This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

Hybrid 2018.05.22.1 - constant crashes using x265
#1
Bug 
Hello

After installing Hybrid 2018.05.22.1 on Windows 7 x64 I can't encode anything using x265. Job cue always crashes upon reaching
x265.exe. There are no specific steps to reproduce this bug as it crashes on normal recode not after some option change.
Reply
#2
Looking into it, seems like something is wrong with x265, since '--max-ausize-factor' but it should be:
https://x265.readthedocs.io/en/latest/cl...ize-factor

The x265 version which comes with Hybrid has these options:
...
Rate control, Adaptive Quantization:
   --bitrate <integer>           Target bitrate (kbps) for ABR (implied). Default 0
-q/--qp <integer>                QP for P slices in CQP mode (implied). --ipratio and --pbration determine other slice QPs
   --crf <float>                 Quality-based VBR (0-51). Default 28.0
   --[no-]lossless               Enable lossless: bypass transform, quant and loop filters globally. Default disabled
   --crf-max <float>             With CRF+VBV, limit RF to this value. Default 0.000000
                                 May cause VBV underflows!
   --crf-min <float>             With CRF+VBV, limit RF to this value. Default 0.000000
                                 this specifies a minimum rate factor value for encode!
   --vbv-maxrate <integer>       Max local bitrate (kbit/s). Default 0
   --vbv-bufsize <integer>       Set size of the VBV buffer (kbit). Default 0
   --vbv-init <float>            Initial VBV buffer occupancy (fraction of bufsize or in kbits). Default 0.90
   --vbv-end <float>             Final VBV buffer emptiness (fraction of bufsize or in kbits). Default 0 (disabled)
   --vbv-end-fr-adj <float>      Frame from which qp has to be adjusted to achieve final decode buffer emptiness. Default 0
   --pass                        Multi pass rate control.
                                   - 1 : First pass, creates stats file
                                   - 2 : Last pass, does not overwrite stats file
                                   - 3 : Nth pass, overwrites stats file
   --[no-]multi-pass-opt-analysis   Refine analysis in 2 pass based on analysis information from pass 1
   --[no-]multi-pass-opt-distortion Use distortion of CTU from pass 1 to refine qp in 2 pass
   --stats                       Filename for stats file in multipass pass rate control. Default x265_2pass.log
   --[no-]analyze-src-pics       Motion estimation uses source frame planes. Default disable
   --[no-]slow-firstpass         Enable a slow first pass in a multipass rate control mode. Default enabled
   --[no-]strict-cbr             Enable stricter conditions and tolerance for bitrate deviations in CBR mode. Default disabled
   --analysis-save <filename>    Dump analysis info into the specified file. Default Disabled
   --analysis-load <filename>    Load analysis buffers from the file specified. Default Disabled
   --analysis-reuse-file <filename>    Specify file name used for either dumping or reading analysis data. Deault x265_analysis.dat
   --analysis-reuse-level <1..10>      Level of analysis reuse indicates amount of info stored/reused in save/load mode, 1:least..10:most. Default 5
   --refine-mv-type <string>     Reuse MV information received through API call. Supported option is avc. Default disabled - 0
   --scale-factor <int>          Specify factor by which input video is scaled down for analysis save mode. Default 0
   --refine-intra <0..4>         Enable intra refinement for encode that uses analysis-load.
                                    - 0 : Forces both mode and depth from the save encode.
                                    - 1 : Functionality of (0) + evaluate all intra modes at min-cu-size's depth when current depth is one smaller than min-cu-size's depth.
                                    - 2 : Functionality of (1) + irrespective of size evaluate all angular modes when the save encode decides the best mode as angular.
                                    - 3 : Functionality of (1) + irrespective of size evaluate all intra modes.
                                    - 4 : Re-evaluate all intra blocks, does not reuse data from save encode.
                                Default:0
   --refine-inter <0..3>         Enable inter refinement for encode that uses analysis-load.
                                    - 0 : Forces both mode and depth from the save encode.
                                    - 1 : Functionality of (0) + evaluate all inter modes at min-cu-size's depth when current depth is one smaller than
                                          min-cu-size's depth. When save encode decides the current block as skip(for all sizes) evaluate skip/merge.
                                    - 2 : Functionality of (1) + irrespective of size restrict the modes evaluated when specific modes are decided as the best mode by the save encode.
                                    - 3 : Functionality of (1) + irrespective of size evaluate all inter modes.
                                Default:0
   --[no-]dynamic-refine         Dynamically changes refine-inter level for each CU. Default disabled
   --[no-]refine-mv              Enable mv refinement for load mode. Default disabled
   --aq-mode <integer>           Mode for Adaptive Quantization - 0:none 1:uniform AQ 2:auto variance 3:auto variance with bias to dark scenes. Default 1
   --aq-strength <float>         Reduces blocking and blurring in flat and textured areas (0 to 3.0). Default 1.00
   --[no-]aq-motion              Adaptive Quantization based on the relative motion of each CU w.r.t., frame. Default disabled
   --qg-size <int>               Specifies the size of the quantization group (64, 32, 16, 8). Default 32
   --[no-]cutree                 Enable cutree for Adaptive Quantization. Default enabled
   --[no-]rc-grain               Enable ratecontrol mode to handle grains specifically. turned on with tune grain. Default disabled
   --ipratio <float>             QP factor between I and P. Default 1.40
   --pbratio <float>             QP factor between P and B. Default 1.30
   --qcomp <float>               Weight given to predicted complexity. Default 0.60
   --qpstep <integer>            The maximum single adjustment in QP allowed to rate control. Default 4
   --qpmin <integer>             sets a hard lower limit on QP allowed to ratecontrol. Default 0
   --qpmax <integer>             sets a hard upper limit on QP allowed to ratecontrol. Default 69
   --[no-]const-vbv              Enable consistent vbv. turned on with tune grain. Default disabled
   --cbqpoffs <integer>          Chroma Cb QP Offset [-12..12]. Default 0
   --crqpoffs <integer>          Chroma Cr QP Offset [-12..12]. Default 0
   --scaling-list <string>       Specify a file containing HM style quant scaling lists or 'default' or 'off'. Default: off
   --zones <zone0>/<zone1>/...   Tweak the bitrate of regions of the video
                                 Each zone is of the form
                                   <start frame>,<end frame>,<option>
                                   where <option> is either
                                       q=<integer> (force QP)
                                   or  b=<float> (bitrate multiplier)
   --lambda-file <string>        Specify a file containing replacement values for the lambda tables
                                 MAX_MAX_QP+1 floats for lambda table, then again for lambda2 table
                                 Blank lines and lines starting with hash(#) are ignored
                                 Comma is considered to be white-space
   --max-ausize-factor <float>   This value controls the maximum AU size defined in specification.
                                 It represents the percentage of maximum AU size used. Default 1.0
....
But the version you used, doesn't.
The current version with Hyrid should be '2.8+1-478bfe2b7673', but the version you used is '2.6+22-ff02513b92c0'.
-> Did you replace the x265 version that came with Hybrid with an older one?

