[Leia's Left Bun]

"Geek.com is reporting that Intel's errata (bug) documentation shows that the Intel Core Duo chip has 34 known issues found in the 20 days since the launch of the iMac Core Duo. (you can read the list) with only plans to fix one of them. While bugs in hardware is nothing new (the P4 has 64 known issues, at this time Intel does not plan to fix a single one) this marks one of the first times that Intel released a processor with known bugs, and some of the bugs are of higher severity than in the past. Also alarming is the rate the flaws have been found, at one and half per day since the launch of the iMac Core Duo."

If Apple wasn't using these chips we would be laughing our asses off.

http://apple.slashdot.org/article.pl.../01/24/1537231

[Super Mario]

Originally Posted by Leia's Left Bun

"Geek.com is reporting that Intel's errata (bug) documentation shows that the Intel Core Duo chip has 34 known issues found in the 20 days since the launch of the iMac Core Duo. (you can read the list) with only plans to fix one of them.

My ex girlfriend had 100 known issues. I had no fix or work around for them

[ShotgunEd]

Originally Posted by Super Mario

My ex girlfriend had 100 known issues. I had no fix or work around for them

wel she is your ex, thats a fix by any standards.

[Leia's Left Bun]

Did the PowerPC's ever have known issues? In terms of bugs that is

[ShotgunEd]

The PowerPC 750gx and 750gl (portable G3s I think) had 9 known errata according to

http://www-306.ibm.com/chips/techlib...256F5C006FF9E6

[Leia's Left Bun]

So how are these fixed? Just though OS updates or do they have some sort of ROM?

[ShotgunEd]

http://www.macnn.com/articles/06/01/23/core.duo.errors/

software usually is able to work around these issues

[Leia's Left Bun]

Is this something Intel will fix with a rev B of the chips or do they just leave it as is?

[Eug Wanker]

Originally Posted by Leia's Left Bun

So how are these fixed? Just though OS updates or do they have some sort of ROM?

Chip updates, if necessary. And there is documentation to work around them.

No big surprise though. For example, as of August 2005, AMD's Opteron had way more than 34 errata (page 13), 7 of which will never be fixed.

EDIT:

Originally Posted by Leia's Left Bun

Is this something Intel will fix with a rev B of the chips or do they just leave it as is?

Yes, and yes. (It depends on the impact of the erratum.)

[Millennium]

Good to know that nowadays, Intel puts as much care and effort into their QA as Apple does.

[rhashem]

Originally Posted by Leia's Left Bun

"Geek.com is reporting that Intel's errata (bug) documentation shows that the Intel Core Duo chip has 34 known issues found in the 20 days since the launch of the iMac Core Duo. (you can read the list) with only plans to fix one of them. While bugs in hardware is nothing new (the P4 has 64 known issues, at this time Intel does not plan to fix a single one) this marks one of the first times that Intel released a processor with known bugs, and some of the bugs are of higher severity than in the past. Also alarming is the rate the flaws have been found, at one and half per day since the launch of the iMac Core Duo."

If Apple wasn't using these chips we would be laughing our asses off.

http://apple.slashdot.org/article.pl.../01/24/1537231

You'd be laughing out of ignorance. All CPUs have errata. The 970FX (G5 single core) has 24 errata, all of which are marked as WONTFIX. The 7457 (old G4) has 32 errata. The 7447A (new G4) has 26 errata. Opterons, which are used in highly mission critical severs, have 87 unique errata between all the versions of the chip.

[demibob]

Originally Posted by rhashem

You'd be laughing out of ignorance. All CPUs have errata. The 970FX (G5 single core) has 24 errata, all of which are marked as WONTFIX. The 7457 (old G4) has 32 errata. The 7447A (new G4) has 26 errata. Opterons, which are used in highly mission critical severs, have 87 unique errata between all the versions of the chip.

but according to the link this is the most flawed chip intel has ever produced

[turtle777]

Originally Posted by demibob

but according to the link this is the most flawed chip intel has ever produced

Which would make Intel's old chips it better than most G4 and G5 ?

-t

[moodymonster]

I had a quick look to compare to PowerPC and couldn't find anything, so thanks for the info rhashem.

[Chuckit]

Originally Posted by Millennium

Good to know that nowadays, Intel puts as much care and effort into their QA as Apple does.

