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

[Ryushin]

Originally Posted by Atak_Snajpera

Works fine for your recently uploaded sample. Chunks are correctly stitched.

VC-1, dammed if you do, damned if you don't.
Looks like we've come back full circle:
https://forum.doom9.org/showpost.php...ostcount=16631

Now I have a good test sample showing the decoder corruption/pixelation.
https://cloud.chrisdos.com/index.php/s/TkXFnas6yMqtnE4

Download the Information file and the Wonder_Woman*. One is the sample and the other is the original source trimmed to 5 minutes.

Maybe now that we have two good test cases, perhaps we should open an issue on github:
https://github.com/FFMS/ffms2/issues

I can open the issue if you like, but I think you have a better understanding of the versions between the years and what is happening.

[sneaker_ger]

So what exactly did you use to decode Wonder Woman? With distributed encoding/chunks? Because indeed I see lots of corruption but all over the place, not limited to exactly 01m:00s, 02m:00s etc. No corruption when decoded using latest ffmpeg. Are you using an old version of ffms2 again?

[tkwou]

Hi Atak, I'm having the same problems with the distributed encoding. I'm running a single NIC on my computer, but because I use Virtualbox and OpenVPN, there are two virtual NIC's created. I have to be very careful to remember to disable both virtual NIC's before I start using the distributed encoding or else it picks up the wrong NIC interface and none of the encoding servers can connect. This occurred just yesterday as I had been connected to work with OpenVPN. When I started running an encode, none of the other encoding servers were connecting. When I checked the other computers I realized they were trying to connect through the OpenVPN NIC address. Once I shutdown OpenVPN and restarted encoding client everything was ok again.

[Atak_Snajpera]

Originally Posted by tkwou

Hi Atak, I'm having the same problems with the distributed encoding. I'm running a single NIC on my computer, but because I use Virtualbox and OpenVPN, there are two virtual NIC's created. I have to be very careful to remember to disable both virtual NIC's before I start using the distributed encoding or else it picks up the wrong NIC interface and none of the encoding servers can connect. This occurred just yesterday as I had been connected to work with OpenVPN. When I started running an encode, none of the other encoding servers were connecting. When I checked the other computers I realized they were trying to connect through the OpenVPN NIC address. Once I shutdown OpenVPN and restarted encoding client everything was ok again.

You can manually specify binded ip in EncodingServer.exe
example
Just run Server with this switch

EncodingServer.exe /IP xxx.xxx.xxx.xxx

You should also do the same in Distributed Encoding tab for EncodingClient.exe

[Atak_Snajpera]

Originally Posted by Ryushin

VC-1, dammed if you do, damned if you don't.
Looks like we've come back full circle:
https://forum.doom9.org/showpost.php...ostcount=16631

Now I have a good test sample showing the decoder corruption/pixelation.
https://cloud.chrisdos.com/index.php/s/TkXFnas6yMqtnE4

Download the Information file and the Wonder_Woman*. One is the sample and the other is the original source trimmed to 5 minutes.

Maybe now that we have two good test cases, perhaps we should open an issue on github:
https://github.com/FFMS/ffms2/issues

I can open the issue if you like, but I think you have a better understanding of the versions between the years and what is happening.

I doubt we will ever have solution for vc-1
Here is how whole vc1 issue looks like.

https://github.com/FFMS/ffms2/releases

• FFMS 2.20 from june 2014 is the last version which is frame accurate and returns correct key frames in text file (video.mkv.ffindex_track00.kf.txt)
  example
  

  # keyframe format v1
  fps 0
  0
  24
  48
  72
  96
  120
  144

  However this version may give you some random macro-blocks in frames like in this example
  
  
  
  We can just hope that this only occurs in animations and not in real film footage...
  
  
• The macroblocking is fixed in latest version of FFMS. Unfortunately the bad news is that anything above FFMS 2.20 is no longer frame accurate and each frame is detected as key-frame(!?).
  This means that in DE mode chunk may start from incorrect frame and on top of that it can be totally corrupted. (JESSE_JAMES sample)
  

  # keyframe format v1
  fps 0
  0
  1
  2
  3
  4
  5
  6

  
  
  

[Ryushin]

Originally Posted by sneaker_ger

So what exactly did you use to decode Wonder Woman? With distributed encoding/chunks? Because indeed I see lots of corruption but all over the place, not limited to exactly 01m:00s, 02m:00s etc. No corruption when decoded using latest ffmpeg. Are you using an old version of ffms2 again?

There is macroblocking all over the place. The minute marks were issues with DE but that issue does not exist with the latest version (though I did see occasional corruption with the Jessie James source).

The latest version of ffmpeg is indeed fixed. Same can be said about the latest version of handbrake (or at least the beta which is what I used).

[Ryushin]

Originally Posted by Atak_Snajpera

I doubt we will ever have solution for vc-1

I'll open on issue on the Git page.

Perhaps the idea to add the extra step to encode vc-1 source to perhaps cq10 (or a user configurable option) AVC using ffmpeg after the demuxing step that will then allow ripbot to correctly process everything.