Cu Selur
Reply
#3
(24.05.2018, 11:50)Selur Wrote: But the version you used, doesn't.
The current version with Hyrid should be '2.8+1-478bfe2b7673', but the version you used is '2.6+22-ff02513b92c0'.
-> Did you replace the x265 version that came with Hybrid with an older one?

Yes I did replaced it but it was few weeks ago, not after 2018.05.22.1 install.
Just reinstalled it again and it still crashes again. I've replaced x265 binary with the version you mentioned above
and it's OK now. Thanks again for help, have a nice encoding day :-)
Reply
#4
I also send you my current dev version via PM. Does the problem also happen with that version?

Cu Selur
Reply
#5
(24.05.2018, 12:12)Selur Wrote: I also send you my current dev version via PM. Does the problem also happen with that version?

Cu Selur

Yes, the dev version crashes as well and I think I know why. Just found a backup of old version 2017.12.27.0 unpacked it and there is a difference in folder
structure vs latest or the dev version. There are no /32bit or /64bit folders on my backup version. I've always replaced x265.exe in root directory and it worked but this time after
installing latest or dev version x265.exe in root folder is not updated. There is 2.8...... version of 265.exe in 64bit folder. I don't know if I'm correct but it looks like my root directory is
very similar to /64bit or /32bit folder. Is it because my install is old and directory structure is now changed and old redundant files in root folder haven't been deleted and are not nedded?  BTW is it safe to delete /32bit folder on 64 bit machine?
Reply
#6
Hybrid changed the folder structure and the installer at the end of 2017.
Which is why it was required then to uninstall Hybrid completely with all profiles, settings, etc.
Aside from Hybrid itself no executable should be inside the Hybrid folder itself.
-> Uninstall Hybrid completely with all it's settings. Make sure your Hybrid folder is deleted, in case it wasn't delete it.
Install the dev version I send you again.

Cu Selur
Reply
#7
Full uninstall compleat. Directory structure is now fixed, all .exe files are pointed to /64bits and i see few minor differences in user interface.
No problems with encoding. Re-installation saved me about ~3 gigs of space on HDD. Only down side is full reset of Hybrid settings.
I wonder how long it will take me to get them to the state before reinstall. Anyway thanks again for help.
Reply
#8
Opss... I did it again, I played with the Hybrid and it crashed again.
My current version is 2018.05.25.1. It looks that it crashes when I try to encode when input is set to "DVD Input"
Reply
#9
I see the problem. Issue is that '-vf zscale=rangein=limited:range=limited' is included two tiles inside the ffmpeg call.
-> will try to fix later today

Cu Selur
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)