Post Your GPU Benchmarks for WPA

N

Ntrain20k

Active member
Feedback: 0 / 0 / 0
Joined
Mar 18, 2020
Messages
459
Reaction score
430
Credits
2,975
BENCHMARK: RX 6900 XT,

OpenCL API (OpenCL 2.1 AMD-APP (3380.6)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #2: AMD Radeon RX 6900 XT, 16256/16368 MB (16112 MB allocatable), 40MCU

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

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#2.........: 1267.5 kH/s (255.24ms) @ Accel:128 Loops:1024 Thr:256 Vec:1
 
Last edited by a moderator:
S

soxrok2212

Active member
Contributor
Feedback: 0 / 0 / 0
Joined
Dec 30, 2019
Messages
2,637
Reaction score
74
Credits
488
Ntrain20k said:
OpenCL API (OpenCL 2.1 AMD-APP (3380.6)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #2: AMD Radeon RX 6900 XT, 16256/16368 MB (16112 MB allocatable), 40MCU

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

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#2.........: 1248.6 kH/s (64.45ms) @ Accel:512 Loops:512 Thr:32 Vec:1
Click to expand...
Impressive! How does this hold under load? Does it throttle?
 
Ranga

Ranga

Active member
Contributor
Feedback: 0 / 0 / 0
Joined
Dec 30, 2019
Messages
442
Reaction score
49
Credits
89
Benchmark RTX 3070Ti

Windows 10 / hashcat (v6.2.6)
ASUS TUF H310-Plus Gaming Motherboard
Intel Core i5 8500K
Corsair Vengeance 300MHz 16GB
ASUS GeForce RTX 3070 Ti TUF Gaming 8GB



C:\hashcat>hashcat.exe -b -m 22000
hashcat (v6.2.6) starting in benchmark mode

CUDA API (CUDA 11.7)
====================
* Device #1: NVIDIA GeForce RTX 3070 Ti, 7119/8191 MB, 48MCU

OpenCL API (OpenCL 3.0 CUDA 11.7.102) - Platform #1 [NVIDIA Corporation]
========================================================================
* Device #2: NVIDIA GeForce RTX 3070 Ti, skipped

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

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#1.........: 644.5 kH/s (74.44ms) @ Accel:8 Loops:1024 Thr:512 Vec:1

Started: Wed Oct 19 23:45:33 2022
 
Cyclone

Cyclone

Active member
Trusted
Contributor
Retired Moderator
Feedback: 3 / 0 / 0
Joined
Dec 30, 2019
Messages
3,176
Reaction score
900
Credits
5,904
BENCHMARK - RTX 4090 (MSI Gaming Trio @ stock 450w)

OS: Linux
Version: Debian 11 (bullseye)
CPU: 2x Intel Xeon E5-2690
Memory: 128gb ECC DDR4

Code: 
hashcat (v6.2.6) starting in benchmark mode

CUDA API (CUDA 11.8)
====================
* Device #1: NVIDIA Graphics Device, 23870/24255 MB, 128MCU

OpenCL API (OpenCL 3.0 CUDA 11.8.88) - Platform #1 [NVIDIA Corporation]
=======================================================================
* Device #2: NVIDIA Graphics Device, skipped

Benchmark relevant options:
===========================
* --optimized-kernel-enable
* --workload-profile=4
-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#1.........:  2489.3 kH/s (208.37ms) @ Accel:32 Loops:1024 Thr:512 Vec:1

Started: Fri Oct 14 17:59:20 2022
 
blandyuk

blandyuk

Active member
Trusted
Contributor
VIP Member
Feedback: 0 / 0 / 0
Joined
Jul 6, 2011
Messages
18,600
Reaction score
428
Credits
11,317
WPA Benchmark 3x RTX 2080 Ti - Managed to beat the RTX 4090, but took 3x RTX 2080 Ti cards to do it. Plus I'm at 3x 250 watts, so 750 watts, way more than 4090. What a beast that 4090 is!

