I use a simple 1x to 16x riser from Delok.
http://www.delock.com/produkte/suche/Ris...41776.html
I changed the wires to be longer. The "UDMA66" wires are the same. You can change them if you want with a wise.
To use 16x16x has no meaning because the pcie bus is runing idle all the time...
I can send a picture if you want.
you don't see performance reduction with pci-e x1 cable? i have 4xhd5970 and my motherboard can do x16 on all 4 slots
The performance reported by the ockHascat-lite (I do not use the rest of the programs) as speed, is the same for all the GPU units. So it seems that the bus speed variation is insignificant. And you can do the math, assuming: A connection between any two PCIe devices is known as a link, and is built up from a collection of 1 or more lanes. BUT:
Capacity Per lane (each direction):
v1.x: 250 MB/s (2.5 GT/s)
v2.x: 500 MB/s (5 GT/s)
v3.0: 1 GB/s (8 GT/s)
v4.0: 2 GB/s (16 GT/s)
Furthermore:
16 lane slot (each direction):
v1.x: 4 GB/s (40 GT/s)
v2.x: 8 GB/s (80 GT/s)
v3.0: 16 GB/s (128 GT/s)
As you can see the fastest 2xGPU on one card at the curent time can not produce the MINIMAL speed of the 1 lane in a normal v2 connector. Maybe in some years we will have such a speedy cards, but not yet. And even in some years CPUs, GPUs, will not be so powerful as the capacity of a 16x connection. Maybe in the era of quantum computing. Just joking.
Ask any more question if you want.
(12-09-2011, 10:24 PM)attila Wrote: [ -> ]I can send a picture if you want.
Yes please !!
I am amazed that a single pcie lane is enough to run a GPU at full speed. I wonder if there is a way to get a single pciex16 to split into more than one card as I have an old computer with only 1 pciex16 slot and it would be cool to be able to run 2 or more cards from it.
Thanks very much for your posts attila, very interesting !
Well all I can say is WOW! What is the purpose of that setup and who owns it? Can we read more on it anywhere? I now know what I want for xmas
Of course 1x can run a GPU at 100%. Think about what is actually being transfered to the GPU, data. 1x PCIe @v2.1 (most cards) allows for 500 megabytes per second. Even if you went over with say a 2gig wordlist the data is cached before pushed, so you would never notice a preload time.
(12-12-2011, 07:05 PM)HRob1n Wrote: [ -> ]Well all I can say is WOW! What is the purpose of that setup and who owns it? Can we read more on it anywhere? I now know what I want for xmas
Purpose of this rig is to unlock mobile phones on hashcat scene known as SL3 (or most 'favorite' algo of radix
). For xmas mhhmmm... maybe too much money involved. If not to count tests complexity and to see only single precision calculation some of GSM scene server farms would go over 1,5 petaflop.
sl3 is not a scene in hashcat.