+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Support (https://hashcat.net/forum/forum-3.html)
+--- Forum: hashcat (https://hashcat.net/forum/forum-45.html)
+--- Thread: 7.1.2 slower than 6.1.2 for SHA512 (/thread-13361.html)
7.1.2 slower than 6.1.2 for SHA512 - zamgold - 08-26-2025
Hi. I'm testing new version of hashcat 7.1.2..
I noticed that algorithm 1800 (sha512) works 3x slower than previos version hashcat 6.1.2...
hashcat (v7.1.2) starting
* Device #2: Outdated or broken Intel OpenCL runtime '495.29.05' detected!
You are STRONGLY encouraged to use the officially supported runtime.
See hashcat.net for the officially supported Intel OpenCL runtime.
See also: https://hashcat.net/faq/wrongdriver
You can use --force to override this, but do not report related errors.
CUDA API (CUDA 11.5)
====================
* Device #01: NVIDIA GeForce RTX 3080, 9576/10018 MB, 68MCU
OpenCL API (OpenCL 3.0 CUDA 11.5.56) - Platform #1 [NVIDIA Corporation]
=======================================================================
* Device #02: NVIDIA GeForce RTX 3080, skipped
Minimum password length supported by kernel: 0
Maximum password length supported by kernel: 256
Minimum salt length supported by kernel: 0
Maximum salt length supported by kernel: 256
Hashes: 1 digests; 1 unique digests, 1 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates
Rules: 1
Optimizers applied:
* Zero-Byte
* Single-Hash
* Single-Salt
* Uses-64-Bit
* Register-Limit
ATTENTION! Pure (unoptimized) backend kernels selected.
Pure kernels can crack longer passwords, but drastically reduce performance.
If you want to switch to optimized kernels, append -O to your commandline.
See the above message to find out about the exact limits.
Watchdog: Temperature abort trigger set to 90c
Host memory allocated for this attack: 1305 MB (56445 MB free)
Dictionary cache hit:
* Filename..: /media/ntfs1/dictionary/dict.dic
* Passwords.: 855284955
* Bytes.....: 10027391827
* Keyspace..: 855284955
Cracking performance lower than expected?
* Append -O to the commandline.
This lowers the maximum supported password/salt length (usually down to 32).
* Append -w 3 to the commandline.
This can cause your screen to lag.
* Append -S to the commandline.
This has a drastic speed impact but can be better for specific attacks.
Typical scenarios are a small wordlist but a large ruleset.
* Update your backend API 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 ypass [c]heckpoint [f]inish [q]uit =>
Session..........: hashcat
Status...........: Running
Hash.Mode........: 1800 (sha512crypt $6$, SHA512 (Unix))
Hash.Target......: $6$rounds=10000$salt$data_hash
Time.Started.....: Tue Aug 26 15:19:03 2025 (12 secs)
Time.Estimated...: Wed Aug 27 03:18:52 2025 (11 hours, 59 mins)
Kernel.Feature...: Pure Kernel (password length 0-256 bytes)
Guess.Base.......: File (/media/ntfs1/dictionary/dict.dic)
Guess.Queue......: 1/1 (100.00%)
[b]Speed.#01........: 19804 H/s (14.87ms) @ Accel:4 Loops:124 Thr:96 Vec:1[/b]
Recovered........: 0/1 (0.00%) Digests (total), 0/1 (0.00%) Digests (new)
Progress.........: 208896/855284955 (0.02%)
Rejected.........: 0/208896 (0.00%)
Restore.Point....: 208896/855284955 (0.02%)
Restore.Sub.#01..: Salt:0 Amplifier:0-1 Iteration:5208-5332
Candidate.Engine.: Device Generator
Candidates.#01...: !mpl@u5!8!l!7y -> !n4d3qu4+3n3ss
Hardware.Mon.#01.: Temp: 55c Fan: 59% Util: 94% Core:1950MHz Mem:9251MHz Bus:16
[s]tatus [p]ause ypass [c]heckpoint [f]inish [q]uit =>
[b]For version 6.1.2:[/b]
hashcat (v6.2.6) starting
* Device #1: WARNING! Kernel exec timeout is not disabled.
