Posts by Poo2

    my point was that even if you dont check that button it still has afterburner enabled because its deeply coded to fdk encoder to be enabled on default if it is not explicitly disable by --afterburner 0 flag

    If the checkbox is ticked, Voukoder is passing --afterburner 1, if it is unticked Voukoder is passing --afterburner 0. either way, it is doing its job, unless I'm totally missing the point poo :/

    Regarding options display, I prefer Daniel's approach to just show everything, just means we have total control even though in most cases we never touch the afterburner flag for example, as long as it is enabled by default on new display then should be good enough, but to be honest I've never encoded with it specifically set as disabled, worth a try for research sake and then pump the results through spectral and see what's what.

    Not sure what to make of this one, it is annoying to say the least.


    If you have a look at the attached image of a sequence, as you can see the audio is shorter, and the workspace marker at the top is aligned to the end of the video I want to export. Premiere's builtin export functionality fills that void with blank audio, but Voukoder fills it with some very psychedelic noises, kind of like a micro buffer of audio from the stream on fast repeat.


    I tried this both with H265 and ProRes444, and in both cases there is psychedelia, in both cases the audio was 24bit 48khz PCM.

    AVIsynth is just one possibiliy of many later down the chain, no relation to Voukoder so totally up to whatever user wants to do.


    I think because you are using the FFMPEG C Api, and 'nut' format is part of FFMPEG, all we'd need is piping support so it can stay as part of Voukoder as it will rely on the FFMPEG C API so will always be integral.


    Its a flexibility option for those who want to do more and shouldn't really impact the existing functionality of the plugin. Keeping it all as one plugin is I think the wisest direction because it is then all one codebase and Voukoder becomes the defacto plugin for all things output, granting a great deal of flexibility which will gain serious recognition as then other developers can write a variety of applications to hook onto the pipe output of Voukoder.


    Let's call Voukoder the mother of exports for the Adobe universe.


    I was just reading up on c++ piping, and it does look rather trivial as a construct... you probably already seen this doc https://docs.microsoft.com/en-…me-library/reference/pipe


    Have a quick snap at piping from Voukoder out to a shell instance running ffmpeg instantiated by voukoder, the instance being :



    ffmpeg -i - -vcodec libx264 -acodec libfdk_aac -vb 1200k -ab 96k testout.mkv



    On the Voukoder side, set the config to match the commandline "-pix_fmt rgb32 -vcodec rawvideo -acodec pcm_s16le -f nut" as an example and analyse the proof of concept, if it works nicely, eureka! :)


    I look forward to seeing the state change to 'investigating'... this would be a very powerful feature with endless possibilities.

    Pretty much indeed. FFMPEG is most popular because of its piping ability, incoming and outgoing. It would be only fitting to grant Voukoder the privilege of becoming the new frameserver as well as the gateway to much needed codecs.


    I believe the '-f nut' is a multiplexed format of audio and video which ffmpeg can serve to another ffmpeg instance nicely at breakneck speed via the pipe, then at that point one could do de-multiplex, split and dice and do whatever wonderful things they need, so we never restrict pipe format as such, we basically give the end-user to then do whatever process they want with a shell ffmpeg instance, and yes you got it, Avisynth can be part and parcel of any user workflow at their end.

    I did some tests, uploaded a vid, grabbed it using a youtube downloader, demuxed it, ran it through spectral layers pro, I can see what they're doing to achieve a better sound, they're actually tweaking the harmonics for the encode, need to do a few more tests to be conclusive. Now this kind of processing will achieve perceptual similarity with very little to no artifacts on any hardware, be it consumer or high-end, but... if you had the original source to listen to, then you do notice the difference.


    For streaming online they've definitely hit the nail on its head... that's the kind of effort all the generic encoders are missing though most do the typical 'cut high frequencies' etc... but still a far-cry from what youtube appear to be doing.


    Now if we had a batch script that we could run on our raw 24 bit uncompressed audio to prepare it for a quality encoder before encoding... then we could in theory emulate their process...


    the above is all guesswork based on the spectral analysis, for all I know it could be their encoder itself that does the wizardy on the fly...


    either way... yup, you said it, too bad it's not for public usage.

