kepler predictions
Does anyone have predictions for the green teams Keplars namely the 680 and hashcat?

I assume we will see them rivalling the red teams cards except for bit allign.

You can calculate this very simple. NOTE: This is -very- inaccurate

The gtx570 (overclocked to 1600Mhz) makes 2170.5 Mhash/s on MD5. It has 480 SP.

480 x 1600 = 768000
2170500000 / 768000 = 2826

The gtx680 has 1536 SP and 1006Mhz = 1545216

1545216 * 2826 = 4366.7 Mhash/s

(The hd7970 makes 8213.6 Mhash/s on MD5)
atom Wrote:NOTE: This is -very- inaccurate

LOL i like that atom, i guess its an improvement for nvidia.

Time will tell. Looking forward to some benches on a keplar card with hashcat soon.
If some asian sources prove to be correct, GTX 680's shader domain will be clocked at 1.411Ghz(stock), which will boost MD5 speed to 6.124B p/s.
Hopefully they are correct!

[Image: NknoC.png]
ForceWare 300.99... Smile
Its still unclear if gtx680 has BFI_INT and / or BIT_ALIGN instruction. This would boost speed by ~ 40%.
Not sure how reliable src(here)


Wow! The claim of beating the HD7970 goes right into the thin air, it seems. Nvidia's new GPU is beaten by the Radeon HD 7970 by an order of magnitude here in double precision floating-point, as well as nearly twice in ordinary single precision floating point. One is speechless here. Even the Radeon HD 7870 with its restricted double precision floating-point still outperforms the GTX680 by a noticeable margin in this department, as you can see here. Only the Radeon HD 7850 is substantially slower.

Looks like RED: 1 Green: 0
Ok, disappointment number 1: shader clock is no more, shaders run at core clock.
Disappointment number 2 (mot confirmed): like FP64 and FP32, integer performance will be 1/2 or 1/4 int ops per shader.
That will mean, if the worst case scenario happens, GTX 580 will be faster than GTX 680.
However, since no reviewers use oclhc-plus or other hashcracking apps, someone has to do it manually.
if someone gets a gtx680 pls PM me