• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

nV: Fermi gaming info forthcoming

"Look, I understand that some people might be feeling anxious because we haven't published detailed information about Fermi-based Geforce cards," said Del Rizzo. "But, I can assure you that data is forthcoming. The wait will be well worth it, especially when people understand what products based on Fermi are capable of."
Click to expand...

Without sarcasm at all, I'm actually waiting with bated breath for Fermi, or "GF100" as the article calls the desktop product (I have to admit it is the first time I've read about it being called "GF100"... before that, all I ever read was "Fermi" and "GT300"). I do hope the wait is worth it.
 
As I was brushing up on Fermi information, I also passed by this: Charlie about Fermi and NV Gaming

I wish that Wreckage had a blog/news site himself, so each night I can read two stories with "facts and logic" so that news consumption would be both exciting and end up balanced, one anti-NV and one pro-NV. Reading Charlie always comes with a caveat that he is rabidly anti-nVidia, whether he is correct or not, so reading an opposing piece would be nice and one could better determine, despite all the bias in both, where the "facts and logic" are more at.
 
nitromullet said:
Big surprise that we'd hear from NVIDIA about Fermi the week ATI "launches" the 5970.
Click to expand...

Yea, not too big of a suprise. What is a bit disappointing is that all we just heard was that we'll hear something later, aka nothing. I think all of us are pretty curious to see if the performance is worth the delay vs. the 58xx launch or if Fermi is the next Radeon 2900XT.
 
I'm thinking if there are no numbers to show-off yet (from nVidia), then Fermi is really a long way off. If their marketing department can't even brag about mega-giga-FPS-per-square-centimeter, or whatever useless marketing jumbo they'd have, then Fermi is not going to be here anytime soon.

I'm hoping Feb 2010, but I've pretty much resigned to expecting it realistically by March or April 2010.
 
I'm pleasantly surprised that 6-7 posts into this thread, no one has horribly spun these statements around to make it sound like nvidia is evil and wrong and has lost the gpu war.

The fact remains that no one outside of nvidia has talked about, or even knows, what the gaming performance will be, and therefore anything that anyone can say right now is speculation and only speculation.

Nvidia has lost a few sales to those who had borderline or below average graphics cards by today's standards and wanted to upgrade now. But the overwhelming of gamers, such as myself, still have great graphics - like a gtx 260 or a hd 4870, and are patiently waiting until ample games needing more performance and all next gen cards have hit the market before making a purchase.

I'm confident that when fermi comes out there will be a very competitive GPU market for which consumers will have plenty of great choices from both AMD and Nvidia.
 
jvroig said:
I'm thinking if there are no numbers to show-off yet (from nVidia), then Fermi is really a long way off. If their marketing department can't even brag about mega-giga-FPS-per-square-centimeter, or whatever useless marketing jumbo they'd have, then Fermi is not going to be here anytime soon.

I'm hoping Feb 2010, but I've pretty much resigned to expecting it realistically by March or April 2010.
Click to expand...

You could be right, but keep in mind there weren't exact performance figures in games with the 58xx series until a few weeks before launch.
 
jvroig said:
As I was brushing up on Fermi information, I also passed by this: Charlie about Fermi and NV Gaming

I wish that Wreckage had a blog/news site himself, so each night I can read two stories with "facts and logic" so that news consumption would be both exciting and end up balanced, one anti-NV and one pro-NV. Reading Charlie always comes with a caveat that he is rabidly anti-nVidia, whether he is correct or not, so reading an opposing piece would be nice and one could better determine, despite all the bias in both, where the "facts and logic" are more at.
Click to expand...

Eh just because you would read both doesn't mean the truth would meet somewhere in the middle. It's like saying you're well informed because you watch Fox News and MSNBC when in reality you're just filled with left/right talking points (and should have been listening to BBC radio 🙂 )
 
scooterlibby said:
Eh just because you would read both doesn't mean the truth would meet somewhere in the middle. It's like saying you're well informed because you watch Fox News and MSNBC when in reality you're just filled with left/right talking points (and should have been listening to BBC radio 🙂 )
Click to expand...

You are a statistician of sorts (at least I know you could not have escaped statistics given your educational background), so I think you'll understand what I mean when I say that the value of being exposed to varied opinions (be they Charlie vs. Wreckage, or Fox vs MSNBC) is that we become all the more aware of possible scenarios and outcomes that we might not have factored into our larger ensemble of conceptual possibilities.

What we have to do though is re-assess the probability of the scenarios as presented by the varied opinion-givers. Naturally those presenting a given opinion feel the probability of their presented possibility is near unity, hence the passion with which they convey the proposed scenario.

So if you deconvolve probability from possibility (the product of the two is plausibility) then you can extract value (knowledge of a new possibility) from exposure to the "extreme" opinions.

Consider that what makes them "extreme" in your assessment is that you feel the probability of the possibility is grossly overstated by the party presenting you with the info regarding the possibility.

(This is also the basis of what we mean to imply when we invoke the "take it with a grain of salt" idiom...we are saying don't rule out the possibility but you need to assign the probability a really low ranking value.)

Assign it your own probability, add the possibility into your ensemble of potential outcomes, and proceed to crank out your maximum likelihood estimation. The result may be unchanged but your confidence limits are improved as the sample size increased by n++.
 
So far, this launch is very similar to NV30. I remember Nvidia talking about NV30 like it was going to totally revolutionize gaming and how it was the dawn of cinematic gaming, and how the GPU was so much more powerful than anything else. Then when it finally came out, it didn't manage to beat ATI's GPU that had been out for over 6 months. Not saying the same thing is going to happen again, just that this reminds me of NV30.

