RipBot264 v1.18.3 - Simple and easy to use GUI -> IPOD . PSP . CONSOLES . BLURAY

[MaistroX]

Hi burt123!


1. No it´s not, it´s a AMD3+ running at originaly @ 4,7Ghz

2. Inside is pretty Clean and case is open all the time.
Also had the idéa of it gettin overheated, but have changed the CPU Fan, Motherboard also tryed differnt ram´s just to figure out what could be wrong, and this only happens with RipBot264 when it runs 100% of all cores on the CPU.


After Writing this Q last night I started a new rip with only 4 cores activated just to see how it worked and adjusted it to 6/8 cores just Before 2-pass started, and PC keeps going without any crash/Freez! :)

http://s17.postimg.org/p0cx9irgf/Rip_Bot264_4_Cores.jpg (http://postimage.org/)

http://s17.postimg.org/ivq09xoqn/Rip_Bot264_2pass.jpg (http://postimage.org/)
image hosting free (http://postimage.org/)


But would be easyer to be able to set the nr. of cores to be used from within RipBot264 otherwise one have to set it manually everytime RipBot264 starts. ;)

Regards MaistroX

A couple of questions for you :-

1st:- is your PC overclocked ???
Ripbot could be considered as a stress test, as it DOES run the CPU @ 100% most of the time.

2nd:- how clean is it inside your case & motherboard etc ??
Sounds like it might be overheating, check & clean the CPU fan(s), and video card(s) fan(s).

I'm pretty sure that Ripbot can't be "throttled".

[Ma]

But would be easyer to be able to set the nr. of cores to be used from within RipBot264 otherwise one have to set it manually everytime RipBot264 starts. ;)

If this crash is related to x265, you can add '--pools 6' option to x265 command line. If it helps, you can try '-F2' option instead of '--pools 6' (if '-F2' helps, you can also try to use x265 compiled by VS 2015 instead of GCC 5.3).

[MaistroX]

But as far as I know I only use the x264 part!

If this crash is related to x265, you can add '--pools 6' option to x265 command line. If it helps, you can try '-F2' option instead of '--pools 6' (if '-F2' helps, you can also try to use x265 compiled by VS 2015 instead of GCC 5.3).

[Ma]

But as far as I know I only use the x264 part!

In x264 there is '--threads X' option. You can try '--threads 6' for example.

[Atak_Snajpera]

@Maistrox
What happens if you run LINX from first post for couple minutes? Check cpu temps during strest test. CPU should be rock stable even at 100%.

[MaistroX]

Just tryed that and the result is here -> http://s11.postimg.org/9gelbqzoj/Lin_X.jpg (http://postimage.org/)
windows screen capture (http://postimage.org/app.php)

Also, just finnished a encode with RipBot264 and set the cores to 6/8, that worked just fine.

Regards MaistroX

@Maistrox
What happens if you run LINX from first post for couple minutes? Check cpu temps during strest test. CPU should be rock stable even at 100%.

[soneca]

nope -> http://forum.doom9.org/showthread.php?p=1699497#post1699497

With the new version of KNLMeansCL, the PC with GTX650 became operational. :)

[Atak_Snajpera]

Just tryed that and the result is here -> http://s11.postimg.org/9gelbqzoj/Lin_X.jpg (http://postimage.org/)
windows screen capture (http://postimage.org/app.php)

Also, just finnished a encode with RipBot264 and set the cores to 6/8, that worked just fine.

Regards MaistroX

Check also with 0.6.5 which uses AVX as well.
https://github.com/sanekgusev/LinX-old/releases/download/0.6.5/LinX-0.6.5.7z

Regarding --threads. By default x264 on your CPU uses 12 threads (8 * 1.5). You may also try --threads 8 instead of 6.

