol:
Morpheus
2005-01-24, 19:01
next question(s) asked:
Does the monster figure relate to release date? or chip type? or other?
other
Do you really believe it is our fate to be here, that it is our destiny?
yes
Morpheus
Programmer had it right. The ogre, Henry J., is WHY Morpheus is keeping mum of late. And being more cryptic than usual. The particular ogre in question has been around 30 years, not 50.
Morpheus
2005-01-24, 22:25
Henry J Waternoose represents the boss of a scary company.
True (who has acused us of having no sense of humor?)
... and something else. :smokey:
Perhaps Morpheus is the one who is scared and this is why he has been quiet lately?
The Lord of Dreams is not scared, but overworked...
Morpheus
Morpheus
2005-01-24, 22:30
so so Morpheus, you seem frightened by the boss of Pixar.
Note: the world does not revolve around TSO. (Oops,.. maybe now I should be afraid... )
Morpheus
Bucolic Old Sir Henry
2005-01-25, 02:30
One wonders whether TSO follows on from GPUL...
In my dreams, perhaps.
Pip pip!
When I smoke too much I tend to trust my magic 8 ball too... :smokey:I always thought you were supposed to smoke too much while snorting your magic 8 ball... maybe LoCash could clarify?
OBTW, welcome back Morpheus. WTF is TSO, BTW?
;)
Note: the world does not revolve around TSO.
TSO - The Smart One (http://en.wikipedia.org/wiki/The_Smart_One) - A modern sect with TSO as its main deity. Fits quite well with our iCEO.
TSO - Time Sharing Option (http://en.wikipedia.org/wiki/Time_Sharing_Option) - Functionality in IBM's OS/390 and z/OS.
hobbit.2
2005-01-25, 05:52
Note: the world does not revolve around TSO. (Oops,.. maybe now I should be afraid... )
Hmmm. I'm sure The Scary One disagrees. ;)
30
30? Lets' see. Apple was officially formed on April 1st 1976 (http://apple2history.org/history/ah02.html). That's 29 years ago. However Steve and Steve worked on the Apple 1 for quite a while before that. It's safe to assume Jobs was involved from 1975 already...
30 is fine then. :)
Morpheus
2005-01-25, 06:00
OBTW, welcome back Morpheus. WTF is TSO, BTW?
;)
TSO: The Supreme One, aka, HS: His Steveness.
:D
Morpheus
Morpheus
2005-01-25, 06:05
30? Lets' see. Apple was officially formed on April 1st 1976 (http://apple2history.org/history/ah02.html). That's 29 years ago. However Steve and Steve worked on the Apple 1 for quite a while before that. It's safe to assume Jobs was involved from 1975 already...
30 is fine then. :)
:rolleyes:
mmm... (http://www.sourcewatch.org/wiki.phtml?title=Microsoft)
Morpheus
Morpheus
2005-01-25, 06:19
OBTW, this one from MOSR:
"The first 65-nm 970's will be arriving in Cupertino soon, but 90-nanometer will be the only viable technology for the majority of 2005 so the problems are considerable. We should know more in the days ahead...."
:eek: :lol:
Who provides banned substances to Ryan?
Morpheus
OBTW, this one from MOSR:
"The first 65-nm 970's will be arriving in Cupertino soon, but 90-nanometer will be the only viable technology for the majority of 2005 so the problems are considerable. We should know more in the days ahead...."
:eek: :lol:
Who provides banned substances to Ryan?
Morpheus
Are you implying no 65-nm's have reached Cupertino yet? Total buzzkill. After taking another hit :smokey:, I now understand you mean to imply that Ryan is wrong because the 65's are ready to roll. ;)
Completely throwing this out as a super wild guess... I always though that in the name 970mp - that the mp meant mobile processor? I'm guessing the true meaning was meant as multiple processor?
And could it be that the 65nm 980 shows it face? FOR REAL :)
OR
Could it be that the cell really will be 65nm - based on the powerpc - and will someway end up in a mac or laptop?
speculation anyone :)
Morpheus: You think that's air you're breathing?
Actually, the "mp" did indeed imply multi-processor from the very start of the Antares (http://www.thinksecret.com/news/antares.html) rumors.
Screed
propropro
2005-01-26, 08:23
News of December from Thinksecret ( http://www.thinksecret.com/news/0412powerpc.html ) :
Confidential documents pertaining to the development of IBM's future single-core PowerPC 970GX and dual-core 970MP processors, code-named Antares, obtained by Think Secret, reveal that Apple has already received a number of prototypes of the processor. Apple is also expected to wrap up Process Verification Testing of the 970MP around the end of January, meaning systems featuring the processor could be ready as early as the Spring including, possibly, the long-awaited PowerBook G5. However, sources continue to report that IBM is working to develop other low-power 970-series processors topping out around 2GHz that could find their way into future PowerBooks instead. Antares will debut at 3GHz.
Maybe this has been already discussed, but I just wanted to know what do you think of this:
1. Could a 970 MP really be put in a Powerbook??!! :eek:
2. Am I the only one that think the Low-Power 970 isn't coming anytime soon? I just think this because of the lack of any news about it, but hopefully I'm wrong...
