PowerBook G5 [Archive] - AppleNova Forums

unixguru

2004-09-30, 04:32

It's gotta be coming. There won't be any new Motorola/Freescale processors until next summer. There's already support for a Powerbook 7,1 and 7,2 with the U3-II (970fx) north bridge in last month's OS release, 10.3.5. 10.3.6 has already been seeded. I don't know exactly what that means, but my guess is there are working prototypes floating around the Apple campus. I think the only remaining thing is for IBM to completely solve their manufacturing problems. A low-voltage 970fx with Powertune should have manageable heat, even in a 1" case. I'm saying January if not sooner. I don't trust the Apple execs -- they aren't going to hurt sales by telling what's coming.

My thoughts:
15" and 17" G5 models ONLY
1.6 and 1.8 Ghz low-voltage PowerPC 970FX with PowerTune
Low voltage 90nm process U3-II system controller (northbridge)
400 Mhz or 450Mhz (x4) front-side bus
latest RADEON Mobility graphics
60 and 80 GB hard drives
possibly some sort of new enclosure
All the typical USB,Firewire,e tc.

Discuss.

Quagmire

2004-09-30, 05:53

I think it will be 1.8 Ghz and 2.0 Ghz G5 with the same or more FSB as the iMac G5. The 12" Powerbook will stay at G4 for at least most of the year and probably get the G5 in Paris '05 or MWSF '06. The HD will be 80 GB standard. There is no problem with your specs. But, the only thing better in the powerbooks will be the GPU.

Matsu

2004-09-30, 07:08

All depends on IBM. There could be one more G4 bump to the 7447/8A whatever they're using now. That would mean IBM is really having problems. Could also be that Freescale is very conservative with their estimates -- they also have 64bit e700 on the map, but not for some time, and furthermore the direction their going seems to be more towards embedded stuff ATM. But there's always room for surprises.

I however, won't expect a major architectural change to the Powerbooks untill the arrival of 10.4.

January maybe? Likely later...

Dave Hagan

2004-09-30, 07:22

The rule of thumb is if we don't see a new PowerBook by the 1st week of November, then it's probably not coming until after New Years.

Mac World San Fran starts Jan. 10th, and my assumption, using history as a guide, is that MWSF has been a totally new PowerBook venue for quite some time. The Titanium and Aluminum were both debuted at Mac World (2001 and 2003 respectively).

Today is an earnings report for Apple, and it will be interesting to see what analysts try to cull from execs, if anything.

I think 1.6 and 1.8GHz models in the 15 and 17 only is probably spot on. I think they would need to have a different enclosure, it's going to be a completely different architecture. Like what we have seen of the Power Mac G5 and the iMac G5, I would hope (and think) that the PowerBook G5 would be much more user serviceable. Right now, the only thing you can replace in an aluminum PB is the memory. You used to be able to replace the hard drive in the Titaniums.

If I had to venture a guess I would say that unixguru is probably on the right path with specs...It would be announced in January at MWSF, shipping in February or early March.

Mac+

2004-09-30, 07:41

DMBand0026

2004-09-30, 11:41

I totally disagree with undertaker, if the 12" doesn't get the G5, none of them will. There is no way that Apple will update 2/3rds of their line. I'm aware of what happened with the 15" TiBooks when the AlBooks came out, but this is entirely different. It's not all that much more difficult to fit all that stuff in a 12" than it would be to fit it in a 15", you'd really start to see an advantage in heat displacement when you got up to 17".

So I think they'll all (12, 15, and 17 inch) get the G5 come January. Probably clocked pretty modestly at first, between 1.4 and 1.8. Same or faster FSB as the iMac G5, 128MB VRAM available on the 12 and 15", standard on the 17". Better displays, 7200RPM HDDs across the board, and all in a sweet 1" thick enclosure (1.25 for the 12", just like it is now).

Those would be sweet machines :)

FFL

2004-09-30, 11:48

I totally disagree with DMB's disagreement. I think it's entirely possible that ONLY the 17" (at first) will get the G5. That would rationalize a larger price delta between the 17" & 15" as well as force power-hungry bleeding-edgers to the most expensive laptop in the line - definitely SOP for Apple Marketing. And of course, starting at 17" and then working down through the line makes life easier for the engineers also.

DMBand0026

2004-09-30, 13:03

I considered that angle because it would follow the business plans of many other computer manufactures. But I know Apple has never been one to just do things halfway. I think Steve Jobs would be personally insulted if they didn't get the whole line with a G5 at the same time. ;)

Mac+

2004-09-30, 13:08

Fellas please ... G5 and the FSB - how does it work itself out exactly? :confused:

spiff

2004-09-30, 16:05

Whatever happens, I just want a G5 Powerbook by the time I'm off to college next September. My guesses:

The entire line will go G5. I agree with DMB. Why upgrade part of the line with a new processor. If you didn't upgrade the 12 inch, then no one would buy it, they'd just get an iBook which will undoubtedly have a faster G4 by then. They only way to make the 12 inch P-book look good then is have an even faster G4 in it which just seems silly to me. They'd be making one special chip for one particular model of powerbook.

Processor at 1.6 and 1.8 but hopefully a 2.0 also.
Faster FSB, faster than the current iMac.
Whatever the best ATI card is out at the time, probably a 9800 mobile
DDR 400 RAM instead of 333
Better screens and a slightly different enclosure.
More upgradability - access to hard drive and what not
Better hard drives - 7200 RPM - 80 gig models - that might be a BTO option. Could be 5400 RPM standard.