Code: 
CUDA API (CUDA 11.7)
====================
* Device #1: NVIDIA GeForce RTX 2080 Ti, 10138/11263 MB, 68MCU
* Device #2: NVIDIA GeForce RTX 2080 Ti, 10138/11263 MB, 68MCU
* Device #3: NVIDIA GeForce RTX 2080 Ti, 10138/11263 MB, 68MCU

OpenCL API (OpenCL 3.0 ) - Platform #1 [Intel(R) Corporation]
=============================================================
* Device #4: Intel(R) HD Graphics 630, skipped

OpenCL API (OpenCL 3.0 CUDA 11.7.57) - Platform #2 [NVIDIA Corporation]
=======================================================================
* Device #5: NVIDIA GeForce RTX 2080 Ti, skipped
* Device #6: NVIDIA GeForce RTX 2080 Ti, skipped
* Device #7: NVIDIA GeForce RTX 2080 Ti, skipped

Benchmark relevant options:
===========================
* --backend-devices=1,2,3
* --optimized-kernel-enable
* --workload-profile=4

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

Speed.#1.........:   927.7 kH/s (298.18ms) @ Accel:32 Loops:512 Thr:1024 Vec:1
Speed.#2.........:   917.1 kH/s (301.41ms) @ Accel:16 Loops:1024 Thr:1024 Vec:1
Speed.#3.........:   917.8 kH/s (301.39ms) @ Accel:32 Loops:512 Thr:1024 Vec:1
Speed.#*.........:  2762.7 kH/s
 
Z

z0mbie

Member
Feedback: 0 / 0 / 0
Joined
Dec 30, 2019
Messages
23
Reaction score
35
Credits
121
7x3080 msi ventux and 1x3080 palit oc
ubuntu 22.04
hashcat (v6.2.6) starting in benchmark mode

CUDA API (CUDA 11.8)
====================
* Device #1: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #2: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #3: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #4: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #5: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #6: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #7: NVIDIA GeForce RTX 3080, 11832/12053 MB, 70MCU
* Device #8: NVIDIA GeForce RTX 3080, 9802/10018 MB, 68MCU

OpenCL API (OpenCL 3.0 CUDA 11.8.88) - Platform #1 [NVIDIA Corporation]
=======================================================================
* Device #9: NVIDIA GeForce RTX 3080, skipped
* Device #10: NVIDIA GeForce RTX 3080, skipped
* Device #11: NVIDIA GeForce RTX 3080, skipped
* Device #12: NVIDIA GeForce RTX 3080, skipped
* Device #13: NVIDIA GeForce RTX 3080, skipped
* Device #14: NVIDIA GeForce RTX 3080, skipped
* Device #15: NVIDIA GeForce RTX 3080, skipped
* Device #16: NVIDIA GeForce RTX 3080, skipped

Benchmark relevant options:
===========================
* --optimized-kernel-enable
* --workload-profile=4

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#1.........: 987.7 kH/s (287.85ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#2.........: 1000.3 kH/s (284.14ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#3.........: 984.9 kH/s (288.53ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#4.........: 1006.6 kH/s (282.37ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#5.........: 1013.1 kH/s (280.51ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#6.........: 997.1 kH/s (285.13ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#7.........: 988.2 kH/s (287.61ms) @ Accel:64 Loops:512 Thr:512 Vec:1
Speed.#8.........: 814.5 kH/s (338.82ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#*.........: 7792.3 kH/s



10x3070
hashcat (v6.2.6) starting in benchmark mode

CUDA API (CUDA 11.8)
====================
* Device #1: NVIDIA GeForce RTX 3070, 7820/7982 MB, 46MCU
* Device #2: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #3: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #4: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #5: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #6: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #7: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #8: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #9: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU
* Device #10: NVIDIA GeForce RTX 3070, 7823/7982 MB, 46MCU

OpenCL API (OpenCL 3.0 CUDA 11.8.88) - Platform #1 [NVIDIA Corporation]
=======================================================================
* Device #11: NVIDIA GeForce RTX 3070, skipped
* Device #12: NVIDIA GeForce RTX 3070, skipped
* Device #13: NVIDIA GeForce RTX 3070, skipped
* Device #14: NVIDIA GeForce RTX 3070, skipped
* Device #15: NVIDIA GeForce RTX 3070, skipped
* Device #16: NVIDIA GeForce RTX 3070, skipped
* Device #17: NVIDIA GeForce RTX 3070, skipped
* Device #18: NVIDIA GeForce RTX 3070, skipped
* Device #19: NVIDIA GeForce RTX 3070, skipped
* Device #20: NVIDIA GeForce RTX 3070, skipped