Morpheus, let's play some mind games ;)
oldmacfan
2005-01-26, 10:28
Morpheus, let's play some mind games ;)
He is already playing mind games on you... :lol:
FallenFromTheTree
2005-01-26, 21:37
Uh Guys, look what I found over at OWC :-)
ATI X800 Mac edition
256 MB single slot card! Happy Happy Joy joy
I'm sitting here watching Graphics and Media State of the Union from WWDC'04 (http://stream.qtv.apple.com/events/jun/wwdc_2004_qt_sotu/wwdc_2004_gm_sotu_ref.mov) and 34 mins into this session the topic is Core image. The presenter talks about Core Image as a "stream based processing model" and "mappable to GPUs, SMT CPUs and any other parallell architectures".
OK.. this got me going on several points:
- Stream based processing model. Last I heard, this is how you describe the Cell and its APUs witch are called a "stream processor".
- GPUs, CPUs and any other parallell architecture? Huh, what other parallell architecture might one have for rendering graphics if not the CPU or GPU? Core Image is certainly optimized for AltiVec, but isn't that part of the CPU? Again.. The APUs in Cell is not considered a part of the CPU but are co-processors, and is certainly some "other parallell architecture"
- SMT CPU? Last I heard no CPU in any Mac to date have symmetric multi threading. Pentium4 have, POWER5 also, but AFAIK neither will ever run Tiger.
Oh well.. this might be me just finding what I'm looking for..
Altivec is a vector processor that does excellent work with streams. Whether the 970 evolves Power5-ish capabilities or an explicit Power5 derivative with Altivec continues the family, SMT is on the way to the desktop and it's just a matter of exactly when.
Programmer
2005-01-28, 11:00
;)
Sometimes it amazes me how long it takes people to notice things.
ablative
2005-02-04, 09:26
I've heard the big W is essentially tri P, Morpheus -- I think you know what that means.
What I don't know is why it is posted in this forum, my understanding is that Apple has nothing to do with this thing. Should I assume that since you mentioned it here that Apple is now involved? I can see why they might want it, but I thought that not everything in it is their desired flavor. I think the original P+x*S would be more up their alley.
eh?
I've heard the big W is essentially tri P, Morpheus -- I think you know what that means.
What I don't know is why it is posted in this forum, my understanding is that Apple has nothing to do with this thing. Should I assume that since you mentioned it here that Apple is now involved? I can see why they might want it, but I thought that not everything in it is their desired flavor. I think the original P+x*S would be more up their alley.
eh?
:confused:
I've heard the big W is essentially tri P, Morpheus -- I think you know what that means.
What I don't know is why it is posted in this forum, my understanding is that Apple has nothing to do with this thing. Should I assume that since you mentioned it here that Apple is now involved? I can see why they might want it, but I thought that not everything in it is their desired flavor. I think the original P+x*S would be more up their alley.
eh?
I just took this as a reason why Morpheus was busy, not necessarily an implication that Apple was involved. If you realize what this is, you also know that signicant lateness is not an option.
Were you guys always so vague at that other place? :D
So Waternoose is Tri P?
Tri-P?? Isn't Xbox 2 supposed to have 3 processors?
Waterhouse could easily be Steve Ballmer.:lol:
How is that for wild speculation.
Programmer
2005-02-04, 23:11
P is for Power?
I know what S is, but I'm not telling.
Morpheus
2005-02-05, 00:19
What I don't know is why it is posted in this forum, my understanding is that Apple has nothing to do with this thing.
Correct, Apple has nothing to do with it. I take the thread as PPC (or even Power), not only Apple related :-)
Morpheus
P is for Power?
I know what S is, but I'm not telling.
I'll take "SPU" for 100, Alex
Bucolic Old Sir Henry
2005-02-05, 02:59
Ok. We have a three core Power processor, and a Power core with x stream processing units attached (which is what Cell is). So Apple is doing something with cell.
But then I know feck all about computer processors, although I do do the Telegraph crossword.
Pip pip!
OBTW, this one from MOSR:
"The first 65-nm 970's will be arriving in Cupertino soon, but 90-nanometer will be the only viable technology for the majority of 2005 so the problems are considerable. We should know more in the days ahead...."
:eek: :lol:
Who provides banned substances to Ryan?
Morpheus
Let's hope that this quote, also from MOSR, is also 'tainted':
"What the grapevine now envisions is a mixed bag -- an entry-level system based on single and dual 2GHz PPC 970 processors, a mid-range machine based on the 2.5GHz 970FX, and a high-end machine with either two 2.8GHz 970GX's or a sole dual-core 2.8GHz 970MP."
...and this one too:
"Oh, and "one last thing": according to reliable sources at ATi Technologies, Apple has bought a considerable volume of new 8X AGP graphics cards for OEM use including Radeon 9800s and X700s. This suggests PCI Express may have to wait for the next revision, because there has been nary a peep about PCI Express cards for Apple OEM.... "
FallenFromTheTree
2005-02-05, 19:44
Let's hope that this quote, also from MOSR, is also 'tainted':
"What the grapevine now envisions is a mixed bag -- an entry-level system based on single and dual 2GHz PPC 970 processors, a mid-range machine based on the 2.5GHz 970FX, and a high-end machine with either two 2.8GHz 970GX's or a sole dual-core 2.8GHz 970MP."
