A note about Intel quality...

Parma_Endorion · Nov 11, 2001

I took a look at the latest CPU specification updates from Intel and Amd. After a quick look I found these interesting results...

Erratas (bugs) found in PIII so far ~ 80
Erratas (bugs) found in PIV so far ~ 57
Erratas (bugs) found in Athlon so far ~ 7

ftp://download.intel.com/design/pentiumiii/specupdt/24445334.pdf
ftp://download.intel.com/design/pentium4/specupdt/24919914.pdf
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/23614.pdf

Is this the quality and stability all Intel fans are talking about? (Sorry, I had to ask.)

/* The more you know, the more you realize how little you know */

Guest · Nov 11, 2001

It certainly says something about quality. It doesn't say much about stability.

rraquib · Nov 11, 2001

Well the reason so many of the intel bugs has been discovered is the sheer number of intel users. AMD user base is tiny in comparision. Intel R&D also dwarfs AMD R&D. I would think given the same amout of exposure AMD will have just as many found bugs.

AmdMELTDOWN · Nov 11, 2001

Intel erratas are insignificant and are minimized with each stepping.

the biggest errata on the AMD chip is the "none or half-a$$" thermal protection that stays with on each stepping, AMD hasn't figured a way around that errata yet! LOL!

"AMD/VIA!...you are still the weakest link, good bye!"

jiffy · Nov 11, 2001

I thought they did, and needed the mother boards to address this in their future mobo.Was this ever an issue, before the Cooling off review came out. I mean how many poeples computer burned up?(So, let's set the record straight: through our video, we revealed the Palomino's major flaw when it is used with modern mainboards. AMD responded to our video and will be setting new guidelines for all mainboard manufacturers. These consist of a modification for all upcoming Socket A motherboards, which are to be integrated with a special electronic switching mechanism to protect the CPU from overheating.

The good news about this solution is that it guarantees that the CPU will survive; the bad news is that data safety is no longer guaranteed. The PC is not shut down properly, and instead it is switched off quite abruptly)

defrage is child's play-fdisk

mr_gobbledegook · Nov 11, 2001

Well Athlon is an older chip than the P4..when was it released again ?. The P4 was released after the Athlon and yet it still has more bugs. Thats all I'm saying.

~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)

MeTaLrOcKeR · Nov 11, 2001

Stop blamming Thermal Protection blah blah and how AMD is crap because of there "Errata"......jesus christ man..
Errata is a Word used by INTEL instead of saying BUGS....Like Processor Flaws......

I'l explian somwething funnier and dont start flamming me saying im a troll becasue im 100% not.....again i dotn use AMD....

Athlon Slot A came out I believe August, 1999.....
P6 Core Design from Pentium Pro Socket 8.....1996.....right MeltDown ??

THE PIII IS NOTHIGN MORE THAN A REVISED PENTIUM PRO.....AND THERE STILL WHAT, 87 errata ?? Please explian this....right now.....explian it, and dont go back to amd and pis poor fragile POS....EXPLAIN TO ME WHY THE INTEL CHIPS HAVE ALL TEH BUGS......Please........after what......5 almost 6 years now Intel STILL cant get there bugs outta there chips ??

Jesus christ dude, and the Athlon core has only got 7.....and its only 2 years old.........

P4......alredy 50 some odd bugs.......only a year old........man...and you gusy complain about how bad AMD is.........go face the corner.......stop judgign befoer you know the facts, just beacsue you cant setup a system properly...or you purposely make the AMD systems you've set-up to fail doesnt mean you have to blame the processor....u nkow what a Chipset is right ???? ok than....stop blamming AMD now......

-MeTaL RoCkEr
My Z28 can take your P4 off the line!

Guest · Nov 11, 2001

lol ya unable to do a micro codeup date on the athlon.

-Spuddy

Being Evil Is Good. Cause I Can Be A Prick And Get Away With It. :lol:

LoveGuRu · Nov 12, 2001

hehe Erreta..funny word..


*******
*K.I.S.S*
*(k)eep (I)t (S)imple (S)tupid*
*******

HolyGrenade · Nov 12, 2001

Quick English lesson, though it's too late to save your embarrassment.

Errata = plural

Erratum = Singular

Thus, no such word as erratas. And I wouldn't use that word anywhere but only to list the typos in a book or report. There are much better words to use elsewhere...

99% of statistics are made up! - Leonardo daVinci

MadCat · Nov 12, 2001

"...
Erratas (bugs) found in PIII so far ~ 80
Erratas (bugs) found in PIV so far ~ 57
Erratas (bugs) found in Athlon so far ~ 7
..."

How do we know the real count of bugs? Maybe AMD don't disclose as many. If this bug count was conducted by an independent third party, then maybe it would be more significant indicator of relative quality between the microprocessor.

FWIW, I don't feel good about AMD because of the messed up JPEG image fiasco (execution speed path problem in MMX in some AMD CPUs). Seems like you have to visit a German site before finding out the AMD admitted that some of the processor made have problems.

Kemche · Nov 12, 2001

I Have to Agree with MadCat. I think everyone knows how state of the Art Intel's QA Department is. I think they would find more bugs then AMD would. I also don't think the ~7 Bug number that AMD came up with is true. If you just look in Toms Hardware Page you will see more then that. Also, I think for AMD the big errata is placing the Heatsink wrong and cracking/breaking your processor. It's too bad it's not documented anywhere in the Link posted above.

Keep it Cool
KG.

AmdMELTDOWN · Nov 12, 2001

"Thus, no such word as erratas. And I wouldn't use that word anywhere but only to list the typos in a book or report. There are much better words to use elsewhere..."

read my post again I don't think I used the word errata's.

btw, it's funny how you pups can't even come back at me with a intelligible retort, ya'll have to settle for the measly day old nitpick don't you, loser.