Benchmark relevant options:
===========================
* --optimized-kernel-enable
* --workload-profile=4

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#1.........: 504.5 kH/s (272.97ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#2.........: 485.8 kH/s (277.88ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#3.........: 508.5 kH/s (278.43ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#4.........: 490.5 kH/s (285.35ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#5.........: 494.2 kH/s (278.85ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#6.........: 505.0 kH/s (277.38ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#7.........: 466.4 kH/s (282.13ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#8.........: 467.5 kH/s (287.33ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#9.........: 518.6 kH/s (282.63ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#10.........: 527.3 kH/s (284.93ms) @ Accel:32 Loops:1024 Thr:512 Vec:1
Speed.#*.........: 4968.3 kH/s
 
N

Ntrain20k

Active member
Feedback: 0 / 0 / 0
Joined
Mar 18, 2020
Messages
459
Reaction score
430
Credits
2,975
6950XT and the iGPU in the 7900X

hashcat (v6.2.5) 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.

hiprtcAddNameExpression is missing from HIPRTC shared library.

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

OpenCL API (OpenCL 2.1 AMD-APP (3444.0)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #1: AMD Radeon(TM) Graphics, 6080/12284 MB (5886 MB allocatable), 1MCU
* Device #2: AMD Radeon RX 6950 XT, 16256/16368 MB (16368 MB allocatable), 40MCU

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

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#1.........: 29321 H/s (69.34ms) @ Accel:256 Loops:128 Thr:256 Vec:1
Speed.#2.........: 1258.7 kH/s (64.49ms) @ Accel:128 Loops:256 Thr:256 Vec:1
Speed.#*.........: 1288.0 kH/s
 
N

Ntrain20k

Active member
Feedback: 0 / 0 / 0
Joined
Mar 18, 2020
Messages
459
Reaction score
430
Credits
2,975
D:\HCBeta>hashcat -b -w 4 -m 22000
hashcat (v6.2.5) starting in benchmark mode

hiprtcAddNameExpression is missing from HIPRTC shared library.

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

ADL2_Overdrive_Caps(): -8

OpenCL API (OpenCL 2.1 AMD-APP (3444.0)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #1: AMD Radeon(TM) Graphics, 6080/12284 MB (5886 MB allocatable), 1MCU
* Device #2: AMD Radeon RX 6950 XT, 16256/16368 MB (16368 MB allocatable), 40MCU

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

-------------------------------------------------------------
* Hash-Mode 22000 (WPA-PBKDF2-PMKID+EAPOL) [Iterations: 4095]
-------------------------------------------------------------

Speed.#1.........: 29288 H/s (277.38ms) @ Accel:1024 Loops:1024 Thr:32 Vec:1
Speed.#2.........: 1281.6 kH/s (253.38ms) @ Accel:128 Loops:1024 Thr:256 Vec:1
Speed.#*.........: 1310.9 kH/s
 
T

thanost

Member
Feedback: 0 / 0 / 0
Joined
Feb 3, 2020
Messages
19
Reaction score
3
Credits
35
Ntrain20k said:
Speed.#1.........: 29288 H/s (277.38ms) @ Accel:1024 Loops:1024 Thr:32 Vec:1
Speed.#2.........: 1281.6 kH/s (253.38ms) @ Accel:128 Loops:1024 Thr:256 Vec:1
Speed.#*.........: 1310.9 kH/s
Click to expand...
Only half the performance of 4090
 
blandyuk

blandyuk

Active member
Trusted
Contributor
VIP Member
Feedback: 0 / 0 / 0
Joined
Jul 6, 2011
Messages
18,600
Reaction score
428
Credits
11,317
I compiled the below, which shows Watt efficiency against WPA speed for numerous cards. I also have 80%, 70% and 60% POWER efficiency speeds. Note 60% is slower, but more efficient. Another note is the GPUs run a lot cooler when reducing the POWER (obviously). Much better way to run jobs I think. More efficient and cooler. I never run at 100% POWER any more. Doing this also extends the lifespan of the GPUs, as less thermal damage due to over-heating.