[MaistroX]

I will try 0.6.5 also, just finnished a 5 step test with ALL RAM avalible and biggest Problem Size, see here ->

http://s21.postimg.org/m2esxm6mv/Lin_X_2016_03_21_19_23_53.png (http://postimage.org/)
image hosting over 5mb (http://postimage.org/)
with the following log:
"
Intel(R) LINPACK 64-bit data - LinX 0.6.4

Current date/time: Mon Mar 21 13:35:50 2016

CPU frequency: 4.690 GHz
Number of CPUs: 8
Number of threads: 8

Parameters are set to:

Number of tests : 1
Number of equations to solve (problem size) : 60083
Leading dimension of array : 60088
Number of trials to run : 5
Data alignment value (in Kbytes) : 4

Maximum memory requested that can be used = 3113540512, at the size = 60083

============= Timing linear equation system solver =================

Size LDA Align. Time(s) GFlops Residual Residual(norm)
60083 60088 4 3864.765 37.4164 3.101542e-009 3.065849e-002
60083 60088 4 3892.257 37.1521 3.101542e-009 3.065849e-002
60083 60088 4 3906.159 37.0199 3.101542e-009 3.065849e-002
60083 60088 4 3906.839 37.0135 3.101542e-009 3.065849e-002
60083 60088 4 3924.515 36.8468 3.101542e-009 3.065849e-002

Performance Summary (GFlops)

Size LDA Align. Average Maximal
60083 60088 4 37.0897 37.4164

End of tests
"

Regards MaistroX

Check also with 0.6.5 which uses AVX as well.
https://github.com/sanekgusev/LinX-old/releases/download/0.6.5/LinX-0.6.5.7z

Regarding --threads. By default x264 on your CPU uses 12 threads (8 * 1.5). You may also try --threads 8 instead of 6.

[Ronski]

I've recently been encoding using H265 but am getting much bigger files sizes than expected, with only a small decrease from the original BR rip. I use CQ and CRF set to 16, profile is [High 4.0] FHD. Progressive. Taking a 20.2GB video file (sound tracks removed) H265 ends up at 24.4GB and H264 ends up at 15.7GB using the same settings.

I thought one of the big improvements with H265 was a reduction in file size, am I misunderstanding or do I just have CRF set to low? I watch on a 120" screen so like to keep good quality.

PS. MaistroX, try lots of different stress tests, recently overclocking my X99 rig it would crash on one particular stress test but not another, and I had to increase the voltage slightly, by reducing the core count your reducing the load and thus the voltage won't drop so much. Video transcoding is considered by many as a form of stress testing.

[soneca]

I've never used the x265 but I understand the standard crf values for the two encoders to generate the same visual quality are quite different. ie, you are using a high value.

x264 default = CRF23
x265 default = CRF28

[Wildfire]

Wait for tomorrow. I think your windows is still using cached update.zip file.

Well, it's been a while but still nothing. The OpenCL tab is still empty.

[Ronski]

I've never used the x265 but I understand the standard crf values for the two encoders to generate the same visual quality are quite different. ie, you are using a high value.

x264 default = CRF23
x265 default = CRF28

Thanks, I'll give it a go at 22 and see how that comes out. Edit: Results in a 5.68GB file (no audio yet), just need to play it on the big screen and see how it compares.

Got to say I've just had one of those moments that makes you smile.

I have an old i750 running at 3.6Ghz and my new X99 5820K running at 4.5Ghz. Now I had 1 encoding server on the i750 and two on the X99, I'm now running 6 encoding servers on the X99, each time I added a server the FPS increased, tried adding another on the i750 and it made no difference, but boy it does work well on the X99. Currently cruising at just over 43FPS between them all in H265, gained about another 10FPS by adding the 4 additional servers.

[guest]

Thanks, I'll give it a go at 22 and see how that comes out. Edit: Results in a 5.68GB file (no audio yet), just need to play it on the big screen and see how it compares.

Got to say I've just had one of those moments that makes you smile.

I have an old i750 running at 3.6Ghz and my new X99 5820K running at 4.5Ghz. Now I had 1 encoding server on the i750 and two on the X99, I'm now running 6 encoding servers on the X99, each time I added a server the FPS increased, tried adding another on the i750 and it made no difference, but boy it does work well on the X99. Currently cruising at just over 43FPS between them all in H265, gained about another 10FPS by adding the 4 additional servers.

