Nvidia 3080 benchmarks
#1
Did anyone had / has access to a Nvidia 3080 to run the hashcat benchmark on it?
I'm curious if the double shaders play a part in a much bigger performance on the new card(s)

Thanks,
V
Reply
#2
Watch @Chick3nman512 on Twitter - he'll likely be posting some in the next day or two.
~
Reply
#3
I see the RTX 3080 has a decent increase in CUDA cores (8704 vs the 4352 on the RTX 2080 TI). Is this one of the main drivers of cracking performance?

Anyways, looking forward to any benchmarks. Smile
Reply
#4
The number of CUDA cores is a little bit more complicated. NVIDIA did a little trick by Ampere. Using Turin you have 1xINT32 and 1xFP32, RTX 2080 TI has in total 4352 INT32 cores and 4352 FP32 cores. Using Ampere it is either 1xINT32 and 1xFP32 or 2xFP32; they can use the INT32 unit as FP32. Thus 3080 can have up to 2x4352=8704 FP32 core but 3080 can only have up to 4352 INT32 cores. The so called CUDA core number for 3080 is under optimal circumstances for FP32.
The base clock of 3080 is only a little bit higher than 2080 TI (1440 MHz vs. 1350 MHz).

Based on AIDA64 pictures I saw 3080 FE is only a little bit faster in SHA1 than 2080 TI FE and a little bit slower than 2080 TI OC. But I do not know how optimized AIDA64 is and if they use OpenCL or CUDA (text mentioned both).
I could not find the article I read, I could only found this: https://babeltechreviews.com/wp-content/...iv3080.jpg
Reply
#5
Thanks for the reply. Sneaky little Nvidia. Smile

I noticed the 3080 FE does very well in the Blender "CUDA Score", but not sure how that translates to cracking.
(https://youtu.be/AG_ZHi3tuyk?t=249)
Reply
#6
Not sure how well this translates, but there are some mining numbers for ETH around: https://videocardz.com/newz/nvidia-gefor...erformance
Reply
#7
ETH heavily relies on memory performance. So it's not the best indicator.
Reply
#8
https://imgur.com/a/Fy4nGAH

courtesy to Monstru from lab501.ro
Reply
#9
thanks for the first benchmark of the NVIDIA RTX 3080

It would be great if they install CUDA from https://developer.nvidia.com/cuda-downloads and perform a further benchmark with CUDA installed (I know, we could figure the cuda speed approximately, but you never know... it doesn't take much effort/time to install CUDA anyway).
Thx again

I converted the screenshot to text to make it easier to search (credits of course go to the author above lab501):
Code:
C:\Users\AORUS Xtreme\Downloads\hashcat-6.1.1\hashcat-6.1.1>hashcat.exe -b
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.

* Device #1: CUDA SDK Toolkit installation NOT detected.
             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 11.1.65) - Platform #1 [NVIDIA Corporation]
=======================================================================
* Device #1: GeForce RTX 3080, 9088/10240 MB (2560 MB allocatable), 68MCU

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

Hashmode: 0 - MD5

Speed.#1.........: 56793.4 MH/s (40.02ms) @ Accel:32 Loops:1024 Thr:1024 Vec:1

Hashmode: 100 - SHA1

Speed.#1.........: 19317.3 MH/s (58.91ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1

Hashmode: 1400 - SHA2-256

Speed.#1.........:  8226.8 MH/s (69.19ms) @ Accel:8 Loops:1024 Thr:1024 Vec:1

Hashmode: 1700 - SHA2-512

Speed.#1.........:  2419.2 MH/s (58.80ms) @ Accel:2 Loops:1024 Thr:1024 Vec:1

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

Speed.#1.........:   955.5 kH/s (72.65ms) @ Accel:8 Loops:512 Thr:1024 Vec:1

Hashmode: 1000 - NTLM

Speed.#1.........:   102.4 GH/s (22.13ms) @ Accel:32 Loops:1024 Thr:1024 Vec:1

Hashmode: 3000 - LM

Speed.#1.........: 57957.5 MH/s (39.13ms) @ Accel:512 Loops:1024 Thr:64 Vec:1

Hashmode: 5500 - NetNTLMv1 / NetNTLMv1+ESS

Speed.#1.........: 57801.4 MH/s (39.17ms) @ Accel:32 Loops:1024 Thr:1024 Vec:1

Hashmode: 5600 - NetNTLMv2

Speed.#1.........:  4296.7 MH/s (66.23ms) @ Accel:8 Loops:512 Thr:1024 Vec:1

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

Speed.#1.........:  2355.5 MH/s (60.30ms) @ Accel:32 Loops:1024 Thr:64 Vec:1

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

Speed.#1.........: 27279.7 kH/s (78.82ms) @ Accel:32 Loops:1000 Thr:1024 Vec:1

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

Speed.#1.........:    79788 H/s (36.21ms) @ Accel:4 Loops:32 Thr:11 Vec:1

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

Speed.#1.........:   417.3 kH/s (66.27ms) @ Accel:8 Loops:256 Thr:1024 Vec:1

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

Speed.#1.........:  1251.1 MH/s (56.84ms) @ Accel:64 Loops:256 Thr:64 Vec:1

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

Speed.#1.........:  1255.0 MH/s (56.67ms) @ Accel:256 Loops:64 Thr:64 Vec:1

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

Speed.#1.........:   162.3 kH/s (71.33ms) @ Accel:4 Loops:1024 Thr:1024 Vec:1

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

Speed.#1.........:    93098 H/s (59.24ms) @ Accel:8 Loops:128 Thr:1024 Vec:1

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

Speed.#1.........:  1166.0 kH/s (55.72ms) @ Accel:16 Loops:63 Thr:1024 Vec:1

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

Speed.#1.........:   937.1 kH/s (68.44ms) @ Accel:4 Loops:4096 Thr:1024 Vec:1

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

Speed.#1.........:   100.9 kH/s (43.04ms) @ Accel:1 Loops:16384 Thr:1024 Vec:1

Hashmode: 13000 - RAR5 (Iterations: 32799)

Speed.#1.........:   103.3 kH/s (81.64ms) @ Accel:4 Loops:1024 Thr:1024 Vec:1

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

clEnqueueNDRangeKernel(): CL_MEM_OBJECT_ALLOCATION_FAILURE

Started: Fri Sep 18 12:05:12 2020
Stopped: Fri Sep 18 12:07:53 2020

C:\Users\AORUS Xtreme\Downloads\hashcat-6.1.1\hashcat-6.1.1>
Reply
#10
(09-18-2020, 02:28 PM)royce Wrote: It would also be helpful if the benchmark was posted as text, instead of as a giant screenshot. This will allow it to be copy and pasted for analysis, picked up by search engines, etc.


i received it like this...  screenshot without cuda toolkit installed

i'm sure the owners will start benchmarking it soon,

i compared to Nvidia 2080FE and the improvement i saw is between 50 to 80% (i compared some hashes i was interested in)
Reply