Various Topics on Videocard ATI

..r.. .. ..e

The difference is that the automotive industry actually test their
prototypes. :-) Consequently the blame lies squarely with Dell and HP for
failing to do proper component integration. It isn't difficult or expensive
to do thermal shock and cycling tests on laptops.

--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."

"Mr.E Solved!" <Iamsingle@askme.out> wrote in message
news:5OmdnS-GyN_Q8QXVnZ2dnUVZ_qDinZ2d@comcast.com...
> So far, every RoHS disadvantage has been demonstrated: cracks,
> brittleness, susceptibility to thermal cycling. All we need is some Tin
> Whiskers and the worst case scenario has come to p***.
>
> Don't blame RoHS however, as other industries besides the Computer
> Manufacturing Industry has used RoHS to advantage, such as the automotive
> industry.
>
> I'm guessing TMSC or whatever foundry made these chips simply had bad
> materials and/or technique from lack of experience, ***uming no malice was
> involved, who can say at this point.

.... ..lve..

First of One wrote:
> The difference is that the automotive industry actually test their
> prototypes. :-) Consequently the blame lies squarely with Dell and HP for
> failing to do proper component integration. It isn't difficult or expensive
> to do thermal shock and cycling tests on laptops.
>

Wow, you want to lay the blame on Dell and HP?

Why not Clevo and Compal and Quanta (who actually make HP and Dell
notebooks) or even Mitac, Asus, Acer, Arima, ECS/Uniwill.

No, you can't blame the laptop maker for a bad video chip package that
they in good faith installed.

Of course, if they knew the chips were faulty and installed them
anyway...that's a whole different concept of culpability.

..r.. .. ..e

"Faulty" isn't as cut and dry as you may think. nVidia doesn't necessarily
design the cooler in a laptop. Moreover, the GPU isn't the only
heat-producing component in the system. The company that designs the whole
laptop is ultimately responsible to ensure all the parts they've selected
can survive the laptop's internal temperatures. I imagine Dell and HP still
design their laptops, then contract out the manufacturing.

Thermal cycle testing is easy to do. Take for example the Dell XPS M1530,
it's advertised to operate in ambient temps of 0 to 35C.
http://support.dell.com/support/edoc....htm#wp1104221
It's easy enough to get a small chamber to alternate between 0C and 35C in
5-minute intervals, then script a couple of burn-in apps such as Toast and
ATiTool to put the CPU and GPU under load when the chamber cycles to 35C. In
two weeks they can ac***ulate over 2000 extreme thermal cycles and validate
the product.

If Dell and HP "trusted" their contractor manufacturers to cobble together
laptops from a paper design, without any qualification testing of
prototypes, then yes it's Dell and HP's own damn fault.

--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."


"Mr.E Solved!" <Iamsingle@askme.out> wrote in message
news:QpCdnaCZAtevMQTVnZ2dnUVZ_uudnZ2d@comcast.com. ..
> First of One wrote:
>> The difference is that the automotive industry actually test their
>> prototypes. :-) Consequently the blame lies squarely with Dell and HP for
>> failing to do proper component integration. It isn't difficult or
>> expensive to do thermal shock and cycling tests on laptops.
>>
>
> Wow, you want to lay the blame on Dell and HP?
>
> Why not Clevo and Compal and Quanta (who actually make HP and Dell
> notebooks) or even Mitac, Asus, Acer, Arima, ECS/Uniwill.
>
> No, you can't blame the laptop maker for a bad video chip package that
> they in good faith installed.
>
> Of course, if they knew the chips were faulty and installed them
> anyway...that's a whole different concept of culpability.

.... ..lve..