    If you haven't already noticed, there is a new option in Premiere when you create a new project:


    [Blocked Image: https://pli.io/Trhdx.png]


    "Preview Cache"


    It's specifically designed for the workstation hypercards apparently.


    This is the blurb: AMD Radeon Pro SSG


    I just wish Adobe would get their rears into gear and optimise their code to start utilising all the cores of modern CPUs efficiently, that has been their biggest failing, an old codebase just dragging its heels.

    Not greatly experienced in forums, IRC background primarily, but I will do that to comply with forum etiquette.

    The mismatch can be a headache, I do hope you can find and solve the issue, it is quite an alarming one really because 10ms delay back or forward can be nervewrecking when it comes to very particular sound/vision trigger syncs, ie a very fast flash with a bip of a noise.

    Ok, I just did the test again.


    The first attachment is MediaInfo analysis of a quicktime output using Adobe's builtin export. = ctest1.mov

    The second attachment is MediaInfo analysis of quicktime output using Voukoder's export. = ctest2.mov

    I take that back, both MP4 and MOV do. But Voukoder is not outputting this in a Quicktime container.

    I have noticed that when opensource software use libav, they for some reason get mismatches in durations in the order of milliseconds, but when you use FFMPEG CLI directly, then they match.

    The info above is from the encode produced by Voukoder, why would I be listing it as a bug here if that were not the case.


    I incorrectly stated x264/x265 when I meant "h264/h265 out of Premiere".


    Please do a quick export of say a minute long video from Premiere using ProRes 4444 and PCM 24 bit in a Quicktime-Mov container then run it through MediaInfo to see a corroboration of the attached image in the first post, and then run an export using Premiere's builtin H264, and compare MediaInfo's analysis to appreciate why I believe this is a bug.

    There is definitely a mismatch in audio and video durations on output from Voukoder no matter what codecs are selected.


    If I export using Premiere's own x264/x265 then the durations match.


    Although it is mere milliseconds out, definitely a bug somewhere, this could impact videos where end frames are lost on playback because audio finishes before video. I have observed this issue with Stan Lee's Mediacoder too.

    This would be absolutely lovely, I recall back in the early days of unix commandline codecs, we had the ultra-flexible option to piping data around through own custom scripts/executables, everything talked to everything by platform design, then Windows came along and began creating brickwalls.


    Ok


    Let's say the ability to pipe out via stdout to a custom application of our choosing. So say in the encoders list we just select 'custom' and then type in a freetext commandline, independent for video and audio, this way advanced users can use their own custom workflow with the plugin.


    Kind of stems off the back of the other feature request which got rejected because it was outside the scope of Voukoder's purpose.


    I guess from a development perspective all it would require is instantiating a stdout to a shell running the commandline entered by user.


    This will make Voukoder very flexible going forward, I have a fair few dozen python scripts which already work off stdout from ffmpeg itself, so I could chain in this way for a very specific workflow straight out of the premiere export door all thanks to Voukoder giving techs backdoor access to the streams.

    Though I can imagine some in the future might request a raw multiplexed outgoing stream, so say FFMPEG just packages and shunts it out the door to stdout where whatever custom program or even say a script running its own ffmpeg process then does wierd and wonderful things for bespoke workflow requirements.

    Oh bummer... sounds quite a serious issue, 122 parameters going all funny every release, that won't be very funny.


    Fingers crossed Zac has a solution and not a new bug report for Adobe.


    I guess you could consider release 8.1 as a beta for 'testers' only, say a dedicated 'beta testers' page or something, just a thought, as I would love to play around with it, on the edge of my seat here.

    Exactly! And this is where it gets annoying, because unless you have a bit-perfect lossless source such as SACD etc, compressed versions sound abysmal on high-end gear. My monitors barely get much use outside of source engineering, indeed they were a worthy investment for master tuning, but playing back anything compressed bar very extreme bitrates really shows how terrible lossy audio codecs still are, it is as they say 'perceptive' and this would be down to accommodating typical consumer grade hardware, for example I'm often having to put on a pair of consumer grade sony headphones to 'hear' what the average person will.


    Unlike the big players, I don't have a team of listeners/viewers to get feedback on edits etc... but I still like to make sure I'm doing all I can to perfect my final outputs even if they may be for a very small closed-door userbase.


    I'm looking forward to FDK Voukoder integration, I can see countless nights of testing coming up ^^