Search found 51 matches

by mrb
Wed Jun 15, 2011 4:06 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Anyone doing bitcoin mining?
Replies: 18
Views: 22308

Re: Anyone doing bitcoin mining?

It's not hard to write your own miner. I wrote mine in 2 days (it is a little faster than the fastest open source ones, on HD 69xx). Writing your own pool is more work. Personally I don't use pools, I mine solo. All nodes in the peer-to-peer network have to speak the same protocol, or else Bitcoin w...
by mrb
Tue Jun 14, 2011 8:47 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Anyone doing bitcoin mining?
Replies: 18
Views: 22308

Re: Anyone doing bitcoin mining?

Don't be sad. You are already one of the top 1% miners with 2.8Gh/s...
by mrb
Fri Jun 10, 2011 7:44 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Anyone doing bitcoin mining?
Replies: 18
Views: 22308

Re: Anyone doing bitcoin mining?

Yep :) I have been mining since December 2010; currently doing 20+ Ghash/s. The largest known public miner is ArtForz with 55 Ghash/s, a third using 5970s and two thirds using custom structured ASICs. Scoobz: correct, $10M would give you 50% of the global hashrate. This may allow you to cause havoc ...
by mrb
Thu Feb 24, 2011 6:08 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 33.1 billion hash/sec

... However I am pretty sure that undervolting is more effective than underclocking. Power consumption goes up linearly with frequency and quadratically with voltage. For a CMOS device P = CV^2f. mrb do you have a forum for the whitepixel project? Ive hit some compile errors and could use some refe...
by mrb
Thu Jan 06, 2011 6:55 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: How will Intel Sandybridge help with GPU programming?
Replies: 1
Views: 5289

Re: How will Intel Sandybridge help with GPU programming?

This answers your question :-)

Intel's Sandy Bridge to Deliver 2% of AMD's Top Graphics Performance
http://blog.zorinaq.com/?e=31
by mrb
Wed Dec 22, 2010 3:20 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 33.1 billion hash/sec

Mysql323 is kind of low on my priority list because it is rarely requested. I have also heard (IIRC) of performance levels of trillions of passwords per second with CPU-based mysql323 brute forcers, so this symbolic barrier has already been broken :-) What is this method you speak of? I tried googl...
by mrb
Mon Dec 20, 2010 10:58 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 33.1 billion hash/sec

Mysql323 is kind of low on my priority list because it is rarely requested.

I have also heard (IIRC) of performance levels of trillions of passwords per second with CPU-based mysql323 brute forcers, so this symbolic barrier has already been broken :-)
by mrb
Sat Dec 18, 2010 10:13 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 33.1 billion hash/sec

I highly favor performance/Watt as opposed to absolute performance, so, yes, I am planning to do many undervolting and underclocking tests... However I am pretty sure that undervolting is more effective than underclocking. My preliminary tests so far show that decreasing the core frequency of one of...
by mrb
Thu Dec 16, 2010 9:51 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 33.1 billion hash/sec

8xHD 6970 should do about 38.6 Bhash/sec under whitepixel v2. Possibly up to ~41 Bhash/sec with some more code optimization ideas I may have... Linux drivers support at least 8 GPUs. Windows drivers are limited to 4 it seems. CPU speed is irrelevant for whitepixel. It does not use the CPU at all. CP...
by mrb
Tue Dec 14, 2010 1:24 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 28.6 billion hash/sec

I released whitepixel v2. The nice surprise is... support for the BFI_INT instruction which allows implementing the MD5 subfunctions F() and G() in 1 clock cycle. My 4x5970 setup now reaches 33.1 billion passwords per second . http://blog.zorinaq.com/?e=43 http://whitepixel.zorinaq.com/example.png
by mrb
Mon Dec 13, 2010 1:10 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 28.6 billion hash/sec

Scoobz is correct (as usual).

But expect a nice surprise with whitepixel v2 :wink:
by mrb
Fri Dec 10, 2010 5:28 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

Re: 28.6 billion hash/sec

Nobody has 8 GTX 580. This is why I wrote "estimated", based on number of ALUs and ALU clock.