Also good to know their QA isn't much worse than it was back when you liked them.

[Love Calm Quiet]

Does having Apple adopt a mfr's CPU put a hex on some aspect of their production/design/or QA abilities.

I'm trying to pronounce "Motorola".

I'm try to spell EYE BEE EM.

[MindFad]

[Leia's Left Bun]

Originally Posted by Millennium

Good to know that nowadays, Intel puts as much care and effort into their QA as Apple does.

How long would you consider Apples QA to be poor?

[leehotti]

Originally Posted by MindFad

AhAhAhahAH. Nice post mindfad. And I've been meaning to say you have a lovely sig.

[macintologist]

Originally Posted by rhashem

You'd be laughing out of ignorance. All CPUs have errata. The 970FX (G5 single core) has 24 errata, all of which are marked as WONTFIX. The 7457 (old G4) has 32 errata. The 7447A (new G4) has 26 errata. Opterons, which are used in highly mission critical severs, have 87 unique errata between all the versions of the chip.

Once again™ Leia's Left Bun gets OWNED

[Spliffdaddy]

Widescale ownageness

[Leia's Left Bun]

Originally Posted by rhashem

You'd be laughing out of ignorance. All CPUs have errata. The 970FX (G5 single core) has 24 errata, all of which are marked as WONTFIX. The 7457 (old G4) has 32 errata. The 7447A (new G4) has 26 errata.

Sorry what am I laughing at? Last time I checked 34 was a higher number than 24.

Not to mention the chip has been out all of 2 weeks to anyone want to take a guess at how many more will be found?

OWNED. You guys kill me

[greenamp]

Originally Posted by Leia's Left Bun

Sorry what am I laughing at? Last time I checked 34 was a higher number than 24.

Not to mention the chip has been out all of 2 weeks to anyone want to take a guess at how many more will be found?

OWNED. You guys kill me

What he's saying is that the number of errata in the Intel Core is normal compared to the number in every other chip out there.

[Drakino]

This is definitely not the most flawed chip Intel has produced. The Pentium 4 has 101 known errata currently. (Source). Keep in mind that doesn't mean every chip has 101 errata, just that 101 different issues have been discovered across now about 5 or so major revisions of the P4 and many minor steppings.

Processor steppings will be how Intel fixes some of these errata in hardware. Other errata will be worked around by firmware microcode for the processor.

The only reason this is big news is because some people are trying to point out any minor flaw in the move to Intel, including people here. Pointing out errata isn't going to reverse the decision.

[rhashem]

Originally Posted by demibob

but according to the link this is the most flawed chip intel has ever produced

The article is full of ****. The Pentium with the FDIV bug was the most flawed chip Intel has ever produced. As the article itself points out, the P4 has almost twice as many outstanding errata as the P-M. Same thing for the Opteron. The G4 and G5 have almost as many. The whole thing is a tempest in a teapot, the result of some journalist who doesn't know enough to know what the errata document represents.

[rhashem]

Originally Posted by Leia's Left Bun

Sorry what am I laughing at? Last time I checked 34 was a higher number than 24.

Do you *really* want to have this argument? Errata documents cannot be compared like that. You have to look at the sevarity of each errata, the existance of workarounds, and the probability of encountering each one. Unless you can give a detailed analysis of those factors showing the P-M is unreliable, you have nothing except FUD, in the truest sense of the word.

[MindFad]

So it's processor FUD, then.

[Leia's Left Bun]

Originally Posted by greenamp

What he's saying is that the number of errata in the Intel Core is normal compared to the number in every other chip out there.

Like I mentioned 34 is a higher number than 27 and this is just after 2 weeks of being released.

rhashem, can you also please post links to your sources.

[wdlove]

Just the reason not to purchase a Rev. A Mac.

[Leia's Left Bun]

Originally Posted by rhashem

Do you *really* want to have this argument? Errata documents cannot be compared like that. You have to look at the sevarity of each errata, the existance of workarounds, and the probability of encountering each one. Unless you can give a detailed analysis of those factors showing the P-M is unreliable, you have nothing except FUD, in the truest sense of the word.

Sure, break it down for us champ. Post links to your sources and let us now how a new CPU out the door already has 34 known bugs and how PowerPC chips are just as bad.