...and this one too:
"Oh, and "one last thing": according to reliable sources at ATi Technologies, Apple has bought a considerable volume of new 8X AGP graphics cards for OEM use including Radeon 9800s and X700s. This suggests PCI Express may have to wait for the next revision, because there has been nary a peep about PCI Express cards for Apple OEM.... "
I understand your viewpoint, but it seems more likely that we would see
a graphics card update before any major processor revision.
PCI-Express is now current technology, while dual core is still in developement.
Either way Apple needs to improve the available pro user upgrade options to remain competitive with other workstation class desktops.
I understand your viewpoint, but it seems more likely that we would see
a graphics card update before any major processor revision.
PCI-Express is now current technology, while dual core is still in developement.
So you're suggesting we're TWO revisions away from dual core (first for PCIe, then for dual core)? Or are you saying they'd be concurrent?
FallenFromTheTree
2005-02-06, 08:03
So you're suggesting we're TWO revisions away from dual core (first for PCIe, then for dual core)? Or are you saying they'd be concurrent?
Dave, I wish I really knew! :-(
My gut feeling tells me that we are more likely to see a minor PowerBook type revision first.
We simply can't have PowerBooks with more base RAM than a G5!
The 1.67 GHz 15" & 17" PB models also now offer a BTO ATI 9700 128MB graphics card capable of running the 30" display.
So the Pro line has a bit of catching up to do.
We can only hope that these two issues would be corrected very soon.
Any speed bumps or price drops would also help to move the existing
chassis configuration without the need for major re-tooling.
Upgrading to 3 GHz, PCI-Express, SLI and Dual Core are far more complicated issues and those may have to wait till WWDC, MWSF and beyond.
I can only hope that all would be included in any major revisions.
Reality is that most of us are SO tired of waiting
that we'll probably leap at any reasonable improvement.
Programmer
2005-02-06, 12:24
Ok. We have a three core Power processor, and a Power core with x stream processing units attached (which is what Cell is). So Apple is doing something with cell.
Just because these things may exist doesn't mean Apple is necessarily doing anything with them. IBM builds the POWER5 server chips too, but Apple doesn't use those.
Of course it doesn't rule it out either.
Dave, I wish I really knew! :-(
Reality is that most of us are SO tired of waiting
that we'll probably leap at any reasonable improvement.
I've decided that I'll be strong - stupid? - and wait for dual core. :grumble:
checksum
2005-02-06, 15:30
Morpheus, should we expect to hear from Trinity soon?
W(h)ither "Morpheus"?
--
I've decided that I'll be strong - stupid? - and wait for dual core. :grumble:
Do what I did put the new Mac money into a CD, I can't touch mine until after WWDC, with the huge amount of interest on in, I might go for an upgrade, like a mouse pad.
FallenFromTheTree
2005-02-06, 17:00
Do what I did put the new Mac money into a CD, I can't touch mine until after WWDC, with the huge amount of interest on in, I might go for an upgrade, like a mouse pad.
Yeah, my bank offers a whopping 1.67% to tie up my money for 3 months.
Funny how that works. :rolleyes:
Originally Posted by Jay
Do what I did put the new Mac money into a CD, I can't touch mine until after WWDC, with the huge amount of interest on in, I might go for an upgrade, like a mouse pad.
So you seriously don't expect anything before WWDC... :eek: :eek: :eek: :eek:
Please... anybody... tell me it's wrong... :(
Do what I did put the new Mac money into a CD, I can't touch mine until after WWDC, with the huge amount of interest on in, I might go for an upgrade, like a mouse pad.
You should have invested in AAPL. ;)
You should have invested in AAPL. ;)
I have some but I was late to the party at $55.15. Anyway it's to easy to get at, I don't even have to leave my computer screen.
Anyway I don't want to buy it until the summer, and they maybe announced in March (which I suspect) but will probably be shipping in June.
Bucolic Old Sir Henry
2005-02-07, 03:48
Just because these things may exist doesn't mean Apple is necessarily doing anything with them. IBM builds the POWER5 server chips too, but Apple doesn't use those.
Of course it doesn't rule it out either.
It's a bit like a buffet, then. IBM lays out its wares, and up come the clients. The ones with the biggest plates get to pile most on.
I wonder how big Apple's plate might be?
Pip pip!
Costique
2005-02-07, 07:35
Just because these things may exist doesn't mean Apple is necessarily doing anything with them. IBM builds the POWER5 server chips too, but Apple doesn't use those.What about stream processing and parallel architectures which Core Image/Video aim at? Cell sounds so good, but can it (I mean the tech itself) really be used in PCs instead of the traditional CPU design? The thing that really bugs me is that nobody ever said Apple is involved. Some say they wish Apple were involved or just find it logical.
I wonder if the following is possible. Cell is indeed highly customizable, which makes it usable anywhere. Apple is not involved in the development because they have no resources, i.e. they have nothing to contribute to the project. Meanwhile, they develop software support for Cell, extend the OS, reimplement frameworks. When IBM is ready for mass production, here comes Apple with a huge order for customized Cell hardware.
The thing that really bugs me is that nobody ever said Apple is involved. Some say they wish Apple were involved or just find it logical.
Did IBM ever say they were building a stripped down Power4 for Apple? I think if Apple is somehow involved in Cell development, they certainly want nobody to know about that untill they've got a product ready.