I don't think there's a whole lot to predict as far as features go. It will all be better.

unixguru

2004-09-30, 20:31

Can one of you please explain to me how a G5 PowerMac can have a FSB that is half the speed of the chip itself, yet the iMac and PB has to have a quarter or other variation of its FSB speed? (for example, using unixguru's spec: a 1.6 G5 with a 400Mhz bus @ 4x) :confused:

TIA
Microprocessors have, for a long time now, supported what is known as a clock multiplier. It allows the microprocessor to run internally at some mulitple of the front-side bus speed, but communicate externally at the same speed as the bus. They typically come in several multipliers; it depends on the CPU. I don't know what all the multipliers the 970fx supports, but apparently it supports 2x and 3x. Other PowerPC chips, like the 750fx, support many more multipliers, up to 9x and beyond. The multiplier is set by several input pins to the processor, in a binary number. For example, the combination 0001 might mean a multiplier of 1 and 0010 might mean a multiplier of 1.5 or of 2, depending on the processor.

unixguru

2004-09-30, 20:40

Whatever happens, I just want a G5 Powerbook by the time I'm off to college next September.

I agree. I want one by next fall too.

The entire line will go G5. I agree with DMB. Why upgrade part of the line with a new processor. If you didn't upgrade the 12 inch, then no one would buy it, they'd just get an iBook which will undoubtedly have a faster G4 by then.
They may not be able to cram the G5 and all the auxiliary stuff into the 12". The 12" machine isn't as much of a Power user machine as the 15" and the 17" anyway. I agree it makes sense to upgrade them all, but from what I've seen for hardware support info in the OS, there are only two G5 powerbook models.

/System/Library/Extensions/AppleMacRISC4PE.kext/Content/Info.plist

<key>IOPlatformPluginTable</key>
<dict>
<key>PowerBook7,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerBook7,2</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerMac7,2</key>
<string>PowerMac7_2_PlatformPlugin</string>
<key>PowerMac7,3</key>
<string>PowerMac7_2_PlatformPlugin</string>
<key>PowerMac8,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerMac9,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>RackMac3,1</key>
<string>RackMac3_1_PlatformPlugin</string>

Quagmire

2004-09-30, 20:41

A little of topic: What in the hell is Powermac 9,1?

oldmacfan

2004-09-30, 21:45

I do believe that if Apple wanted to have a G5 portable computer they could sacrafice something to get it done. Apple doesn't like sacrafices.

Apple could could come out with an XBook portable computer, 17" or 20" models only. With G5's clocked at 1.4 and 1.6 Ghz, I think it is possible to get it down around 1.5" thick. I just can't see Apple doing it.

oldmacfan

2004-09-30, 21:47

A little of topic: What in the hell is Powermac 9,1?

That is the Quad 3Ghz 970MP XStation... ;)

:lol: :lol: :lol: :lol:

scratt

2004-09-30, 22:45

... but from what I've seen for hardware support info in the OS, there are only two G5 powerbook models.

/System/Library/Extensions/AppleMacRISC4PE.kext/Content/Info.plist

<key>IOPlatformPluginTable</key>
<dict>
<key>PowerBook7,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerBook7,2</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerMac7,2</key>
<string>PowerMac7_2_PlatformPlugin</string>
<key>PowerMac7,3</key>
<string>PowerMac7_2_PlatformPlugin</string>
<key>PowerMac8,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerMac9,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>RackMac3,1</key>
<string>RackMac3_1_PlatformPlugin</string>

That may be the case.. but surely Apple would just release a new OS with the new Powerbook.. Much like they have with recent G5 releases.. and then upgrade globally shortly after.

Hey maybe they'll roll out a G5 Portable with the next OS update?

unixguru

2004-10-01, 01:43

A little of topic: What in the hell is Powermac 9,1?
I have not a clue what PowerMac 9,1 is. The PowerMac G5 is PowerMac 7,x and the iMac G5 is PowerMac 8,x. The fact that the major is 9 suggests a big change. It may be a new desktop model. Or it might just be a 970MP based machine. Who knows...

unixguru

2004-10-01, 01:48

That may be the case.. but surely Apple would just release a new OS with the new Powerbook.. Much like they have with recent G5 releases.. and then upgrade globally shortly after.

Hey maybe they'll roll out a G5 Portable with the next OS update?
I'm not quite sure why that stuff is in the plist, because the actual SMU_Neo2 plug-in isn't included. It may just simplify OS upgrades for prototype machines. Whatever reason, the only conclusion I can draw is that there is actual hardware in existence that is a Powerbook of some sort and uses the G5-II System Management Unit (SMU).

Mac+

2004-10-01, 02:34

Microprocessors have, for a long time now, supported what is known as a clock multiplier. It allows the microprocessor to run internally at some mulitple of the front-side bus speed, but communicate externally at the same speed as the bus. They typically come in several multipliers; it depends on the CPU. I don't know what all the multipliers the 970fx supports, but apparently it supports 2x and 3x. Other PowerPC chips, like the 750fx, support many more multipliers, up to 9x and beyond. The multiplier is set by several input pins to the processor, in a binary number. For example, the combination 0001 might mean a multiplier of 1 and 0010 might mean a multiplier of 1.5 or of 2, depending on the processor.Thanks unixguru - I was aware of the clock multiplier. My question was more geared toward the fact that I thought a G5 chip would share the same multiplier features with *every incarnation* of itself. So the new iMac would function the same as the PM G5 series - and the new PB (hopefully) would be able to do so also. But, I guess, the G5 CPU comes in various incarnations for each Apple product it has to serve, with different multiplier and FSB capabilities. Is that right?

bassplayinMacFiend

2004-10-01, 08:01

Thanks unixguru - I was aware of the clock multiplier. My question was more geared toward the fact that I thought a G5 chip would share the same multiplier features with *every incarnation* of itself. So the new iMac would function the same as the PM G5 series - and the new PB (hopefully) would be able to do so also. But, I guess, the G5 CPU comes in various incarnations for each Apple product it has to serve, with different multiplier and FSB capabilities. Is that right?

