BRIGHT SIDE OF NEWS About | Advertise | Contact

Users profile

Username:
Password:

Lost password

Please enter your email address

Send

New user

Proceed to registration page.

| Register

SUBSCRIBE Newsletter | RSS Feeds

HOME

APPLE

GRAPHICS

CLOUD COMPUTING

ENTERPRISE

SOFTWARE

BUSINESS

ENTERTAINMENT

SECURITY

News Analysis Interviews Reviews Rumors

Thursday, March 18, 2010

Email this to a friend.

Your friend's e-mail:
Your Name:
Your e-mail:
Message subject:

GT300 Demo card was a Dummy; Does that really matter?

10/5/2009 by: Sean Kalinich - Get more from this author

So we hear that the Tesla card shown at GTC by NVIDIA was a dummy; no GPU, no Memory, just a PCB and cooling solution on it. This information was "leaked" from a source inside AMD and sent around the internet until it found a home with someone willing to write about it as a sign that the GT300 is in trouble and that NVIDIA was/is lying about it.

The question I want to know is; is this really news worthy? Ask around and you will find that 99% of products shown to the press and intended for them to handle or even be displayed are simply dummies with either dead components or nothing under the "hood" at all. After all these products are going to be shipped, handled and [more likely than not], killed by accidental ESD [electrostatic discharge].

Given this, why would any company risk a working chip[set] at a press event? They wouldn't. NVIDIA [as of this morning] responded to the news and stated that yes the card shown off was a dummy but that there was a live card in the system used to demo the Tesla performance and that the performance numbers shown were done on a live and working GT300 card and not a simulated environment. The company failed to disclose the pictures of the real GT300 card, though.

So next you will hear that this means they must not have enough GPUs to go around [due to bad yields of course] but is this true? After all the 5870 [and 5870 X2] cards that AMD passed around for the press to handle were not live and working cards and 90% of the mainboards/Graphics cards shown at CES were not live and working either. Also, whenever you see a vendor claiming a breakthrough in the world of SSDs, bear in mind that those shiny casings or PCIe cards aren't functional, and yes - don't feature memory chips at all. Does this mean that those companies were having supply problems too? Does it mean they did not have working versions? No it does not. It simply means that a display demo does not need to be working to be used for its intended purpose.

What this news [and how it got out] really shows is that AMD is very concerned about the GT300 and what it might be able to do. Given the inside look we have on both architectures, we simply don't understand why AMD is doing so. True, they don't have some of the features "Furbie" has [we heard that the competitors are using "Furbie" to describe Fermi architecture and the GT300/NV70 chip] When the mud starts being slung over a common practice in trade shows it is pretty strong evidence that all is not as it appears. Having frequented many auto trade shows where concept and future model cars were shown off I can tell you than many of them were only shells. They had no engine, no transmission, and would have only moved if pushed down a hill. It is the same thing here.

So you can listen to the FUD or you can wait and see how things play out in this one. AMD seemed happy to leave things when the world had no information on NVIDIA's new DX11 solution; now that they have pretty much announced it, AMD is coming out to the press with FUD. To me this is just plain dirty pool and also flies in the face of their public position on a fair and open market to compete in.

Personally I would be highly suspicious of any information "leaked" from a competing company. After all, do you really think they would have something complimentary to say about any product that is not theirs?

Tags:

NVIDIA, Fermi, cPUG, GT300, Press, Marketing, Trade Shows

Share and enjoy :)

Comments:

well duh by: Anonymous on 2/27/2010

Of course it was a dummy and if you didn`t figure that out then you are too.

nVidiot Site Of News by: Anonymous on 1/21/2010

Why don´t you be honest and call this site nVidiot Site Of News, when this is what this site really is about.

The fakes have it. by: Anonymous on 1/5/2010

You see Sean,

That card isn't due out until march now!
Nvidia knew they couldn't get a card out and needed to slow ATI down. In the end it was TSMC that dropped the ball, but thats besides the point.
Anyway, Speculation is trying to glean the truth from the lies, the cause from the effect, the purpose from the marketing.

Nvidia's response was based on negating sales loses and trying to convince people they have something under the hood.

I'll be surprised if I see a bench mark at CES.

GT300 Where Art Thou? by: Greg442 on 11/9/2009

Well, we've seen the paper launch, now lets see the real thing.

by: Anonymous on 11/2/2009

"Ask around and you will find that 99% of products shown to the press and intended for them to handle or even be displayed are simply dummies with either dead components or nothing under the "hood" at all."

[citation needed]

@ Ironic by: Sean Kalinich on 10/29/2009

actually nVidia did admit it was a dummy card. They called it a Mock up. but have stated over and over the demo was on a real Fermi Card.