First of One wrote:
> "Faulty" isn't as cut and dry as you may think. nVidia doesn't necessarily
> design the cooler in a laptop. Moreover, the GPU isn't the only
> heat-producing component in the system. The company that designs the whole
> laptop is ultimately responsible to ensure all the parts they've selected
> can survive the laptop's internal temperatures. I imagine Dell and HP still
> design their laptops, then contract out the manufacturing.
>
> Thermal cycle testing is easy to do. Take for example the Dell XPS M1530,
> it's advertised to operate in ambient temps of 0 to 35C.
> http://support.dell.com/support/edoc....htm#wp1104221
> It's easy enough to get a small chamber to alternate between 0C and 35C in
> 5-minute intervals, then script a couple of burn-in apps such as Toast and
> ATiTool to put the CPU and GPU under load when the chamber cycles to 35C. In
> two weeks they can ac***ulate over 2000 extreme thermal cycles and validate
> the product.
>
> If Dell and HP "trusted" their contractor manufacturers to cobble together
> laptops from a paper design, without any qualification testing of
> prototypes, then yes it's Dell and HP's own damn fault.
>


Holy cow, you are hell bent on defending nvidia and part of that defense
is your imagination? "Imagining" Dell and HP still design their
laptops? I expect better from you, at least say: "I don't know which
engineers are responsible for which part of the process".

I know quite a deal about this process and I don't know which engineers
***emble/design precisely which parts, as they are very complex and take
place over multiple locations. But I do know that the bulk of the entire
process takes place in Asia, not stateside at Dell or HP with Dell or HP
personnel.

It's probably that type of non-rigorous thinking that caused the release
of the faulty substrate in the wild to begin with. As in: but for the
faulty substrate, the chips, video subsystem and laptop would work
normally with the components that are in them with their existing
cooling solutions.

Since I did not read the supplied thermal or electrical specifications,
I do not know if the manufacturers were told: "By the way, this stuff
has a worse performance envelope than all the stuff before it" or "No
changes needed."

Either way, the supplier is usually contracted in writing to certain
specifications and other implied warranties are in force depending upon
geographic location and mutual agreement. So they (Nvidia) can't take
the attitude you are seemingly taking: Haha, we gave you junk and you
didn't test it thoroughly enough to find our hidden flaws, eat it.

I'm sorry if your nvidia stock is weaker, but so far all reports point
to a persistent state of knowledge of a faulty product by nvidia. So
much so that nvidia partners and co-dependents are clamoring for a
recall by nvidia so they don't go bankrupt trying to fix the growing
problem.

..n

"Mr.E Solved!" <Iamsin...@askme.out> wrote:

> I'm sorry if your nvidia stock is weaker, ...

Could get weaker yet. Fuddy is speculating:
"Nvidia M84 & M86 problems to cost more"
<http://www.fudzilla.com/index.php?
option=com_content&task=view&id=8782>
"Monies set aside will not be enough to cover cost
...."

And it's not just the parts, labor, S&H.
If that story of the laptop graphics being excluded from some
laptop warranty is true, cl*** actions are a certainty.
Any number of lawyers will be found to own affected LTs.

Fud closes with:
"All of these issues could be a very big advantage for AMD ..."
Only if the back-alley Chinese fab at the root cause didn't
build stuff the same way for other clients.

--
Regards, Bob Niland mailto:name@ispname.tld
http://www.access-one.com/rjn email4rjn AT yahoo DOT com
NOT speaking for any employer, client or Internet Service Provider.

..S

"rjn" <email4rjn@yahoo.com> wrote in message
news:5c96bb07-345b-484f-82bf-7e95075faa18@d77g2000hsb.googlegroups.com

[...]

> Fud closes with:
> "All of these issues could be a very big advantage for AMD ..."
> Only if the back-alley Chinese fab at the root cause didn't
> build stuff the same way for other clients.

TSMC ain't no back-alley operation! Which makes their side of the story all
the more interesting.

..r.. .. ..e

"Mr.E Solved!" <Iamsingle@askme.out> wrote in message
news:G5qdnVNVM4gh1AfVnZ2dnUVZ_t7inZ2d@comcast.com. ..
> Holy cow, you are hell bent on defending nvidia and part of that defense
> is your imagination? "Imagining" Dell and HP still design their laptops? I
> expect better from you, at least say: "I don't know which engineers are
> responsible for which part of the process".