No, every G5 chip can have a variety of FSB speeds. If you sent the right commands to the G5 chips in the PowerMacs, they too could connect to a buss running at 1/3 CPU clock (like the current iMacs). Likewise, if you could alter the settings in the new iMac, you could have the iMac's FSB running at 1/2 CPU instead of the 1/3 CPU they run at now.

Basically it wouldn't make sense for IBM to create a different G5 CPU for each FSB multiplier that Apple wants to use. They just put the ability for the CPU's clock unit to communicate with busses that run at 1/2 CPU, 1/3 CPU, 1/4 CPU, etc. Apple just sends the proper 'command' to the CPU letting the CPU know what speed the FSB is running.

Does that clear things up a bit?

bassplayinMacFiend

2004-10-01, 08:02

Mac+,

Now you've got me wondering if I can overclock the FSB in my new Mac! :D Heck it's a 20" model, does that mean it could disperse the extra heat generated by the faster FSB?

DMBand0026

2004-10-01, 11:12

<key>IOPlatformPluginTable</key>
<dict>
<key>PowerBook7,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerBook7,2</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerMac7,2</key>
<string>PowerMac7_2_PlatformPlugin</string>
<key>PowerMac7,3</key>
<string>PowerMac7_2_PlatformPlugin</string>
<key>PowerMac8,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>PowerMac9,1</key>
<string>SMU_Neo2_PlatformPlugin</string>
<key>RackMac3,1</key>
<string>RackMac3_1_PlatformPlugin</string>

It's very possible that the current line is only regarded as two machines. Keep in mind, the chip only has two clock speeds @ 1.33 and 1.5, and the system architecture is the same as far as bus speeds and everything else. The different size enclosure may not make for a different model if the internals are all the same.

chucker

2004-10-01, 12:22

A 20 inch laptop? What's next, a 50 inch cinema display?

curiousuburb

2004-10-01, 12:47

The peecee world has taken the 'portable workstation' concept another direction:

http://gadgets.tonytalkstech.com/images/MaxPac.JPG (http://gadgets.tonytalkstech.com/archives/2004/05/05/maxpac8200x-dual-xeon7505-rugged-portable-workstation-multiple-19-lcd-pci-x-5-fixed-scsi-hd/)

oldmacfan

2004-10-01, 14:33

For good reason I didn't call it a 20" laptop... lol

Robo

2004-10-02, 01:39

DMBand0026

2004-10-02, 10:04

Kill the 12" PowerBook and you've killed a small piece of me.

stoo

2004-10-02, 11:35

PowerMac 9,1 probably isn't an XServe, as they are referred to (so far) as RackMac.

Luca

2004-10-02, 15:55

I think it's possible that Apple could go G5 but have trouble getting it into the 12". They'd have a riot on their hands if they didn't announce something about the PBG5 at MWSF, so i think they'll probably have it in the 15" and 17" (not 17" only, though.) As for the 12"? Well, if they can't get the G5 in there, I could see them making like a "PowerPad" tablet sorta thing - I think people would like it, G4 and all (of course, lots of people would then wait for the 2006 G5 version of it). I know, I know, wishful thinking, but hey, it'd explain the whole 9,1 bit. Just speculation.
Yeah uh... what are you smoking?

Apple will put G5s in the PowerBooks once they can put G5s in ALL the PowerBooks. Until then, they'll just keep putting faster G4s in them. I don't see what's so hard to understand about that. They've never shown any indication of doing something weird like that in the past, so why would they start now?

applenut

2004-10-02, 16:45

Yeah uh... what are you smoking?

Apple will put G5s in the PowerBooks once they can put G5s in ALL the PowerBooks. Until then, they'll just keep putting faster G4s in them. I don't see what's so hard to understand about that. They've never shown any indication of doing something weird like that in the past, so why would they start now?

um... i think what's hard to understand about that is the fact that the G4 is all but dead as far as increases in clockspeed. so, they better do something....even if its "weird" in your opinion. I've already had several people ask me about when a G5 powerbook would be available. the G5 is getting a lot of attention and changing people's perception of the mac being slow.

Luca

2004-10-02, 17:18

Why's the G4 dead for increases in clock speed? Where did you get the information that the G4 will never get any faster? Because I heard that Motorola has announced the 7448 G4, which will go to 2.0 GHz. That's pretty good.

People were totally convinced that the G4 was stuck after the 1.42 GHz PowerMac G4 was scrapped in favor of the G5s. Then what? PowerBooks get 1.5 GHz G4s. So I don't see why they won't simply keep bumping the PowerBooks' G4 processors until they can get a G5 in all of them. It would still be a temporary measure, but then again, what isn't temporary in the computer world?

stoo

2004-10-02, 19:29

The G4 is regarded as dead for speed increases because the next model won't be available for seven months in a product, and doesn't appear to be especially great. 2GHz isn't a huge leap over 1.5GHz, and a 200MHz MPX bus is a little on the slow side. There might be some more minor revisions of existing G4s between now and then, but the current G4 is beginning to look long in the tooth.

Luca

2004-10-02, 19:50

I think the G5 is dead for speed increases. After all, there are all these delays with them! :lol:

In reality, I think PowerBook G5s could be just around the corner, but I see them as being about a 50/50 chance next to the 7448 at 2.0 GHz. You say that 2 GHz isn't a huge leap over 1.5 GHz, but you and I both know that if the PowerBooks do in fact get G5s soon, they will certainly not exceed 2 GHz. In fact, it's very likely they'll go with 1.8 GHz for the faster units.