This may cause "CL_OUT_OF_RESOURCES" or related errors.
To disable the timeout, see: https://hashcat.net/q/timeoutpatch
* Device #2: WARNING! Kernel exec timeout is not disabled.
This may cause "CL_OUT_OF_RESOURCES" or related errors.
To disable the timeout, see: https://hashcat.net/q/timeoutpatch
CUDA API (CUDA 11.5)
====================
* Device #1: NVIDIA GeForce RTX 3080, 9576/10018 MB, 68MCU
OpenCL API (OpenCL 3.0 CUDA 11.5.56) - Platform #1 [NVIDIA Corporation]
=======================================================================
* Device #2: NVIDIA GeForce RTX 3080, skipped
Minimum password length supported by kernel: 0
Maximum password length supported by kernel: 256
Hashes: 1 digests; 1 unique digests, 1 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates
Rules: 1
Optimizers applied:
* Zero-Byte
* Single-Hash
* Single-Salt
* Uses-64-Bit
ATTENTION! Pure (unoptimized) backend kernels selected.
Pure kernels can crack longer passwords, but drastically reduce performance.
If you want to switch to optimized kernels, append -O to your commandline.
See the above message to find out about the exact limits.
Watchdog: Temperature abort trigger set to 90c
Host memory required for this attack: 281 MB
Dictionary cache hit:
* Filename..: /media/ntfs1/dictionary/dict.dic
* Passwords.: 855284955
* Bytes.....: 10027391827
* Keyspace..: 855284955
Cracking performance lower than expected?
* Append -O to the commandline.
This lowers the maximum supported password/salt length (usually down to 32).
* Append -w 3 to the commandline.
This can cause your screen to lag.
* Append -S to the commandline.
This has a drastic speed impact but can be better for specific attacks.
Typical scenarios are a small wordlist but a large ruleset.
* Update your backend API 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 ypass [c]heckpoint [f]inish [q]uit =>
Session..........: hashcat
Status...........: Running
Hash.Mode........: 1800 (sha512crypt $6$, SHA512 (Unix))
Hash.Target......: $6$rounds=10000$salt$data_hash
Time.Started.....: Tue Aug 26 15:21:20 2025 (18 secs)
Time.Estimated...: Tue Aug 26 19:08:44 2025 (3 hours, 47 mins)
Kernel.Feature...: Pure Kernel
Guess.Base.......: File (/media/ntfs1/dictionary/dict.dic)
Guess.Queue......: 1/1 (100.00%)
[b]Speed.#1.........: 62680 H/s (6.16ms) @ Accel:256 Loops:128 Thr:128 Vec:1[/b]
Recovered........: 0/1 (0.00%) Digests (total), 0/1 (0.00%) Digests (new)
Progress.........: 1146880/855284955 (0.13%)
Rejected.........: 0/1146880 (0.00%)
Restore.Point....: 1146880/855284955 (0.13%)
Restore.Sub.#1...: Salt:0 Amplifier:0-1 Iteration:1664-1792
Candidate.Engine.: Device Generator
Candidates.#1....: ,ecmrf1985 -> ,fcntn05
Hardware.Mon.#1..: Temp: 56c Fan: 59% Util: 89% Core:1905MHz Mem:9251MHz Bus:16
[s]tatus [p]ause ypass [c]heckpoint [f]inish [q]uit =>
[b]I tested on different GPUs NVidia 3080 and GPUs NVidia 4090..this algorithm on version 7.1.2 works slower.[/b]
RE: 7.1.2 slower than 6.1.2 for SHA512 - atom - 08-26-2025
It should not, please update to latest CUDA version 12.9 or higher.