I am not sure how that is FUD. When a company releases a new product they do not say "here is a mock up of the new Prduct X" they say "here is product X"

It is an important thing to take note of as it is a common practice in all industries.

It was the FUD that followed claiming that the dummy (fake) card was proof of a conspiracy by NV that is the issue.

by: Anonymous on 10/28/2009

This is ironic, given the whole Fermi launch was Nvidia providing FUD against the 58xx launch.

AMD has product out there and selling now - Nvidia just has promises and very badly made dummy product that they won't admit to. Nvidia are the only people producing FUD in this matter.

lol by: Anonymous on 10/13/2009

this information was "leaked" from within amd...

every single person who looked at the card or the pictures of it and knows the basics of pc hardware could tell it was a dummy card... it was so obvious... alleging that amd "leaked" this or started a pr campaign against nvidia for showing a dummy card is ridiculous...

@Sean sean sean by: Sean Kalinich on 10/11/2009

Again, this is a normal business practice. AMD does it to Intel. Do you know that AMD typically rents out a suite close to where Intel is having IDC and pulls people in there to show them the virtues of AMD?

Intel "leaks" information to dampen AMD announcements. ATi did it to NV. NV does it to AMD/ATi.

I agree that NV should have shown more, they should have shown a DX11 demo. They should have shown a game running. That is what I think was missing out of the Demo.

I personally do not care if there was a piece of pizza under the hood. I have seen more mock ups, demo shells and simulated software than I can to write about.

The fact remains that the fake card is not the real news. It is a diversion, and one that AMD would like people to focus on right now.

58xx is a good card, I have not said otherwise. But with this being pushed and talked about it does show that they could be very concerned with Fermi and NV right now.

I would like to see people get over the faked card and look at what the demo showed and what it could mean for the future of the GPU (and not just from NV).

Sean Sean Sean by: Anonymous on 10/9/2009

People have been awaitng the 5800 series for months, and also seeking news of the GT300. September 22nd ATI released the 5870 in an amazing fashion, the eyeinfinity, 24 monitors thing was very impressive. In all the fanfaire of the 5870 release approximately 1 week later Nvidia demo's a "mocked up" GT300. Coincidence? I think not. Okay demo cards are a normal industry standard, we get that. What we have a hard time with is that this was done so close to the released date of the ATI 5870. In my opinion it "appears" Nvidia was trying to tarnish the launch of the 5870, thats why the whole fake demo card is causing such and uproar. While demo cards maybe normal, its the timing of the fake demo that's problem here. It was poor judgement to show a "mocked up" demo card in this climate, with the 5870 newly released. They should've just shown the actual card, (if they have one) wires hanging out, duct tape and all, then they wouldn't have this mud on thier face, for showing the "dud."

The root of the issue by: Sean Kalinich on 10/9/2009

"Lord Jesus, what is WRONG with you? How many sites ran articles on the Fermi architecture on that day? Only MOST of them. To the point I was getting pissed because I was more interested in reading the 5850 reviews and that was getting crowded out by the Nvidia Fermi story. "

You were more interested in reading about the 5850....

This pretty much explains your perspective on the issue. This is an agnostic view of thing and not an anit-nVidia one.

It is not hard to sell out when there is a short supply one that even AMD admitted to.

here is a short list of other people doing the same thing....

The Fermi Demo was a Demo.
They did not show the card that it was run on..

The AMD DX11 Demo at Quakecon was a Demo.
AMD did not show the card that it was run on.

Phenom I was demoed prior to launch at 3GHz, the launch clock speeds were much lower. Yet AMD claimed they would scale to 3GHz at launch.

The Fermi Card was a Teaser Photo Op done very poorly. The Dell Showing was a product release/announcement of a product launching this month. The Amado XPS was waived around then put away. Engadget ran the story about the Adamdo only being waived around. It compares as there is no proof that Dell has this ready for launch or that it is anything other than a pipe dream at this stage.

Raytracing Demos by Intel are of relevance as they can be run on a CPU. The actual system was closed and people were not allowed to see into it. SO the Raytracing Demo might have been faked and only run on the CPU.

There are four examples of other companies doing the same type of thing.

You have already stated your alliance yet you accuse me of being biased. Something that does not fit with my comments nor with the article.

by: Anonymous on 10/9/2009

***So,as with my other article, I guess that is why Dell just did the same thing. They are worried that other companies are selling better so they "flashed" the Adamo XPS without showing it off or even opening it.***

?? How is this remotely pertinent? OR significant? Comparing a computer makers new model compared to a graphics company next generation chip ... ... man, that makes NO sense, doesn't even belong on the same planet.