The memory bus is a disadvantage of the G4, but the G5 has plenty of disadvantages of its own; for example, the G5 has a 20+ stage pipeline (like the Pentium 4) while the G4 has a 7 stage pipeline. I think my point is that while everyone seems to think that Apple's doomed unless they come out with a G5 PowerBook, I believe Apple will figure something out and give us a good update whether they can get the G5 into a PowerBook or not. They won't split them and give some of them G4s and others G5s.

Another interesting thing: mobile processors these days are using the strategy of putting lots of L2 cache on their chips to increase performance while keeping the clock speed (and therefore, power requirements and heat) low. Mobile Athlon 64s are basically the desktop versions, but with double the L2 cache (1 MB instead of 512 kb). So if you take a mobile A64 rated at 3200+, that's a 2.0 GHz processor with 1 MB of cache. The desktop 3200+ is 2.2 GHz, with half the cache. And there you have it - the laptop chip generates less heat and requires less power, because it's running 200 MHz slower, and yet the speed rating is the same, all because of a little bit of cache. If Apple were to put 1 MB of L2 cache in the PowerBooks, they could use a processor with a lower clock speed and still have them perform as well as (or more likely, better than) the current iMacs. Whether they use a G4 or a G5 remains to be seen, but I don't think it'll matter much. They're about equal clock-for-clock.

Mac+

2004-10-02, 19:59

No, every G5 chip can have a variety of FSB speeds. If you sent the right commands to the G5 chips in the PowerMacs, they too could connect to a buss running at 1/3 CPU clock (like the current iMacs). Likewise, if you could alter the settings in the new iMac, you could have the iMac's FSB running at 1/2 CPU instead of the 1/3 CPU they run at now.

Basically it wouldn't make sense for IBM to create a different G5 CPU for each FSB multiplier that Apple wants to use. They just put the ability for the CPU's clock unit to communicate with busses that run at 1/2 CPU, 1/3 CPU, 1/4 CPU, etc. Apple just sends the proper 'command' to the CPU letting the CPU know what speed the FSB is running.

Does that clear things up a bit?Yes it does - thanks bpMF. I thought it would be a waste of resources on IBM's part to create different G5 chips (with different multipliers) for each machine.

However, this confirms to me that Apple is crippling their machines on some arbitrary way with the commands sent to the chip for the FSB multiplier. I guess they have always done this - but it only really dawned on me now with the talk about the G5 FSB in the PowerMacs and iMacs.

Btw, I would *love* to own a G5 PB running at 2GHZ with a 1Ghz FSB! :cool:
(Ooh the bandwidth - the bandwidth!!) :drool:

Luca

2004-10-02, 20:34

Apple's not "crippling" anything by changing the FSB multiplier. See, when they built the iMac's motherboard and processor, they used parts that meet or exceed the required specifications (that being a processor that runs at 1.6/1.8 GHz and a FSB that runs at 533/600 MHz).

On a PC, for instance, you can set the FSB, core voltage, and multiplier to whatever you want, using your motherboard's BIOS (assuming your motherboard supports it). That's just one of the many ways to overclock things. But the parts are only rated for a certain speed when they leave the factory. IBM just makes a bunch of G5s, some of them turn out really well and can run at 2.5 GHz stably, and some of them only run stable at 1.8 GHz. They sort them out and put them in the appropriate machines.

You can be assured that in a very large majority of iMacs, they are going at their full potential. Assuming you had a way to easily manipulate the FSB, voltage, and multiplier in an iMac, you could certainly try to get it running at a 2:1 FSB instead of 3:1, but your chances of success would be low. You'd probably have better luck getting a 1.6 GHz iMac to run at 1.8 GHz with a 3:1 bus speed, but even then, that iMac's parts just aren't supposed to run that fast.

How does this relate to the PowerBook? Well... I don't know. Some people have said that Motorola is overclocking the G4 in order to get the 1.5 GHz speeds that the PowerBooks currently use, but that's not true because overclocking, by definition, is done by someone other than the chip maker. And if Moto isn't supplying chips rated at 1.5 GHz, then Apple wouldn't be able to build nearly as many 1.5 GHz PowerBooks as they have.

To keep this post on topic, back in February of this year, MacRumors posted an article (http://www.macrumors.com/pages/2004/02/20040223113112.shtml) saying that the newest version of the 7447 would run at 1.5 GHz and would probably make it into the PowerBooks. They were right - some two months later, the PowerBooks got the chips. I think it's likely that the next PowerBooks could get 7448 processors running at 1.8 or 2.0 GHz in the next two or three months. It's entirely possible that Motorola's "early 2005" announcement was slightly conservative. But I guess we'll just have to wait and see.

Mac+

2004-10-02, 22:09

Yeah, I guess "crippling" wasn't the right word to use. I was basing my comment on the (perhaps misunderstood) understanding that G5 chips were the same and could all handle various multiples of the FSB. Thus, the ones in the iMac were being limited by what commands Apple was sending them (ie: what multiplier ratio to be working at), due to the way they "specced" their machines.

But, I'm still a little confused. Why can Apple release a PM G5 @ 1.8Ghz with a FSB of 900 MHz and an iMac G5 @ 1.8 GHz with a FSB of 600 MHz? Why didn't they maintain the FSB speed for the iMac at 900 MHz? Wait a tick - it's all to do with price points and marketing isn't it? To keep the iMac spec from eating into the potential PM sales. D'oh! :grumble:

OK - back on topic: I'm curious about the FSB and the G5 b/c on the next PB I'd like to see a huge throughput - and a 1 GHz bus appeals to me more than a 600 MHz bus (or whatever bus speed they're using now on the PB - what is it 167 MHz?).