Ultimately Dell or HP's logo appears on the laptop. They are the final
system integrators. They may choose to delegate a portion of the integration
to a third party to cut costs, but they are still liable to ensure the end
product works.

> It's probably that type of non-rigorous thinking that caused the release
> of the faulty substrate in the wild to begin with. As in: but for the
> faulty substrate, the chips, video subsystem and laptop would work
> normally with the components that are in them with their existing cooling
> solutions.

The truly non-rigorous thinking exists where the system builder no longer
tests its own products in an attempt to rush to market.

> Since I did not read the supplied thermal or electrical specifications, I
> do not know if the manufacturers were told: "By the way, this stuff has a
> worse performance envelope than all the stuff before it" or "No changes
> needed."

The latter is a pretty bold statement to make, since nVidia doesn't have
direct control over the cooling solution. Different laptop ch***is will have
vents in different places. The heatpipes may run through both the CPU and
GPU plates, etc. In fact, if nVidia were to make a statement like "no
cooling changes needed", it would be poor engineering judgement for system
builder to not validate it for themselves. The system builders are not
innocent victims in this case.

> Either way, the supplier is usually contracted in writing to certain
> specifications and other implied warranties are in force depending upon
> geographic location and mutual agreement.

Specifications won't cover every possible failure mode. And you can be sure
things like thermal fatigue cycles don't make their way into contracts.

As for implied warranties, don't confuse consumer protection laws with
supplier/integrator relationships. In fact, for new products or design
changes, the integrator often defines the [re]qualification tests that the
supplier has to do.

> So they (Nvidia) can't take the attitude you are seemingly taking: Haha,
> we gave you junk and you didn't test it thoroughly enough to find our
> hidden flaws, eat it.

No, it's more like: My GPU technically meets my published specs, but since I
didn't design your laptop, you are responsible for making sure it works in
your particular application.

This is true even in automotive industry, where suppliers handle much of the
design and subsystem testing. Yet the automaker still puts the prototypes in
the desert, in the arctic, on the Nurburgring...

> I'm sorry if your nvidia stock is weaker,

Don't ***ume. I shuffle my money between oil and gold, and leave the tech
stocks alone. As "Mr. Tony" would say, I don't **** where I eat.

> but so far all reports point to a persistent state of knowledge of a
> faulty product by nvidia. So much so that nvidia partners and
> co-dependents are clamoring for a recall by nvidia so they don't go
> bankrupt trying to fix the growing problem.

So Dell and HP ***** loudly and p*** the buck. The "journalists" at the Inq
then take little bits of information and spin them into FUD. And all of a
sudden it becomes a "persistent state of knowledge"?

--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."

.... ..lve..

First of One wrote:

> So Dell and HP ***** loudly and p*** the buck. The "journalists" at the Inq
> then take little bits of information and spin them into FUD. And all of a
> sudden it becomes a "persistent state of knowledge"?

All interesting points but I don't think that they apply. I think your
position "here are our chips, we wash our hands of them" is not going to
hold up either in the boardroom or courtroom and based on recent events
it has not. Supplier-Integrator responsibilities go far beyond the
"consumer protections" end-user requirements you think I'm confusing
with the 'fitness of purpose' requirements that are usually negotiated,
I haven't seen the contract so I can't comment on specifics, but I've
seen enough of them to know what is typical and manufacturing defects
such as this are typically covered.

Especially when the problem encomp***es many vendors in many different
designs and best practice cooling solutions do not result in other chips
failing. But for the bad chip packaging, there would be no fault. No
problems.

Especially when this was discovered in early 2007, a year and a half ago
and reported in April of 2007. Plenty of time to fix things, but nothing
was fixed. This didn't happen over a week or a day, it took months and
months.

http://www.theinquirer.net/gb/inquir...-g846-problems

Oh, you don't think the Inq can report the truth? why not take it from
nVidia themselves:

NVIDIA president and CEO Jen-Hsun Huang stated:

"Although the failure appears related to the combination of the
interaction between the chip material set and system design, we have a
responsibility to our customers and will take our part in resolving this
problem. The GPU has become an increasingly important part of the
computing experience and we are seeing more interest by PC OEMs to adopt
GPUs in more platforms. Recognizing that the GPU is one of the most
complex processors in the system, it is critical that we now work more
closely with notebook system designers and our chip foundries to ensure
that the GPU and the system are designed collaboratively for the best
performance and robustness."


