Where are RTX 3060 Benchmarks?
#1
Where are RTX 3060 Benchmarks? Can't seem to find them anywhere online.

On a separate note. It would be nice if the hashcat community did a table of benchmaeks for all graphics cards so one can see all this info in one place when making a decision of what card to purchase for hashcat in terms of value (performance/cost). TIA.
Reply
#2
I finally got my greasy hands on an RTX 3060 (NOT the Ti version) without having to pay scalpers, but overclockers.co.uk still charges substantially higher than the MSRP. It cost £650 ($895)! This took me about a month of watching online stocks through various websites to finally get a hold of one. And to be honest I am rather disappointed as in some benchmarks it seems to do slightly worse than an RTX 2060, eg -m 11300 is a measly 5135 H/s vs 5302 H/s on RTX 2060

Does anyone know if this is because of the hash limiter enforced by Nvidia by any chance? This card has 3584 vs the RTX 2060's 1920 cores, it boggles the mind. Or am I doing something wrong? I honestly hope it's the latter.


The test was done in Ubuntu 20.04 i7-6700 with 16GB RAM and Nvidia driver version 460:

$hashcat -b -w 4
hashcat (v6.1.1-120-g15bf8b730) starting in benchmark mode...

clGetDeviceIDs(): CL_DEVICE_NOT_FOUND

clGetDeviceIDs(): CL_DEVICE_NOT_FOUND

Successfully initialized NVIDIA CUDA library.

Failed to initialize NVIDIA RTC library.

* Device #1: CUDA SDK Toolkit installation NOT detected or incorrectly installed.
CUDA SDK Toolkit installation required for proper device support and utilization
Falling back to OpenCL Runtime

* 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
OpenCL API (OpenCL 1.2 CUDA 11.2.153) - Platform #1 [NVIDIA Corporation]
========================================================================
* Device #1: GeForce RTX 3060, 11072/12050 MB (3012 MB allocatable), 28MCU

OpenCL API (OpenCL 1.1 Mesa 21.2.0-devel (git-376c878 2021-04-15 focal-oibaf-ppa)) - Platform #2 [Mesa]
=======================================================================================================

Benchmark relevant options:
===========================
* --workload-profile=4

Hashmode: 0 - MD5

Speed.#1.........: 10000.4 MH/s (187.59ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1

Hashmode: 100 - SHA1

Speed.#1.........: 4909.0 MH/s (382.40ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1

Hashmode: 1400 - SHA2-256

Speed.#1.........: 2237.4 MH/s (419.54ms) @ Accel:32 Loops:1024 Thr:1024 Vec:1

Hashmode: 1700 - SHA2-512

Speed.#1.........: 946.7 MH/s (247.96ms) @ Accel:32 Loops:256 Thr:1024 Vec:1

Hashmode: 22000 - WPA-PBKDF2-PMKID+EAPOL (Iterations: 4095)

Speed.#1.........: 398.8 kH/s (287.02ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1

Hashmode: 1000 - NTLM

Speed.#1.........: 13858.5 MH/s (135.46ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1

Hashmode: 3000 - LM

Speed.#1.........: 21180.3 MH/s (88.00ms) @ Accel:1024 Loops:1024 Thr:64 Vec:1

Hashmode: 5500 - NetNTLMv1 / NetNTLMv1+ESS

Speed.#1.........: 13257.2 MH/s (141.49ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1

Hashmode: 5600 - NetNTLMv2

Speed.#1.........: 1438.4 MH/s (326.22ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1

Hashmode: 1500 - descrypt, DES (Unix), Traditional DES

Speed.#1.........: 979.7 MH/s (479.41ms) @ Accel:256 Loops:1024 Thr:64 Vec:1

Hashmode: 500 - md5crypt, MD5 (Unix), Cisco-IOS $1$ (MD5) (Iterations: 1000)

Speed.#1.........: 898.3 kH/s (253.79ms) @ Accel:8 Loops:1000 Thr:1024 Vec:1

Hashmode: 3200 - bcrypt $2*$, Blowfish (Unix) (Iterations: 32)