Luca

2004-10-02, 22:43

The reason the FSB on the new iMac is 600 MHz and not 900 MHz is to reduce cost. It's not just a marketing decision - 90% of iMac buyers don't even know what the frontside bus is. But it generates less heat and it costs less to make a motherboard that is only capable of using a 600 MHz bus. If they wanted to use a 900 MHz bus in the iMac, they probably could, but they'd have to use a more expensive motherboard that can handle the extra speed without overheating or getting unstable. That's how pretty much all these things work - the chip, FSB, memory, video card, and so on, can only handle a certain frequency. The higher that frequency is, the more it costs.

What I don't know is whether IBM actually makes G5s meant for all different frequencies, or if they simply make G5s, test them, and then stamp them with a label declaring the fastest they can run while being stable. I'm not totally sure, really. But it works at least something like that. You'll see phrases like "yields" when talking about processors. IBM may set out to make a bunch of 2.5 GHz G5s, but the yield of processors that can actually run that fast might be low. They end up with a bunch that can only run at 2.0 GHz or 1.8 GHz or something. The lower the clock speed, the higher the yields. They're constantly working to improve their yields, which in turn increases supply, fulfills demand, and reduces prices. Once the yields are high enough at 2.5 GHz, they'll try working on getting good yields at 2.8 GHz or 3.0 GHz or something. And so it continues.

While the FSB is a factor in how a chip performs, it's certainly not the only factor. Comparing the G4 to the G5 solely on 64-bit vs. 32-bit, or only by virtue of their bus speeds, is not a good idea. There are a zillion tiny components that all go into making a chip perform a certain way. There's cache: L1, L2 and L3. There's the number of pipeline stages: 4 for the G3, 7 for the G4, 20+ for the G5 and older P4s, 30 for the Prescott P4. There are all sorts of different ways the processor makes calculations: floating point functions, integer functions, vector processing (i.e. AltiVec). I can't begin to understand how all these things work together, but I know that you can't just look at one thing. The super-short four stage pipeline in the G3 is sometimes really good, but it also prevents it from reaching very high clock speeds. Meanwhile, some may see the G5's 20-stage pipeline as a big problem, but it IS running at more than twice the clock speed of that little 4-stage G3. Strength in some areas makes up for weaknesses in other areas.

Sorry I got a bit off topic, I'm just trying to explain as best I can the different factors that come into play when looking at how a computer performs. Remember that all those things I mentioned are just the ones having to do with the processor and motherboard! You can't forget how those interface with the rest of the system, the speed of other components (like the video card), and most importantly, how well the software can take advantage of the various strengths and weaknesses of the entire system.

BenRoethig

2004-10-03, 07:26

PowerMac 9,1 probably isn't an XServe, as they are referred to (so far) as RackMac.

My guesses: xMac/xStation, new eMac, or (hopefully) a new prosumer machine.

Henriok

2004-10-03, 10:38

the G5 has a 20+ stage pipeline (like the Pentium 4) while the G4 has a 7 stage pipeline..Not entirely tue. Some of the AltiVec queues of the G5 might be 20+ stages long but the INT and FP queues are not nearly that long. The AltiVec queues in the G4 aren't 7 stages short, they are longer. I also think that the G5 uses micro ops so the pipes are not easily compared.

Some good articles over at Ars cmparing the G5 to G4:
Inside 970 pt 1 (http://arstechnica.com/cpu/02q2/ppc970/ppc970-1.html)
Inside 970 pt 2 (http://arstechnica.com/cpu/03q1/ppc970/ppc970-0.html)

Nullpunkt

2004-10-03, 19:24

Gah, not much of a "spec whore" per se........

all i'd ask for in a PB G5 is that it would be priced on "iMac-ian price points"......

you know, seeing as the price of the G5's is less to fabricate than the G4's.... and well, the market is slowly beginning to drop on LCD's.....

I think it's plausible.

Make it a little fatter if need be. Just ASAP, okay, Mr. Jobs, Ives and co.......

Mac+

2004-10-03, 20:52

Thanks for the informative post Luca. I was aware of the other factors that help contribute to a machine's speed, but I appreciate the time and effort you took to type it all up and help clarify the discussion. Cheers. :)

FFL

2004-10-03, 21:05

The G5 PowerBook design will contain a lot of cooling innovations, using what they learned with the xServe and the iMac. The slow but steady movement of the G5 down the line is what makes me predict a 17" PowerBook only to start. A 15" as well is a toss-up, and a G5 in a 12" is a long shot - probably at least a revision away.

I hope the PB doesn't go any lower than the iMac in bus speed.

onlyafterdark

2004-10-03, 22:22

Only time will tell.

gyroscope

2004-10-04, 00:40

Yeah, but he doesn't post so much these days.

unixguru

2004-10-04, 02:07

unixguru

2004-10-04, 02:42

Umm... yeah... I'm gonna have to kind of disagreeee with you on that one. Yeah... Motorola's been kind of flaky lately and I'm not sure if....

In reality, I think PowerBook G5s could be just around the corner, but I see them as being about a 50/50 chance next to the 7448 at 2.0 GHz. while everyone seems to think that Apple's doomed unless they come out with a G5 PowerBook, I believe Apple will figure something out and give us a good update whether they can get the G5 into a PowerBook or not. They won't split them and give some of them G4s and others G5s.

to put 1 MB of L2 cache in the PowerBooks, they could use a processor with a lower clock speed and still have them perform as well as (or more likely, better than) the current iMacs. Whether they use a G4 or a G5 remains to be seen, but I don't think it'll matter much. They're about equal clock-for-clock.
So what company is this that is going to bring out a 7448 at 2.0Ghz? It's not Motorola/Freescale, at least not the one I'm familiar with. Last time Apple wanted to ship faster Powerbooks, they were held up because Moto was unable to produce enough of the 7447 models to allow a production run. I wouldn't be unhappy with a 2.0Ghz 7448; I just don't see it happening.