The WPA speeds are RAW, so not using kH/s. Also, "Eff" = Efficiency, which is: (Speed / Watts) = Eff. The "% orig" is the speed from the %100 original. At 60% power, I get 81.2% of the original speed for my RTX 2080 Ti.

Shows how CRAP the RTX 3000 series cards were. They are beaten by my old RTX 2080 Ti on efficiency.

The GTX 4090 I'm not sure on actually wattage, so it's at 450 watts for now. Even at 600 watts, it scores 4200 efficiency, still beating the others. If you can give me actual GPU speeds and wattage, I'll update but please note, I want RAW speeds, no OVER-CLOCKING!

1676366836188.png
 
Last edited:
N

Ntrain20k

Active member
Feedback: 0 / 0 / 0
Joined
Mar 18, 2020
Messages
459
Reaction score
430
Credits
2,975
blandyuk said:
I compiled the below, which shows Watt efficiency against WPA speed for numerous cards. I also have 80%, 70% and 60% efficiency speeds. Note 60% is slower, but more efficient.

The WPA speeds are RAW, so not using kH/s. Also, "Eff" = Efficiency, which is: (Speed / Watts) = Eff. The "% orig" is the speed from the %100 original. At 60% power, I get 81.2% of the original speed for my RTX 2080 Ti.

Shows how CRAP the RTX 3000 series cards were. They are beaten by my old RTX 2080 Ti on efficiency.

The GTX 4090 I'm not sure on actually wattage, so it's at 450 watts for now. Even at 600 watts, it scores 4200 efficiency, still beating the others. If you can give me actual GPU speeds and wattage, I'll update but please note, I want RAW speeds, no OVER-CLOCKING!

View attachment 29586
Click to expand...
2371 mhz and 325 watts
 
T

thanost

Member
Feedback: 0 / 0 / 0
Joined
Feb 3, 2020
Messages
19
Reaction score
3
Credits
35
Good information, Blandy
Agree hash per Watt is the most important for running costs
All my posts now have to be approved by mods, so I am less inclined to give the figures for my 4090 rig right now
 
T

thanost

Member
Feedback: 0 / 0 / 0
Joined
Feb 3, 2020
Messages
19
Reaction score
3
Credits
35
Ntrain20k said:
2371 mhz and 325 watts
Click to expand...
You have half the hash rate of 4090, so 4090 only needs to consume less than 650W to be more efficient?
Hard to have a conversation here when all posts need to be approved by mods. Tedious.
 
N

Ntrain20k

Active member
Feedback: 0 / 0 / 0
Joined
Mar 18, 2020
Messages
459
Reaction score
430
Credits
2,975
thanost said:
You have half the hash rate of 4090, so 4090 only needs to consume less than 650W to be more efficient?
Hard to have a conversation here when all posts need to be approved by mods. Tedious.
Click to expand...

None of my posts have to be approved. Looking at your post history I can see why they might have you on approval required.....
 
N

Ntrain20k

Active member
Feedback: 0 / 0 / 0
Joined
Mar 18, 2020
Messages
459
Reaction score
430
Credits
2,975
thanost said:
You have half the hash rate of 4090, so 4090 only needs to consume less than 650W to be more efficient?
Hard to have a conversation here when all posts need to be approved by mods. Tedious.
Click to expand...

It's also a generation behind the 4090.
 
blandyuk

blandyuk

Active member
Trusted
Contributor
VIP Member
Feedback: 0 / 0 / 0
Joined
Jul 6, 2011
Messages
18,600
Reaction score
428
Credits
11,317
Anyone got the AMD RX 7900 XT and XTX WPA speeds?

Also updated my GPU speed comparison text above with this. Worth reading!

"Another note is the GPUs run a lot cooler when reducing the POWER (obviously). Much better way to run jobs I think. More efficient and cooler. I never run at 100% POWER any more. Doing this also extends the lifespan of the GPUs, as less thermal damage due to over-heating."
 
You must log in or register to reply here.
Top