***I see no mass defections to the 5800 that you speak of, Many OEMs are still reporting shortages of the GPU. Remember that at Quakeconn AMD did not show the press the card or the rest of the parts in the system being used for the demo. They did the same thing with Eyefinity.***

?? Again, wtf? You're not making sense. It's a new card, with the initial run already sold out and production ramping up. There can be no serious question AMD can sell all the 5870 and 5850 cards they can produce through the rest of the year. How about instead of strawman arguments you try to make honest rational points?

***Again I am not saying anything bad about AMD, these are examples of how this is nothing new and only being used to spread bad PR and FUD about another company while doing exactly the same thing themselves.***

You have given NO valid examples of AMD doing 'exactly' what Nvidia just did.

***It is wrong to apply one set of standards to one company and not to all. Yet I did not see a large number of articles about the "fake" Adamo used at Dell's Launch of it. Nor did I see any about the locked cases for the AMD DX11 and Eyefinity demos.***

First, it was not a 'launch', it was a teaser photo op. Second, it's JUST A LAPTOP MODEL. Third, the whole IDEA of such a product is HOW IT LOOKS, actual RELATIVE performance is secondary. Fourth, You didn't see any other articles about the Adamo being 'fake' because no one else is ******** enough to CARE. 5. The AMD demo's were done months BEFORE any next generation cards were released, there was a REASON to keep some things under wraps. But by Nvidia's presentation, the AMD cards were already released, there was no REASON for Nvidia to cover anything up.

***The same can be said about Raytracing demos given by Intel...***

Again, no relevence. It was PURELY a technical demonstration of ray tracing capability, and so understood as such ... by ALMOST everyone.

***It is only a small number of sites with a history of anti-NVIDIA articles that are running this as a major story. My question above and still is "why?"***

The only site it was a MAJOR story on is Semi-Accurate. It was a story on several other tech sites BECAUSE IT IS A RELEVANT STORY?

***Why is it the one thing they are covering?
Why does anyone feel that showing off this dead and faked card means anything?***

Lord Jesus, what is WRONG with you? How many sites ran articles on the Fermi architecture on that day? Only MOST of them. To the point I was getting pissed because I was more interested in reading the 5850 reviews and that was getting crowded out by the Nvidia Fermi story.

***You could conversely say that showing a disabled card means that the BOM used must not be working out and needs to be rebuilt.

That would not be true, but it could be speculated also.***

Utterly no proof. Whatsoever. Or any reason for them to do so. Maybe AMD didn't want someone smuggling one of the cards out and pre-releasing benchmarks? That thought ever cross your mind?

Same can't be said of Nvidia, can it? Wasn't even a possibility their card was a 'needing to be rebuilt' BOM card was there? Because THEIR card was a crude chop job of some OTHER card they had laying around that they CLAIMED was a real Fermi card.

Dude, you are Wile E Coyote 30 feet past the edge and nothing but air under you feet. I know it sucks to be so wrong, but so wrong you are and I suggest you do some introspection to find out how it is you got there.

For a reactionary fanbois say what you said is one thing, but you write articles for a tech site ... that is something altogether different. The two should never cross paths like they did here.

Ryan Shrout over on PC Perspectives addressed this topic in an article and podcast and shows the way for an AGNOSTIC tech site to do it right.

Ok by: Sean Kalinich on 10/9/2009

So,as with my other article, I guess that is why Dell just did the same thing. They are worried that other companies are selling better so they "flashed" the Adamo XPS without showing it off or even opening it.

I see no mass defections to the 5800 that you speak of, Many OEMs are still reporting shortages of the GPU. Remember that at Quakeconn AMD did not show the press the card or the rest of the parts in the system being used for the demo. They did the same thing with Eyefinity.

Again I am not saying anything bad about AMD, these are examples of how this is nothing new and only being used to spread bad PR and FUD about another company while doing exactly the same thing themselves.

It is wrong to apply one set of standards to one company and not to all. Yet I did not see a large number of articles about the "fake" Adamo used at Dell's Launch of it. Nor did I see any about the locked cases for the AMD DX11 and Eyefinity demos.

The same can be said about Raytracing demos given by Intel...

It is only a small number of sites with a history of anti-NVIDIA articles that are running this as a major story. My question above and still is "why?"

Why is it the one thing they are covering?
Why does anyone feel that showing off this dead and faked card means anything?

You could conversely say that showing a disabled card means that the BOM used must not be working out and needs to be rebuilt.