Yes, I run multiple servers on the one PC, and yes it DOES make a lot of difference.

We need more servers, Atak :)

[soneca]

I do not have conditions to run that way then do my conversions using my old 980X, or 4790K. Lately I have used only the 4790K for this kind of work because it is slightly faster and consumes less power.
Let them work and go to sleep.

[jthekk2]

Atak, is there any way to include a setting to have the first pass handled by one machine for the entire file and then split the file (and stats file) into chunks for distributed encoding? I've been testing with the current distributed encoding but am seeing massive bitrate drops in high action scenes where the scene fell into different chunks. I'm hoping that a first pass pre-chunking would allow for proper bitrate determination relative to the file as a whole before encoding. This will obviously bottleneck the process to the speed of the server doing the first pass, but is more ideal since it yields a better encode in the end.

[Atak_Snajpera]

I think It would be better if you increased chunk size from 1 min to old 10 min value. This method would be much faster than first pass on single machine.

[jthekk2]

I think It would be better if you increased chunk size from 1 min to old 10 min value. This method would be much faster than first pass on single machine.I'll give it a try but the problem would still remain if the action scene falls at the 10 minute mark and is split and the second chunk has an otherwise low level of action (so lower average bitrate). Also, since most shows aren't exactly a multiple of 10, the last chunk will always be smaller and might present the same issue. Unless I'm missing something and there is some error handling for that.

[Atak_Snajpera]

Thanks, I'll give it a go at 22 and see how that comes out. Edit: Results in a 5.68GB file (no audio yet), just need to play it on the big screen and see how it compares.

Got to say I've just had one of those moments that makes you smile.

I have an old i750 running at 3.6Ghz and my new X99 5820K running at 4.5Ghz. Now I had 1 encoding server on the i750 and two on the X99, I'm now running 6 encoding servers on the X99, each time I added a server the FPS increased, tried adding another on the i750 and it made no difference, but boy it does work well on the X99. Currently cruising at just over 43FPS between them all in H265, gained about another 10FPS by adding the 4 additional servers.

It is odd that you have troubles with saturating 6C / 12T CPU.
I have no problems with 8C / 16T CPU even with cropped frame (1920x800).

http://i.cubeupload.com/VxwnRJ.png

[Atak_Snajpera]

I'll give it a try but the problem would still remain if the action scene falls at the 10 minute mark and is split and the second chunk has an otherwise low level of action (so lower average bitrate). Also, since most shows aren't exactly a multiple of 10, the last chunk will always be smaller and might present the same issue. Unless I'm missing something and there is some error handling for that.


Last chunk is always the biggest. Otherwise We could end up with 1 frame chunk ;).
Longer chunk sizes should help in better bitrate distribution (less aggressive)

[Ronski]

Yes it is odd, just trying it the other way round with the X99 as the client and running more than one on it doesn't seem to make much difference. I'll have to test again with the older PC as the client and see what happens. Could it of just been that the FPS happened to ramp up when I tried it as the content of the video changed, thus purely coincidence?

[defalopii]

Hello, RipBot developer, how much You can add distributed server at maximum?
Log time ago ive seen ure using 4 DE, then 8DE, and now 10DE. Is that possible to make it unlimited DE server?
I imagine that i can add more and more DE server, this will speedup encoding, especially for the next codec HEVC

[guest]

Hello, RipBot developer, how much You can add distributed server at maximum?
Log time ago ive seen ure using 4 DE, then 8DE, and now 10DE. Is that possible to make it unlimited DE server?
I imagine that i can add more and more DE server, this will speedup encoding, especially for the next codec HEVC

Hear, hear....................

[gabbett1]

I just recently used Ripbot to convert Insurgent and I'm having an issue. The file I created was a .mp4. When I try and play the movie it takes about 30 seconds for it to begin play, and when it does I only get sound, no video. If I simply rename the file to a .mkv it will play just fine. Rename back to .mp4 and same issue. Takes a while to play and no video. What could be the problem? I could use some help please.