Speed.#1.........: 31866 H/s (151.06ms) @ Accel:16 Loops:32 Thr:11 Vec:1

Hashmode: 1800 - sha512crypt $6$, SHA512 (Unix) (Iterations: 5000)

Speed.#1.........: 52080 H/s (438.54ms) @ Accel:4 Loops:1024 Thr:1024 Vec:1

Hashmode: 7500 - Kerberos 5, etype 23, AS-REQ Pre-Auth

Speed.#1.........: 463.4 MH/s (506.48ms) @ Accel:512 Loops:256 Thr:64 Vec:1

Hashmode: 13100 - Kerberos 5, etype 23, TGS-REP

Speed.#1.........: 449.6 MH/s (522.06ms) @ Accel:128 Loops:1024 Thr:64 Vec:1

Hashmode: 15300 - DPAPI masterkey file v1 (Iterations: 23999)

Speed.#1.........: 67393 H/s (290.16ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1

Hashmode: 15900 - DPAPI masterkey file v2 (Iterations: 12899)

Speed.#1.........: 38093 H/s (477.24ms) @ Accel:16 Loops:512 Thr:1024 Vec:1

Hashmode: 7100 - macOS v10.8+ (PBKDF2-SHA512) (Iterations: 1023)

Speed.#1.........: 487.4 kH/s (415.56ms) @ Accel:64 Loops:127 Thr:1024 Vec:1

Hashmode: 11600 - 7-Zip (Iterations: 16384)

Speed.#1.........: 381.1 kH/s (300.41ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1

Hashmode: 12500 - RAR3-hp (Iterations: 262144)

Speed.#1.........: 45607 H/s (314.16ms) @ Accel:64 Loops:16384 Thr:128 Vec:1

Hashmode: 13000 - RAR5 (Iterations: 32799)

Speed.#1.........: 42448 H/s (337.17ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1

Hashmode: 6211 - TrueCrypt RIPEMD160 + XTS 512 bit (Iterations: 1999)

Speed.#1.........: 304.4 kH/s (348.49ms) @ Accel:8 Loops:1024 Thr:1024 Vec:1

Hashmode: 13400 - KeePass 1 (AES/Twofish) and KeePass 2 (AES) (Iterations: 24569)

Speed.#1.........: 49355 H/s (387.21ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1

Hashmode: 6800 - LastPass + LastPass sniffed (Iterations: 499)

Speed.#1.........: 2704.0 kH/s (221.50ms) @ Accel:64 Loops:249 Thr:1024 Vec:1

Hashmode: 11300 - Bitcoin/Litecoin wallet.dat (Iterations: 200459)

Speed.#1.........: 5135 H/s (455.78ms) @ Accel:32 Loops:512 Thr:1024 Vec:1

Started: Thu Apr 15 14:31:40 2021
Stopped: Thu Apr 15 14:36:35 2021
Reply
#3
The only other thing worth mentioning is that The motherboard is Asus Prime Z-270P with Gen 3 PCI-e, whereas the GPU is a PCI-e gen 4 card. So that could be the bottleneck. Is anyone able to confirm this? Unfortunately this is the only motherboard I have, so I can't do the benchmarks in gen 4.
Reply
#4
It's kinda impossible that PCIe 3.0 x16 will be a a bottleneck in any scenario. In benchmarking it doesn't even matter at all, since the performance purely depends on the work, that is done on the GPU directly.
Reply
#5
It seems that the hashrate is limited by the linux drivers too as Nvidia planned. Bechause for -m 11300, the hash rate starts at around 30,000 H/s then can go down to around 6,000 H/s depending on how big the wordlist is. So not a suitable graphics card for hashcat it would seem. Can anyone else with this card confirm this at all?
Reply
#6
The Hashcat OpenCL performance was looking good except for the RTX 3060 not providing much uplift in some of the benchmarks compared to the RTX 2060.

https://www.phoronix.com/scan.php?page=a...pute&num=2
Reply