That would not be true, but it could be speculated also.

by: Anonymous on 10/9/2009

It is obvious to anyone not an utter imbecile it was a rather desperate attempt to slow down the defections to the 5800 cards by any means possible, up to and including a fake pcb and outright lying = they don't even HAVE an actual sample of a Fermi pcb board to show off. And their self described frankenstein looking working prototype as the reason no photographs were allowed, they are not IN AMDs positiion, handing out actual populated final spec pcb boards and chips all over the place. They needed to provide some PROOF they aren't up to their eyeballs in sheepdip if they wanted to stop the defectons to AMD. That they provided NO certified proof, ALL the hardware except for one single hand numbered chip was either fake or beind the green curtain, there remain only smoke and mirrors to try to delay the defections.

How you can possibly think this is of no consequence boggles the mind. It shows they aren't even remotely close to being where AMD was in their similar show four months before they released the 5800 cards.

Even IF a respin isn't necessary it's going to be long months before Fermi see the light of day in numbers big enough to matter.

And when that does finally happen, AMD with months to optimize their yields on an inherently cheaper die and pcb will be able to undercut Fermi to a point that after satisfying their few thousand hard core fanbois, Nvidia will have to cut their prices to razor thin margins to compete at all.

Final note: it was Nvidia that INITIATED the FUD here with their lies and fake card. All the sites you're complaining about did was report on the fact based ACTUALITY of the situation and make logical deductions based on the data so revealed.

That that makes them FUD spreaders in your mind suggests a need for you to clear you head and reassess your operating basis.

Sean sounds like a fan boy!!!!! by: Anonymous on 10/8/2009

Dude you sound just like a "fanboy". Fact remains that your beloved Nvidia time is running out. Again they have no x86 license and Intel is going "Larabee, Laughabee" whatever you want to mock it and AMD has "Fusion". Nvidia has......ION LOL what a joke!!! I dont see them lasting another year, maybe 2 if they are lucky. Someone is going to buy them out I say Intel. So sorry to sound so prude against you bro but I am not buying your story, the facts speak for themselves you cannot rebut the obvious.

opinion by: Anonymous on 10/8/2009

Okay Sean,

I think I understand you now. I think you have also clairified, at least to me, you position on this article and why/what was important about it.

I think something along the lines of the last response should have been included in the article as it would have helped clear things up a little.

thanks for your response.

112 by: Anonymous on 10/8/2009

ATI HD 5870X2 лучшее!

by: Anonymous on 10/8/2009

ATI лучше!

FUD by: Sean Kalinich on 10/7/2009

Looking over the comments here it does not appear that people are concentrating on the information given out at GTC.

It is also part of my job to look at things like this and find the reasons behind them.

After receiving the e-mails and then seeing the articles pop up a day or so later I began to wonder about the reason for the "news" and why it was done.

I checked my information about dead/deactivated products at press events (even AMD later confirmed they do this), checked the time lines with my sources, and wrote the article. We have other articles that were written by our staff that cover the the parts of the conference we were able to get into.

Even if I do not sway the fans one way or the other, a few people might take a closer look at the issue and reconsider their stance on this. If that happens with even a handful of people then the article was a success and worth all of the insults that have been hurled at me (although I have to say I got a kick out of the cheerleader one)

Sean

by: Anonymous on 10/7/2009

So why are you perpetuating the story you say should not even be front page news? Why even write this article?

People are fickle and if "news agencies" let it go, we'd forget about in a matter of weeks. Old news. Who cares?

The smart people to whom this information is important to do not need these types of articles that are spreading FUD, as you put it. They have already formed their own opinions on the information presented at GTC and articles like this are not likely going to sway them one way or the other. Your article, though you are attempting to debunk the FUD, is in fact perpetuating it.

@Opinion by: Sean Kalinich on 10/7/2009

My information does not come from NV. My only good contact there is with the Tegra team at the moment. Yes I can e-mail them and get basic answers, but not much more. I have not had a good relationship with NV PR (GPUs and Chipsets) for many years. (you can even ask CD about that)

My information comes from sources with different OEMs and other companies that NV has to work with to produce their products.

This is where the launch dates I talk about come from. Also where I got the information that all companies do this. Speaking with them they were also at a loss as to why this was a big deal.

NV did not show the card at the demo, yes that is true. Remember AMD did not when they demoed Evergreen DX11 performance at Quakconn either. The system had a frosted window and the press were not allowed to see inside. This is not an attack on AMD, just showing that this is common at this stage of the game.

The reason behind the article is that out of all the things that could have (and should have) been covered at the event, the fake card becomes the big one. It makes no sense I could understand attacking the demo, the way it was done, the lack of DX11 performance being showcased. all of that, but a fake card?

Why that one topic, it is out of place. As you said my job is to find things like that and point them out. That stood out like a sore thumb. There were so many other things that could have been covered, but all we see is an uproar over a fake card...

See all comments

Leave a comment:

Author:

Title:

Comment:

Enter the code shown above:

(Note: If you cannot read the numbers in the above
image, reload the page to generate a new one.)