NV30 : not so fast?

[Galilee]

Well I am not familiar with the term tape-out, but they have som hardware samples prior to the final version.

 

[Mulciber]

i don't think there's such thing as :

"Before they tape-out the final version
they have made a few production samples."

you're talking about revisions not tape-out

and you're argueing semantics which can change from person to person and place to place

I think the real deffinition of the word "tape-out" is a little lost because of its origin

 

[pikkachu]

Mulciber wrote:

"I think the real deffinition of the word
"tape-out" is a little lost because of its origin"

anyway,the meaning maybe is lost on you?
but don't forget nvidia's CEO was an engineer
before he was the CEO and not just some suit
and he know exactly what it means...

 

[Qroach]

Just saw this posted at Nvnews

http://www.nvnews.net/

I appreciate the comments as they were provided as an attempt to clarifiy certain points. I guess there are times when one reads enough nonsense and can set the "record straight" without breaking a non-disclosure agreement. Of course I could be way off here, but my gut-feeling is that this person is telling the truth.


The graphics processor throughput of 200 million triangles per second is not correct. The sources may have mistaken the figure.

The 3DMark performance was exaggerated. As Scott mentions below the results are more than likely based on using extreme graphics settings. However, there's a good chance that the NV30 will significantly outperform the GeForce4 Ti 4600 in the next version of 3DMark.

The Doom 3 performance estimate was based on using the latest NV30 beta board and was not done via simulation.

NV30 beta chips and boards have been around for some time and the final chip should come back any day.

The 48GB/sec of memory bandwith is based on a calculation which includes factoring in overdraw removal (occlusion culling).

Now, I find that comments in bold interesting. It really makes me wonder if Carmack would bother sending comments (that Ben recently posted) to Nvidia regarding the NV30 and how he plans to base his next 3D engine on it, if he's never seen the card? That doesn't sound like somehitng he'd do IMO.

 

[Galilee]

I am guessing NV30 in october It's a bold guess, but the "hollidays" can theoretically start around that time.

 

[Mephisto]

That's the difference.

PR people, fanboys, so called "sources close to Nvidia" are free to lie to MikeC, Ben6, ReactorCritcal and 3DGPU, as long it is "off the record".

But Nvidia's CEO, Jen-Hsun Huang, would get into serious trouble if he is lying to investors. That's why all this new "damage control rumours" of fansites are not true.

 

[MfA]

Unless they use deferred shading (pretty much infeasible without using a tiler) I dont see bandwith savings due to occlusion culling making a huge impact anymore. For anything more sophisticated than HyperZ diminishing returns would hit hard IMO.

 

[RussSchultz]

Why don't we all just shut up about all this senseless theorizing and particularly the namecalling.

The board will hit the market at a time that is not determined by when anybody here thinks, predicts, knows, etc. It will show up when its ready, and not because enough people believe it will show up then.

Just as reciting matras and prayers to the sun god doesn't make it rise and set each day, people wishing or not wishing this damn chip to come out won't change anything either.

In otherwords, winning this argument, like most, is completely pointless.

But, to do my part, I think it will come out sooner, than later.
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"

Has it come out yet? NO? Damn, must keep repeating....

"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"
"sooner, than later"

 

[Mulciber]

Mulciber wrote:

"I think the real deffinition of the word
"tape-out" is a little lost because of its origin"

anyway,the meaning maybe is lost on you?
but don't forget nvidia's CEO was an engineer
before he was the CEO and not just some suit
and he know exactly what it means...

You're just side-stepping the issue

 

[Guest]

I can assure you that I did not break my NDA by giving that minimal piece of info. It is quite a pain if you break an NDA, even accidentally. As to the reference board, it may very well be .You`ll have to wait and see. And to people that think that it is absolutely impossible to have hard info on the NV30, think again. All official partners of nVidia have been briefed and are up to date to the status of this chip. Do you think that a marketing campaign can be concocted in the short time from the FINAL tapeout to availability on the market?

No you couldn't have done that I mean we already know that the nv30 has not been taped out. Can you tell us how many metal revs have been done?

