Thursday, December 25th 2008
NVIDIA GT216 Specifications Trickle-In
The AMD RV740 has been gaining quite some attention lately. Following news of it being the first GPU to be built on the 40nm manufacturing process, there is some information trickling in about NVIDIA's offering to combat it. The GT216, unlike what most speculations talked about being a next generation high-end GPU, is supposed to be a mainstream GPU whose specifications make it a rival to the AMD RV740. This, according to specifications of the said GPU emerging from Beyond3D, part of which they are sure of, part speculation.
For starters, the GT216 will be built on the 40nm manufacturing process. It will feature a 192-bit wide GDDR3 memory bus. Graphics cards built on this would feature 768 MB of memory (with possible OEM variants having just 384 MB). The memory is clocked at 1200 MHz (2400 MHz DDR). There is speculation linked to the rest of the specs at this point in time, notably that it might have 7 ALU clusters (168 SPs). The manufacturing process will make sure the GPU is cost-effective. With this, the next round of mainstream GPU competition sounds even more exciting.
Sources:
Beyond3D Forums, Expreview
For starters, the GT216 will be built on the 40nm manufacturing process. It will feature a 192-bit wide GDDR3 memory bus. Graphics cards built on this would feature 768 MB of memory (with possible OEM variants having just 384 MB). The memory is clocked at 1200 MHz (2400 MHz DDR). There is speculation linked to the rest of the specs at this point in time, notably that it might have 7 ALU clusters (168 SPs). The manufacturing process will make sure the GPU is cost-effective. With this, the next round of mainstream GPU competition sounds even more exciting.
15 Comments on NVIDIA GT216 Specifications Trickle-In
Then I think would be at least almost enough.
But if it has more shaders then the G80-GTX, probably it will request more bandwidth. (Why they would do a more powerful shader gpu that is weak because have less texture power?)
What can I do if I have a poor english? : P
The 8600GT(G84) had more shader power than the 7900GTX(G71) but only a 128-bit memory bus, about half the memory bandwidth of the 7900GTX.
The 7600GT(G73) had more shader power than the 6800Ultra(NV45) but again only a 128-bit memory bus, about half the memory bandwidth of the 6800Ultra.
The HD4670 has the same shader power as the HD3870, but again only a 128-bit memory bus, about half the memory bandwidth of the HD3870.
The HD2600XT has a lot more shader power than the X1950XTX, but again only a 128-bit memory bus, about half the memory bandwidth.
Of all the letters and numbers available to them in the Western world, WHY OH WHY make it so confusing for everyone?
On topic, I don't know how accurate and true are these numbers. It could use GDDR5 probably (earlier rumors said Nvidia would use 40nm and GDDR5 at the same time) or even have a 384 bit memory bus. The amount of memory would be the same and could have confused whoever listened the rumors first. 168 SP and only 12 ROPs just doesn't make sense to me either.
Came across this post on ABT
alienbabeltech.com/main/?p=2240
Looks like Nvidia will also be using GDDR5 on upcoming chips.
- More shader power than 8800GTX with the same architeture, but more SPs
- Less die size (because 40nm)
- But less bandwidht
They are using more die size to put more useless shaders?!
Even with GDDR5, it would have the SAME bandwidht as the 8800GTX, but with much more Shader Power (+SPs, probably +Clock).