I wonder if the following is possible. Cell is indeed highly customizable, which makes it usable anywhere. Apple is not involved in the development because they have no resources, i.e. they have nothing to contribute to the project. Meanwhile, they develop software support for Cell, extend the OS, reimplement frameworks. When IBM is ready for mass production, here comes Apple with a huge order for customized Cell hardware.
I find that scenario highly likely. Sony, Toshiba and IBM should all be very eager to license this technology to whomever wants to use it. It just fits into the core application model of the Cell technology that they would want as many as possible to adopt Cell. IBMs strides to make an open community arround their Power effrorts hints to this also. IBM have seen the benefits of making a large and open base of Power, and that should be true in regards of Cell too. Even more so.
But.. Sony, Toshiba and IBM might want to get a head start and be the early adopters that show the world how insanely great this technology is and reap the initial benefits of this and get a strong foothold into a potential new market. They might not want to bring Apple along in the beginning as Apple carries a certain stigma. Apple might steal the limelight, and some might not even want to use Cell if there is some connection to Apple (since the only makes mice with one button).
But.. if I was Apple I would just sit back, let Sony, Toshiba and IBM do all the hard work and then just tag along for the fun of it.. 'caus I think Cell will be really really great, and certain applications of it fits just right into Apple's products.
Costique
2005-02-07, 10:00
Which leads to the main question: can Cell be used instead of a traditional CPU and what sort of changes to kernel/memory management/etc. would that require?
Did IBM ever say they were building a stripped down Power4 for Apple? I think if Apple is somehow involved in Cell development, they certainly want nobody to know about that untill they've got a product ready.You're right, of course. I was thinking that Apple is almost the only customer for PPC970, which is easier to conceal. Cell is, like, Trinity = IBM + Sony + Toshiba, a project involving engineers from at least 3 huge companies, a lot of patented stuff, which is broader.
This is beginning to feel like a twilight zone episode...
Is any thing going to ever be released?
Product:
PowerMac
Last Release:
June 09, 2004
Days Since Update:
243 (Average = 176)
Deeper into the Twilight Zone...http://www.macsimumnews.com/index.php/archive/2619/#comments :D
:err: The idea of cell actually scares me.... :no:
checksum
2005-02-07, 17:24
This CELL or another: opinions will invariably oscillate, right?
A: http://www.qualia.sony.us/movies/qualia_phase2_high.mov - "that's that’s both eclectic and artistic in nature"
B: "The idea of cell actually scares me ..."
C: "What's happening right now may be beyond anyone's understanding, that goes for all of these properties at first." [don't think "spotlight" or "qualia" are some sort of accidental motivational codenames.]
And so on.
The formation of these various alliances (OSGi, GLOBUS, Power.org, etc.) is encouraging, as they will tend to regulate each other in interesting ways. As long as they won't coalesce into one big crowd.
Side note to self:
1. Remember Le Bon - crowds are reactionary (translation: "stupid")
2. Never do anything without a clear purpose.
3. Get in touch with reality: go for a good walk.
--
Programmer
2005-02-08, 00:04
Which leads to the main question: can Cell be used instead of a traditional CPU and what sort of changes to kernel/memory management/etc. would that require?
Well considering Sony is talking TVs & game consoles, and IBM is talking about workstations and supercomputers, I'm sure the Mac could fit in there somewhere. They say it has a Power core and it can run Linux (and is OS neutral). That bodes pretty well for being able to run MacOS X. There would probably be some OS kernel and driver changes necessary, but that isn't unprecidented for a new PPC processor. The new SPU cores would need software support, but they are the kind of thing that Apple is creating the CoreAudio/CoreImage/etc APIs to support.
Frankly I think Apple would be silly not to take this seriously. Lets see, 2 cores at maybe 3 GHz, or 9 cores at 4 GHz. I don't know, what do you think?
:err: The idea of cell actually scares me.... :no:
Wasn't there a whole saga about the horrors of Cell on Dragonball Z ? :lol:
wizard69
2005-02-08, 02:26
Well considering Sony is talking TVs & game consoles, and IBM is talking about workstations and supercomputers, I'm sure the Mac could fit in there somewhere.
I hope it fits in somewhere. Maybe not the Cell we are hearing about tonight but a variant at the very least.
In any event that PPC core just about has my pants wet.
They say it has a Power core and it can run Linux (and is OS neutral). That bodes pretty well for being able to run MacOS X.
This is what we are hearing, a full PPC core in there someplace. I do hope that accessibility to the rest of Cell will be good. I'd like to see the market move back to opening up hardware so that things like Linux can fully exploit what is there.
There would probably be some OS kernel and driver changes necessary, but that isn't unprecidented for a new PPC processor. The new SPU cores would need software support, but they are the kind of thing that Apple is creating the CoreAudio/CoreImage/etc APIs to support.
Apple could have OS/X up on the PPC core in a very short time considering how the chip looks at the moment. It probally wouldn't take them long to get Core Audio working on one of the SPE/U's either. The SIMD units may not be AltVec but it appears that AltVec instruction streams could map into them pretty quick.
Frankly I think Apple would be silly not to take this seriously.
Maybe CoreAudio/CoreImaging is Apples way of telling us that they are taking Cell seriously.
Lets see, 2 cores at maybe 3 GHz, or 9 cores at 4 GHz. I don't know, what do you think?