[Atak_Snajpera]

Are you using MPC-HC player?

[omegaeel]

I have been getting Socket Errors on a 2nd PC I use in DE in Ripbot. The log in EncodingServer first says "Socket Error 10060" and after clicking ON within EncodingClient the server logs return "Socket Error 10038". As well as an "Access violation" error.

I tried deleting and re-established the Firewall rules and disabling Windows Firewall all together to no effect.

Using v1.19.3 on both PCs.

https://www.dropbox.com/s/2fmi3do6054v8ib/Screenshot%202016-03-30%2000.43.18.png?dl=0

[dv8r]

gabbett, insurgent has a lot of scrambled titles. There are a couple of long threads about it on makemkv forum. Maybe this is the issue?

[gabbett1]

Are you using MPC-HC player?

I'm using Windows Media Player.

[gabbett1]

gabbett, insurgent has a lot of scrambled titles. There are a couple of long threads about it on makemkv forum. Maybe this is the issue?

I don't think so. I'm using the title that AnyDVD said was good. And like I said, it plays just find when I rename it to a .mkv

[Atak_Snajpera]

I'm using Windows Media Player.

Bad idea. Use something more advanced like MPC-HC.

[Atak_Snajpera]

I have been getting Socket Errors on a 2nd PC I use in DE in Ripbot. The log in EncodingServer first says "Socket Error 10060" and after clicking ON within EncodingClient the server logs return "Socket Error 10038". As well as an "Access violation" error.

I tried deleting and re-established the Firewall rules and disabling Windows Firewall all together to no effect.

Using v1.19.3 on both PCs.

https://www.dropbox.com/s/2fmi3do6054v8ib/Screenshot%202016-03-30%2000.43.18.png?dl=0

Most likely some software on your main pc (where you have EncodingClient running) is blocking connection from EncodingServer. Connection must be established in both ways

http://i.cubeupload.com/N1A0QZ.png

[fatpizza]

Are there any known issues with using OpenCL on integrated intel GPU's? I have a Core i7 6700 and with Intel 530 and whenever I enable OpenCL in the Ripbot options it comes up with this:

C:\>"C:\Program Files\Ripbot\RipBot264v1.19.2\tools\avs2yuv\avs2yuv.exe" "C:\Temp\RipBot264temp\job105\job105.avs" -o - | "C:\Program Files\Ripbot\RipBot264v1.19.2\tools\x264\x264_x64.exe" --opencl --opencl-device 0 --crf 20 --fps 24000/1001 --force-cfr --min-keyint 24 --keyint 240 --frames 197017 --sar 1:1 --level 4.1 --preset slow --tune film --aud --nal-hrd vbr --vbv-bufsize 62500 --vbv-maxrate 62500 --stdin y4m --output "C:\Temp\RipBot264temp\video.264" -
y4m [info]: 1920x804p 1:1 @ 10000000/417083 fps (cfr)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 AVX2 LZCNT BMI2
x264 [info]: OpenCL acceleration enabled with Intel(R) Corporation Intel(R) HD Graphics 530
x264 [info]: profile High, level 4.1
x264 [error]: OpenCL: Invalid value.
x264 [error]: OpenCL: fatal error, aborting encode
x264 [error]: OpenCL: Invalid value.
x264 [error]: OpenCL: fatal error, aborting encode
x264 [error]: OpenCL: Invalid value.
x264 [error]: OpenCL: fatal error, aborting encode
x264 [error]: x264_encoder_encode failed

I tried installing the 2015 vc_redist mentioned a couple of pages ago but with no luck.

