Poor performance Mac OS and ATI Radeon HD 5870
#1
I am getting horrible performance with Mac Pro / Radeon 5870 under Mac OS, but on the same machine under Windows everything works fine. Running vBulletin>v3.8.5 hashes with an input mask.

Mac OS Open CL info:

Quote:OpenCL Info:

Platform ID #1
  Vendor  : Apple
  Name    : Apple
  Version : OpenCL 1.2 (Jan  4 2017 22:35:59)

  Device ID #1
    Type           : CPU
    Vendor ID      : 4
    Vendor         : Intel
    Name           : Intel(R) Xeon(R) CPU           W3565  @ 3.20GHz
    Version        : OpenCL 1.2 
    Processor(s)   : 8
    Clock          : 3200
    Memory         : 1536/6144 MB allocatable
    OpenCL Version : OpenCL C 1.2 
    Driver Version : 1.1

  Device ID #2
    Type           : GPU
    Vendor ID      : 2147483648
    Vendor         : AMD
    Name           : ATI Radeon HD 5870
    Version        : OpenCL 1.2 
    Processor(s)   : 20
    Clock          : 850
    Memory         : 256/1024 MB allocatable
    OpenCL Version : OpenCL C 1.2 
    Driver Version : 1.2 (Jan 10 2017 22:25:08)

Mac OS startup:

Quote:OpenCL Platform #1: Apple
=========================
* Device #1: Intel(R) Xeon(R) CPU           W3565  @ 3.20GHz, skipped
* Device #2: ATI Radeon HD 5870, 256/1024 MB allocatable, 20MCU

Hashes: 6 digests; 6 unique digests, 6 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates

Applicable Optimizers:
* Zero-Byte
* Precompute-Init
* Early-Skip
* Not-Iterated
* Brute-Force

Watchdog: Temperature abort trigger disabled
Watchdog: Temperature retain trigger disabled


Sample Mac OS status:

Quote:Session..........: hashcat
Status...........: Running
Hash.Type........: vBulletin > v3.8.5
Hash.Target......: hash_in.txt
Time.Started.....: Wed Mar 29 00:20:23 2017 (1 min, 5 secs)
Time.Estimated...: Fri Aug  4 00:04:44 2017 (127 days, 23 hours)
Input.Mask.......: ?l?l?l?l?l?l?l?d?d?d?d [11]
Input.Queue......: 1/1 (100.00%)
Speed.Dev.#2.....: 43579.1 kH/s (236.91ms)
Recovered........: 0/6 (0.00%) Digests, 0/6 (0.00%) Salts
Recovered/Time...: CUR:0,N/A,N/A AVG:0.00,0.00,0.00 (Min,Hour,Day)
Progress.........: 2818703360/481908610560000 (0.00%)
Rejected.........: 0/2818703360 (0.00%)
Restore.Point....: 0/4569760000 (0.00%)
Candidates.#2....: lpfanan1234 -> xlrumfm1234


Windows Open CL info:

Quote:OpenCL Info:

Platform ID #1
  Vendor  : Advanced Micro Devices, Inc.
  Name    : AMD Accelerated Parallel Processing
  Version : OpenCL 2.0 AMD-APP (1800.11)

  Device ID #1
    Type           : GPU
    Vendor ID      : 1
    Vendor         : Advanced Micro Devices, Inc.
    Name           : Cypress
    Version        : OpenCL 1.2 AMD-APP (1800.11)
    Processor(s)   : 20
    Clock          : 850
    Memory         : 512/1024 MB allocatable
    OpenCL Version : OpenCL C 1.2
    Driver Version : 1800.11 (VM)

  Device ID #2
    Type           : CPU
    Vendor ID      : 128
    Vendor         : GenuineIntel
    Name           : Intel(R) Xeon(R) CPU           W3565  @ 3.20GHz
    Version        : OpenCL 1.2 AMD-APP (1800.11)
    Processor(s)   : 8
    Clock          : 3192
    Memory         : 2047/6131 MB allocatable
    OpenCL Version : OpenCL C 1.2
    Driver Version : 1800.11 (sse2)

Platform ID #2
  Vendor  : Intel(R) Corporation
  Name    : Intel(R) OpenCL
  Version : OpenCL 1.2

  Device ID #3
    Type           : CPU
    Vendor ID      : 8
    Vendor         : Intel(R) Corporation
    Name           : Intel(R) Xeon(R) CPU           W3565  @ 3.20GHz
    Version        : OpenCL 1.2 (Build 25)
    Processor(s)   : 8
    Clock          : 3200
    Memory         : 1532/6131 MB allocatable
    OpenCL Version : OpenCL C 1.2
    Driver Version : 6.4.0.25

Windows startup:

Quote:OpenCL Platform #1: Advanced Micro Devices, Inc.
================================================
* Device #1: Cypress, 512/1024 MB allocatable, 20MCU
* Device #2: Intel(R) Xeon(R) CPU           W3565  @ 3.20GHz, skipped

OpenCL Platform #2: Intel(R) Corporation
========================================
* Device #3: Intel(R) Xeon(R) CPU           W3565  @ 3.20GHz, skipped

Hashes: 6 digests; 6 unique digests, 6 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates

Applicable Optimizers:
* Zero-Byte
* Precompute-Init
* Early-Skip
* Not-Iterated
* Brute-Force

Watchdog: Temperature abort trigger disabled
Watchdog: Temperature retain trigger disabled

Sample Windows status:

Quote:Session..........: hashcat
Status...........: Running
Hash.Type........: vBulletin > v3.8.5
Hash.Target......: hash_in.txt
Time.Started.....: Wed Mar 29 00:37:57 2017 (56 secs)
Time.Estimated...: Tue Apr 04 09:54:48 2017 (6 days, 9 hours)
Input.Mask.......: ?l?l?l?l?l?l?l?d?d?d?d [11]
Input.Queue......: 1/1 (100.00%)
Speed.Dev.#1.....:   873.3 MH/s (381.29ms)
Recovered........: 0/6 (0.00%) Digests, 0/6 (0.00%) Salts
Progress.........: 48087695360/481908610560000 (0.01%)
Rejected.........: 0/48087695360 (0.00%)
Restore.Point....: 0/4569760000 (0.00%)
Candidates.#1....: auranan1234 -> uzasqlw4569


The device order is flipped between Mac OS and Windows, but they each appear to be using the video card properly.

By manually trying different tuning parameters I was able to get the Mac version time estimate down to 30 days, which is still way off the Windows time of 6-7 days.

I ran a Luxmark benchmark on both OS and the Windows one was about 10% faster than the Mac, not many multiple times faster.

At this point, I'm stuck and could use some help. Thanks

edit: forgot to mention Windows 7, and have tried Mac OS 10.10, 10.11, and 10.12
#2
The GPU is ways too old, the drivers are outdated. I'm sure you had to use --force to get it working. There's a reason why we put that flag there. Solution: Trash the GPU, buy a new one.
#3
(04-09-2017, 08:54 AM)atom Wrote: The GPU is ways too old, the drivers are outdated. I'm sure you had to use --force to get it working. There's a reason why we put that flag there. Solution: Trash the GPU, buy a new one.

I don't need to run --force on it except when testing different parameters. It runs perfectly fine on Windows.

But OK.