Allright.. Cell has a Power core with VMX (AltiVec?), mutithreading and running faster than 4 GHz. Seems pretty much like Nr.9 was very wrong in what he said was a fact in the industry. The question I'm asking is if the +4 GHz is erfering to the processor as a whole or some part of it. Is it really the Power core that's suddenly decided to increase its speed two fold, or is is just the SPUs that's runing that fast? When I first saw the figure "4.6 GHz" i immediately thought that this was the speed of the streaming processors and that the core was half of that. a Power core running in 2.3 GHz isn't that much of a deal.
FallenFromTheTree
2005-02-08, 04:59
Skimming through this info left me with the impression that the G5 processor
would still be the controller or you might say the primary CPU.
With that in mind, I then considered the possibility that each of the dual G5 processors could then be used to control each of the 8 Cell cores.
So right or wrong, my impression is that the cell cores might then be used as a secondary GPU rather than the primary CPU.
If cell was used to process graphics, video and sound, then
what happens to PCI, PCI-X and PCI-Express ?
I'm very confused
Either way, my Mac fund is going back in the bank. :rolleyes:
Costique
2005-02-08, 05:02
They say it has a Power core and it can run Linux (and is OS neutral). That bodes pretty well for being able to run MacOS X. There would probably be some OS kernel and driver changes necessary, but that isn't unprecidented for a new PPC processor. The new SPU cores would need software support, but they are the kind of thing that Apple is creating the CoreAudio/CoreImage/etc APIs to support. Frankly I think Apple would be silly not to take this seriously. Lets see, 2 cores at maybe 3 GHz, or 9 cores at 4 GHz. I don't know, what do you think?If Cell does provide the claimed performance increase over, say, PPC970, then Apple simply does not have a viable alternative. If the damn thing can indeed run Linux, then it can run Mac OS X too.
Cell has a Power core with VMX (AltiVec?), mutithreading and running faster than 4 GHz.I don't think so. If I understood Hannibal (http://arstechnica.com/articles/paedia/cpu/cell-1.ars/2) correctly:
Finally, the instruction set for the SPEs is not VMX compatible or derivative, because its execution hardware doesn't support the range of instructions and instruction types that VMX/Altivec does.Anyway, this can be solved with a well-designed framework abstracting the hardware.
The SPEs might not be VMX-compatible (i won't question Hannibal on this topic) but the Power core does have VMX as stated by Sony's spec-sheet (http://www.scee.presscentre.com/imagelibrary/detail.asp?MediaDetailsID=25555). I wan't to have it confirmed that this VMX is indeed AltiVEc comptatible. It seems to me that this Power core (sans SPEs) is a Power5 derivative, quite suitable for Mac use in its own right, and such it would be interessting to know if it is the Power core that is running in +4 GHz or if it's some other component in the Cell processor.
Baron Munchausen
2005-02-08, 07:12
FallenFromTree,
I read the article in the same way as you did - a dual core hub processor controlling 8 cell elements being handed tasks.
It would not surprise me if we see an initial 'cut' with just a couple of
cell cores to get the platform out there, but a multi-core device would be sensible as a home theatre platform that can handle multiple HDTV streams at once.
I wonder if Cell support spawns OS/XI...
Costique
2005-02-08, 07:57
The SPEs might not be VMX-compatible (i won't question Hannibal on this topic) but the Power core does have VMX as stated by Sony's spec-sheet (http://www.scee.presscentre.com/imagelibrary/detail.asp?MediaDetailsID=25555).Yes, you're right. I've missed it somehow. So Cell appears to be a PowerPC-like core + x vector-processing units, rather than x PowerPC-like cores. This is good, IMHO, because: a) they mainly leave the core alone and b) SPEs don't have to be fully VMX/Altivec-compatible since they are just additional computational units.
BTW, Ars forums do seem enlightening, though it's hard for me to understand everything.
wizard69
2005-02-08, 20:41
That is what Cell appears to be at this time. I'm expecting another report today (hopefully) or Ars that will go into more details. If the core is indeed running faster than 4 GHz this could be a very good thing indeed.
As to Nr.9, I will repeat in thought what I've said before. 1. The 970 in all its appearances seems to have been a rushed job to get a product to market for Apple. Little tought was put into power usage. 2. Many manufactures have indeed gotten excellent power scaling by going ot 90nm. You do have a serious problem with static power being a large proportion of the total power due to leakage. That however does not mean that different organizations have not attempted to address the issue or not meant with success. It is certainly a case that many currrent processes have bumped up against a wall of sorts, that just gives a vendor the incentive to enhance the process.
As to the PPC, it does appear from the Sony announcement that it does indeed run that fast. It will be interesting to see how they accomplished that and how much it impacts performance. Personally I'm very interested in the power performance of this chip, it could be that the chip is impractical to run at high speed even if it is reliable. Or this core could be very energy stingy and ideal for a fast laptop. I can't help but to think that this core will show up elsewhere, possibly without the Cell baggage.
Dave
Allright.. Cell has a Power core with VMX (AltiVec?), mutithreading and running faster than 4 GHz. Seems pretty much like Nr.9 was very wrong in what he said was a fact in the industry. The question I'm asking is if the +4 GHz is erfering to the processor as a whole or some part of it. Is it really the Power core that's suddenly decided to increase its speed two fold, or is is just the SPUs that's runing that fast? When I first saw the figure "4.6 GHz" i immediately thought that this was the speed of the streaming processors and that the core was half of that. a Power core running in 2.3 GHz isn't that much of a deal.