[Ryushin]

Originally Posted by Ryushin

I'll open on issue on the Git page.

Issue opened: https://github.com/FFMS/ffms2/issues/345

[tkwou]

Thanks Atak, my OpenVPN and VirtualBox are not always running so when I go into Settings, the adapter option was greyed out. Once I turned them both on, it gave me to option to lock to a specific NIC. Works fine now.

[byteshare]

Originally Posted by Viper714

Okay, will do. I have a feeling that it is going to take them a while, but I will definitely contact them. I agree with your view. Thanks!!!

I would recommend that you try using Logitech G Hub instead since they're moving to that.
I use it and haven't had any issues with my servers being blocked that I know of.

[soneca]

@Atak

Could you please help with this Media Player Classic error?

E:\Temp\RipBot264temp\job2\job2_preview.avs::Avisynth video #1

Media Type 0:
--------------------------
Video: YV12 3840x2160 29.97fps

AM_MEDIA_TYPE: 
majortype: MEDIATYPE_Video {73646976-0000-0010-8000-00AA00389B71}
subtype: MEDIASUBTYPE_YV12 {32315659-0000-0010-8000-00AA00389B71}
formattype: FORMAT_VideoInfo {05589F80-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 12441600
cbFormat: 88

VIDEOINFOHEADER:
rcSource: (0,0)-(0,0)
rcTarget: (0,0)-(0,0)
dwBitRate: 0
dwBitErrorRate: 0
AvgTimePerFrame: 333667

BITMAPINFOHEADER:
biSize: 40
biWidth: 3840
biHeight: 2160
biPlanes: 1
biBitCount: 12
biCompression: YV12
biSizeImage: 12441600
biXPelsPerMeter: 0
biYPelsPerMeter: 0
biYPelsPerMeter: 0
biClrUsed: 0
biClrImportant: 0

pbFormat:
0000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0020: 00 00 00 00 00 00 00 00 63 17 05 00 00 00 00 00 ........c.......
0030: 28 00 00 00 00 0f 00 00 70 08 00 00 01 00 0c 00 (.......p.......
0040: 59 56 31 32 00 d8 bd 00 00 00 00 00 00 00 00 00 YV12.ؽ.........
0050: 00 00 00 00 00 00 00 00                         ........

[byteshare]

Originally Posted by Atak_Snajpera

I doubt we will ever have solution for vc-1

Could we use something other than FFMS for VC-1/VP9 sources?
I don't have a source to test with but I've heard LWLibav can handle importing those sources correctly.

[Atak_Snajpera]

Originally Posted by byteshare

Could we use something other than FFMS for VC-1/VP9 sources?
I don't have a source to test with but I've heard LWLibav can handle importing those sources correctly.

LWLibav can't even correctly decode above anime sample. Lot's of dancing grey blocks everywhere.

[duffbeer]

Still no problem with any of the encodes I've done from VC-1 sources. I've checked quite a few of them now.
A friend of mine has problems with VC-1 and the only difference from my setup is that he is using Windows 10 - I'm still on Windows 7. COuld that be part of the problem?

[Atak_Snajpera]

Originally Posted by duffbeer

Still no problem with any of the encodes I've done from VC-1 sources. I've checked quite a few of them now.
A friend of mine has problems with VC-1 and the only difference from my setup is that he is using Windows 10 - I'm still on Windows 7. COuld that be part of the problem?

I do not think so because I use FFMS in ripbot264 instead of built-in video decoder in OS.

[byteshare]

Originally Posted by Atak_Snajpera

LWLibav can't even correctly decode above anime sample. Lot's of dancing grey blocks everywhere.

Oh, that is too bad Thank you for letting me know.

[byteshare]

Originally Posted by Atak_Snajpera

LWLibav can't even correctly decode above anime sample. Lot's of dancing grey blocks everywhere.

Have you tested with a newer version?
A friend of mine tested in StaxRip and did see the dancing grey blocks with v1.9 but not with v2.0

[Atak_Snajpera]

Originally Posted by byteshare

Have you tested with a newer version?
A friend of mine tested in StaxRip and did see the dancing grey blocks with v1.9 but not with v2.0

I just used this version
https://forum.doom9.org/showthread.p...64#post1870564

Result

[byteshare]

Originally Posted by Atak_Snajpera

I just used this version
https://forum.doom9.org/showthread.p...64#post1870564

I'll look to see what version StaxRip 2.0 uses. Maybe StaxRip is using a specific version that works.

Update:
I can't tell the exact version Staxrip 2.0.0.0 is using but the size is very different and the date modified as well.
The newest version that you linked is 27.1MB and a date modified of 2019-03-31 and the one Staxrip uses is 11.1MB and a date modified of 2018-12-31.
Link to the Staxrip version: https://mega.nz/#!N9QX2Q7a!pBmNmgQIf...2KMTVt4aCMJX7A

[sneaker_ger]

That version gives me the gray problems. It also still has the "old" ffmpeg bugs. Every program still built from old ffmpeg is totally useless for VC-1 decoding - too many bugs.

https://github.com/VFR-maniac/L-SMASH-Works/issues/58