Hi Sylvain,

> 
> But is it fast enough ?
> 
> I had a shot at writing a bitslincing impl for NVidia using opencl for
> my laptop card. (9600M GT). But it turned out that a bitslicing 128
> bits impl using SSE x86_64 would be just as fast ...
> 

Not sure. I tried different setups.
I used my defect HD 4850 which is causing to crash the computer on high
loads and used that for testing A5Ati on win32/x64. Most of the results
with that 4850 were found faster than the A5Cpu.
But there were also searches that took longer using A5Ati than A5Cpu.
So I am not sure if comparing just a few numbers cracked with Kraken is
generally a good idea.

But i have run a few tests and these are the results:

 Used bursts:
--------------
001010101000111010100001001000001111110010010001101110000010010111000101001011010110111011011011000001101111011110
010011101100000111111101011111011100101111101100101101101010001100111100111011010101111111100100101100010100101000
101100000101010100010010110011100101101111000111001010101100000001101001100010011110000111101101000101111000000111
010110010011111100100111101001110101000101001100001011001110011101001011110001110010111101011100111111100011011001

For all bursts, the first table hit was the correct one.

 Results:
----------
[1] HD4850:  28 sec, HD3870: 101 sec, CPU: 184 sec
[2] HD4850: 150 sec, HD3870: 153 sec, CPU: 174 sec
[3] HD4850:  37 sec, HD3870:  60 sec, CPU: 131 sec
[4] HD4850: 135 sec, HD3870: 117 sec, CPU:  82 sec

My PC:
 - Intel Core 2 Duo E8500 3.16GHz
 - Vista x64
 - 8GB RAM
 - 2TB Western Digital WD20EARS for the tables


>From these numbers I would say, the A5Ati is performing better than
A5Cpu even on the old card.


BR,
Georg
_______________________________________________
A51 mailing list
[email protected]
http://lists.lists.reflextor.com/cgi-bin/mailman/listinfo/a51

Reply via email to