Again, if nvidia knew about the chip material sets different performance
characteristics and did not tell anyone, that is their fault. If they
did and the makers ignored their warnings, then it's the makers
fault...for building a broken machine and using a broken component.

HP has split the difference in cost with nvidia, but no one else
has...the rest are demanding recall and restitution and are jumping
ship. Hard to say why, but HP can afford the $150 out of $300 liability
each repair is estimated to cost. Are you saying HP knew the chip was
faulty as well and crossed their fingers?

So far it looks like nVidia left everyone swinging in the breeze and
nothing you have said indicates otherwise, are you SURE you don't have
any nvidia stock?!

..inendle.. ..ail.c..

On Aug 8, 11:21*am, "Mr.E Solved!" <Iamsin...@askme.out> wrote:
> First of One wrote:
> > So Dell and HP ***** loudly and p*** the buck. The "journalists" at theInq
> > then take little bits of information and spin them into FUD. And all ofa
> > sudden it becomes a "persistent state of knowledge"?
>
> All interesting points but I don't think that they apply. I think your
> position "here are our chips, we wash our hands of them" is not going to
> hold up either in the boardroom or courtroom and based on recent events
> it has not. Supplier-Integrator responsibilities go far beyond the
> "consumer protections" end-user requirements you think I'm confusing
> with the 'fitness of purpose' requirements that are usually negotiated,
> I haven't seen the contract so I can't comment on specifics, but I've
> seen enough of them to know what is typical and manufacturing defects
> such as this are typically covered.
>
> Especially when the problem encomp***es many vendors in many different
> designs and best practice cooling solutions do not result in other chips
> failing. But for the bad chip packaging, there would be no fault. No
> problems.
>
> Especially when this was discovered in early 2007, a year and a half ago
> and reported in April of 2007. Plenty of time to fix things, but nothing
> was fixed. This didn't happen over a week or a day, it took months and
> months.
>
> http://www.theinquirer.net/gb/inquir...there-are-no-m...
>
> Oh, you don't think the Inq can report the truth? why not take it from
> nVidia themselves:
>
> NVIDIA president and CEO Jen-Hsun Huang stated:
>
> "Although the failure appears related to the combination of the
> interaction between the chip material set and system design, we have a
> responsibility to our customers and will take our part in resolving this
> problem. The GPU has become an increasingly important part of the
> computing experience and we are seeing more interest by PC OEMs to adopt
> GPUs in more platforms. Recognizing that the GPU is one of the most
> complex processors in the system, it is critical that we now work more
> closely with notebook system designers and our chip foundries to ensure
> that the GPU and the system are designed collaboratively for the best
> performance and robustness."
>
> Again, if nvidia knew about the chip material sets different performance
> characteristics and did not tell anyone, that is their fault. If they
> did and the makers ignored their warnings, then it's the makers
> fault...for building a broken machine and using a broken component.
>
> HP has split the difference in cost with nvidia, but no one else
> has...the rest are demanding recall and restitution and are jumping
> ship. Hard to say why, but HP can afford the $150 out of $300 liability
> each repair is estimated to cost. Are you saying HP knew the chip was
> faulty as well and crossed their fingers?
>
> So far it looks like nVidia left everyone swinging in the breeze and
> nothing you have said indicates otherwise, are you SURE you don't have
> any nvidia stock?!

My 780i still has vc after 4 bios updates. Failure is nvidia's main
product

..ri..

kainendless@gmail.com wrote:

>My 780i still has vc after 4 bios updates. Failure is nvidia's main
>product

Ever hear of trimming your posts?