"AMD/VIA!...you are still the weakest link, good bye!"

Kelledin · Nov 12, 2001

How do we know the real count of bugs? Maybe AMD don't disclose as many. If this bug count was conducted by an independent third party, then maybe it would be more significant indicator of relative quality between the microprocessor.

FYI, the majority of CPU bugs are reported by hardcore users with engineering and assembly-language knowledge--and of course, there have been many instances where Intel tried to cover up these bugs. FDIV, anyone?

I personally think the Athlon just hasn't been out long enough to have all its bugs uncovered.

That doesn't really explain the P4's bug count, though, but really, the P4 was a poor hack-job that Intel was rushed to release. Probably a great many of the loose ends were held in with the wafer equivalent of duct tape and a prayer. The first P3 1.13GHz showed us just how far Intel might cut corners when desperate.

AmdMeltdown:

I don't think i used the word erratta's

Apparently, you don't think, period. Read the post again. :lol:

Kelledin
<A HREF="http://www.linuxfromscratch.org/" target="_new">LFS</A>: "You don't eat or sleep or mow the lawn; you just hack your distro all day long."

Guest · Nov 12, 2001

You are a rude hemit like man i would think you would be a tad nicer. But no yer a power tripping white guy with too much time on yer hands pity

.

-Spuddy

Being Evil Is Good. Cause I Can Be A Prick And Get Away With It. :lol:

vader1882 · Nov 12, 2001

according to Parma's numbers when intel releases the P6 it will only have eleven errors in it.

intel amd the ultimate deathmatch

Parma_Endorion · Nov 12, 2001

AFAIK more Athlon’s has been sold than P4's. So that means Athlon should have more bugs discovered since it has a larger user base, right?

Intel’s R&D better? Please, they just have an easier job finding the bugs since there are a lot more of them...

/* The more you know, the more you realize how little you know */

Parma_Endorion · Nov 12, 2001

Insignificant? I think releasing a CPU that can't calculate floating point numbers correctly isn't insignificant. (P5 core aka 585.9996432)

/* The more you know, the more you realize how little you know */

Guest · Nov 12, 2001

>AFAIK more Athlon’s has been sold than P4's. So that means
>Athlon should have more bugs discovered since it has a
>larger user base, right?

Good point. Although its also probably true the Athlon appeals to a different public then the P4. I guess more Athlonn users are home users and gamers, while more p4 users are corporate users. I dont think gamers, tweakers and overclockers are the typical users that will discover a cpu flaw. I think that really takes a low level assembly developpers.

I kinda doubt most bugs (errata) are discovered by end users. I think its more likely they are found by companies like chipset vendors or developpers. Hence, the size of the 'installed' base is not so important. Its rather the number of partners, developpers, etc.

finally, while the numbers given are something to to consider, I think the nature of the bug could be much more important than the number. Most of these errata will probably never ever show up or affect anything. One nasty bug could be much worse than 20 irrelevant ones.
Still, the comments on the age of the P6 core, and its staggering ammount of errata is indeed surprising.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

girish · Nov 12, 2001

Erratum = a note of error
Errata = a list of notes of errors

So Erattas could be said to be a set of notes of errors, which are pretty high in number. A Volume I, Volume II of errata I guess ;-)

Any Ph.D English to resolve this???

No system is fool-proof. Fools are Ingenious!

Kelledin · Nov 12, 2001

First-year Latin might help. Or it may just give you a headache, which I consider to be rather fun. I'm a bit sadistic like that.

"erro" is a Latin verb meaning "make a mistake"; "erratum" is taken from its infinitive form and treated as a neuter-form Latin noun, second declension. Suffixes for different uses are as follows:

singular nominative: erratum
singular genetive: errati
singular dative: errato
singular accusative: erratum
singular ablative: errato
plural nomitative: errata
plural genetive: erratorum
plural dative: erratis
plural accusative: errata
plural ablative: erratis

The only ones that actually translate into English are the singular and plural genetive (genetive use being the subject of a clause). Yo u could say the accusative form also translates, but only because it is not distinguishable from genetive when taken out of context.

Ok, that's a wrap. Tempus could probably explain it a bit better, I suppose; four semesters of high-school latin 5 years ago can only go so far.

Kelledin
<A HREF="http://www.linuxfromscratch.org/" target="_new">LFS</A>: "You don't eat or sleep or mow the lawn; you just hack your distro all day long."

Parma_Endorion · Nov 12, 2001

Another "insignificant" erratum:

http://www.theinquirer.net/12110106.htm

/* The more you know, the more you realize how little you know */Edited by Parma_Endorion on 11/12/01 09:10 AM.

Parma_Endorion · Nov 12, 2001

Wrong! Micro-code updates has been possible since the first K6 chip.

/* The more you know, the more you realize how little you know */

mr_gobbledegook · Nov 12, 2001

He is another bug with Intel Chips or more correctly with Win XP. Apparently PIII-M processors are over heating because of Win XP.

<A HREF="http://www.theinquirer.net/12110104.htm" target="_new">http://www.theinquirer.net/12110104.htm</A>

Microsoft are bad !

~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)

girish · Nov 12, 2001

true!

but they are usually updated with BIOS updates, you dont get them separately to flash them into your processor!

No system is fool-proof. Fools are Ingenious!

A note about Intel quality...

Distinguished

Guest

Guest

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Guest

Guest

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Guest

Guest

Distinguished

Distinguished

Distinguished

Guest

Guest

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Share this page