Jetson Nano Benchmarks
#1
After quite a bit of work, CUDA devices can now be used without OpenCL, so we can do things like this:

https://twitter.com/Chick3nman512/status...5873122305

https://gist.github.com/Chick3nman/49813...ca4f26ef21
Reply
#2
Hey, awesome - a nice start to a new (retro?) era for hashcat!
~
Reply
#3
Not the Jetson Nano but here are some benchmark results from its bigger brother the Jetson AGX Xavier (set to „nvpmodel -m 0“ - 30W):

Code:
hashcat (v5.1.0-1397-g7f4df9eb+) 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.

clGetPlatformIDs(): CL_PLATFORM_NOT_FOUND_KHR

CUDA API (CUDA 10.0)
====================
* Device #1: Xavier, 15698 MB, 8MCU

Benchmark relevant options:
===========================
* --optimized-kernel-enable

Hashmode: 0 - MD5

Speed.#1.........:  4666.5 MH/s (57.07ms) @ Accel:32 Loops:1024 Thr:1024 Vec:1

Hashmode: 100 - SHA1

Speed.#1.........:  1548.2 MH/s (86.25ms) @ Accel:32 Loops:512 Thr:1024 Vec:1

Hashmode: 1400 - SHA2-256

Speed.#1.........:  670.5 MH/s (49.61ms) @ Accel:8 Loops:512 Thr:1024 Vec:1

Hashmode: 1700 - SHA2-512

Speed.#1.........:  217.4 MH/s (76.74ms) @ Accel:8 Loops:256 Thr:1024 Vec:1

Hashmode: 2500 - WPA-EAPOL-PBKDF2 (Iterations: 4095)

Speed.#1.........:    72302 H/s (56.39ms) @ Accel:8 Loops:256 Thr:1024 Vec:1

Hashmode: 1000 - NTLM

Speed.#1.........:  7027.2 MH/s (75.98ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1
Reply