hashcat v6.2.0 SCRYPT

[hashzero]

There are also several other scrypt related improvements including some of the most in-depth sections of the salsa algorithm having been optimized.
...
I've posted a write-up on how to find the ideal tuning settings for your device here: https://github.com/hashcat/hashcat/blob/...#L388-L474
...

Am I correct in my understanding that these improvements and tunings are not applicable when scrypt N is very high?

I have an ethereum wallet with N=262144 and therefore limited to CPU for the foreseeable future.


Also,when I benchmark using hashcat.exe -b -m 15700, I get slightly worse H/s with 6.2.1 than I do with 6.1.1. Is this expected?


6.2.1 = 10 H/s

Code:

❯ .\hashcat.exe -b -m 15700 -D 1 -d 2
hashcat (v6.2.1) starting in benchmark mode...

Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.

* Device #1: Unstable OpenCL driver detected!

This OpenCL driver has been marked as likely to fail kernel compilation or to produce false negatives.
You can use --force to override this, but do not report related errors.

OpenCL API (OpenCL 3.0 ) - Platform #1 [Intel(R) Corporation]
=============================================================
* Device #1: Intel(R) HD Graphics 530, skipped

OpenCL API (OpenCL 2.1 WINDOWS) - Platform #2 [Intel(R) Corporation]
====================================================================
* Device #2: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz, 48765/48829 MB (12207 MB allocatable), 8MCU

Benchmark relevant options:
===========================
* --backend-devices=2
* --opencl-device-types=1
* --optimized-kernel-enable

Hashmode: 15700 - Ethereum Wallet, SCRYPT (Iterations: 262144)

Speed.#2.........:      10 H/s (1.65ms) @ Accel:8 Loops:1024 Thr:1 Vec:1

6.1.1 = 13 H/s

Code:

❯ .\hashcat.exe -b -m 15700 -D 1 -d 2
hashcat (v6.1.1) starting in benchmark mode...

Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.

Kernel ./OpenCL/m15700-optimized.cl:
Optimized kernel requested but not needed - falling back to pure kernel

* Device #1: Unstable OpenCL driver detected!

This OpenCL driver has been marked as likely to fail kernel compilation or to produce false negatives.
You can use --force to override this, but do not report related errors.

OpenCL API (OpenCL 3.0 ) - Platform #1 [Intel(R) Corporation]
=============================================================
* Device #1: Intel(R) HD Graphics 530, skipped

OpenCL API (OpenCL 2.1 WINDOWS) - Platform #2 [Intel(R) Corporation]
====================================================================
* Device #2: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz, 48765/48829 MB (12207 MB allocatable), 8MCU

Benchmark relevant options:
===========================
* --backend-devices=2
* --opencl-device-types=1
* --optimized-kernel-enable

Hashmode: 15700 - Ethereum Wallet, SCRYPT (Iterations: 1)

Speed.#2.........:      13 H/s (627.44ms) @ Accel:1 Loops:1 Thr:1 Vec:1

[atom]

No, it's faster. You can try to increase -n value. Higher Scrypt parameter work much better with v6.2.x than older ones.