Ok, sorry. Couldn’t make a new vid till this weekend. Here is what you asked for showing Project start time. The sample rates are the same (48k) on both machines and there is no display offset entered. Framerate is set at 23.97fps in Cubase and Logic…24fps in VSTP.
https://www.dropbox.com/s/4a88ahzqgttbhq6/VST%20Performer%20MTC%20sync%20issue%20pt2.mov?dl=0
As mentioned in the video, the closer your start time is to 0, the closer the MTC is to the correct time. The further you start from 0, the more the MTC timecode is ahead of VSTP and Cubase.
confused, could it be that this is spread over several threads? I replied but don’t see that here.
Sorry for the confusion I have two separate threads for separate issues in VSTCP:
this one started out addressing the Video Hosting issues in VSTCP, But morphed into addressing the MTC Output issues.
the other one is for the issues with the Video Hosting in VSTCP
https://forums.steinberg.net/t/vstc-performer-video-hosting-issues/687386/4
These are two separate issues, but this thread started out as something different. Should I start a separate one for MTC Output problems and repurpose this one back to the Hosted Video issues?
Sorry about that:grimacing:
Ok, sorry. Thing is, these issues are probably related so to not have to search forever and postpone other tasks, could you pls test both issues with a ‘regular’ frame rate like 24, 25, or 30? It would save a lot of time.
Ah - but MTC offset also had a bug (fixed with next version), still it would be very helpful to know if the video file start time issue is related to drop frames or the like. Thanks!
I suppose I could use some video off my phone for 30fps, but 23.98 is standard frame rate across virtually ALL broadcast and film productions.
A frame rate issue would only affect the sync for 1 second at a time though. Every 1 second, things would lock back up. Neither of these issues seem very related to frame rate as the MTC sync drifts further and further, the farther one’s start time gets from 0.
And with the Video Hosting issue, the video stays in sync if both the Cubase start time AND movie start time are the same. If they are not, the video still stays in RELATIVE sync, it is just offset by however much the Cubase session start time is different from the Video file start time.
yeah I got that. Ok, I’ll check myself.
It all works now here. imported 23.9760239760… fps video (now doesn’t sound that professional in the digital age) and imported same video in Performer. Video and all time displays incl. bars+beats are in sync automatically.
However, this is with the version that we have here. It fixes that the bars+beats included project offset and some more. This all only works when you don’t change the Video File Start item, this is set automatically when you move a video event or change project offset etc.
Unfortunately, it will take a while (mid February as of now) until it will be released, sorry for that.
Ok, will wait (kinda) patiently
Have tried v5.5.
-
MTC sync at 24fps still drifts the further you get from 00:00:00:00
-
Loading a video locally in VSTCP still syncs the first frame of the video to the start of the Cubase file. This happens irregardless of what the Cubase SMPTE offset or VSTCP “Video File Start” are set to.
VSTCP’s internal timecode display is always locked correctly to Cubase’s Timecode, which makes me clueless as to what the VSTCP “Video File Start” setting is supposed to accomplish?
The way it seems to me is that you should be able to set the VSTCP “Video File Start” to (for example) 1:00:00:00 (if that is the first frame of the video file) and then if your Cubase file’s start time is 1:20:00:00, the movie will start playback 20 mins into the video file when you start playback at the beginning of your Cubase file. No?
- will double-check
yes.
That would be true if the video in Cubase/Nuendo starts at that time, or the project has that time as Project Start time and the video starts at 00:00:00.
Video Start Time defines the timecode where on the DAW timeline (!) the first frame of the video file plays, as said before. That is when the timeline displays timecode (which may differ from seconds when the frame rate is not an integer, such as 23.98).
Also note that when you add, move, or modify a video file in Cubase/Nuendo, the Performer Video Start Time is adjusted automatically. The whole idea is to replace the video playing in the DAW by a local file such that the only difference is quality. For that, you don’t have to adjust anything, as that alignment is applied automatically.
Ok, so I think I understand now how “Video file Start" is currently designed to work.
If I am understanding correctly, then currently the only way to record multiple cues for a film would be to have a separate video for each cue and each of these videos would need to be cut so they start at the exact same SMPTE time as the corresponding Cubase project. Is this correct?
This would also mean if I wanted to “relax” the start time of a cue a couple of frames to picture during a session, I would need to have the local video file recut so that it also starts 2 frames later?
It would be fantastic if we could get the following functionality:
-Load a local VSTCP video
-tell VSTCP what timecode to associate with the first frame of that video (ie: 00:59:30:00)
-the video would start playback corresponding to the SMTPE timecode coming from Cubase (which is also displayed on VSCTP overlay)
(Ie: if the Cubase file SMPTE time started at 1:20:00:00 for bar 1, the video would start playback at 1:20:00:00 when Cubase starts at bar 1)
This would allow for one video to be loaded once. Multiple cues (multiple Cubase sessions) could be recorded (each with a different SMTPE start time) without having to cut a bunch of videos and have the performer reload a new video for each new cue.
VSTCP’s SMPTE overlay is already locked precisely to whatever the Cubase SMPTE is regardless of the Cubase start time. Could it be possible to have the video adjust to where it starts playback accordingly? (if VSTCP is told what the SMPTE is for the first frame of the video file)
Thanks!!!
No, not at all.
Not sure if you know Nuendo, I strongly suggest you have a look at its taker. This is what VST Connect ADR is entirely integrated. You have a cue (cycle marker) list and there are rehearse, free run, record etc modes and you select a cue and then go with preroll, swipes, count-in, take length indicator and of course dialogue. This is usually across one entire movie. You may also have multiple video files, but for local Performer only one video is selectable. You are better off with streaming then.
That is pretty much exactly how it works so I’m a bit confused as to what doesn’t work for you?
