Use makeMKV it’s free and lossless.
3 Likes
on the web of MakeMkv doesn´t put anything free, according to the website, its price is 60 euros.
Its changed then.
Its beta and requires a beta key you get from the makeMKV forum for windows.
Never used it in android or anything iOS though.
Still says it’s free on its site.
Doom9 is a great site for video and audio tools also.
1 Like
MakeMKV is free to use as long as it’s in beta, which it still is and has been for years. Every month or two you have to enter a new registration key (which is posted in the MakeMKV forums) to keep it active. If you get bored of doing that, you can pay to activate it in perpetuity, but I’ve been using temporary keys for years.
2 Likes
Ok. Thanks for your answers! Now I have a Beta Version with Beta Key.
3D frame packed works fine in my Vero 4K +. Only one movie dont work correctly, Terminator 2, everything else fine!
Hello the team. Was just wondering some questions. Is all the work you did specific to s90x or lets dream of a future vero 4k pro based on s922 (why not dream) will the work on 3d Still be ok ?
This is still the latest model and we will not be releasing any S922X solution in the future. We aren’t sure if it will work on newer AMLogic SoCs and AMLogic no longer consider MVC to be supported.
Our focus is to get this working on the 4K/4K+.
Ok thanks was just wondering.
Thanks a lot for this 3D version.
I just installed according to instructions in first post.
The mayority of 3D blu ray play back fine with ‘hardware accelerated’. Alas, some of them expose every 2? seconds or so jumpyness. Hard to describe. Picture jumps back halt a second or so.
HSBS and HTAB videos i tried play back fine. Nice that they are played back indicating SBS/TAB HDMI modes. It is not possible to play them back “hardware accelerated” (frame packed) though.
Hence, full-SBS or full HBTAB videos can not be played back in full resolution, these would need to be ‘converted’ to play back frame packed.
Btw: Two example of 3D Blu Rays that cause problems are “Upside Down” (german BD3D) and “Metallica Through the Never”.
The worrysome part of this jumpyness is that i had the same jumpyness on Kodi Intel MVC build running on PC with AMD graphic card, so that i had to add the Intel MVC software decoder. With an older version of the SW decoder it is decoding fine, but with a newer one, it exhibits the samejumpyness. I ca not remember how it performs with HW-decoder on Intel GPU, need to check.
I could not make out from the prior postings if i am reporting the same issue already reported.
Have you checked the MediaInfo to see if this file has an NTSC flag?
enlighten me…
mediainfo as in the CLI tool ?
File as in the left eye h264 M2TS file on the ISO ?
medianfo on vero4 did not produce any output sigh.
Otherwise i think the answer to your question is yes:
File with problems:
Video
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.1
Format settings, CABAC : Yes
Format settings, ReFrames : 4 frames
Codec ID : 27
Duration : 1h 48mn
Bit rate mode : Variable
Maximum bit rate : 31.0 Mbps
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 fps
Standard : NTSC
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Color range : Limited
Color primaries : BT.709
Transfer characteristics : BT.709
Matrix coefficients : BT.709
File without problems:
Video
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.1
Format settings, CABAC : Yes
Format settings, ReFrames : 2 frames
Codec ID : 27
Duration : 1h 55mn
Bit rate mode : Variable
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
It’s a NTSC video; so yes, that is a known problem. I’ve already implemented a workaround for the upcoming 4.9 kernel and if I’ll find more time I’ll do it right 
.
Thanks a lot.
Mind explaining how this would be a fix to the kernel ? I would have expected it to be a fix in kodi, so its quite curious to hear it would be a kernel change.
I’m just guessing, here, but I suspect it’s not so much that the change requires a new kernel, it’s just that we’re overdue for a release based on the new kernel, and there’s a moratorium on doing any significant new development work based on the old one.
No, it doesn’t require a new kernel. I could’ve fixed it for the old (3.14) one, too. But you’re right, we’re concentrating now to get the new one out of the door.
Well, … it’s complicated 
. When I started my odyssey with that NTSC issue I’ve thought it’s a Kodi bug. Then after investigating the related Kodi code I’ve thought it’s a kernel bug, then it’s a Kodi bug again and so on. That issue drove me crazy. But then, by chance, I’ve found the bug in the MVC decoder driver.
The root cause is a timing issue. The MVC decoder doesn’t decode frames at a (somehow) fixed interval of about 40 ms as other decoders do. So it might happen that 5 frames are decoded within a couple of ms and then, the next frame takes 0.2 seconds. So it could happen that Kodi asks for the next frame but there’s none available. In that case Kodi just continues and freezes the old frame. When the next frame is due then the previous (now late) one is ready. Kodi drops it because it’s a late one and displays the next frame (if there’s one).
Dropped frames are now causing that “stuttering” issue in the MVC driver. To make the long story short, there’s a bug in the MVC decoder which causes currently active frames to be reused instead of the dropped ones.
This bug is only triggered by NTSC videos, but it could have happened to other MVC videos, too. It seems that NTSC videos take a bit longer to decode as non-NTSC videos do.
To add to this, we have contacted AMLogic to ask them to explain the status of 3D MVC support. There is no reference implementation, even under Android, which suggests to us that this was integrated in the SoC as an IP block and probably had little verification beyond VHDL.
They’ve told us that 3D MVC support is no longer maintained – which seems to be true as the 3D MVC driver in 4.9 uses the old ring buffer approach, while other decoders use KFIFOs. We’ve asked for more documentation as well as whether this functionality is EOL, because confusingly, AMLogic are still updating the microcode for the MVC decoder.
Thanks folks for the explanations. So kernel means the (open source)? kernel level driver that feeds the SoC’s MVC decoder.
Yeah, removal of MVC decodingfrom future SOCs would be worrysome.
I was also curious because as said in before, i have seen the same issue on Intels reference software MVC decoder (windows only of course), and only older versions of that decoder work in the MVC branch of kodi. So either intel did introduce a bug and didn’t want to fix it anymore (which i would be a bit surprised, because that was back in ca. 2018), oder the NTSC flag encoding on the BD’s is actually wrong when used in conjunction with MVC, and older code was just not falling trap to it.