[Wildfire]

Are there any known issues with using OpenCL on integrated intel GPU's? I have a Core i7 6700 and with Intel 530 and whenever I enable OpenCL in the Ripbot options it comes up with this:

C:\>"C:\Program Files\Ripbot\RipBot264v1.19.2\tools\avs2yuv\avs2yuv.exe" "C:\Temp\RipBot264temp\job105\job105.avs" -o - | "C:\Program Files\Ripbot\RipBot264v1.19.2\tools\x264\x264_x64.exe" --opencl --opencl-device 0 --crf 20 --fps 24000/1001 --force-cfr --min-keyint 24 --keyint 240 --frames 197017 --sar 1:1 --level 4.1 --preset slow --tune film --aud --nal-hrd vbr --vbv-bufsize 62500 --vbv-maxrate 62500 --stdin y4m --output "C:\Temp\RipBot264temp\video.264" -
y4m [info]: 1920x804p 1:1 @ 10000000/417083 fps (cfr)
x264 [info]: using SAR=1/1
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 AVX2 LZCNT BMI2
x264 [info]: OpenCL acceleration enabled with Intel(R) Corporation Intel(R) HD Graphics 530
x264 [info]: profile High, level 4.1
x264 [error]: OpenCL: Invalid value.
x264 [error]: OpenCL: fatal error, aborting encode
x264 [error]: OpenCL: Invalid value.
x264 [error]: OpenCL: fatal error, aborting encode
x264 [error]: OpenCL: Invalid value.
x264 [error]: OpenCL: fatal error, aborting encode
x264 [error]: x264_encoder_encode failed

I tried installing the 2015 vc_redist mentioned a couple of pages ago but with no luck.

The obvious question here is, of course, are you using the most recent drivers for the Intel 530 GPU?

[Taurus]

@fatpizza:
did you verify your opencl installation with a tool like "GPU Caps Viewer" or similar?
I had a bad time some month ago finding the right drivers at Intel.
The "normal" drivers did not support opencl.
After some fiddling around and try and error, I finally got it working.
and happily living ever after...

[fatpizza]

The obvious question here is, of course, are you using the most recent drivers for the Intel 530 GPU?

Sorry, that should have been the first thing that I mentioned. :P Yes I did download the latest drivers.

@fatpizza:
did you verify your opencl installation with a tool like "GPU Caps Viewer" or similar?
I had a bad time some month ago finding the right drivers at Intel.
The "normal" drivers did not support opencl.
After some fiddling around and try and error, I finally got it working.
and happily living ever after...

I just used the OpenCLInfo.exe and that seemed to think everything was ok. The drivers I installed was just the normal download however in the release notes it says it includes the OpenCL package. I'll try out the Caps Viewer thing now though.

[fatpizza]

OK, I've tried the GPU Caps Viewer and I can run all the OpenCL tests on both CPU and GPU without issue.

[omegaeel]

Most likely some software on your main pc (where you have EncodingClient running) is blocking connection from EncodingServer. Connection must be established in both ways

http://i.cubeupload.com/N1A0QZ.png


Thanks for the advice. I had forgotten I installed VirtualBox a while back. It installed its own virtual adapter, once disabled, everything worked fine.

[rekd0514]

I'm having an issue where when I encode with a locked sized at 23830 to fit on a BD-R disc it no longer will fit correctly. The encoded file size is almost always a little too large to fit on the disc when the job is complete. Ripbot has only been doing this when I upgraded to the newest 1.19.3 version. I believe I was on 1.18.3 before.

Any ideas? Thank you for the help in advance!:cool:

[Atak_Snajpera]

Just reduce target size to 23800 if you get too big file. Maybe newer x264 builds changed something in 2-pass mode.

[rekd0514]

Just reduce target size to 23800 if you get too big file. Maybe newer x264 builds changed something in 2-pass mode.

Ahhh I didn't realize you could just type in the size that you wanted! I thought the locked sizes were the only options. Very cool, thank you!

[gabbett1]

Bad idea. Use something more advanced like MPC-HC.

Currently I use WMP to stream my files throughout my home. Can I do same thing with MPC-HC and if so, will I need to?

Thanks

[chane2k1]

Hello Atak,
I dont know if it's been mentioned before, but it looks like downmixing to stereo on DTS tracks with 3/1 mapping doesn't occur. When I have the original demuxed 3/1 track loaded in Ripbot the only available audio profiles show as being 2.0 with different bitrates, however the resulting encode file comes out as 4 channel and it isn't 3/1 it's 2/2. Flac, AAC, and Opus don't understand 3/1 channel mapping. I can correct this in eac3to by adding the -down16 and -downdpl switches, seems to be a simple bug. The particular movie was Edward Scissorhands and there isn't an alternate english ac3 track to switch to. Thanks for all the hard work you do, much appreciated.

