NVIDIA GF100 & Friends speculation

The launch of the GeForce GTX 400 series is just weeks away and we finally got our samples. We can't share any more specific details on the card or architecture, but we will try to make up for it at launch day. During the day we received some information from a big Swedish distributor that could confirm the pricing of the GeForce GTX 400 family.

The MSRP will be around 5100 SEK (divide by ten to get USB/EUR) for the top model GeForce GTX 480 and around 3600 ($360) for GeForce GTX 470. The prices on launch day may vary a bit depending the store and even if the prices are averaged it shows how NVIDIA intends to price the cards.
http://www.nordichardware.com/en/co...hics/10925-nvidia-geforce-gtx-400-prices.html
 
XFX GTX 480 & GTX 470

xfx_02.jpg


xfx_03.jpg


xfx_04.jpg


xfx_05.jpg



http://www.expreview.com/9888.html
 

Surely they are wrong about core clock frequency, correct? I mean, sure, it is theoretically possible that core clock frequency = half hot clock frequency, but it's odd that NVIDIA would do this given that they intentionally designed the chip to have certain functional units operating at half hot clock freq (and, by intention, higher than core clock freq).
 
Surely they are wrong about core clock frequency, correct? I mean, sure, it is theoretically possible that core clock frequency = half hot clock frequency, but it's odd that NVIDIA would do this given that they intentionally designed the chip to have certain functional units operating at half hot clock freq (and, by intention, higher than core clock freq).
Those core clock frequency are correct.GF100's core clock is just GPC clock,not ROP/L2 clock.
 
So, given the rumoured RAM clocks, it would put Fermi just slightly above Cypress in bandwidth department (177 vs 153 GB/s).
 
What I believe Chal is saying is that it's not the number of decimals per se that is the issue with SPFP, but rather when you do operations on SPFP numbers you end up with a much lower precision end result, which can be problematic.

You seem determined to simply not want to understand this issue. That's not constructive.

Its hard to believe that such a minor precision difference would have any impact in actual game. Something like a difference of 1/1000000 will make zero impact in game.
The biggest problem right now is the amount of calculation and not preccision. Objects jerk around in games because they have such a limited physics model in game. Even if u would end always with 1/1000 error (which would be quite shame in year 2010) It would be imposible to see the diference.
I think people here are too much focused on the actual numbers and precision. The difference betwen force of 1001 and 1000 would be imposible to see.
 
Its hard to believe that such a minor precision difference would have any impact in actual game. Something like a difference of 1/1000000 will make zero impact in game.
The biggest problem right now is the amount of calculation and not preccision. Objects jerk around in games because they have such a limited physics model in game. Even if u would end always with 1/1000 error (which would be quite shame in year 2010) It would be imposible to see the diference.
I think people here are too much focused on the actual numbers and precision. The difference betwen force of 1001 and 1000 would be imposible to see.
You might think so, but the differences tend to add up rather rapidly. Consider, for instance, if you have 60 physics updates per second, it only takes about 11.5 seconds for a 1/1000 error to be as large as the number in question.

For example, if your system adds energy due to numerical errors at a rate of 1/1000 every physics update, at 60 updates/sec, the energy gets doubled after only 11.5 seconds just due to errors.
 
So at this point, we still don't know the amount of Stream Processors enabled on the GTX 480...:LOL:

If the TDP difference between it and the GTX 470 remains, either the GTX 480 has much higher frequencies or it really does have the full 512 Stream Processors.
 
I love it when some decide which Fudzilla's info is good or not...

Usually it's "Fudzilla sucks", but when some purposes are suited for that particular info, then it's probably correct.
 
I love it when some decide which Fudzilla's info is good or not...

Usually it's "Fudzilla sucks", but when some purposes are suited for that particular info, then it's probably correct.

In case of Fudo, who's almost part of nV PR, in general the more negative it's towards nV it is, the more likely it's to be true
 
Back
Top