I only see two possible reasons for why we don't have any performance data yet, and neither is good:

1. It's still too early for even Nvidia to know how it will perform. This means the card is still far from release.

2. Nvidia have seen the numbers, and they know it's not pretty.

If Nvidia knew the performance level, and knew it to be above or at least on par with that of the 5870, then I see no reason why they should not tell everyone. This would be similar to AMD's CPU division - they had a working Phenom at 3 GHz, but they wouldn't reveal any benchmark results because they knew it would barely beat the Q6600.
 
Nvidia is evil and wrong and has lost the gpu war.

Happy now? 😉

I'm pretty excited to see some new silicon. Just wish it was sooner. I'm not snowboarding this season so I need something to blow my money on.
 
tviceman said:
But the overwhelming of gamers, such as myself, still have great graphics - like a gtx 260 or a hd 4870, and are patiently waiting until ample games needing more performance and all next gen cards have hit the market before making a purchase.
Click to expand...
I'm running SLIed 512MB 8800GTs and other than Crysis (der) I've been able to play just about everything I want at 1920x1200 with everything full blast. A couple games require backing off the AA or possibly dropping to 1680, but in general unless Fermi cards offer 50% more performance than this for $300 or less, I'm inclined to keep on rolling with this setup until a game that looks like Crysis, but is actually fun, comes along.
 
JimmiG said:
So far, this launch is very similar to NV30. I remember Nvidia talking about NV30 like it was going to totally revolutionize gaming and how it was the dawn of cinematic gaming, and how the GPU was so much more powerful than anything else. Then when it finally came out, it didn't manage to beat ATI's GPU that had been out for over 6 months. Not saying the same thing is going to happen again, just that this reminds me of NV30.

I only see two possible reasons for why we don't have any performance data yet, and neither is good:

1. It's still too early for even Nvidia to know how it will perform. This means the card is still far from release.

2. Nvidia have seen the numbers, and they know it's not pretty.

If Nvidia knew the performance level, and knew it to be above or at least on par with that of the 5870, then I see no reason why they should not tell everyone. This would be similar to AMD's CPU division - they had a working Phenom at 3 GHz, but they wouldn't reveal any benchmark results because they knew it would barely beat the Q6600.
Click to expand...

Pretty much sums up what I think. Though I will wait for the cards to show up first and make my purchase decision then (I don't game for now).
 
SlowSpyder said:
Yea, not too big of a suprise. What is a bit disappointing is that all we just heard was that we'll hear something later, aka nothing. I think all of us are pretty curious to see if the performance is worth the delay vs. the 58xx launch or if Fermi is the next Radeon 2900XT.
Click to expand...

You're sorta into a delay regardless of what nV does, says, presents, etc.. since AMD won't provide, no?
 
With no "A list" DX11 titles and with the competition barely able to make cards, I think NVIDIA has plenty of time.
 
Idontcare said:
You are a statistician of sorts (at least I know you could not have escaped statistics given your educational background), so I think you'll understand what I mean when I say that the value of being exposed to varied opinions (be they Charlie vs. Wreckage, or Fox vs MSNBC) is that we become all the more aware of possible scenarios and outcomes that we might not have factored into our larger ensemble of conceptual possibilities.

What we have to do though is re-assess the probability of the scenarios as presented by the varied opinion-givers. Naturally those presenting a given opinion feel the probability of their presented possibility is near unity, hence the passion with which they convey the proposed scenario.

So if you deconvolve probability from possibility (the product of the two is plausibility) then you can extract value (knowledge of a new possibility) from exposure to the "extreme" opinions.

Consider that what makes them "extreme" in your assessment is that you feel the probability of the possibility is grossly overstated by the party presenting you with the info regarding the possibility.

(This is also the basis of what we mean to imply when we invoke the "take it with a grain of salt" idiom...we are saying don't rule out the possibility but you need to assign the probability a really low ranking value.)

Assign it your own probability, add the possibility into your ensemble of potential outcomes, and proceed to crank out your maximum likelihood estimation. The result may be unchanged but your confidence limits are improved as the sample size increased by n++.
Click to expand...

The track record of predicting the outcome, is really the important statistic. Charlie though twisted has broken some very big stories. Such as no Fermi this year.

Wreckage is also very entertaining, but not sure if his connections are up to snuff.
 
Wreckage said:
With no "A list" DX11 titles and with the competition barely able to make cards, I think NVIDIA has plenty of time.
Click to expand...

NVIDIA dude#1:"So what is the strategy for 2009?"

NVIDIA dude#2: "Lets make no cards because the opposition can barely do cards."

Nothing like answering to few with nothing. 🙂
 
Wreckage said:
With no "A list" DX11 titles and with the competition barely able to make cards, I think NVIDIA has plenty of time.
Click to expand...

If Fermi GeForce parts come out before a bunch of A list DX11 games than that will certainly be a good thing for Nvidia. But the 5870's horsepower is still useful for DX10.1/10/9 games too. I'm willing to bet most people who buy it are buying it for what it can do with today's games. DX11 is just a little icing on the cake.

But I agree with you (wow) about AMD's supply problems. It looks like they will have several months all to themselves to sell next gen hardware, with no competition from Nvidia. But they aren't meeting demand, and if they continue to struggle with getting cards out than this big lead really won't amount to much.

On the other hand we have to wonder if Fermi might be a flop from a gaming perspective. I'm not saying it will, but it's very possible that it won't 'kill' the 5870 in performance, and it has to compete with the multi-GPU 5970 already. I'm really curious to see what Nvidia brings...
 
Last edited:
yeah really pathetic than ati can't produce 10 times the number of cards

tsmc 40nm or not. they should start running them out at 40nm from GF.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top