I have been reading that it is the SPEs, or SPUs, that run at 4 GHz or over. The PPC core, which appears to be a Power5 derivative, runs at just 1 GHz. These are second hand quotes, however. Anyone see this in a reliable report? :confused:
The PPE core is reportedly not a Power5 derivative, but an extension of the "1GHz PPC" project from 5-6 years ago and presented at this same conference in 2000. A project from days when 1GHz was still a long way off in PPC land. I have also read reports the PE only runs at 1GHz and the SPEs run at 4GHz.
Programmer
2005-02-08, 23:06
This is not correct -- the whole chip runs at 4 GHz (well, if you can believe their claims it well). The in-order dual issue, long pipeline design of the Power core is clearly designed to get maximum clock rate. The Power core may have been an offshoot of the "1 GHz PPC" project, but it has far exceeded the initial demonstration.
I don't doubt the PPE is vastly superior to the old demo core. There seems to be more than a few conflicting reports on its speed though and I don't know what to take to the bank yet, hence the "read reports" method of passing along tidbit info. Are you saying you have a definitive source on the whole chip being 4Ghz? Anything definitive at this point is better than the scatterblasts all over the web right now, although the variance is reducing slowly.
Hannibal's Part II (http://arstechnica.com/articles/paedia/cpu/cell-2.ars) is up, fyi.
Don't know if you saw this....
STI =
STI cell processor
Next generation processors
Just as the cells in a body unite to form complete physical systems, a "Cell" architecture will allow all kinds of electronic devices (from consumer products to supercomputers) to work together, signaling a new era in Internet entertainment, communications and collaboration.
The Vision:
Breakthrough microprocessor architecture that puts broadband communications right on the chip.
Markets:
·**
Next-generation communications
·**
Consumer multimedia applications
STI cell processor defined
Two years ago, Sony and Toshiba and IBM (STI) announced that they had teamed up to design an architecture for what is termed a system-on-a-chip (SoC) design. Code-named Cell, chips based on the architecture will be able to use ultra high-speed broadband connectivity to interoperate with one another as one complete system, similar to the way neural cells interoperate over the brain's network.
Market demand for STI cell processor
IBM expects Cell to define an entirely new way of operating. Cell's underlying architecture will enable it to manifest itself into many forms for many purposes, helping to open up a whole new set of applications. Incorporating this architecture, chips will be developed for everything from handheld devices to mainframe computers.
IBM strategy with STI cell processor
IBM has an unmatched history and capability of building custom chips and believes the one-size-fits-all model of the PC does not apply in the embedded space; embedded applications will require a flexible architecture, like Cell. Cell also brings together, for the first time, many leading-edge IBM chip technologies and circuit designs developed for its servers.
STI cell processor benefits
Cell will take advantage of IBM's most advanced semiconductor development and process technologies.
http://www-1.ibm.com/businesscenter/venturedevelopment/us/en/featurearticle/gcl_xmlid/8649/nav_id/emerging
Morpheus
2005-02-10, 21:00
Don't know if you saw this....
Just as the cells in a body unite to form complete physical systems, a "Cell" architecture will allow all kinds of electronic devices (from consumer products to supercomputers) to work together, signaling a new era in Internet entertainment, communications and collaboration.
Yes, Virginia. Re-read the above paragraph as many times as needed until it sinks in.
(Very few in the press have)
:)
Morpheus
- so my ps3 and mac will interface to give me 'cell' power
- and if they can give me more computer 'cell' power, then apple could use the cell!
and "work together"
"signaling a new era in Internet entertainment"
"communications and collaboration."
and it's all run on apple software? as the president of sony told steve 'just make software not hardware'
Morpheus
2005-02-11, 02:28
...
and it's all run on apple software? ...
Apple software? Who mentioned Apple software? :grumble:
Morpheus
:lol:
Apple software? Who mentioned Apple software? :grumble:
Morpheus
:lol:
So basically its all over for Apple, right? All we get is the pukey Antares. Cell destroys it and everything else in it's path. :eek: Sometime in 2007, when Cell running Longhorn has dominated, Apple will get another shot. 666 :devil:, indeed. :lol:
Apple software? Who mentioned Apple software? :grumble:
Morpheus
:lol:
Power PC ISA, plus additional sets of instructions should run Cell? For IBM, Sony, Toshiba would that be Linux? Wouldn't it be possible for BSD to be run on Cell??
I swear, I am pertetually in a constant state of confusion. :confused:
checksum
2005-02-11, 11:40
Just as the cells in a body unite to form complete physical systems, a "Cell" architecture will allow all kinds of electronic devices (from consumer products to supercomputers) to work together, signaling a new era in Internet entertainment, communications and collaboration.