[Ronski]

Currently I use WMP to stream my files throughout my home. Can I do same thing with MPC-HC and if so, will I need to?

Thanks

Use something like Emby, Mediaportal, Kodi, Plex etc.

[gabbett1]

Use something like Emby, Mediaportal, Kodi, Plex etc.

Plex will stream from my PC to my PS3?

[Ronski]

Come on, a simple Google will tell you that!

https://plex.tv/playstation

[fatpizza]

I've noticed something that's a bit worrying when using Ripbot, in that there seems to be some stereo downmixing going on even when choosing the 5.1 ac3 audio profiles. I've compared 5.1 ac3 files created by ripbot to ones created by eac3to and there is a massive difference in audio track separation. So basically as a workaround whenever I'm using surround audio in ripbot I'm just manually replacing what's in the audio encode cmd files with an eac3to script.
Has anyone else noticed this?

[lemaireus]

Atak, is it possible to have some sort of protection in the case of an error in chapter formats? I encode in x265, and it is annoying to have a 12-13 hour encode result only in an error with no output at all. The entire file has to be re-encoded all over again, and the chapter file manually edited to display either the correctly formatted chapters, or no chapters at all.

There seem to be two rather straightforward ways out of this: either RipBot provides an option which instructs the encoder and muxer to go ahead with the job as they normally would, in the case of an error in chapters. The chapter error can be noted in the job log, but the encoding process should finish as it normally would if there was no error, the only difference being that the output video would have no chapter markers.

The second way to is do what is outlined above manually: in case of a chapter error, RipBot shouldn’t terminate the encode and delete the encoded video.x265 file. The user can then manually mux that x265 video with the audio and save 12 or 13 hours of work which otherwise will have to be done all over again.

I get these chapter errors once every 10 to 12 encodes, depending on the source file being encoded, and browsing the forum one can see that there are other users who have reported the same issue. So it might be time to address the problem as, speaking from a layman’s perspective, I don’t think programming-wise it should be too hard to address this issue in the next RipBot version.

[AMiR9!WV]

hey atak , all my jobs getting Errors after they done
i think they got a problem with audio encoder , this error was in logs
"Error: The file 'C:\Temp\RipBot264temp\job1\Encoded_Audio_1.aac' could not be opened for reading: open file error."
http://8pic.ir/images/7qb6xz0xmeu2bx77far9.png

[slalom]

Atak, is it possible to have some sort of protection in the case of an error in chapter formats? I encode in x265, and it is annoying to have a 12-13 hour encode result only in an error with no output at all. The entire file has to be re-encoded all over again, and the chapter file manually edited to display either the correctly formatted chapters, or no chapters at all.

There seem to be two rather straightforward ways out of this: either RipBot provides an option which instructs the encoder and muxer to go ahead with the job as they normally would, in the case of an error in chapters. The chapter error can be noted in the job log, but the encoding process should finish as it normally would if there was no error, the only difference being that the output video would have no chapter markers.

The second way to is do what is outlined above manually: in case of a chapter error, RipBot shouldn’t terminate the encode and delete the encoded video.x265 file. The user can then manually mux that x265 video with the audio and save 12 or 13 hours of work which otherwise will have to be done all over again.

I get these chapter errors once every 10 to 12 encodes, depending on the source file being encoded, and browsing the forum one can see that there are other users who have reported the same issue. So it might be time to address the problem as, speaking from a layman’s perspective, I don’t think programming-wise it should be too hard to address this issue in the next RipBot version.
YOU can do that manually
Run CombineAllChunks.cmd in Chunks folder to get your video

Then run mkvmerge with that one and the original video to get your result

[Atak_Snajpera]

@lemaireus
Can you show me how corrupted chapter file looks like?

@Amir
What happens if you run ENCODE_AUDIO_1.cmd in console?