+- Selur's Little Message Board (https://forum.selur.net)
+-- Forum: Hybrid - Support (https://forum.selur.net/forum-1.html)
+--- Forum: Problems & Questions (https://forum.selur.net/forum-3.html)
+--- Thread: [HELP] Jobs just stops and the next one needs to be resumed manually. (/thread-2833.html)
Pages:
1
2
Jobs just stops and the next one needs to be resumed manually. - SkyNET - 18.07.2022
Hello!
Been using this wornderful tool for quite sometime for single encodes and it has been wonderfull.
Thank you so much, Selur!
Info:
Hybrid rev 2022.03.20.1
Windows 10 Education x64 21H2 19044.1826
Lately, I been doing batch encodes, but, I have run into a bit of a problem.
After each job in the queue, the queue would just stop.
With the current "finished" but not deleting a file, thus, not reaching 100%. (Only up to 83 and something %)
The final encode is even muxed, but, the job is stuck.
I have to manually clean (using the "broom" icon) for it to continue to the next job.
What I did:
Set the Output path.
Add the files to encode: ProRes 4:2:2 Mov files
Set the encode parameters: x264. VapourSynth Color Tweak.
Start the encode
What I have already tried:
Reinstalling Hybrid.
Reseting to defaults.
Deleting the Hybrid forlder in Roaming to start clean.
I am right now waiting for an encode so I can screen shot what I mean and post a proper debug log.
Thank you!
RE: Jobs just stops and the next one needs to be resumed manually. - Selur - 18.07.2022
I'll send you a link to my current dev version.
Might be that I fixed the problem already, if not create a debug output of the problem (analysis, job creation and processing) so I can see in details what is happening and why Hybrid simply stops.
Cu Selur
RE: Jobs just stops and the next one needs to be resumed manually. - SkyNET - 18.07.2022
Hi Selur!
I was waiting for the encode to end so the debug log is generated for the current version I am using.
See attached files.
And sure, I'd like to try the Dev build.
RE: Jobs just stops and the next one needs to be resumed manually. - i990049 - 18.07.2022
It sounds like my situation.shanaencoder and xmedia recode works fine with batch mod.
https://forum.selur.net/thread-2474-page-2.html
rename to 06.mkv 07.mkv 08.mkv 09.mkv still the same
RE: Jobs just stops and the next one needs to be resumed manually. - Selur - 18.07.2022
@SkyNet: I already send you a link to the dev version when I last posted, you should check your PMs.
@i990049: Since shanaencoder and xmedia recode work so wonderful for you, why use Hybrid?
Also I would have written more to your problem if I could, but with the infos you provided I have no clue why Hybrid is not allowed to access the files. I'm closing your thread now since you keep nagging and I can't help.
RE: Jobs just stops and the next one needs to be resumed manually. - SkyNET - 18.07.2022
(18.07.2022, 18:29)Selur Wrote: @SkyNet: I already send you a link to the dev version when I last posted, you should check your PMs.
Yes, I have installed it and its doing the same. I uploaded that log since I already had it.
Here is a new log with the dev version.
RE: Jobs just stops and the next one needs to be resumed manually. - Selur - 18.07.2022
Okay, problem seems to be that Hybrid expects:
"C:\Users\SkyNET\AppData\Local\Temp\x264_lookahead_2022-07-18@08_07_31_6932_02.clbin"x264 --preset veryslow --pass 2 --bitrate 8000 --profile high422 --level 4.1 --ref 3 --sync-lookahead 18 --psy-rd 1.00:0.15 --vbv-maxrate 50000 --vbv-bufsize 187500 --deadzone-inter 6 --deadzone-intra 6 --sar 1:1 --deblock -1:-1 --non-deterministic --range tv --colorprim bt709 --transfer bt709 --colormatrix bt709 --stats "C:\Users\SkyNET\AppData\Local\Temp\E06 - Chemistry_1920x1080_2022-07-18@08_07_31_6932_02.stats" --demuxer y4m --input-range tv --fps 30/1 --output-csp i422 --opencl --opencl-device 1 --opencl-clbin "C:\Users\SkyNET\AppData\Local\Temp\x264_lookahead_2022-07-18@08_07_31_6932_02.clbin" --output-depth 10 --output "C:\Users\SkyNET\AppData\Local\Temp\2022-07-18@08_07_31_6932_04.264" -No job creation etc.
The surprising thing is that, I would have expected the second pass to fail if the clbin file did not exist.
(As a side note: using opencl there is usually not adviced, since there are only a few rare cases where it actually has a positive effect.)
Okay, I found the issue:
x264 [warning]: OpenCL: not compiled with OpenCL support, disablingHybrid checks the output of "x264 --fullhelp" and only enables opencl if the output contains:
--opencl Enable use of OpenCL
--opencl-clbin <string> Specify path of compiled OpenCL kernel cache
--opencl-device <integer> Specify OpenCL device ordinalOkay, I just tested with the same files from the last dev and here the opencl support works fine on my system,x264 isn't complaining here.
So I suspect taht you either exchanged the x264.exe or that the x264.exe looks for something in the system it is used on.