[ShotgunEd]

Originally Posted by rhashem

The article is full of ****. The Pentium with the FDIV bug was the most flawed chip Intel has ever produced. As the article itself points out, the P4 has almost twice as many outstanding errata as the P-M. Same thing for the Opteron. The G4 and G5 have almost as many. The whole thing is a tempest in a teapot, the result of some journalist who doesn't know enough to know what the errata document represents.

Why is everyone flaming SWG?

This may well be the most flawed chip intel has released because errata can be discovered throughout the life of a chip.

This chip has 34 known errata on launch. This is what the article is pointing out and SWG admitted his ignorance a few posts later and asked for other opinions.

Some people read a username and completely dismiss the topic before they even read the content.

[Lateralus]

Originally Posted by Leia's Left Bun

Sure, break it down for us champ. Post links to your sources and let us now how a new CPU out the door already has 34 known bugs and how PowerPC chips are just as bad.

And while he's at it, a break down on the severity of the compared eratta wouldn't hurt either. Since it isn't the number of eratta that count, it's the severity of them.

[Eug Wanker]

Originally Posted by ShotgunEd

Why is everyone flaming SWG?

Because his conclusion is not correct.

This may well be the most flawed chip intel has released because errata can be discovered throughout the life of a chip.

They can also be fixed throughout the life of a chip.

To put it another way, this "revelation" would in no way deter me from buying a Yonah-based Mac. What deters me is the lack of Intel-native software and drivers.

[ShotgunEd]

Originally Posted by Eug Wanker

Because his conclusion is not correct.

His conclusion was superceded by another few posts where he asked questions.

I redeemed him because of the progression of the topic.

Others continued to flame based on the initial post.

[ghporter]

I should point out here that not only is the Core Duo the most complex CPU that Intel has yet released, it's also significantly more complex than the G5 (at least twice as complex, due to having two cores that communicate with each other on the same chip).

Drakino has good info-Intel will produce a new stepping of the Core Duo that will address the important issues and microcode will be released to work around the less significant ones that need addressing.

Further, not all "issues" with a particular CPU need to be addressed on every platform that uses it. If an OS doesn't touch a particular portion of that CPU, it is like an issue with that portion doesn't exist.

[Chuckit]

Originally Posted by ShotgunEd

Why is everyone flaming SWG?

This may well be the most flawed chip intel has released because errata can be discovered throughout the life of a chip.

This chip has 34 known errata on launch. This is what the article is pointing out and SWG admitted his ignorance a few posts later and asked for other opinions.

So rhashem offered an opinion. SWG responded by flaming him, complete with "OWNED." This has been the template for every part of the conversation since.

By rhashem: "Do you *really* want to have this argument? Errata documents cannot be compared like that. You have to look at the sevarity of each errata, the existance of workarounds, and the probability of encountering each one. Unless you can give a detailed analysis of those factors showing the P-M is unreliable, you have nothing except FUD, in the truest sense of the word."

SWG replies: "Sure, break it down for us champ. Post links to your sources and let us now how a new CPU out the door already has 34 known bugs and how PowerPC chips are just as bad."

This is very patronizing if SWG admits to not knowing much. Not only that, but SWG just keeps repeating "Intel's QA department has found a bunch of errata internally, so the chip must suck" no matter how many times people in this thread or, heck, the entire comments section of the article he posted point out that the number isn't really indicative of the processor's fitness for use and is only brought up to inspire FUD in people who don't know about processors.

The article really, really looks like FUD to me. It counts the number of errata known so long after release, and then tries to spin that as almost two being found per day. They assume there's a constant stream of errors being found instead of, as it appears, a bunch of errors that were mostly found and properly documented before anybody got his hands on a computer with one of these things.

[analogika]

Originally Posted by Leia's Left Bun

Sorry what am I laughing at? Last time I checked 34 was a higher number than 24.

Not to mention the chip has been out all of 2 weeks to anyone want to take a guess at how many more will be found?

OWNED. You guys kill me

1.) I notice nobody has posted ANY data on DUAL-CORE G5 chips.

2.) It stands to reason that the single-core G5 has been through a dozen revisions since it was first introduced almost two and a half years ago. AFTER ALL THOSE REVISIONS, there are still 24 unfixed errata. How many do you think might have been in the very first revision?