--
"Yes, Virginia. Re-read the above paragraph as many times as needed until it sinks in. (Very few in the press have)
Morpheus"
--
... And when you're done, cycle through A-C on the [2005-02-07, 17:24] post, until C really kicks in. In 2020 perhaps.
checksum
--
So basically its all over for Apple, right? All we get is the pukey Antares. Cell destroys it and everything else in it's path. :eek: Sometime in 2007, when Cell running Longhorn has dominated, Apple will get another shot. 666 :devil:, indeed. :lol:
These are interesting times. Maybe we get Cell in Macs in place of Antares, or in addition to Antares? Cell's PPE has a full Power implementation, including the vector unit, so it should run code for OS X and its applications. Existing code will not benefit from all Cell has to offer, however, until the application's code is updated. If Cell is indeed in Apple plans, I'm sure Tiger will be ready for it.
gyroscope
2005-02-11, 13:35
There is a lot of information about Cell at RealWorldTech (http://www.realworldtech.com/page.cfm?ArticleID=RWT021005084318), but I can't reach the site right now. WTF ? :confused:
wizard69
2005-02-11, 13:48
Very interestig times! However I'm not sure about the need to update all existing applications. I could see Apploe setting up system services such as sound to be handled by one of the SPE's. That could provide a significant boost for some application right away.
At one point I though we heard about an Appple employee commenting on future machine being able to take advantage of Core Audio and Video in a dramatic fashion. Many assumed that they where talking about exploitation of Video cards. I'm willing to bet the thought was more along the lines of Cell or a Cell like processor coming down the pike.
As to the PPE, this is possibly the most interesting part for Apple users. Its real world performance would make or break Cell for Apple users. It is interesting that information seems to be embargoed with respect to this processor core. Since PPC is nothing new one would have to think that other parties have an interest in this CORE.
DAVe
These are interesting times. Maybe we get Cell in Macs in place of Antares, or in addition to Antares? Cell's PPE has a full Power implementation, including the vector unit, so it should run code for OS X and its applications. Existing code will not benefit from all Cell has to offer, however, until the application's code is updated. If Cell is indeed in Apple plans, I'm sure Tiger will be ready for it.
wizard69
2005-02-11, 13:58
His article was very good but unfortunately still leaves out a lot material with respect to exactly what the machine is capable of. I think we will have to wait even longer for a definitive explanation of the processor and the capabilities of the PPE and the SPE's.
There is more there than first thought, in the SPE's, but it still is not clear as to how capable they are. It would be nice to have a few manual to download ;) In any event there seems to be more compatibility with PPC and altvec in the cores than first expected.
Dave
There is a lot of information about Cell at RealWorldTech (http://www.realworldtech.com/page.cfm?ArticleID=RWT021005084318), but I can't reach the site right now. WTF ? :confused:
Bucolic Old Sir Henry
2005-02-11, 16:11
Earlier in this thread, Morpheus broke the news about the 970GX and 970MP, sampling last year and probably in production now. We have also heard suggestions that a "low power" laptop chip was in testing last year (at rather unexciting clock rates), but we have heard very little more than that. Now we have a huge amount of speculation about Cell, and what it might mean for Apple/OS X. In particular, there seems to be some sort of consensus that Tiger (with Core Audio, Core Image, etc) is pretty well placed to take advantage of "cell" attributes.
So here's my question: do the chips coming down the pipe have cell-type features that require Tiger in order to really fly? That would mean that getting Tiger out of the door was a pre-requisite for the release of new machines - PowerMacs and PowerBooks in particular.
I'm no expert, but Apple has clearly been in the loop on cell development (witness the VMX/Altivec unit in the cell), and could easily have cherry-picked cell features for its own chips. This might be particularly true for the laptop chip. Given the relatively low clock rates quoted for last year (1.8GHz tops, was it? - not much of an advantage over the G4 there), does it perhaps have two low-power cores (Power5 derived?), with cell-like SPUs tacked on. That would mean it could deliver real power at low clock rates.
Once again, I put myself in the hands of the God of Dreams...
Pip pip!
Programmer
2005-02-12, 14:06
I'm no expert, but Apple has clearly been in the loop on cell development (witness the VMX/Altivec unit in the cell)
That is an awfully big and unwarranted assumption. If Sony wanted IBM to add a good vector unit to the PPE, why would IBM invent a new one instead of just using the one they already had lying around? Alternatively, perhaps this Power core will show up in an Apple chip that is not a Cell chip and naturally Apple wanted VMX... so IBM had already added it anyhow to the core going into the Cell.
I think a lot of people are unclear on this: this Power core is not the Cell processor. The Cell processor is an on-chip network with a collection of devices -- in the first implementation there is 1 Power core, 8 SPEs, a memory controller, and an IO module or two. That Power core could be used alone as its own chip, or in a different multi-chip configuration more like the POWER4/POWER5, and that would not be a Cell chip.
From the Real World Technologies article:
"Members of the CELL processor family share basic building blocks, and depending on the requirement of the application, specific versions of the CELL processor can be quickly configured and manufactured to meet that need."
Maybe the chip that was presented is not intended for Apple. But it seems pretty obvious to me that the G6 will be a member of the cell family.
Bucolic Old Sir Henry
2005-02-12, 15:00
That is an awfully big and unwarranted assumption.
And we never make those, do we?
Alternatively, perhaps this Power core will show up in an Apple chip that is not a Cell chip and naturally Apple wanted VMX... so IBM had already added it anyhow to the core going into the Cell.
Which was my point, really. What might Apple have asked for, and for what machines?
I think a lot of people are unclear on this: this Power core is not the Cell processor.... in a different multi-chip configuration more like the POWER4/POWER5, and that would not be a Cell chip.
I understand that. My own feeling is that Apple won't be using the chip unveiled this week, but the presence of the vector unit suggests that they will use some "flavour" of cell: what about 2 cores+VMX, 4SPUs. Clocked well below 4GHz, that could be a low-power chip with real grunt.
