Member
Join Date: Jan 2006
|
I've just read, that's there's a estimate of 34 errors on the new Core Duo - Solo chip form Intel.
"Releasing a brand new processor with 34 known errors seems almost criminal to me, especially..." - said RickGeek on January 23th. Link to Geek.com http://geek.com/news/geeknews/2006Ja...0123034350.htm I've read it at www.geek.com, but is that reliable source - anyone? Even though they might be easy to fix, I'm glad that I decided not to buy a MacBook Pro - .. yet. 34 errors is quite a lot, that's not a "whoops, we missed that one." Quite a start for Intel. Oops, wrong number in the headline - whoops, missed that one - sorry. |
quote |
‽
|
I'm not sure it's a lot. Can we have a comparison to other major CPUs?
|
quote |
Member
Join Date: Jan 2006
|
Quote:
|
|
quote |
Rocket Surgeon
Join Date: Feb 2005
Location: The Canadark
|
The comparison on the site is that the Pentium 4 has been found to have 65 errors in total in the time it's been out. No idea how fast they were found, though.
|
quote |
Veteran Member
Join Date: May 2004
Location: Clayton, NC
|
According to the linked article, the Pentium 4 has 65 known errata with no plans for correction. There's a comparison. At least 34 is less than 65.
[edit] Bryson beat me. Anyway, only 14 of the 34 errors are "show stoppers". Ugh. |
quote |
Member
Join Date: Jan 2006
|
Quote:
Still - I don't get it. Why release a chip with flaws, 64 and 34 is quite a lot to miss.. Not a "whoops, missed that one - again" |
|
quote |
Veteran Member
|
As people have already said... It depends on the errors.. Heck every floating point operation on a modern computer is an 'error' in real terms.
But I have read that some of these are quite fundamental flaws. That is dissapointing. Out of interest how do Intel chips measure against PowerPC chips in terms of errors? Not stirring (well not much), just curious... I may yet snag one of the last 12" PowerPC PBs before they discontinue them.... Hang having a machine I can test Universal Binaries on - I want one of them 12" suckers! 'Remember, measure life by the moments that take your breath away, not by how many breaths you take' Extreme Sports Cafe | ESC's blog | scratt's blog | @thescratt |
quote |
‽
|
Quote:
Quote:
|
||
quote |
Rocket Surgeon
Join Date: Feb 2005
Location: The Canadark
|
For comparison, the PPC970FX (Or G5) has 24 errors and the G4 (7447) has 36.
(God bless Google!) |
quote |
Member
Join Date: Jan 2006
|
Ok Bryson, didn't you read the reply above you ?
Was anyone of them fundamental flaws ? Or just "easy to fix flaws ? |
quote |
Member
Join Date: Jan 2006
|
Quote:
|
|
quote |
Veteran Member
|
Read the article.. Two things stick in my mind..
Quote:
|
|
quote |
Rocket Surgeon
Join Date: Feb 2005
Location: The Canadark
|
Quote:
Quote:
|
||
quote |
Member
Join Date: Jan 2006
|
I'm starting to feel sorry for them who just bought / ordered a new Intel Mac.. This is not quite the "Intel Start" I was hoping for.. But, let's see, this whole thing might just be .. bs ? Does anyone know if Geek.com is a reliable source ?
|
quote |
Member
Join Date: Jan 2006
|
Quote:
|
|
quote |
Rocket Surgeon
Join Date: Feb 2005
Location: The Canadark
|
Quote:
Quote:
|
||
quote |
Member
Join Date: Jan 2006
|
Never mind Bryson .. my bad.. Not quite sure what I was thinking .. Just blaim me
|
quote |
Member
Join Date: Nov 2005
Location: Las Vegas, NV
|
NOTHING is released without flaws. Processors released with errata is nothing new and this web site knows it. The geek.com article is taking advantage of the situation. Using this to throw mud on Apple and Intel(or just Intel). Keep in mind Intel's transition to 65nm and the fact that its processors are FAR more complex than AMD or IBM.
If your computer works what does it matter? Are you going to pick a fight with someone? My G5 has 24 errors and your CoreDuo has 34 ha! This reminds me of a thread a few days ago about a guy who thought his Powerbook G4 was crap because the new Intel notebooks will be shipping soon. If you are waiting for perfection you will be waiting an eternity. Real artists ship. |
quote |
Veteran Member
|
Was it Intel or AMD (I think it was Intel) who had that fundamental maths error on their chips which was catastrophic? This is ancient history I am talking about now..
If I remember correctly AMD cleaned up then, as their version of the chip did not have the error.. Proving once and for all in a double whammy that they were not just copying silicon... Quote:
'Remember, measure life by the moments that take your breath away, not by how many breaths you take' Extreme Sports Cafe | ESC's blog | scratt's blog | @thescratt |
|
quote |
‽
|
Quote:
|
|
quote |
‽
|
Quote:
|
|
quote |
ಠ_ರೃ
Join Date: May 2004
Location: Minnesota
|
I think that was the Pentium II, scratt. EDIT: Whoops, guess it was the original Pentium.
That said, I think it's disappointing how little context there is in this thread. Even though we do have numbers for some other processors, I really have no idea how serious these errors are or how serious processor errata are in general. |
quote |
Veteran Member
|
Havng read through the erratta I would say a few years ago these problems would have bugged me.. I was working on beta silicon, and on first gen chips in machine code, writing right up from driver level to high end system level packages. That's because I could see these problems impacting on my code at a low level...
However, now that I have become this monster of OOP and C++ (God I hate myself) I really don't give two hoots, as other's are going to find and experience these problems way before I do. With the beauty of the internet you can pretty much guarantee that it'll be out in the ether within an hour from some anorak in Minsk who's got more gear than the NSA. He'll have fixed it, not released the fix, made a few thousand billion rubles from selling a virus that exploits it to retired KGB, and then posted the fix and a laa dee daa account of how clever he is on P2P networks and IRC, and promtply been blown up by Mossad, working for the US Secret Service, funded by M$. So, no, in real terms I don't think it's too much to get worked up about, but its a bit embarressing for Intel, and I suspect we'll find more. To balance that a bit most of the erratta have been owned up to, and only experienced by Intel themselves. Got to give them points for honesty I suppose... 'Remember, measure life by the moments that take your breath away, not by how many breaths you take' Extreme Sports Cafe | ESC's blog | scratt's blog | @thescratt |
quote |
Cat's Dreamlands
Join Date: May 2004
|
Yes the G4 and G5 where not bug free.
At the difference of IBM, Intel has a more transparent communication. Intel is more willing to communicate on such flaws, than some others companies. |
quote |
Posting Rules | Navigation |
|
Thread Tools | |