Hahahaha! Sagi, with the pace and amount of cuts many of us get in the film and tv world this is generally not going to fly. In tv and advertisement scoring you get a cut in the morning, and then in the after noon, and then probably another one before bed. This is the reality of working in this industry (film is another story and I don’t have this issue with film clients), so you get the job done. Somehow, Avid, one of the slowest evolving companies and one most entrenched to shareholders has been able to fix this issue. They see that many of their post clients have to deal with this same reality. Steinberg, being more nimble, should be light years ahead in this regard. But here we are. As for the video I attached, that was what was asked for from support…and I imagine it was compressed many times before it made its way to my hands. It came from a large edit house working for one of the largest advertising agencies in the world….
Now I’m curious - have all the files you deal with been like this and shown the same ‘bad bug’ behaviour in Cubase (you said ‘any’ H264 file you had).? What’s been your solution thus far.? How have you coped.? Do you output a new video from Cubase when you send back your results and hope they don’t notice.!? Or just send the audio, for the client to re-mux into the video their end.?
Yeah, you’d think - but maybe, in their defence, no-one (not many) have seen the issue or bothered to report it so far, along with a reproducible file they can use to investigate…?
Same/similar capability for DNxHD in your ER ToolKit, on a Windows machine.?
Same/similar capability for DNxHD in your ER ToolKit, on a Windows machine.?
AFAIK Should work more or less the same, but don’t take my word, try it yourself with the above video sample and let me know if importing the separate audio and video files into Cubando gives you a perfect sync.
Just note that DNxHD is a bit limited compered to the ProRes codec, and ProRes works perfectly well even on Windows without the need for the extra license that you need in Cubando to playback Avid DNxHD codecs.
@jjb0737 ,
Obviously, I’m also getting new cuts three times a day, and yes, even I receive tons of h264 files from production houses I work with. But still, there is a limit to what I am willing to tolerate. I will never accept an h264 with a bitrate of 1 kb/s or less, I will never use it in Cubase (Nuendo in my case), and I will never trust the audio that comes embedded with an H264 video. Everything I get I convert to ProRes before importing into my DAW project. My trusted audio is always an AAF I get with the video reference, the audio that came with the video serves as a limited reference to hear what the video editor premix sounds like, but nothing more.
I can reproduce this in C14.0.10 by importing the supplied video clip ‘as is’ (extract audio checked).
Further detail:-
I see the ‘2’ pop video frame right at the start (for 1 frame); I don’t get any black frame (or 1/2 frame) ‘added’ at the head of the file as is reported, but clearly the audio is similarly offset from the beginning, by approx 1 frame. Overall file length reported as 1s 904ms (by MediaInfo).
Converted into ProRes, the file imports with the audio lining up at the start (similar to OP’s first screengrab showing results from ProTools). File length reported as 1s 919ms
A further test involving extracting audio from video to load each separately turned out a waste since Cubase does not support importing the *.m4a audio file type.
I bear in mind the goal throughout all this, is for the OP to load his video as sent, straight into Cubase, without prior hassles, to get on with the work.! ProTools seemingly allows him to do that; Cubendo currently, won’t.
@m.chulek - clearly, some behaviour here worthy of scrutiny.
[EDIT:- Using a DNxHD conversion, as @Sagi points out, Cubase requires a purchased license of the codec for playback (which I don’t have). I tried it anyway; the video imports ‘blank’ the audio intact but offset, similar to first pic above. So perhaps nothing meaningful to add really… File length reported as 1s 919ms.]
[EDIT 2:- disclaimer; zero use of any AI in generating this response.]
@Puma0382
Thank you for the detailed walkthrough!
As expected, Cubando is not equipped with h264 audio offset correction. But you should know that PT is also very incapable when it tries to correct some h264 codecs.
I’m not sure if anyone here is aware, but there are several h264 encoding flavors (such as isommp42, qt, mp42mp41, isomiso2avc1mp41, mp42isom, iso6avc1mp41, and more). Each codec uses a different amount of audio offset, and PT only Correctly recognizes one or two of them. With the others, it makes such a terrible correction that it would be better if it did nothing at all.
About two years ago, I sent Steinberg an extremely comprehensive test I conducted on that issue, pointing out how each leading DAW and NLE handles the H264 audio offset. I hope this report is taken seriously and there is a ‘fix’ in the works.
About the ER Media ToolKit audio offset correction, in some cases, the free FFmpeg corrects the audio offset during its conversion to ProRes or DNxHD, so the user shouldn’t check the ‘Auto fix audio offset issue’ in the app’s settings. In other cases, FFmpeg will not do this audio offset correction and the user must check the ‘Auto fix audio offset issue’ setting in the app in order to force this extra process. Note that FFmpeg on Mac and PC are using a different code, so the user should be vigilant and if the 2-pop does not align as it should, one should try both with and without the ‘Auto fix audio offset issue’ setting option.
