-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
The first "proper" Kepler news Fri 17th Feb?
- Thread starter dave_beast
- Start date
Caporegime
Anyone know Japanese:
http://www.4gamer.net/games/120/G012093/20120306077/
I wonder if Nvidia is using something similar to HT with Kepler??
It might explain why driver optimisations are supposed to be important.
http://www.4gamer.net/games/120/G012093/20120306077/
I wonder if Nvidia is using something similar to HT with Kepler??
It might explain why driver optimisations are supposed to be important.
Caporegime
Here is the relevant part translated(badly) by Google Translate:
"Our users, from conventional products by improving performance "only" 1.5 times as much would not be satisfied. Details of Kepler is not clear, and performance per unit of power consumption is up significantly, it should also significantly improve the absolute performance compared with conventional products of the same price range. "
NVIDIA has been seen, from the end of 2011, has begun the publication of element technology adopted by the GPU in the future, some of which are to be adopted and in fact even Kepler.
When you place an example one, NVIDIA will have to introduce the approach in the technical conference of semiconductor-related was held at the end of 2011, in order to improve the efficiency of use of CUDA Core, to be able to handle more threads on each core was. It was spoken there, increase the number of threads (batch = a bundle of thread) "Warp" which is controlled by 32 thread in the current generation of products, by raising the utilization efficiency of the GPU combined with improvements in scheduling, such is the content, trying to balance the optimization and higher performance with low power consumption.
I think the details are want to tell once again the opportunity, as can be seen if Moraere Think of a Hyper-Threading "" of Intel "approach is NVIDIA, to be able to use threading to other free resources CUDA Core It is, and shall raise the processing performance per CUDA Core, be reduced so that the power consumption per performance "that (official NVIDIA).
In addition, officials close to NVIDIA, "in Kepler, and to improve the processing performance of the DirectCompute PhysX" but also said, If we take the approach NVIDIA officials say certainly, than the arithmetic processing graphics processing "light" is not assigned to the free resources of CUDA Core, a, would not be surprised as to improve the performance be measured.
In any case, future-based products than actually running at the venue of CeBIT 2012 thus, not so far, including the model for the desktop PC, details of Kepler is supposed to be clear. I would not have it be nice to see and finally went into the position for takeoff Kepler."
"Our users, from conventional products by improving performance "only" 1.5 times as much would not be satisfied. Details of Kepler is not clear, and performance per unit of power consumption is up significantly, it should also significantly improve the absolute performance compared with conventional products of the same price range. "
NVIDIA has been seen, from the end of 2011, has begun the publication of element technology adopted by the GPU in the future, some of which are to be adopted and in fact even Kepler.
When you place an example one, NVIDIA will have to introduce the approach in the technical conference of semiconductor-related was held at the end of 2011, in order to improve the efficiency of use of CUDA Core, to be able to handle more threads on each core was. It was spoken there, increase the number of threads (batch = a bundle of thread) "Warp" which is controlled by 32 thread in the current generation of products, by raising the utilization efficiency of the GPU combined with improvements in scheduling, such is the content, trying to balance the optimization and higher performance with low power consumption.
I think the details are want to tell once again the opportunity, as can be seen if Moraere Think of a Hyper-Threading "" of Intel "approach is NVIDIA, to be able to use threading to other free resources CUDA Core It is, and shall raise the processing performance per CUDA Core, be reduced so that the power consumption per performance "that (official NVIDIA).
In addition, officials close to NVIDIA, "in Kepler, and to improve the processing performance of the DirectCompute PhysX" but also said, If we take the approach NVIDIA officials say certainly, than the arithmetic processing graphics processing "light" is not assigned to the free resources of CUDA Core, a, would not be surprised as to improve the performance be measured.
In any case, future-based products than actually running at the venue of CeBIT 2012 thus, not so far, including the model for the desktop PC, details of Kepler is supposed to be clear. I would not have it be nice to see and finally went into the position for takeoff Kepler."
Caporegime
I think so far only the 7950 wasn't a paper launch?
Saw 7970 on offer for *No competitor price hinting* earlier today.
I genuinely feel so sorry for ATI, why have they not taken full advantage of their lead ? Everyone would have brought those cards if the price was right !
Depending what Nvidia do pricewise, ATI could be in trouble and they may be against peoples mindset "Bad value card" going cheap, because you only get one chance to create a that first impression...if you follow
I genuinely feel so sorry for ATI, why have they not taken full advantage of their lead ? Everyone would have brought those cards if the price was right !
Depending what Nvidia do pricewise, ATI could be in trouble and they may be against peoples mindset "Bad value card" going cheap, because you only get one chance to create a that first impression...if you follow
Caporegime
Good find Cat. Now all I need is Xsistor or Duff-Man to explain that to me in lamen terms 
The memory is probably rated much higher anyway, I can't seem to remember any exact figures though.
My 5850 had a stock memory of 1000 MHz and I remember them clocking to 1200-1300 MHz so perhaps that's why you didn't notice it at all.
The memory overclock always gained a little in benchmarks like Heaven or 3DMark, up until the memory correction kicked in when the scores were virtually lower than with lower frequencies.
