v62 is up
#31
Can you send here all data for reproduce all bugs you found ?

Thanks

(09-02-2012, 09:54 PM)M@LIK Wrote: System: Windows7 64-bit, nVidia, 304,79 beta, oclHashcat-plus-0.09b68.

oclHashcat-plus-0.08
-m0:
hc64p -n320 --gpu-loops=512 -a3 -1 ?l?u?d?s --runtime=120 ..\M\1mfakemd5.hash ?1?1?1?1?1?1?1?1
Speed........: 164.9M c/s Real, 164.8M c/s GPU
-m2811:
hc64p -n320 -m2811 -a3 -1 ?l?u?d?s --runtime=60 5kfakeipb.txt ?1?1?1?1?1?1?1?1
Speed........: 52526.1k c/s Real, 52576.8k c/s GPU

oclHashcat-plus-0.09b68
-m0:
hc64p -n320 --gpu-loops=512 -a3 --bf-min=8 --runtime=60 ..\M\1mfakemd5.hash ?a?a?a?a?a?a?a?a
Speed........: 104.4M c/s Real, 104.6M c/s GPU
-m2811:
hc64p -n320 -m2811 -a3 --bf-min=8 --runtime=60 5kfakeipb.txt ?a?a?a?a?a?a?a?a
Speed........: 62915.0k c/s Real, 63064.1k c/s GPU

IPB's speed is very nice compared to the previous version. That leaves MD5 questionable.

These algos might not be very important, but in case you want this to be perfect:
-m3100: When I use -a3, all is being cracked but if len(pass) > 8 it truncate the passes to 8.
-m3000: LM is behaving much better! BUT still:
-a0: does NOT crack any 6-character-long passwords!
-a6: Sometimes adds a char in the end of a cracked password. (Very similar to: https://hashcat.net/forum/thread-1314.html )

Going to re-test everything. Hopefully today.

All tested @ Windows7 64-bit, nVidia, 304,79 beta, oclHashcat-plus-0.09b68
Everything works flawlessly EXCEPT the two algos mentioned above (-m3000, -m3100).

Will try to do some speed\performance tests later.
#32
oclHashcat-plus-0.09b68 running under Windows 7 64 12.8

A simple test to see the evolution of the speed using 2811 algos,
the last test was with b53, and I'm getting after every second or so
the known pause/stop/resume options but with the message:
WARN: ADL_overdrive5_temperature_get(): -5

Changed to a MD5 kind, to see if this was a problem, and got the
same message, but time followed by windows message oclhashcat
stoped working etc.

Have to restart to be sure. Didn't change anything since my last
test about 2811 performance, so, eventually, after a few days
on, this must be shutdown for a few minutes.

Will update this as soon as possible.
#33
I am currently running the full test on Nvidia. It's not finished but I noticed that the percent progress bug still occurs. This was on -a 7 for SHA512(unix).


Attached Files Thumbnail(s)
   
#34
Hi atom, have created a graph, seems b62 is only slower with alot of hashes, see graph below:

[Image: oclhc-plus_graph.jpg]

It's faster with fewer hashes, anything over 200000 say its slower than v0.081.

Hope this helps.
#35
(09-03-2012, 01:41 AM)proinside Wrote: oclHashcat-plus-0.09b68 running under Windows 7 64 12.8

A simple test to see the evolution of the speed using 2811 algos,
the last test was with b53, and I'm getting after every second or so
the known pause/stop/resume options but with the message:
WARN: ADL_overdrive5_temperature_get(): -5

Changed to a MD5 kind, to see if this was a problem, and got the
same message, but time followed by windows message oclhashcat
stoped working etc.

Have to restart to be sure. Didn't change anything since my last
test about 2811 performance, so, eventually, after a few days
on, this must be shutdown for a few minutes.

Will update this as soon as possible.

The ADL_overdrive5_temperature_get Error is always a good indice to show that there is some problem due to driver installation. To make sure your drivers are correct installed please do the following EXACTLY this way:

1. Reboot
2. uninstall catalyst using catalyst
3. reboot
4. uninstall catalyst using driver sweeper software
5. reboot
6. Manually remove remaining opencl.dll from windows directories (one in 32 bit and on ine 64 bit dir).
7. reboot
8. install catalyst 12.8
9. reboot
10. try again, it should work
#36
(09-03-2012, 10:12 AM)blandyuk Wrote: Hi atom, have created a graph, seems b62 is only slower with alot of hashes, see graph below:

[Image: oclhc-plus_graph.jpg]

It's faster with fewer hashes, anything over 200000 say its slower than v0.081.

Hope this helps.

Oh yes, that explains why I have noticed an increase in speed while you noticed a decrease. I am testing with the example hash (~7k hashes).

so if this is slower for larger hashlists, it might be an memory problem... need to think about that
#37
(09-02-2012, 09:54 PM)M@LIK Wrote: System: Windows7 64-bit, nVidia, 304,79 beta, oclHashcat-plus-0.09b68.

oclHashcat-plus-0.08
-m0:
hc64p -n320 --gpu-loops=512 -a3 -1 ?l?u?d?s --runtime=120 ..\M\1mfakemd5.hash ?1?1?1?1?1?1?1?1
Speed........: 164.9M c/s Real, 164.8M c/s GPU
-m2811:

oclHashcat-plus-0.09b68
-m0:
hc64p -n320 --gpu-loops=512 -a3 --bf-min=8 --runtime=60 ..\M\1mfakemd5.hash ?a?a?a?a?a?a?a?a
Speed........: 104.4M c/s Real, 104.6M c/s GPU
-m2811:

how many hashes did you use for testing?

I tried with old 295.x devdriver and with cuda 4.2.9, but it was just getting slower and slower. I am back at 304.x and CUDA 5.0 RC but still it has not the speed of 0.081 on high number of hashes. On low number of hashes, new version is faster. Can you verify?
#38
Quote:-m3100: When I use -a3, all is being cracked but if len(pass) > 8 it truncate the passes to 8.

reproduced. will be fixed with b69.

Quote:-m3000: LM is behaving much better! BUT still:
-a0: does NOT crack any 6-character-long passwords!
-a6: Sometimes adds a char in the end of a cracked password. (Very similar to: https://hashcat.net/forum/thread-1314.html )

see what i did here, i can not reproduce:

-a 0 with len 6:

Code:
root@ht:~/xy/oclHashcat-plus-0.09# echo -n 078407               | ./cudaHashcat-plus64.bin ${OPTS} -a 0 -m 3000 'bce83d454d81d1ce'
cudaHashcat-plus v0.09 by atom starting...

Device #1: Kernel ./kernels/4318/m3000_a0.sm_21.ptx

Starting attack in stdin mode...

bce83d454d81d1ce:078407                      
                                            
Status.......: Cracked

not adding random char at end:

Code:
root@ht:~/xy/oclHashcat-plus-0.09# echo 0784 > dict

root@ht:~/xy/oclHashcat-plus-0.09# ./cudaHashcat-plus64.bin ${OPTS} -a 0 -m 3000 'bce83d454d81d1ce' -a 6 dict ?d?d
Device #1: Kernel ./kernels/4318/m3000_a1.sm_21.ptx

Scanned dictionary dict: 5 bytes, 1 words, 100 keyspace, starting attack...

bce83d454d81d1ce:078407                      
                                            
Status.......: Cracked
Input.Base...: File (dict)
Input.Mod....: Mask (?d?d)
#39
b69 is up (all nvidia stuff)
#40
b70 is up !

@mastercracker:

- fix > 100% progress bug
- fix -m 3200 clEnqueueNDRangeKernel() -54 bug