As far as the G5 goes, I really think that Apple will be putting this into Powerbooks as soon as possible. You get a lot of "wow" factor that you just don't get with G4. People assume a G5 is faster than any G4 (and that's currently true), plus there's the 64-bit wow-ness.

oldmacfan

2004-10-06, 19:28

What if Apple could today release PowerBook G5's at 1.0GHZ 12", 1.2GHZ 15", & 1.4 GHZ 17", Would you buy?

Quagmire

2004-10-06, 19:33

What if Apple could today release PowerBook G5's at 1.0GHZ 12", 1.2GHZ 15", & 1.4 GHZ 17", Would you buy?

The price will have to be right and prove it is still faster then todays 1.5 Ghz 7447A. Which I doubt. Then I say I will stick with the crappy G4 even though it beats the G5 with the same speed.

Mac+

2004-10-06, 19:36

No - and I'm not sure most people would either - I think there is still something to be said for the majority of the buying public following the "MHz myth" and they prefer to see their computer clock speeds increase with each new product release. :\

Luca

2004-10-06, 21:34

The G5 is no faster than the G4 at equal clock speed. Hell no, I wouldn't buy one. I'd get a used 1.5 GHz G4 as it would be faster.

oldmacfan

2004-10-07, 16:10

Ok, then what if the 15" and 17" were 1.6GHZ and the 12" was 1.4GHZ But each unit was .25" thicker and $150 more expensive?

Luca

2004-10-07, 17:24

That would still be pretty lame. Why'd you pay MORE for a laptop that is not only thicker, but has such a small improvement in speed that no one would notice it? Plus, if they put a 3:1 bus in the iMac, they'd probably put a 3:1 bus in the PowerBook. And that would also suck because as the new iMac has demonstrated, a G5 without an 800 MHz+ frontside bus speed is pretty unimpressive.

I'd want 1.8 GHz. Don't care which processor they use, but it should jump to 1.8, with 1.5 or 1.6 GHz on the low end. And there should be a 7200 RPM hard drive option.

onlyafterdark

2004-10-07, 19:33

Mac+

2004-10-07, 23:08

Yeah, I third Luca's post too, but for some reason I'm prepared to wait for the speed to reach 2Ghz. :D

oldmacfan a 1.6 GHz G5 in a casing that is .25" thicker and $150 more expensive? No way - I actually want the PowerBook to remain as sleek as possible ... btw (and I know this splitting hairs) isn't the Albook slightly thicker than the original TiBook? :bummer:

oldmacfan

2004-10-08, 08:06

709

2004-10-08, 08:12

The 17" Al is the same thickness as the old TiBooks (1"). The Al series gets progressively thicker as they get smaller (to 1.18" for the 12").

oldmacfan

2004-10-08, 08:24

The 17" Al is the same thickness as the old TiBooks (1"). The Al series gets progressively thicker as they get smaller (to 1.18" for the 12").
Thanks for the information.

I don't believe I would like my Powerbook any thinner than it is.

DMBand0026

2004-10-08, 15:32

Thanks for the information.

I don't believe I would like my Powerbook any thinner than it is.

But I wouldn't like it any thicker than it is either.

oldmacfan

2004-10-09, 10:13

But I wouldn't like it any thicker than it is either.
That is my point above.

:)

Canoe112

2004-10-10, 16:25

Has anyone heard if the new PBs will be built so they can run the new 30" Cinema display.

I think the 30" Display requires a new video card and two dvi ports?

Luca

2004-10-10, 16:32

No, it requires a dual-link DVI port. It's a confusing naming scheme, but that refers to a new type of DVI port that can handle higher resolutions. It doesn't mean you have to hook it up to two DVI outputs. So I wouldn't be surprised if the next PowerBooks have dual-link DVI.

And yes, that means that you can connect TWO 30" Cinema displays to a single PowerMac with the GeForce 6800 Ultra. *explodes*

Dave Hagan

2004-10-10, 19:34

If we don't see a new PowerBook by the first week of November, we will see a redesigned PowerBook G4 debuted at MacWorld in January. I don't think the G5 is quite ready for the PowerBooks yet, and so I think Apple will revamp the PowerBook G4 by way of improved displays, battery life, heat, and more. Those are my 2�. By the way, I hope I am wrong about it not having a G5 :)

DMBand0026

2004-10-10, 19:35

And yes, that means that you can connect TWO 30" Cinema displays to a single PowerMac with the GeForce 6800 Ultra. *explodes*

If I had money, I tell you what I'd do, go downtown and buy a Mac or two ;)

Quagmire

2004-10-10, 19:50

Here is my $.02, if ibooks get upgraded to 1.2 Ghz and 1.33-1.4 Ghz, before a Pbook update, the G5 Pbook is coming. That is way to close to Pbook speeds and no G4 out there avaiable to pbooks, G5's are coming then. The 15" and 17" pbook G5 will get a GPU that can drive 2 30" displays in my opinion.

DMBand0026

2004-10-10, 19:59

PowerBooks first, iBooks second. There is absolutely no way in the world that there would be an iBook with a faster clocked G4 than in a PowerBook. No way, no how. It will not happen, not ever. An iBook update will not be indicative of a PowerBook update, the PowerBook update will be indicative of what to expect in the next generation of iBooks.

Luca

2004-10-10, 20:26

It's happened before. In 2000, the iBooks were available in 366 MHz and 466 MHz speeds, while PowerBooks were 400 MHz and 500 MHz. The 400 MHz PowerBook was still significantly faster than the 466 MHz iBook though, and had a much nicer screen. I think Apple could pull off a 1.2/1.4 GHz iBook with 1.33/1.5 GHz PowerBooks, but I don't think it's necessary, or a good decision. iBooks are in pretty good shape right now.

