hashcat Forum

Hi,

I ran cudahashcat v1.37 on a small dic, and I had the message "ATTENTION!  The wordlist or mask you are using is too small....."

Ok. however it tooks ~10 seconds :

Then I piped with hashcat 0.50  ; it tooks 2 minutes with a huge drop speed (sometimes 0 H/s, sometimes 300kH/s, instead of Mh/s without pipe)

Hum, what am I doing wrong ?

Thank you.

Windows 8 / GTX 980 / regedit patch applied.

You're running a straight dictionary attack against a fast hash, so it is impossible to gain GPU acceleration as you do not have an amplifier. You then made an already-slow attack even slower by using pipe.

OK, new try with :
- wordlist : 38M words (more than the 10M advised in the FAQ)
- [amplifier] rules list of 2000 rules applied on GPU (more than the 1000 advised in the FAQ)

Give :

- Straigh use of cudahashcat :
"ATTENTION! The wordlist or mask you are using is too small....."
Duration : 1 minute 10 seconds

- piping hashcat :
hashcat-cli64.exe --stdout *.dic | cudaHashcat64.exe -w 3 -r 2k.rule hash.txt
Duration : ~3 minutes

Quote:For fast hashes such as MD5, it is crucial to expand the candidates on the GPU with rules or masks in order to achieve full acceleration

It seems that I did what the FAQ advise, why can't I get full speed ?

For fast hashes, I would rather not using pipe, right ?

It's a potential workaround, not a magic fix. Pipe incurs more host-device transfers and is more sensitive to PCI-e bus speed & latency.

You should be using cpu hashcat for this attack.

(12-03-2015, 10:30 AM)epixoip Wrote: [ -> ]It's a potential workaround, not a magic fix.

Ok, got it.

Quote:You should be using cpu hashcat for this attack.

Hum, Cpu Hashcat gives 33Mh/s while cudahashcat give 500Mh/s, so even with the message 'attention..' I would continue to use GPU

(12-03-2015, 02:45 PM)Mem5 Wrote: [ -> ]

(12-03-2015, 10:30 AM)epixoip Wrote: [ -> ]It's a potential workaround, not a magic fix.

Ok, got it.

Quote:You should be using cpu hashcat for this attack.

Hum, Cpu Hashcat gives 33Mh/s while cudahashcat give 500Mh/s, so even with the message 'attention..' I would continue to use GPU

Calling experts:  Is the above true or is the rate an indicated rather than actual rate?  I've noticed the same thing in my limited testing and wondered why GPUs aren't used even for sub-optimal GPU tasks if they run through many more hashes than a CPU (albeit inefficiently).  Can anyone clarify?

It depends on the hardware and the workload. There's no "one size fits all" answer, you just have to know your tools. In this case, it looks like Mem5 has a really underpowered CPU if he's only pulling 33 MH/s on MD5, so even an inefficient attack on GPU is faster in his case.

Exactly, I use hashcat on a lowcost i3, and cudahashcat on a high-class GPU that's why my I would always prefer using GPU tools.