Search Results
|
Post |
Author |
Forum
[asc]
|
Replies |
Views |
Posted |
|
|
Thread: GPU memory considerations
Post: RE: GPU memory considerations
It doesnt matter because the OpenCL runtime limits a single GPU memory block to 512mb. In other words you will get out-of-memory Problems regardless of GPU memory size. |
|
atom |
Hardware
|
6 |
17,561 |
10-26-2012, 10:03 AM |
|
|
Thread: GPU memory considerations
Post: RE: GPU memory considerations
Well I should say this is for AMD OpenCL runtime only. But the reasons behind? Only AMD can know :) |
|
atom |
Hardware
|
6 |
17,561 |
10-26-2012, 12:03 PM |
|
|
Thread: GPU memory considerations
Post: RE: GPU memory considerations
Yes, but it wont help since we need to copy the hashlists to each device on its own to have fast access. |
|
atom |
Hardware
|
6 |
17,561 |
10-27-2012, 08:28 AM |
|
|
Thread: Help to start OclHashcat-lite on ati4870x2
Post: RE: Help to start OclHashcat-lite on ati4870x2
oclHashcat works with OpenCL. OpenCL runtime from AMD do not allow access to required fast memory for hd4xxx cards.
For SL3 on hd4xxx cards use ighashgpu. |
|
atom |
Hardware
|
13 |
20,489 |
11-26-2012, 11:11 AM |
|
|
Thread: Why NVidia gpus are slower in computing
Post: RE: Why NVidia gpus are slower in computing
The main reasons for AMD cards are faster than NV cards in general are:
- AMD GPU has more raw integer power (more SP), floating point is not important in crypto.
- AMD GPU/XOP CPU have a much better... |
|
atom |
Hardware
|
4 |
13,928 |
12-09-2012, 11:29 AM |
|
|
Thread: AMD 5570 & HD4650 can use them together was not?
Post: RE: AMD 5570 & HD4650 can use them together was no...
Mixed GPU support added to latest version, thread closed. |
|
atom |
Hardware
|
4 |
17,901 |
01-02-2013, 12:30 PM |
|
|
Thread: MultiGPU Verify Usage on Windows
Post: RE: MultiGPU Verify Usage on Windows
Mixed GPU support added to latest version, thread closed. |
|
atom |
Hardware
|
3 |
12,383 |
01-02-2013, 12:32 PM |
|
|
Thread: Does HC treat Infiniband-connected servers as a single machine?
Post: RE: Does HC treat Infiniband-connected servers as ...
afaik you can do tcpip-over-infiniband |
|
atom |
Hardware
|
3 |
8,391 |
03-18-2013, 05:50 PM |
|
|
Thread: 5970 slow?
Post: RE: 5970 slow?
I dont get what the problem is here. The numbers shown are the ones to expect. |
|
atom |
Hardware
|
12 |
22,929 |
04-19-2013, 10:49 AM |
|
|
Thread: 5970 slow?
Post: RE: 5970 slow?
uzomoshi Wrote: (04-19-2013, 12:43 PM)
--
Just curious, why is salt + pass significantly slower than pass + salt ?
Could it be done faster for single salt situations?
--
No, that is because of the... |
|
atom |
Hardware
|
12 |
22,929 |
04-19-2013, 04:37 PM |
|
|
Thread: HD7950, HD5870, or HD5970
Post: RE: HD7950, HD5870, or HD5970
personally I would not buy the hdXX50 versions. If you really want it as cheap as possible, the hd5870 is not bad. |
|
atom |
Hardware
|
4 |
9,569 |
05-18-2013, 10:16 PM |
|
|
Thread: Radeon HD 7990 Review
Post: RE: Radeon HD 7990 Review
What its speed on MD5? Somewhere at 18000 MH/s? |
|
atom |
Hardware
|
11 |
33,154 |
06-17-2013, 12:37 PM |
|
|
Thread: First build, need opinions.
Post: RE: First build, need opinions.
2x7870 are around 1.35 times faster than 1x7970, but they produce 350W and the 7970 only 230W TDP |
|
atom |
Hardware
|
9 |
14,591 |
06-20-2013, 03:49 PM |
|
|
Thread: oclhashcat: recommended amount of RAM
Post: RE: oclhashcat: recommended amount of RAM
As RAM isn't that expensive anymore, I recommend 16gb of RAM. It's not really required but if you do a lot of cracking job it helps to have the dictionaries cached in RAM (automatically done by the OS... |
|
atom |
Hardware
|
3 |
9,036 |
11-11-2013, 04:41 PM |
|
|
Thread: Low GPU load
Post: RE: Low GPU load
|
atom |
Hardware
|
8 |
13,403 |
11-25-2013, 12:02 PM |
|
|
Thread: Review - AMD Radeon R9 290X
Post: RE: Review - AMD Radeon R9 290X
There is 13.12 stable now, you might want to try again with that one |
|
atom |
Hardware
|
75 |
216,074 |
12-21-2013, 04:57 PM |
|
|
Thread: 8K Budget
Post: RE: 8K Budget
Yeah, latest info on this is that 10x GPU work on X11, however AMD's OpenCL runtime still only supports a maximum of 8. |
|
atom |
Hardware
|
7 |
14,385 |
03-11-2014, 07:50 PM |
|
|
Thread: [Beta] cudaHashcat-1.20 - GTX 750 TI Benchmark
Post: RE: [Beta] cudaHashcat-1.20 - GTX 750 TI Benchmark
Hey, thanks, can you please post all the other algorithms as well? |
|
atom |
Hardware
|
10 |
28,998 |
03-21-2014, 09:53 AM |
|
|
Thread: Stupid Question
Post: RE: Stupid Question
Yes, the 2xhd6990 is faster, but not much:
1x hd7990 @ ntlm : 32986 MH/s
2x hd6990 @ ntlm : 39151 MH/s
But you save the pcie slot and the power consumption |
|
atom |
Hardware
|
1 |
4,471 |
03-27-2014, 10:21 PM |
|
|
Thread: Is a setup with 128 GPUs possible?
Post: RE: Is a setup with 128 GPUs possible?
I'm not aware of a single board with that much pci-express slots. But at the time when VCL was still working it was faking to the OpenCL client that it has X numbers of OpenCL capable devices so if yo... |
|
atom |
Hardware
|
3 |
7,941 |
04-14-2014, 06:40 PM |