Deep Dreamer Crashing



  • Hello, i have Deep Dreamer on an iMac 14,2 which has 10.11.2. It crashes on any setting I can offer it. I have tried changing from the OSX driver to the NVIDIA Web Driver and have the same problem. here are some errors from console when it happens.

    12/9/15 9:58:47.478 PM Deep Dreamer[1154]: [CL_INVALID_VALUE] : OpenCL Error : Failed to retrieve device information! Insufficient memory to hold value! Need 24 bytes.
    12/9/15 9:58:50.452 PM Deep Dreamer[1154]: endSheet:returnCode: requires a non-nil sheet
    12/9/15 9:59:02.780 PM Deep Dreamer[1154]: [CL_INVALID_VALUE] : OpenCL Error : Failed to retrieve device information! Insufficient memory to hold value! Need 24 bytes.
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: GPU error detected!
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: Calling gpusKillClient for task <ptr>
    12/9/15 9:59:02.000 PM kernel[0]: NVDA(Compute): Channel exception! Exception type = 0x1f Access Violation Error (MMU Error 2)
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: Channel Info: [23, 0x16, 0x8, 0x4b6]
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: Version Info: [com.nvidia.web.GeForceWeb, 10.0.9, 0x5e5f4c655b, 20192841, 346.03.04f01, 1]
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: MMU Error: FAULT_PTE at 0x103e9000
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: Resource Manager Info:
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 4443564e 000000ac 569777cd 3d0cea67 00000001 00000014 d3793533 46d3a4a6
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 4614f297 e71edccf 00088301 00000075 149a030a 820c0a00 01180913 00482050
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 0a010190 3a13923d 0024380a 000e0000 e0010000 15490000 00010000 12490000
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 00000000 04470000 00ff0001 00ff0000 00ff0000 00ff0000 00ff0000 00ff0000
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 210a0000 081e13c2 18001000 f7e080a8 2816200b 3803301f 9ff05000 08d3da96
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 48028040 00000004 4443564e
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: Accelerator Event History:
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 0a0808001a04080010010a0808001a04080310010a2b0800122708bc80021080c89bba
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 84f0ffffff0118c182c08efc82800b20e3a38080f0bc0428b789808080010a23080012
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 1f08c080021080c89bba84f0ffffff0118c182c08efc82800b208180800828000a0808
    12/9/15 9:59:02.000 PM kernel[0]: NVDA: 001a04080310000a0808001a0408001000
    12/9/15 9:59:02.838 PM Deep Dreamer[1154]: Handling crash with signal 6...
    12/9/15 9:59:02.846 PM DumpGPURestart[1182]: Saved gpuRestart report for kernel to /Library/Logs/DiagnosticReports/Kernel_2015-12-09-215902_Big-Mac.gpuRestart
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: GPU error detected!
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: Calling gpusKillClient for task <ptr>
    12/9/15 9:59:03.000 PM kernel[0]: NVDA(Compute): Channel exception! Exception type = 0x45 Graphics Engine Error (GR Class Error)
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: Channel Info: [23, 0x16, 0x8, 0x4b7]
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: Version Info: [com.nvidia.web.GeForceWeb, 10.0.9, 0x5e5f4c655b, 20192841, 346.03.04f01, 1]
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: Resource Manager Info:
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 4443564e 000000cc ac867988 37a3479f 00000001 00000014 d3793533 46d3a4a6
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 4614f297 e71edccf 00088301 00000096 12f21f0a 081a0a1c 18201000 28002000
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 38003000 80400280 50004802 60165800 9a030a00 0c0a0014 18091382 48205001
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 01019000 13923d0a 24380a3a 0e000000 01000000 490000e0 01000015 49000000
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 00000012 47000000 ff000104 ff000000 ff000000 ff000000 ff000000 ff000000
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 0a000000 1e13c221 00100008 e080a818 16200bf7 03304528 a8500038 d4cbf092
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 02804008 00000548 4443564e
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: Accelerator Event History:
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 0a0808001a04080010010a0808001a04080310010a290800122508bc80021080c89bba
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 84f0ffffff0118c182c08efc82800b20818080805028b889808080010a230800121f08
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: c080021080c89bba84f0ffffff0118c182c08efc82800b208182800828000a0808001a
    12/9/15 9:59:03.000 PM kernel[0]: NVDA: 04080310000a0808001a0408001000


  • administrators

    Thanks for reporting the problem with the error info. We have seen that Deep Dreamer can bump into a driver problem on NVidia graphics cards which is usually solved by moving from GPU to CPU mode. Can I confirm that CPU mode is also crashing as well?

    Thanks

    Mike



  • On CPU, it doesn't crash, but I get a different error instead. This same dialog appears with both CPU Normal and CPU Fast.

    Screen Shot 2015-12-10 at 7.18.59 AM.png
    12/10/15 7:21:20.787 AM Deep Dreamer[1942]: [CL_INVALID_VALUE] : OpenCL Error : Failed to retrieve device information! Insufficient memory to hold value! Need 24 bytes.
    12/10/15 7:21:23.469 AM Deep Dreamer[1942]: endSheet:returnCode: requires a non-nil sheet
    12/10/15 7:21:34.877 AM Deep Dreamer[1942]: CoreAnimation: warning, deleted thread with uncommitted CATransaction; set CA_DEBUG_TRANSACTIONS=1 in environment to log backtraces.


  • administrators

    Got it. Getting that message indicates that for some reason the Neural Net is becoming corrupt. This is something we are looking into and I'll get back to you ASAP.

    Regards

    Mike



  • Thanks. I'm looking forward to it.



  • Is there further progress on this?

    I have the same problem.


Log in to reply
 

Looks like your connection to Forum was lost, please wait while we try to reconnect.