Tried dependency walker (https://www.dependencywalker.com/) on the .exe which comes with Hybrid, but there are no dependencies:
![[Image: dependencies.png]](https://i.ibb.co/S6SFFY4/dependencies.png)
which shouldn't be on a normal Windows version.
If you didn't exchange the binary, this might be related to the video drivers on your system.
I attached a zip with a OpenCLinfo.exe, if you execute that exe inside a windows command prompt it should report whether opencl support is available on your system.
I get:
C:\Users\Selur\Desktop>OpenCLinfo.exe
Platform 0.
Name : NVIDIA CUDA
Vendor : NVIDIA Corporation
Version : OpenCL 3.0 CUDA 11.6.127
Profile : FULL_PROFILE
Extensions : cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_device_uuid cl_khr_pci_bus_info cl_khr_external_semaphore cl_khr_external_memory cl_khr_external_semaphore_win32 cl_khr_external_memory_win32
0.0 Device name : NVIDIA GeForce GTX 1070 Ti
Hardware version : OpenCL 3.0 CUDA
Software version : 512.15
OpenCL C version : OpenCL C 1.2
Compute units : 19Cu Selur
Ps.: Not sure whether I should rewrite forcefully abort if "OpenCL: not compiled with OpenCL support, disabling" appears in the logs,..
RE: Jobs just stops and the next one needs to be resumed manually. - SkyNET - 18.07.2022
Alright.
First of, I am so sorry, when I installed the dev build I resumed the job I had made instead of doing it from scracht.
I will attach a new log with a full job creation, encode, etc. as soon as the encode finishes.
About, x264.exe is the same as supplied, never changed the binary of it.
Here is the output for OpenCL
Platform 0.
Name : AMD Accelerated Parallel Processing
Vendor : Advanced Micro Devices, Inc.
Version : OpenCL 2.2 AMD-APP (3417.0)
Profile : FULL_PROFILE
Extensions : cl_khr_icd cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_amd_event_callback cl_amd_offline_devices
Platform 1.
Name : Intel(R) OpenCL HD Graphics
Vendor : Intel(R) Corporation
Version : OpenCL 3.0
Profile : FULL_PROFILE
Extensions : cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_intel_command_queue_families cl_intel_subgroups cl_intel_required_subgroup_size cl_intel_subgroups_short cl_khr_spir cl_intel_accelerator cl_intel_driver_diagnostics cl_khr_priority_hints cl_khr_throttle_hints cl_khr_create_command_queue cl_intel_subgroups_char cl_intel_subgroups_long cl_khr_il_program cl_intel_mem_force_host_memory cl_khr_subgroup_extended_types cl_khr_subgroup_non_uniform_vote cl_khr_subgroup_ballot cl_khr_subgroup_non_uniform_arithmetic cl_khr_subgroup_shuffle cl_khr_subgroup_shuffle_relative cl_khr_subgroup_clustered_reduce cl_intel_device_attribute_query cl_khr_suggested_local_work_size cl_intel_spirv_media_block_io cl_intel_spirv_subgroups cl_khr_spirv_no_integer_wrap_decoration cl_intel_unified_shared_memory cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_intel_planar_yuv cl_intel_packed_yuv cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_3d_image_writes cl_intel_media_block_io cl_khr_gl_sharing cl_khr_gl_depth_images cl_khr_gl_event cl_khr_gl_msaa_sharing cl_intel_dx9_media_sharing cl_khr_dx9_media_sharing cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_intel_d3d11_nv12_media_sharing cl_intel_sharing_format_query cl_khr_pci_bus_info cl_intel_subgroup_local_block_io cl_intel_split_work_group_barrier cl_intel_simultaneous_sharing
0.0 Device name : gfx1010:xnack-
Hardware version : OpenCL 1.2 AMD-APP (3417.0)
Software version : 3417.0 (PAL,LC)
OpenCL C version : OpenCL C 1.2
Compute units : 18
1.0 Device name : Intel(R) UHD Graphics 750
Hardware version : OpenCL 3.0 NEO
Software version : 30.0.101.1994
OpenCL C version : OpenCL C 1.2
Compute units : 32As you can see, both gpus support OpenCL.
The first card is a Radeon RX 5600XT
RE: Jobs just stops and the next one needs to be resumed manually. - Selur - 18.07.2022
Quote:About, x264.exe is the same as supplied, never changed the binary of it.Okay, so it's probably some driver issue.
-> Try whether it works if you change the device from 1 to 0.
Cu Selur
RE: Jobs just stops and the next one needs to be resumed manually. - SkyNET - 18.07.2022
(18.07.2022, 20:09)Selur Wrote:Quote:About, x264.exe is the same as supplied, never changed the binary of it.Okay, so it's probably some driver issue.
-> Try whether it works if you change the device from 1 to 0.
Cu Selur
I will just disable OpenCL altogheter.
It's more headache than anything.
And, I agree with you, it would be smart to implement a forcefully abort if "OpenCL: not compiled with OpenCL support, disabling" appears in the logs.
Thank you so much for figuring what I was doing wrong. In the end, Hybrid was working fine!
Edit:
100% confirmed, it was the OpenCL setting not creating the file, thus, halting the queue.
Selur, thank you so much for your time and patience.
Best regards,
SkyNET