Soldato
- Joined
- 28 May 2007
- Posts
- 10,165
Saw 7970 on offer for *No competitor price hinting* earlier today.
I genuinely feel so sorry for ATI, why have they not taken full advantage of their lead ? Everyone would have brought those cards if the price was right !
Depending what Nvidia do pricewise, ATI could be in trouble and they may be against peoples mindset "Bad value card" going cheap, because you only get one chance to create a that first impression...if you follow
Ahh because its amd its gonna hurt them. It never really hurt nvidia over the years. Infact its probably done the opposite because people think high price premium brand. You get people saying you get what you pay for with nvidia. Amd are in control atm as nvidia is late to the playing field. If kepler is good but overpriced i dont think we will see as much moaning.
Soldato
Good find Cat. Now all I need is Xsistor or Duff-Man to explain that to me in lamen terms
Heh - don't look this way, I can't make heads or tails of it!
The translation is really bad, but I THINK that they're saying Nvidia has focused on allowing a larger number of parallel threads to be executed at once. Also that they have reduced the performance per CUDA core as a consequence of improving performance per Watt.
I'm guessing really though... It's hard to read anything through that translation!
Associate
I'm kinda hoping a 4GB variant will come out I really don't get why they release a 3GB GTX580 but only a 2GB GTX680 (if the info is correct)...
I really don't get why they release a 3GB GTX580 but only a 2GB GTX680 (if the info is correct)...
Caporegime
Soldato
I have said it numerous times now, but when you price something up (anything, even houses) you price them up by comparing them to what is available and what it cost.
It's not rocket science, it's actually very simple. I mean crap, even I could have priced up the 7950 and 7970.
7950, right on par with a 1.5gb 580 only faster, has more vram, and overclocks like a nut case. Hmmm.. What price? I know ! simply make it the same price as the 580. That way any one looking for a 580 will be made to think twice. Quite cunning.. Those who have already bought a 580 won't be buying one, so you don't need to make the price so low as to tempt them as they wouldn't bother.
The 7970. Much faster than a 580. Not only that, but also has 3gb vram. Let's see.. Hmmm.. What card can we compare that to? Oh yeah ! I know ! the 3gb 580. Not only is it faster than the 3gb 580 but it also uses less power and overclocks like a nut case.
This is what makes me laugh at Nvidiots who come along whining about AMD's pricing.
They got it from Nvidia you clowns !
If the 580 was £300 the 7950 would have been £300. It doesn't need to be cheaper it's faster. So at identical pricing the AMD cards are good value for money.
Again, not going to make any one but the die hards rush out and buy one, but any one in the market for a high end card would be a fool to buy Nvidia.
Then they release the 7850 and 7870 and give prospective 560 ti and 570 buyers something to think about.
It's all amazingly clever. Now maybe if Nvidia had like, "got it" then they would have dropped their prices and forced AMD to drop theirs. But even they don't seem capable of working it out.
Soldato
I'm kinda hoping a 4GB variant will come out
Because Kepler will cache ram. Now back in the days of my 9800GT a fair lump of memory used to go missing. That didn't happen with my 470 as Nvidia probably felt they had enough to comfortable take care of any game without caching.
Face it. Nvidia are tight with ram. Their vram costs a lot more than AMD's I hear, so they are going to give you as little as they can get away with.
Ed. If you needed something to go by...
6970 1gb. about £25 less than the 2gb.
580 3GB about £90 more than the 1.5gb.
Associate
Ok but what does that exactly mean that it can cache ram, will it need less memory or...? I don't get it sorry
Soldato
Ok but what does that exactly mean that it can cache ram, will it need less memory or...? I don't get it sorry
When it runs out of vram it will use your physical memory.
Current cards use your pagefile.sys which sits on your hard drive. Hence the lag and stuttering when that happens.
Caching from physical ram is obviously faster, but it's still not ideal. Vram runs far faster than physical ram.
Nvidia's ram is far more expensive as they chose to use expensive stuff. I would imagine it's because that is what they use for their Tesla and server cards and it needs to be more reliable and stable.
@ALXAndy,
Yes, this works for houses, but by that logic Pentium 2 would cost more than Pentium, Pentium 3 more than Pentium 2, etc etc... i5 should cost somthing stupid - but it doesn't!
7970 is faster than 580, but its 2 years younger, its new generation, new tech. At very least it should cost the same imo.
Yes, this works for houses, but by that logic Pentium 2 would cost more than Pentium, Pentium 3 more than Pentium 2, etc etc... i5 should cost somthing stupid - but it doesn't!
7970 is faster than 580, but its 2 years younger, its new generation, new tech. At very least it should cost the same imo.
Soldato
In computer parts your comparison products are the ones already out. Every $ of the price tag comes from each FPS you gain in gaming.
P2 and P3 cost less than the original MMX because IMO by that time AMD had turned up and were knocking out cheap as chips (pardon the pun) K62 3DNOW ! chips. Up until that point AMD had been very quiet and understated. Once the Athlon came along Intel were forced to drop prices.
It costs about the same as a 3GB 580, last I checked anyway.