Bucolic Old Sir Henry

2004-10-10, 20:40

Over in the Power PC thread, the esteemed Morpheus suggests that PowerBooks will be updated to G5 in the near future, January at the latest. Low-power chips, with speeds 1.6 & 1.8GHz, with a possibility of 2.0GHz.

Sounds good to me.

Pip pip!

wizard69

2004-10-10, 21:38

It always amazes me when we here about "absolutes" in the world of business. Apple could very well thake the iBook in a totally diffrrent direction where competition with the PowerBook isn't a concern.

One possibility is to push the enevelope as far as expected battery life. Apple may very well be able to do this with some of the announced Freescale processors. That is if they every come into production.

thanks
Dave

PowerBooks first, iBooks second. There is absolutely no way in the world that there would be an iBook with a faster clocked G4 than in a PowerBook. No way, no how. It will not happen, not ever. An iBook update will not be indicative of a PowerBook update, the PowerBook update will be indicative of what to expect in the next generation of iBooks.

tradman

2004-10-11, 11:43

Guys, I'm all up in a heap now. I am travelling to the sates from Ireland before the end of the year and planned to buy a top speced G4 Powerbook.The price difference between the states and ireland is huge. From reading all the threads, I think I would be daft not to wait for the G5.
Is there any idea's as tho the price of the new range of PowerBooks.
Should we not be seeing some big price drops on the current range from now on to coincide with a release date of early next year??

skid_boy_99

2004-10-12, 03:03

Don't mean to throw a spanner in the works - I'm waiting for updates too - but check the title at the PowerBook homepage (http://www.apple.com/powerbook/).

Wouldn't it be a little lame if Apple released Newer New Powerbooks this month? That kind of behaviour would remind me (vaguely) of the "best bargain" "closing down" sales that have even better bargains each week and actually never close... does that make sense to anyone?

Kurt

2004-10-12, 04:04

I am pretty sure the iMac was still refered to as the 'New iMac' somewhere on Apple's site right before the iMac G5 was announced. When you compare it to how long it took Apple to get to the new design from the original, it was still relatively 'new'. ;)

wannabe mac user

2004-10-19, 07:33

why not kill the 12" powerbook, the new ibook's are pretty much the current 12" powerbook.

Luca

2004-10-19, 08:10

Nah, the PowerBook has plenty of features that set it apart from the 12" iBook:

Processor is 133 MHz faster (not really enough to notice though)
Superdrive is an option
Monitor spanning is built in
DVI output instead of VGA means you can hook up a cinema display
Video card is faster and has double the VRAM
Overall build quality is higher
Comes with Bluetooth built in (a $50 option on the iBook)
Comes with double the hard drive capacity (60 GB vs. 30 GB, $75 option on the iBook)
Smaller and lighter. Not by much, but it's .3 pounds lighter and about .1 inches thinner. On such a small machine, that can make a bit of a difference.

SteveS

2004-10-19, 10:42

The G5 is no faster than the G4 at equal clock speed. Hell no, I wouldn't buy one. I'd get a used 1.5 GHz G4 as it would be faster.

I'm not sure of the basis for this claim. For example, let's look at some of the benchmarks over at barefeats.com

iMac G5 benchmarks at barefeats.com (http://www.barefeats.com/imacg5.html)

For starters, there isn't an easy one to one comparison. That is, from what I've seen, INT performance is about the same (per clock speed) between the processors, Altivec is slightly faster on the G4 (per clock speed), but FP is MUCH faster on the G5.

This test compares a 1.8 Ghz G5 to a 1.5 Ghz G4. There is a 20% difference in clock speed, so the G5 should score 20% better on these tests to be exactly linear. In reality, actual performance usually scales less than proportional to the CPU speed increase, so I'd expect a 1.8Ghz G4 to score closer to 17% faster than a 1.5Ghz G4.

In the tests, the G5 was...

27% faster in Appleworks
17% faster in iMovie 4 render
31% faster in iTunes4
18% faster in Filemaker

In all of the tests above, the actual performance of the G5 was equal to or greater than where the G4 would be at the same clock speed.

Photoshop seems to be an anomoloy at this point as the G5 was only marginally faster than the G4, despite the larger clock speed difference. We'll skip the Motion preview benchmark as that's largely GPU dependent.

More tests here:

Barefeats 2Ghz G5 vs 1.42Ghz G4 (http://www.barefeats.com/g5sum.html)

In this test, the clock speed is 40% difference in favor of the G5. In real worl tests, that should translate close to 35% difference in actual performance. Let's see how the G5 did.

The G5 was...

68% faster in Bryce.
110% faster in Cinebench 2003
64% faster in After Effects
117% faster in UT Bot match (though may be partially GPU bound)
86% faster in Speedtest CPU rating.
37% faster in Altivec fractal test, etc.

So, all the G5 had to be was about 35% faster in real world tests to be equivalent to the G4 at the same clock speed. From these test, we can conclude that Altivec heavy apps such as Photoshop and Altivec Fractal, benefit little if any between G4 and G5 (per clock speed). INT based apps are somewhat faster on the G5 (per clock speed) and FP based apps are MUCH faster (per clock speed).

That said, I challenge your notion the G5 is not faster than the G4 at the same clock speed. Overall, it is. Yes, there are a few anomalies with Altivec heavy tests. Most benchmarks and applications do not fit into this category. That said, I'd definitely take the G5 over the G4 if the clock speeds were equal. Further, the difference will only be more in favor of the G5 once applications are tuned accordingly.

Steve

rickag

2004-10-19, 13:11

I'm not sure of the basis for this claim. For example, let's look at some of the benchmarks over at barefeats.com......

....Further, the difference will only be more in favor of the G5 once applications are tuned accordingly.