3.) I have a number of friends/colleagues who have been hit FAR harder by APPLE'S ****ed-up mis-implementation of PCI on two generations of dual-core G5 PowerMac towers than ANYBODY will likely ever be hit by these CPU errata.

[Eug Wanker]

Originally Posted by analogika

APPLE'S ****ed-up mis-implementation of PCI on two generations of dual-core G5 PowerMac towers

[Leia's Left Bun]

Originally Posted by Eug Wanker

Because his conclusion is not correct.

My "conclusion"? Sorry what was that again? The fact that if Apple was still PowerPC exclusive we would be laughing at Intel? Ya, that is one heck of a conclusion

[Lateralus]

Originally Posted by analogika

2.) It stands to reason that the single-core G5 has been through a dozen revisions since it was first introduced almost two and a half years ago. AFTER ALL THOSE REVISIONS, there are still 24 unfixed errata. How many do you think might have been in the very first revision?

You say that as if most eratta are fixed. They're not.

[analogika]

There was a massive cock-up in two whole series of dual 2.0/2.2-GHz G5 PowerMacs that made them unusable with two or more PCI cards of certain types, IIRC.

This affected a WHOLE LOT of PCI audio solutions and made a lot of people very angry, though it didn't get a lot of publicity.

[analogika]

Originally Posted by Lateralus

You say that as if most eratta are fixed. They're not.

'Twas said above that many errata ARE fixed over the different revisions of a processor.

I would have thunk that is one of the reasons for stepping in the first place.

[Leia's Left Bun]

Originally Posted by ShotgunEd

Why is everyone flaming SWG?

This may well be the most flawed chip intel has released because errata can be discovered throughout the life of a chip.

Because they switch into full fanboy/steve's reality/marketing mode the second Apple does anything. I love apple but I am not blind to their flaws.

Like I said 1 year ago you know there would have been the same link to the story but it would be about how much intel sucks, how PowerPC is better and nobody would stick up for intel.

[analogika]

Can we all drop the ego bullsh¡t?

This stuff is actually interesting.

And it seems quite clear that there are no real differences between processor manufacturers as regards errata, save that one monumental division error back in the Pentium.

[Eug Wanker]

Originally Posted by Leia's Left Bun

Like I said 1 year ago you know there would have been the same link to the story but it would be about how much intel sucks, how PowerPC is better and nobody would stick up for intel.

We've known about chip errata for just about forever. I've always thought the Opteron is an excellent chip (and still do), and it has more errata than Yonah.

Originally Posted by analogika

There was a massive cock-up in two whole series of dual 2.0/2.2-GHz G5 PowerMacs that made them unusable with two or more PCI cards of certain types, IIRC.

This affected a WHOLE LOT of PCI audio solutions and made a lot of people very angry, though it didn't get a lot of publicity.

Interesting, I didn't know that, and yes, that would be a bazillion times more significant than most chip errata.

[Leia's Left Bun]

Originally Posted by analogika

And it seems quite clear that there are no real differences between processor manufacturers as regards errata, save that one monumental division error back in the Pentium.

According to what sources? So far NOBODY has posted a link to their claims other than me.

[Chuckit]

What claims, specifically, would you like a link to? (Also, with regards to "there are no real differences between processor manufacturers" — it would be up to you to show that there is a big difference. The burden of proof lies with the one making the positive claim.)

[Leia's Left Bun]

Originally Posted by Chuckit

What claims, specifically, would you like a link to? (Also, with regards to "there are no real differences between processor manufacturers" — it would be up to you to show that there is a big difference. The burden of proof lies with the one making the positive claim.)

Links showing the PowerPC has just as many problems and the severity of them.

Is that so much to ask when people make these claims?

You don't see me on hear saying "As far as I remember the power PC had 100 billion bugs and they were all as bad as Intels".

[moodymonster]

list of 27 errata:

http://www-306.ibm.com/chips/techlib...256B0600038768

for the Power PC 750FX aka G3

powerpc 970FX aka G5: 24 errata

http://www-306.ibm.com/chips/techlib...256E93006C957E

powerpc 7457/7447 aka G4: 36 errata

http://www.freescale.com/files/32bit.../MPC7457CE.pdf

The only thing I'm worried about with the Intel chips is how fast photoshop etc will go.

[Leia's Left Bun]

So far the only chip that is worse based on your links is the G4. But we don't know how major/minor they were. Many of them were fixed in later revs.