hashcat v3.00 - Is this the reasonable speed on Mac?

[ukyang]

Just compiled hashcat v3.00-7-gedb5028 on osx 10.11.5. Here is the result I got from benchmark:

ukyang@MBP [:][/Forfun/hashcat] ./hashcat.app -d 2 -b
hashcat (v3.00-7-gedb5028) starting in benchmark-mode...

OpenCL Platform #1: Apple
=========================
- Device #1: Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz, skipped
- Device #2: GeForce GTX 760, 512/2048 MB allocatable, 6MCU
- Device #3: Iris Pro, skipped

Hashtype: MD5

Speed.Dev.#2.:  6341.0 GH/s (0.05ms)

6431 GH/s, this speed is amazing. But it seems too good to be true. Can anyone tell me if this benchmark is valid or not?

Thank you.

[rico]

6.43 TERA hashes per second. Too good to be true indeed...

Here's my 760 for comparison (slower in v3 than v2.0.1):
cudaHashcat v2.01 starting in benchmark-mode...

Device #1: GeForce GTX 760, 2048MB, 1137Mhz, 6MCU

Hashtype: MD5
Workload: 1024 loops, 256 accel

Speed.GPU.#1.: 2864.4 MH/s


v3.00:
C:\hashcat-3.00>hashcat64.exe -m 0 -b -d 1 -w 3
hashcat (v3.00-1-g67a8d97) starting in benchmark-mode...

OpenCL Platform #1: NVIDIA Corporation
======================================
- Device #1: GeForce GTX 760, 512/2048 MB allocatable, 6MCU

OpenCL Platform #2: Intel(R) Corporation
========================================
- Device #2: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz, skipped

Hashtype: MD5

Speed.Dev.#1.: 1799.5 MH/s (96.00ms)

Started: Thu Jun 30 23:24:57 2016
Stopped: Thu Jun 30 23:25:00 2016

[atom]

This is wrong, for sure, and for sure it's an OpenCL runtime error. We have no influence on this, you need to report to NVidia. Until they release an updated version you can use -D 1 to use your CPU.

[kolokos]

hashcat (v3.00-1-g67a8d97) starting...

OpenCL Platform #1: NVIDIA Corporation
======================================
- Device #1: GeForce GTX 1080, 2048/8192 MB allocatable, 20MCU

OpenCL Platform #2: Intel(R) Corporation
========================================
- Device #2: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz, skipped