Quanta sues AMD over defective chips

[savantu]

Quanta Computer Inc. , the world’s largest contract maker of notebook computers, sued Advanced Micro Devices Inc. (AMD) for breach of contract, alleging the chipmaker sold defective products.

AMD and its ATI Technologies Inc. unit sold chips that didn’t meet heat tolerances and were unfit for particular purposes, Taoyuan, Taiwan-based Quanta claimed yesterday in a federal court filing in San Jose, California. The chips were used in notebooks Quanta made for NEC Corp. and caused the computers to malfunction, according to the filing.

http://www.bloomberg.com/news/2012-0...computers.html

I wonder why this hasn't been settled quietly; it's a blow for AMD to upset Quanta and NEC.

[SoF]

It will get really bad for AMD when they fail at the court to proof these news wrong...

[Hornet331]

Upsetting quanta... thats a bad idea...

[informal]

Hmm

“AMD is aware of no other customer reports of the alleged issues with the AMD chip that Quanta used, which AMD no longer sells,” Silverman said. “In fact, Quanta has itself acknowledged to AMD that it used the identical chip in large volumes in a different computer platform that it manufactured for NEC without such issues.”

So either chips weren't defective and quanta failed to meet the cooling requirements or chips were defective but only in this one series of products (which is really strange to say the least).

[god_43]

i call bs. there is no way that an entire series of defective chips would be released from amd or intel. these things are heavily tested. burden of proof lies with quanta, i think they are trying to money grab.

[Lanek]

Look like Quanta have been sued by some client, and they try to return the sue against AMD.

[savantu]

i call bs. there is no way that an entire series of defective chips would be released from amd or intel.

Quanta claims basically that AMD chips burned more power than TDP which caused NEC to fire back to Quanta. Since Quanta designed the cooling solution for the TDP given by AMD, the fault lies with AMD which either did not bin the chips correctly or variation is significant and out of control.

these things are heavily tested. burden of proof lies with quanta, i think they are trying to money grab.

Quanta to grab money from AMD when Quanta revenue is $39B vs. AMD's $6B a year , seems a bit far fetched ? Thirdly, NEC was the final customer. Most likely, NEC got a lot of complains and returns, blamed Quanta and asked for refunds, Quanta said it's not our fault, but AMDs, NEC said I don't care, you two sort it out, but first pay me the money. AMD didn't bulged to Quanta's request and now you have the saga in court...

[[XC] gomeler]

i call bs. there is no way that an entire series of defective chips would be released from amd or intel. these things are heavily tested. burden of proof lies with quanta, i think they are trying to money grab.

Didn't NVIDIA just have a spat of mobile chips that had issues? Yeah.. mistakes still slip through the cracks. Now the problem is whether this was greed or a misunderstanding.

[Nikolasz]

I see lots of defective chips over net nowdays. And wtf companies sell defective chips nowdays as ok to ppl, cmon!!
In old times that was really humiliating when some one sell half defective chips remember zx spectum sinclair?
Nowdays amd and nvidia maybe intel now doing same stuff.

I say this has to stop now! Wtf im spending 300 euros on broken gpu and i found out that days after i spend money for it!

[gallag]

Hmm

So either chips weren't defective and quanta failed to meet the cooling requirements or chips were defective but only in this one series of products (which is really strange to say the least).

i call bs. there is no way that an entire series of defective chips would be released from amd or intel. these things are heavily tested. burden of proof lies with quanta, i think they are trying to money grab.

ADF (amd defence force) any time there is negitive news (or positive for intel) the ADF will be there lol.

[Manicdan]

if they use the same chips from other batches with no issues, then the cooling system might not be at fault. but theres not nearly enough details for us to know exactly whos to blame.

[STaRGaZeR]

What "chips" exactly?

[Lightman]

What "chips" exactly?

The ones AMD no longer sells!
That would imply 45nm Turions and Athlons II, because I'm unaware of any 32nm products being EOLed.

[Russian]

Also, AMD essentially sells most of these chips with a very short OEM warranty and as a result, if they were indeed defective such an event would have probably seen much more widespread failures on the part of consumers.

Furthermore, Quanta has to prove that AMD KNOWINGLY sold them defective processors. If AMD sold them without knowing there was a defect or that they were already bad, AMD is really not on the hook legally unless they wish to save their relationship with Quanta. This lawsuit seems like it took quite a bit of time to make, but I just don't think that any evidence so far has proven that AMD has done so with malice or knowledge of defect.