Has anyone on the forum tried my method to down-plug a PCIe x16 graphics card in an x1 slot?
by mrb
Thu Dec 09, 2010 3:11 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: 33.1 billion hash/sec
Replies: 25
Views: 32101

33.1 billion hash/sec

Edit: now reaching 33.1 billion password hashes per second... see later comment. I wrote software and built a machine capable of cracking 28.6 billion MD5 password hashes/sec , handily beating D3ad0ne's challenge and his 17.2 billion hashes/sec :-) My setup is 1.7x faster , consumes half the power ...
by mrb
Thu Dec 02, 2010 6:04 pm
Forum: General Talk
Topic: OMG DNS
Replies: 2
Views: 5023

Re: OMG DNS

It is in these times that we see those who are so addicted to 3.14.by that they still manage to post to the forum :lol:
by mrb
Wed Jun 02, 2010 6:41 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: RAR archive encription
Replies: 27
Views: 27136

Re: RAR archive encription

I finally did an official release of the unrarhp brute forcer I mentioned 1+ year ago in this thread:
http://blog.zorinaq.com/?e=15

Better late than never :-)
by mrb
Wed May 12, 2010 8:59 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: GeForce GTX 480
Replies: 5
Views: 7326

Re: GeForce GTX 480

To add some practical numbers to Scoobz's post, a GTX 480 is not much faster... [*] A GTX 480 reaches about 1590 million passwords per sec with BarsWF [*] A GTX 295 reaches about 1410 million passwords per sec with BarsWF If you want to crack MD5s, buy ATI. For example the 5870 is cheaper and two ti...
by mrb
Sat Jan 09, 2010 7:45 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: patched ati version
Replies: 10
Views: 7489

Re: patched ati version

Even though the SSE path is obviously faster on all processors supporting SSE. This is not true. Which I was talking about. It is true in the vast majority of the cases. It makes no sense at all to care about, and optimize the very rare cases (P4, VIA in combination with some rare SSE instruction s...
by mrb
Sat Jan 09, 2010 7:42 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: patched ati version
Replies: 10
Views: 7489

Re: patched ati version

I doubt that ICC intentionally generates slow code for AMD CPUs. Yes, ICC will intentionally select a slower code path for non-Intel CPUs. For example I remember seeing code that was compiled to 2 different code paths, an SSE one, and a non-SSE one, in the same binary/executable file; but during ru...
by mrb
Fri Jan 08, 2010 3:34 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: patched ati version
Replies: 10
Views: 7489

Re: patched ati version

I've just heard few days ago that Intel intentionally chooses slower codepath for AMD. Doubt it's true. And AMD can always make their own compiler if ICC that bad ;). Yes it is true. It is overly documented in all the lawsuits Intel is currently facing and has been known in technical circles for ye...
by mrb
Wed Dec 02, 2009 6:22 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: How high can you get
Replies: 260
Views: 160512

Re: How high can you get

Again, it's not a LDS issue. We don't need LDS for single hash at all. It's "memory access model changed with 5000 but still undocumented because ATI don't care" issue. I know. Single hash doesn't need LDS and ighashgpu performs extremely well in this case like I said in the beginning. This time I ...
by mrb
Wed Dec 02, 2009 12:18 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: How high can you get
Replies: 260
Views: 160512

Re: How high can you get

Actually it's not Brook+ issue (never used it, only CAL/IL). It's not a problem to access LDS even with Brook+ but it won't helps a lot. Main problem right now that HD 5000's memory fetching working worse than HD 4000's one. And there no explanations from ATI why it happens and will they fix this w...
by mrb
Tue Dec 01, 2009 4:58 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: How high can you get
Replies: 260
Views: 160512

Re: How high can you get

It's possible to reach 4.3B on 5970 for single md5 hash but for list with 75K hashes speed dropped 3x times to only 1.4B. And with larger lists it becoming even worse We were talking about single-hash perf. You are right about multi-hash which is a whole different thing. However we all know it suck...
by mrb
Tue Dec 01, 2009 7:58 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: How high can you get
Replies: 260
Views: 160512

Re:

I can only imagine 12 virtual CPU cores and 7 GTX 380's.. GTX 380 = 1955.38 Million H/sec each card i9 = ~40 Million H/sec each core at a speed of about 3.6Ghz 1955.38 X 7 = 13687.64 40 X 12 = 480 480 + 13687.64 = total 14167.64 Million hashes a sec for single hash speed. I usually get around 3.4 B...
by mrb
Sat May 02, 2009 4:32 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Nvidia or ATI
Replies: 14
Views: 8429

Re: Nvidia or ATI

knapish wrote: As soon i get my hands on the report i think next week, i could upload it. And my mate had some weird connection to ATI i could ask him about that if you like.
Please do, I would love to know. Thanks in advance.
by mrb
Fri May 01, 2009 3:45 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Nvidia or ATI
Replies: 14
Views: 8429

Re: Nvidia or ATI

Its a long story but me and my mates from school had a project to build a high-end comp. And yes we noticed about the drivers so we had to build our own drivers into our linux based OS. [...] EDIT: We got some help from ATI with the drivers :sad: Great ! Can you give more details ? I think a lot of...