For your comments following about multiple cues I would strongly suggest that you have a look at Nuendo (download the manual, or at least read the VST Connect 5.5 manual), as it all is there for you exactly how you require, as far as I understand your workflow. Which is what professional ADR studios are using already very sucessfully with Talents beeing very exited about its simplicity.
As such work is usually done in the professional environment, Nuendo shouln’t be too much to pay for, given what you get. Also the base is the same as Cubase, so there is no learning curve at all (except for said ADR functionality). If you rather want to stick with Cubase nevertheless, you should create a Marker Track with your cues and text etc, and locate to each cue for the takes.
Hope that helps.
Thanks for the suggestions, I have downloaded Nuendo before, but could not figure out how it was different with regard to my workflow.
Currently, if I have 20 music cues, each cue is a separate Cubase Project. This way, adjusting tempo, SMPTE start, meters, etc. for cue-3 does not affect cue-18
During a recording session, Each separate Cubase Project is opened. We record/work on the cue, close it and move on to the next one. We have an external Daw (Logic or Pro Tools) that is hosting the video playback and is synced via MTC.
You’re saying that currently I could have this workflow with Nuendo (separate projects for each cue) and use VSTCP hosting a local video and it will sync properly even is the start time of a Nuendo project is different than the the start time of the local video in VSTCP?
For that workflow Nuendo is not any different, as you have separate projects for each cue.
Nevertheless, it should just work if you make sure that a video is placed at the desired point in the timeline. I’d add the video to each project and align it but I get you don’t want that, though it would be just as simple as adjusting Project Start Time which you’d have to do anyway (or move the entire content).
If you place a video in Cubase and connect to VST Performer, it will check the first video in the project and report its start time (including possible offset) such that a local Performer video adjusts the Video Start Time such that it matches TC in the Cubase project. If you have no video at all, it will be Project Start Time by default and you may change that to what you need. In your case, that would probably have to be negative which I guess is not currently supported. Your options are then to either adjust start times of your cue projects, or move its content to the desired position.
I did a similar thing with a band live performance (click based), placed the video in all of the Cubase projects and adjusted it to fit the audio. It’s reasonable to want to watch the video anyway during the recording process. Then using streaming with VST Connect made repairing takes a snitch.
MTC does not define pull rates at 24 frames (like 23.98). Next version features an “AUTO” setting (default, replacing “OFF”), which takes the project frame rate for MTC speed control, this should work for you.
Thank you for the patience on this and continuing to help me out.
Alright, so I revisited hosting video in Cubase to see if I could make it work the way I need.
Here’s what I found:
-
Create master template file to copy for individual cue files.
-set start time of project (bar 1) to 00:59:30:00
-import video to bar 1.
(Cubase sets the default origin for both video and the audio track to 1.1.1.0)
-select video audio and choose Audio:Update Origin
(this changes the Audio file origin in the Pool to 1787.4.1.35. I’m assuming this is the bar offset from bar 1 at 00:00:00:00 to bar 1 at 00:59:30:00 considering 120bpm which is what my file is at.)
It would be help
-In the Pool, change the video file origin to 1787.4.1.35 to match the audio file.
-
Do a “Backup Project” to create a new cue (2m10)
-
With the cursor at bar 5, do a "Set Timecode at Cursor to set the music start (bar 5) of the cue to 1:20:00:00
-(OPTION 1) if on the following alert I choose YES for “keep project content at current timecode postion”, then the video correctly adjusts, the extra video and audio that are now happening before the Cubase file starts are cut off, and at bar 5 the video is on frame 1:20:00:00 and Cubase is on frame 1:20:00:00.
-(OPTION 2) If I choose NO, then the video stays at it’s current bar position (this is expected behavior).
Now at bar 5, Cubase is on frame 1:20:00:00 but the video is still on frame 00:59:30:00.
It would be easy enough to select the video and audio and choose Move to Origin since we had set that in the Pool, but Cubase does not know how to handle moving items before the project start (even though it did it correctly in OPTION 1) and gives the error:
"Event cannot be moved. Origin is before project start)
If I only needed to adjust the start time of a Cubase file in relation to the video once, OPTION 1 would be fine. However after a cue has been written and/or to adjust for picture changes, I will need to use “Set Timecode at Cursor” many more times. At this point, OPTION 1 is not usable as everything that has been recorded would get shifted as well. OPTION 2 is also not usable since Cubase won’t move the video using “Move to Origin” to correctly resync with Cubase’s new start time.
This is basically where things end because not being able to adjust the music start of a cue (with parts which maintain their bar positions) to a new frame in the picture is a deal breaker. If I was able to do this, I still wouldn’t be able to load video locally in VSTCP, but I could at least stream from Cubase to VSTCP.
Am I missing something with this scenario?
Just one question…does that happen often? otherwise, I’d just clip the video file such that its start aligns with the cue project start.
Will ask the guys about setting origins, but still I feel there might be a simpler solution. Off my head I did
- get the movie and start cutting cues. movie starts at 1:00 or wherever so that’s the Project Start offset, and also cue 1, save as template
- cue 2 starts at 1:00:20:00, new Cubase Project from template. Adjust movie clip (offset 20 sec) and Project start offset to cue 2, stream and you’re done, no? Even a Performer local video should be in sync then w/o even adjusting its Video Start Time (it does that automatically).
- cues 3, 4, 5 etc
- cue 4 needs to be adjusted. Now we do have a problem if (!) we need to work on subsequent cues, as those need to be adjusted to the changed movie. But you need to change ‘something’ anyway, right?
It all comes down to why it would be a problem to adjust the movie clips (offset) instead of using Performer as an MTC Synchroniser plus an additional video player. But it may well be that I’m still missing a vital point, bear with me 