 

[Gerry]

Unless they use deferred shading (pretty much infeasible without using a tiler) I dont see bandwith savings due to occlusion culling making a huge impact anymore. For anything more sophisticated than HyperZ diminishing returns would hit hard IMO.

I have to admit that whilst others dismiss nv30 being a tiler out of hand, I wouldn't be THAT surprised. Some comments/rumours to hint at the possibility - others don't. I guess we'll just have to wait and see.

 

[MikeC]

PR people, fanboys, so called "sources close to Nvidia" are free to lie to MikeC, Ben6, ReactorCritcal and 3DGPU, as long it is "off the record".

What's ironic is that the information I just posted at nV News was paraphrased by me based on a reply in this very thread. It's also interesting that nobody has questioned the information, other than Geek_2002 who did just a few moments ago.

 

[LeStoffer]

Unless they use deferred shading (pretty much infeasible without using a tiler) I dont see bandwith savings due to occlusion culling making a huge impact anymore. For anything more sophisticated than HyperZ diminishing returns would hit hard IMO.

Exactly my point/speculation. You cannot really give out an effective bandwidth number (hype or not) unless you know exactly how efficient the architecture is at any given time (and you cannot do that with HyperZ or LMA).

 

[Guest]

PR people, fanboys, so called "sources close to Nvidia" are free to lie to MikeC, Ben6, ReactorCritcal and 3DGPU, as long it is "off the record".

What's ironic is that the information I just posted at nV News was paraphrased by me based on a reply in this very thread. It's also interesting that nobody has questioned the information, other than Geek_2002 who did just a few moments ago.

Err,, where is that posted? I don't see it on the news page.. I have been questioning this heavily for some time now. I am unsure of just why it is that a blind eye is turned to claims made by certain people that turn out to be false. Now if this is comming from nvidia then ... no surprise. Otherwise I may be opening a can of worms that I didn't intend on.

 

[MikeC]

Otherwise I may be opening a can of worms that I didn't intend on.

As for opening a can of worms, I expect that NVIDIA may contact me and even ask for the source of the information. They may even ask me to remove the post. But I had a hunch that the information that was provided seemed to be accurate. Otherwise I probably wouldn't have posted it. If it's not accurate, then I'll look like the fool.

 

[Guest]

Otherwise I may be opening a can of worms that I didn't intend on.

As for opening a can of worms, I expect that NVIDIA may contact me and even ask for the source of the information. They may even ask me to remove the post. But I had a hunch that the information that was provided seemed to be accurate. Otherwise I probably wouldn't have posted it. If it's not accurate, then I'll look like the fool.

I would like to have a link to your posting if you could provide it.(I can't seem to find it.) Thanks.

EDIT: I can see it now... http://www.d-silence.com/news.shtml#newsitem1028606395,65,

 

[Guest]

PR people, fanboys, so called "sources close to Nvidia" are free to lie to MikeC, Ben6, ReactorCritcal and 3DGPU, as long it is "off the record".

What's ironic is that the information I just posted at nV News was paraphrased by me based on a reply in this very thread. It's also interesting that nobody has questioned the information, other than Geek_2002 who did just a few moments ago.

It would seem that nvidia is not to upset about what you have written Mike. At any rate at this point I am upset about some supposed "facts" that were posted here some time ago. In fact some had claimed that the NV30 had been "taped out". Now I took these persons word that they had in fact heard that the NV30 was taped out and that it was "public information." So as I understood it nvidia had said that the NV30 was taped out. But I just didn't believe that this was the case as often companies put up "window dressing" to maintain investor confidence. What makes me upset is that I have used this as proof positive that nvidia did in fact have the nv30 taped out, I was sceptical of it, but unless nvidia claimed otherwise ....

At any rate the last CC that nvidia held the CEO made it clear that the nv30 was not taped out. My question is why is that particular matter overlooked? Then the debacle with pics that were clearly not done on the nv30.(but that is another matter.)