@Puma0382 , it would be very helpful if you could try converting with the ProResER Pro app (as shown above) with and without the ‘Auto fix audio offset issue’ setting and post the results here.
I’m asking this because, at the moment, I don’t have access to my Windows testing machine and it would be great to know if it works as it should, not only on macOS (each I already checked and works perfectly well).
Hello @Sagi - good extra info/food for thought there, thanks.! Jeez, never straightforward with these things, is it… LOL.!
I could do as you ask when/if I decide to purchase your app - I feel I’ve probably run out of goodwill from you, (by seeking yet another Trial.!), as I have previously demoed the toolkit on a couple of occasions already. I was impressed each time. Its very good and very convenient with the post-processing facility inside the Cubase Export window too. However, for my limited use/humble needs, I decided Shutter Encoder proved itself satisfactory enough, so far… 
Interesting you saying about preparing the detailed report for Steinberg… That was diligent and good of you to put the work in. Did you get any thanks/acknowledgement they at least received it.?
Its kind of a given for them to have all the myriad Cubase/Nuendo ‘features’ orbit around like satellites, with some not coming back into view for several releases hence (think VariAudio, Expression Maps, etc…). Years upon years will roll by without another sighting…
Maybe, with that limited interaction from SB staff earlier in the thread, it will soon be time again and the ‘video engine’ feature will roll round into view for attention…
Circling back on this one…
After some offline chat with @Sagi in the interim, I’ve now managed to test further as was suggested above. Results shown here…
Top track = video
Middle track = audio with the ‘Auto-Fix Audio’ checkbox enabled (no audio produced at all.!)
Bottom track - audio with that checkbox disabled (audio intact and aligns perfectly with the beginning of video).
In summary, I confirm on a Windows machine that the ER Toolkit produces expected results (no audio offset/video padding) using the settings shown above in earlier post, converting to ProRes and with ‘Auto-fix Audio’ unchecked.
Thank you very much @Puma0382 for taking the time to test this.
As I said before, sometimes FFmpeg is smart enough to calculate the audio offset by itself and checking the ‘Auto-Fix Audio offset’ is actually ‘wrong’ as FFmpeg already adjusted the audio, so using the ER audio adjustment shouldn’t happen. But not all videos are correctly recognized by the FFmpeg process. So, one must try both options to ensure an aligned audio output.
Either way, the ER Media ToolKit solves the problem on both Mac and PC. 
Thank you for the video. We are working on this issue.
Nice, thanks - though credit goes to @jjb0737 who supplied the video sample.!
Also, in my conversations with @Sagi , he’s your man for deep technical knowledge should you need extra discussion. He seems very approachable and has been very helpful…
@Puma0382
re: Your profile avatar (?), I used to drive London Transport routemasters in the 70s, back in the mists of time.
I’m experiencing this: I import the video, one that’s 24 fps for example. The burned in timecode shows up at the very beginning of the video as 01:00:00:00, as it should. The video is placed at the start of the project on bar 1. However, when I go to setup and add an offset of say, 1 bar, to have the project start at bar 0 and the video starts at bar 1, there is now some black added to the beginning of the video, like a second of so, before the timecode appears and runs. This is not the case when I play the video in Quicktime player so somethings happening in Cubase. The project is set at 24fps and the video is a .mov. This also happens with h264 videos too.
@Fujibass I think your mistake is that you think videos can align with the musical bars grid, while they only work correctly when placed on a video’s FPS (frames per second) grid.
Hence, you must always place videos using the video Timecode grid to ensure the video will playback properly. If you place a video in between the TC grid, you will always get black frames at the start, and more importantly, you will never have complete synchronization between your project and the video.
The problem is that most times the video TC grid does not align with most musical (BPM) grids. To align (synchronize) both TC and BPM grids, I always place the video at TC 01:00:00:00, then switch to musical bars grid, add and enable a Tempo track, switch to the Warp Grid tool, and drag to align the closest bar start to the start of the video by placing the cursor precisely on the video start and locking the Snap tool to the Cursor position.
It is not ideal, but at least it ensures correct work. Other DAWs such as PT, will not even allow users to place videos in between the project’s video frames. Steinberg, for some reason, doesn’t do that, but they are still bound by video rules like everyone else. So you get the odd anomalies you described above.
I hope this helps.
Thank you for your detailed reply Sagi. I did some further tests and the culprit I believe is the video itself. After trying your suggested steps the problem persists but other videos are behaving correctly. I’m not very knowledgeable about video specs other than the fact that the fps have to match the Cubase project. The videos differ as far as aspect ratios and data rates but I’m now getting into the weeds here, lol. Anyways, I found a video that does work. Thanks again
@Fujibass . It is good to hear you found a working video. At least now you know how to make sure it is not Cubase fault and ask for a better video.
All the best,
Sagi