RTX 2080 Low performance
#1
I'm running a single RTX 2080 and getting ~1k MH/s on MD5 even though benchmark is at ~38k.
I'm running ubuntu 18.04, nvidia drivers were installed according to wiki, i've tried -w 3, --force, -0 without any results. The day before i was running the same cotmmand and it worked at full speed. What could cause the problem? Here's my output:


Code:
vox@ubuntu:~$ sudo hashcat -a 0 -m 0 ~/Documents/HASH1/Res_MD5_1.txt ~/MEGAsync\ Downloads/weakpass_no_dubles.txt -r ~/password_cracking_rules/OneRuleToRuleThemAll.rule
hashcat (v4.0.1) starting...

* Device #2: Not a native Intel OpenCL runtime. Expect massive speed loss.
             You can use --force to override, but do not report related errors.
OpenCL Platform #1: NVIDIA Corporation
======================================
* Device #1: GeForce RTX 2080, 1988/7952 MB allocatable, 46MCU

OpenCL Platform #2: The pocl project
====================================
* Device #2: pthread-Intel(R) Core(TM) i5-8400 CPU @ 2.80GHz, skipped.

Hashes: 208826 digests; 203474 unique digests, 1 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates
Rules: 51995

Applicable optimizers:
* Zero-Byte
* Early-Skip
* Not-Salted
* Not-Iterated
* Single-Salt
* Raw-Hash

Password length minimum: 0
Password length maximum: 256

ATTENTION! Pure (unoptimized) OpenCL kernels selected.
This enables cracking passwords and salts > length 32 but for the price of drastical reduced performance.
If you want to switch to optimized OpenCL kernels, append -O to your commandline.

Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger disabled.

INFO: Removed 1666 hashes found in potfile.

* Device #1: build_opts '-I /usr/share/hashcat/OpenCL -D VENDOR_ID=32 -D CUDA_ARCH=705 -D AMD_ROCM=0 -D VECT_SIZE=1 -D DEVICE_TYPE=4 -D DGST_R0=0 -D DGST_R1=3 -D DGST_R2=2 -D DGST_R3=1 -D DGST_ELEM=4 -D KERN_TYPE=0 -D _unroll'
Dictionary cache hit:
* Filename..: /home/vox/MEGAsync Downloads/weakpass_no_dubles.txt
* Passwords.: 8653507966
* Bytes.....: 105044799160
* Keyspace..: 449939146692170

- Device #1: autotuned kernel-accel to 64                 
- Device #1: autotuned kernel-loops to 128
[s]tatus [p]ause [r]esume [b]ypass [c]heckpoint [q]uit => [s]tatus [p]ause [r]esCracking performance lower than expected?                 

* Append -O to the commandline.
  This lowers the maximum supported password- and salt-length (typically down to 32).

* Append -w 3 to the commandline.
  This can cause your screen to lag.

* Update your OpenCL runtime / driver the right way:
  https://hashcat.net/faq/wrongdriver

* Create more work items to make use of your parallelization power:
  https://hashcat.net/faq/morework

[s]tatus [p]ause [r]esume [b]ypass [c]heckpoint [q]uit => s

Session..........: hashcat
Status...........: Running
Hash.Type........: MD5
Hash.Target......: /home/root/Documents/HASH1/Res_MD5_1.txt
Time.Started.....: Mon Apr  8 14:18:18 2019 (1 min, 24 secs)
Time.Estimated...: Sat Apr 13 21:57:35 2019 (5 days, 7 hours)
Guess.Base.......: File (/home/root/MEGAsync Downloads/weakpass_no_dubles.txt)
Guess.Mod........: Rules (/home/root/password_cracking_rules/OneRuleToRuleThemAll.rule)
Guess.Queue......: 1/1 (100.00%)
Speed.Dev.#1.....:   979.1 MH/s (12.14ms)
Recovered........: 1666/203474 (0.82%) Digests, 0/1 (0.00%) Salts
Recovered/Time...: CUR:0,N/A,N/A AVG:0,0,0 (Min,Hour,Day)
Progress.........: 81544937472/449939146692170 (0.02%)
Rejected.........: 0/81544937472 (0.00%)
Restore.Point....: 1507328/8653507966 (0.02%)
Candidates.#1....: clydegwendoly0215 -> peacekamilisidoretapi
HWMon.Dev.#1.....: Temp: 56c Fan: 41% Util: 98% Core:1905MHz Mem:6800MHz Bus:16
Reply
#2
did you actually read what the output says ?

it literally said that you should use -w 3 and -O
Reply
#3
(04-08-2019, 09:10 PM)philsmd Wrote: did you actually read what the output says ?

it literally said that you should use -w 3 and -O

Yeah, i'm kind of new to this. I did that and speed went up to 7-9k but still not 38k, anything else i can do here?
Reply
#4
now i also get this
Code:
* Device #1: Kernel m00000_a0-optimized.4e53f2bc.kernel not found in cache! Building may take a while..
Reply
#5
The kernel building is a once-per-algorithm-per-hashcat-version thing, and will only slow you down on the first run.
~
Reply
#6
(04-08-2019, 10:14 PM)royce Wrote: The kernel building is a once-per-algorithm-per-hashcat-version thing, and will only slow you down on the first run.

ok, so what are possible reasons of the slowdown? It was running at full speed without -w 3 and -O
Reply
#7
(04-08-2019, 09:17 PM)thomasfinch Wrote: Yeah, i'm kind of new to this. I did that and speed went up to 7-9k but still not 38k, anything else i can do here?

Normal speed for recovering >200k passwords. You would get benchmark speed if you try to recover 1 password.
Reply
#8
(04-08-2019, 11:07 PM)Flomac Wrote:
(04-08-2019, 09:17 PM)thomasfinch Wrote: Yeah, i'm kind of new to this. I did that and speed went up to 7-9k but still not 38k, anything else i can do here?

Normal speed for recovering >200k passwords. You would get benchmark speed if you try to recover 1 password.
How did i get 38k MH/s yesterday on the same file and why do i get the same speed on different files now? Had 1/3M of md5s recovered in about 40 minutes at ~38k MH/s without -w 3 and -O. Something's clearly wrong..
Reply