Pip pip!
morpheus - laughing cause it's true?
i figure what does apple have to bring to the table? software + osx+ quicktime? so, sony commputers running mac osx and perhaps playstation? and ibm with low end servers and (others) running osx etc.. Toshiba gets ... just the chip or mac osx and benefits of quicktime for their tvs
then the cell format could fly.. it needs a os and software base to take flight...
morpheus - laughing cause it's true?
i figure what does apple have to bring to the table? software + osx+ quicktime? so, sony commputers running mac osx and perhaps playstation? and ibm with low end servers and (others) running osx etc.. Toshiba gets ... just the chip or mac osx and benefits of quicktime for their tvs
then the cell format could fly.. it needs a os and software base to take flight...
I think we will see the starts of the software evolution in Tiger with Core Audio/Image/Video. It seems like those are things that could almost immediately make use of Cell.
wizard69
2005-02-12, 22:47
That is an awfully big and unwarranted assumption.
Well I wouldn't go that far as we aren't even clear as to what IP Apple owns with respect to AltVec/VMX/Whatever.
If Sony wanted IBM to add a good vector unit to the PPE, why would IBM invent a new one instead of just using the one they already had lying around? Alternatively, perhaps this Power core will show up in an Apple chip that is not a Cell chip and naturally Apple wanted VMX... so IBM had already added it anyhow to the core going into the Cell.
I suspect this core is what Apple is waiting for to put into a 64 bit laptop and maybe a few of their lower cost machines. I geuss it would be an open question with respect to actually using CELL. The only thing we have to go on is Apple touting dramatic performance increases in the near future to support "core" functionality. Many people took this to mean GPU acceleration, but that could very well have been a leap of faith. Sure it is an equal leap of faith to suggest that Cell has a future in Apple hardware, but this new PPC core seems to be no leap at all at the moment.
I think a lot of people are unclear on this: this Power core is not the Cell processor. The Cell processor is an on-chip network with a collection of devices -- in the first implementation there is 1 Power core, 8 SPEs, a memory controller, and an IO module or two. That Power core could be used alone as its own chip, or in a different multi-chip configuration more like the POWER4/POWER5, and that would not be a Cell chip.
I will agree with you totally here, there does seem to be alot of misconception out there with respect to Cell. One can not underestimate how important its modular structure is. With respect to Apple I would have to say it is pretty clear that they could easily leverage a cell with 2 PPE and a fewer number of SPE's.
Dave
FallenFromTheTree
2005-02-13, 08:16
The problem for me with all this, is trying so wait while wondering
how much improvement we might see realistically in the nearest amount of time.
Days, weeks, months?
Major, minor revision when? Argggggh!
So it looks like we might see a small PowerBook type bump any day now or
have to wait till they release Tiger, with hope that they would also
debut a "new improved" PowerMac capable of serious new head turning benchmarks.
Right now June seems like an eternity.
I'm losing it. :err:
Right now June seems like an eternity.
I'm losing it. :err:
Perhaps you could have yourself put into a coma for 5 months. Imagine waking up, apparently tomorrow, and being able to watch the WWDC Keynote in Quicktime 7 to see in your dreams have come true. Then again, from Morpheus' hints, you might want to stay under until 2007. ;)
As I once said, the revolution always seems to be 2 years away.
FallenFromTheTree
2005-02-13, 15:32
I could always use a nice long nap. ;)
macgeek2004
2005-02-13, 16:09
The PowerPC first came out a LONG time ago with the PowerPC 601.
Even now, 10 years later apple's newest computer is still running on a PowerPC proccesor.
I think that very soon, like end of 2006 the latest, apple's new chip will be released. Something like the Super PowerPC, or sumthin like that.
PowerPC is gettin old.....
. . . PowerPC is gettin old.....
How old is Unix? It just keeps getting better in the right hands. :)
How old is Unix? It just keeps getting better in the right hands. :)
UNIX was developed at Bell Labs between 1969 and 1974, by Ken Thompson and Dennis M Ritchie, mainly as a response to the failed MULTICS OS. Most of the grunt work was done in 1969, on a borrowed DEC PDP-7. Bell Labs, however, wasn't impressed, and only agreed to keep funding the development if the two programmers came up with a text deditor for them. They duly did, and the rest is history.
- Campbell-Kelly, Martin and Aspray, William, 2004: Computer: A History of the Information Machine, Boulder, CO.: Westview Press.
I know bugger-all about CELL, but at least there was one question I could answer! And an annotated answer, at that! :D
Jouster, I am going to have to mark you down for the spelling mistake, but your answer still scores highly.
96% - High distinction.
oldmacfan
2005-02-14, 17:35
The PowerPC first came out a LONG time ago with the PowerPC 601.
Even now, 10 years later apple's newest computer is still running on a PowerPC proccesor.
I think that very soon, like end of 2006 the latest, apple's new chip will be released. Something like the Super PowerPC, or sumthin like that.
PowerPC is gettin old.....
:smokey: So what are you smoking 20+ year old x86... :no:
Does everyone know about this report yet? Someone on AI posted a link late last night. :)
http://homepage.mac.com/dke/FileSharing2.html
vBulletin® v3.6.4, Copyright ©2000-2010, Jelsoft Enterprises Ltd.