Because I am sceptical of nvidia and what they claim at this point it makes me a f@nboy in some horrible way that I don't fully understand. This may explain why others do not question the prospects of the nv30 even though it is still not taped out. On Beyond 3D its ok to be bias as long as you are a "software guy" or "demo-coder" and the like.

I am upset that I took the words of certain individuals to be the truth. Now it seems clear that the information is incorrect why are they not upset with nvidia for giving false information? BTW I really don't expect a straight answer as I have really gotten none yet. Also I am not trying to upset anyone here I would however like to get to the bottom of this..

http://www.beyond3d.com/forum/viewtopic.php?t=1587&postdays=0&postorder=asc&start=0

NV30 has taped out. That's public information.

Kyle, its been spoken about in some of the recent investors talks NVIDIA have been doing.

Hey Kyle , was nice meeting you at San Francisco, though you didn't know who I was in the Cirque room. Was a blast meeting some of the people who I talk to in forums once in a while. Did you make your filight? when me and doc overclock were leaving the Fairmount, you hadn't come down for the waiting car... Nvidia has had a investor's meeting last week or this week where it was mentioned NV30 was taped out.

Ben

 

[Qroach]

Geek,

Would you please give it up? Why do you feel we need to see you post the same quotes over and over again? I don't think you're a fanboi just becuase you are "skeptical" nvidia. I think your a fanboi because you're only skeptical about nvidia. If you looked at ATI, Matrox, and others through the same googles then perhaps people wouldn't think you were such a "fan" along with two or three others.

I agree, what's been said doesn't make any sense. We're getting conflicting informaiton all over, and you'll find out the truth soon enough, as will everyone else. Why can't you wait?

 

[Guest]

Geek,

Would you please give it up? Why do you feel we need to see you post the same quotes over and over again? I don't think you're a fanboi just becuase you are "skeptical" nvidia. I think your a fanboi because you're only skeptical about nvidia. If you looked at ATI, Matrox, and others through the same googles then perhaps people wouldn't think you were such a "fan" along with two or three others.

I agree, what's been said doesn't make any sense. We're getting conflicting informaiton all over, and you'll find out the truth soon enough, as will everyone else. Why can't you wait?

I am sorry I disagree. Nvidia is known for its "aggressive" PR. This has little or nothing to do with me being a f@nboy. BTW I have only posted these particular remarks one other time with NO explanation. Thanks for agreeing though its the only response that I have had with regards to the subject at all. Not all things are equal and on the same coin not all companies are the same either.

What if I were to post elsewhere that the NV30 had been taped out based on the comments quoted above, and then come the last CC the CEO of nvidia clearly stated that it had not.. Wouldn't you be a tad embarrassed? or even just a little upset? for being made a fool of... If the information comes from nvidia then fine no surprise but there is no way to prove it as their is no record of it that I can find. Which logically should bring you back to the source of the information....

 

[Ty]

I am sorry I disagree. Nvidia is known for its "aggressive" PR. This has little or nothing to do with me being a f@nboy. BTW I have only posted these particular remarks one other time with NO explanation. Thanks for agreeing though its the only response that I have had with regards to the subject at all. Not all things are equal and on the same coin not all companies are the same either.

What if I were to post elsewhere that the NV30 had been taped out based on the comments quoted above, and then come the last CC the CEO of nvidia clearly stated that it had not.. Wouldn't you be a tad embarrassed? or even just a little upset? for being made a fool of... If the information comes from nvidia then fine no surprise but there is no way to prove it as their is no record of it that I can find. Which logically should bring you back to the source of the information....

I certainly agree that nVidia is known for its aggressive PR. That was certainly the case a couple of years ago when it was headlined across a few tech sites. Whether or not that has changed I can't say for sure but I would bet that much of the same attitude remains. With that said however, understand that people make mistakes or unintentionally espouse information that is incorrect or misunderstood. I believe HardOCP had stated that the NV30 came back at .15um and that even Doomtrooper said the same so you certainly can't claim that only nVidiots are claiming it taped out already.

As for me I'm still unclear as to what state NV30 development is in but frankly I actually don't care much since it only matters to me when I could possibly get one in my grubby little hands.