[trans am]

What "chips" exactly?

clawhammers and newcastles

[god_43]

ADF (amd defence force) any time there is negitive news (or positive for intel) the ADF will be there lol.

lol informal thanks you. you know i could have come into this thread and started crapping on intel shills and what not....but i didn't because that is not constructive. take your trolling else where please.

[zanzabar]

i dont know if any1 has had the pleasure of using an nec box, but if u look at something like lenovo that keeps the fan always on and has copper heat pipes, nec is the opposite and has aluminum blobs on the chips and they only turn the fan on low affter the bottom burns u to the touch then it keeps getting hotter and hotter until it shuts off (so dont try and run a game or occt on it.)

and yes, when i get a new laptop i occt it.

[gallag]

Lighten up man, lets face it, if someone had to bet who would post first in defence of amd it would be you guys lol, and as much as it would please me for informal to like my post if i was betting i would imagine the report and thank buttons were confused lol.

On topic, this is pretty bad for amd, even if they win in court this will hurt their image and annoy a big customer.

[SKYMTL]

i call bs. there is no way that an entire series of defective chips would be released from amd or intel. these things are heavily tested. burden of proof lies with quanta, i think they are trying to money grab.

One multi billion dollar company with teams of lawyers and product specialists sues another multi billion dollar company over a potentially defective product and you call BS? Is that a knee jerk reaction or something?

Read the lawsuit. It isn't BS at all. If you check out the documents filed and read the official press releases, this is about AMD misreporting the thermal characteristics of their chips. They sent information to Quanta, Quanta used that information to design thermal management solutions that were supposed to (if AMD's information was accurate) efficiently disperse the chip's heat. Instead, AMD's chips went OVER their stated TDP envelope, overwhelmed the cooling solution and essentially caused a large amount of RMA due to malfunctioning notebooks.

That's pretty serious stuff IMO but as the article stated, the fact that this couldn't be settled out of court is very surprising.

[god_43]

One multi billion dollar company with teams of lawyers and product specialists sues another multi billion dollar company over a potentially defective product and you call BS? Is that a knee jerk reaction or something?

Read the lawsuit. It isn't BS at all. If you check out the documents filed and read the official press releases, this is about AMD misreporting the thermal characteristics of their chips. They sent information to Quanta, Quanta used that information to design thermal management solutions that were supposed to (if AMD's information was accurate) efficiently disperse the chip's heat. Instead, AMD's chips went OVER their stated TDP envelope, overwhelmed the cooling solution and essentially caused a large amount of RMA due to malfunctioning notebooks.

That's pretty serious stuff IMO but as the article stated, the fact that this couldn't be settled out of court is very surprising.

i just meant that it's hard to believe that amd or intel would let this happen knowingly, the response from amd states that those same chips were used in other systems just fine, so if that's true then quanta must not have done there job. what quanta has to prove is that amd knowingly misreported the facts, if amd did then they should be punished....but i really don't see that being the case, amd cant screw over partners as easily as intel or nvidia could (nv mobile gpu's anyone).

[gallag]

Tin foil hat time, why are quanta doing this? Say what they are sueing for is indeed correct, why bother? It is probaly not big money, especilay to them, i would imagine having a healthy amd stopping an intel monopoly would be worth a lot more to the largest notebook builder. its not just a case of paying damages that will hurt amd, this would damage their already frail reputation. Time for the tinfoil hat part. sue them, devalue them then buy them.

[SKYMTL]

i just meant that it's hard to believe that amd or intel would let this happen knowingly, the response from amd states that those same chips were used in other systems just fine, so if that's true then quanta must not have done there job. what quanta has to prove is that amd knowingly misreported the facts, if amd did then they should be punished....but i really don't see that being the case, amd cant screw over partners as easily as intel or nvidia could (nv mobile gpu's anyone).

Under reporting of TDP happens all the time but likely not with any intent to misrepresent one's products. However, as we have all seen, certain batches of the same CPU can exceed TDP quite easily.

[K404]

For everyone saying big companies thoroughly test parts before release, who remembers Intels "B2" screwup?

[highoctane]

If AMD understated the tdp of a product and it ended up exceeding the capacity of the cooling solution engineered around AMD's claimed specifications then I see reasonable fault with AMD's specification.

Company X makes a train advertised to travel at a max speed of 200mph
Company Y manufactures and installs the track for the train which has a rated max safe speed of 200mph
Who would be at fault if the trains running on the track derailed because the trains where capable of exceeding their max claimed 200mph speed, the max speed for which the rails where designed to operate safely.

[zalbard]

Intel TDP != AMD TDP != Real Life TDP