Steve

Good post. While the G5 may not be faster clock for clock in all aspects, the fact that it has 2 FP units as opposed to 1, with much greater FSB speed, it only makes sense that it clobbers the G4 in many many applications. As IBM and Apple tweak the architecture and software the gap will only grow.

Luca

2004-10-19, 15:59

My main assertion is that a lot of Apple fans recently have been getting into this G5 circle-jerk frenzy when the processor really isn't tons better than the G4. It's very different from the G4 of course, so it will have advantages in some areas and disadvantages in others. I just don't think that a G5 PowerBook is necessary as long as Motorola can continue updating the G4 fast enough. A lot of people seem to be saying "G5 or bust!" about the PowerBook, but I'd be happy as long as it gets a faster processor. If it gets a 1.8 GHz G4 instead of a 1.8 GHz G5, and the G4 gets longer battery life, then I'm all for using a G4. A lot of people would gladly pay extra and sacrifice battery life or even compactness for the G5, but I wouldn't.

onlyafterdark

2004-10-19, 17:03

Me either.

curiousuburb

2004-10-19, 17:12

Depends if the Barefeats tests were real reviews or he just walked into an Apple Store again. :rolleyes:

iCreate

2004-10-23, 18:36

Rhumgod

2004-10-23, 22:22

Yes quite easy to understand. That is why they are called clock multipliers. This go round, they use the U3 Lite bridge which is 1/4 CPU speed (400 or 600MHz in the iMac G5). The PowerBook G5 will have similar technologies, FWIW.

wizard69

2004-10-24, 13:28

The one thing that would make a G5 based PowerBook a requirement in the 64 bit addressing capability of the chip. It is pretty obvious from the performance standpoint that the little extra one would get from a G5 isn't worth the power drain. Apple could stretch out the useful life of the 32 bit processors by using the extned addreessing capability of the 32 bit PPC line.

Extended addressing won't allow processes to have more than 4 gigs of memory but that would be an improvement for many. The question in my mind is how compatible is the virtual addressing of the G4 and the mode Apple currently uses on the G5? Could apple easily move the system code that allows the current implementation of the G5's to the 32 bit systems? At this moment in time it would not be dramatically different from the users perspective. I would love to hear form people more up to speed on the G4's physical and virtual addressing modes comment on this possibility.

The other interesting thing with respect to the G4 is that the new core that Freescale is promising has some improvements to various parts of the processor. So there is a reasonable potential that the processor will more than scale if provided with a fast integrated memory interface. Of course Freescale has yet to deliver those improvements. I would not be surprised to see the G5 in the PowerBook simply because it is the only alternative available for Apple to implement.

Dave

It's very different from the G4 of course, so it will have advantages in some areas and disadvantages in others. I just don't think that a G5 PowerBook is necessary as long as Motorola can continue updating the G4 fast enough.

runner91786

2004-10-24, 18:25

I'm not sure of the basis for this claim. For example, let's look at some of the benchmarks over at barefeats.com

iMac G5 benchmarks at barefeats.com (http://www.barefeats.com/imacg5.html)

For starters, there isn't an easy one to one comparison. That is, from what I've seen, INT performance is about the same (per clock speed) between the processors, Altivec is slightly faster on the G4 (per clock speed), but FP is MUCH faster on the G5.

This test compares a 1.8 Ghz G5 to a 1.5 Ghz G4. There is a 20% difference in clock speed, so the G5 should score 20% better on these tests to be exactly linear. In reality, actual performance usually scales less than proportional to the CPU speed increase, so I'd expect a 1.8Ghz G4 to score closer to 17% faster than a 1.5Ghz G4.

In the tests, the G5 was...

27% faster in Appleworks
17% faster in iMovie 4 render
31% faster in iTunes4
18% faster in Filemaker

In all of the tests above, the actual performance of the G5 was equal to or greater than where the G4 would be at the same clock speed.

Photoshop seems to be an anomoloy at this point as the G5 was only marginally faster than the G4, despite the larger clock speed difference. We'll skip the Motion preview benchmark as that's largely GPU dependent.

More tests here:

Barefeats 2Ghz G5 vs 1.42Ghz G4 (http://www.barefeats.com/g5sum.html)

In this test, the clock speed is 40% difference in favor of the G5. In real worl tests, that should translate close to 35% difference in actual performance. Let's see how the G5 did.

The G5 was...

68% faster in Bryce.
110% faster in Cinebench 2003
64% faster in After Effects
117% faster in UT Bot match (though may be partially GPU bound)
86% faster in Speedtest CPU rating.
37% faster in Altivec fractal test, etc.

So, all the G5 had to be was about 35% faster in real world tests to be equivalent to the G4 at the same clock speed. From these test, we can conclude that Altivec heavy apps such as Photoshop and Altivec Fractal, benefit little if any between G4 and G5 (per clock speed). INT based apps are somewhat faster on the G5 (per clock speed) and FP based apps are MUCH faster (per clock speed).

That said, I challenge your notion the G5 is not faster than the G4 at the same clock speed. Overall, it is. Yes, there are a few anomalies with Altivec heavy tests. Most benchmarks and applications do not fit into this category. That said, I'd definitely take the G5 over the G4 if the clock speeds were equal. Further, the difference will only be more in favor of the G5 once applications are tuned accordingly.

Steve

Well if you think thats a fair comparison so then you could compare any processor of lesser power by a percentage? So a pentium III 1ghz is 33.3% of a p4 3gz? Its not the same perfomance and the comparison is not correct. You might as well throw that out as it seems the g4(even by your comparison) is pretty dang close. By unequal comparison at 17% for anything would be more like 15% in a fair comparison persay, so then its actually slower right? "There is a 20% difference in clock speed" & " 17% faster in iMovie 4 render" Since 15< 20 Think about that a bit.