Midi timing suggestion on recorded VST instruments.

as typical MIDI interfaces such as USB are often not very good for even basic tasks, hence my using such things as Parallel port based MIDI interfaces and PCI.

Padawan, please do a search before you put info on this board, or don’t state it as fact but as an assumption from your side. Because it’s simply not true.

I think there is more relevant information on problems like these on the Sound on Sound site. I can get answers to problems in minutes that would take years to see here. There’s some serious system experts over there and they always explain things very clearly.
And they don’t visit this forum.

Try setting the “Priority” to midi?

Thanks Como, I’ve read those articles before I began this thread and lots of
facts are covered there. Still I believe that this easy sequencer tweak
of moving forward the mididata is the answer to the problem. (the
problem that I talk about) I can’t see why it should work in any other way.

I agree, thanks. That has to be the best work around.

And that seems to me like a really simple tweak, which I think
they should implement in Cubase.

Would even Cubase be the first DAW to have a playback timing correct recording
environment of VST instruments if they did?

I’m tired of moving stuff after I record. And even if I record on my lowest
latency and the sound plays only a few milliseconds earlier, I’m tired of
knowing it’s not exact, especially when believing the tweak would be so simple.

I hear you!

If I were to put a name to what you propose, I’d call it MDRC … Midi Delay Record Compensation. It should be easy to accomplish this with incorporation of the same calculations used in Plugin Delay Compensation as an after you stop recording auto adjustment.

If you are recording against other playing tracks, presumably any midi latency would already be factored in by the compensation you, the player, makes to stay with the beat of the other tracks while monitoring your playing and what your hearing from the program.

I think there also must be an issue here similar to ‘direct monitoring’ as used by a vocalist tracking against a project.

Depending on your sound source, with a hardware module couldn’t you use a midi splitter with one out going into Cubase for recording and the other directly to your module. That way … I think … the alignment would by off no more than the minimal playback latency your system is capable of.

If you also recorded in with record quantization set around 1/64th or 1/28th, shouldn’t your recording pretty much exactly reflect your performance?

Como

PS I love midi!

What gives you that impression?

So as I understand is that this problem has to do with Softsynths? Not hardware midisound generators etc?
That explains a lot. 15 years ago I made miditracks in digital orchestra pro on a computer and a roland Soundcanvas.
That system didn’t had that problem :smiley:

Mostly, yes. But it’s the same for VST effects. Cubase PDC (plugin delay compensation), adjusts output latency to the level of whatever is the longest summed delay of audio through a channel. If you had a softsynth with latency and then had it bussed to an effect rack or had an insert rack on the VSTi output channel, all those consecutive delays are summed. The longest delay chain sets the total latency and all the other channels have some fraction of that delay … the difference between their delay and that of the longest … added, so all the audio comes out in sync.

Depending on the VSTi, some are very CPU demanding and will require a higher buffer, which results in longer latency. If I’m recording midi on my system using hardware and without running any VST or VSTi plugins, I can set the buffer at 32 samples … for all practical purposes undetectable latency.

Como

Guys this is a lost battle.

5 pages and amongst all the posts Steinberg denying the facts, saying this is not a bug. The only workaround is to work at the lowest possible latency and create a keycommand for advancing MIDI some ticks, then hit it everytime you record something.

If someone got to something better than that please tell us here.

Don’t use midi equipment that doesn’t come with a proper (non class-level) driver.

:unamused:
You are getting really annoying Padawan. You just fail to see the problem.
It’s not about the midi timing. What happens is that after recording midi notes appear EARLIER in the editor.
Better midi drivers will not let the midinotes appear where they should be.

It has to do with the latency of VST instruments. That latency is compensated for in Cubase, so because of that the midi notes appear earlier because they are not taken in the equation.

Can you please read carefully all replies and join constructively or shut up?

Greetz Dylan.

No greeting required Dylan, for I am not even sure that you are experiencing the problem?

I have never experienced the problem so it is quite likely one of those issues that are infinitely unsolvable without the purchase of improved hardware.

The only person whom as I see things (enemy or not) who really has any idea about this at all would be Como, as he has bitten the bullet long ago and purchased the proper kit as I have said myself in another post.

Oh and PDC was never invented or invisioned for use on the recording side of the DAW, it was, is and always will be a mixing paradigm.

Cheers

Okay,
Turns out you can’t read , didn’t realize that, sorry.

This thread is getting confusing because different problems are adressed but I want to stick with the original one.

Joel the OP wrote: And this is the part I am interested in:

When recording on a VST instrument’s track, Cubase
should move the timing of notes forward - equal amount of ms as is the
current output latency. This to give the player and the recorded take justice.

AFAIK, this would explain why midi notes appear earlier in the editor.
This also points out that mididrivers, sytem timestamp, hardware configuration are irrelevant in this scenario.

In my case, I have this behaviour since I start working with VST instruments.
When I used a Korg Triton as sound generator the midi was always in place.
Nowadays I solely work with softsynths.

All I want to know is if Joel has a point or not.
Arguing about midi hardware clutters up this important thread.

Greetz Dylan.

What I think should not be forgotten is systems are not systems unless their timing is accurate and any form of “compensation” in the record path is an indication that something is not right somewhere.

There are many examples of hardware, in particular USB type interfaces that have tried to overcome the inherint timing innacuracies of the USB (buss) but with limited success.

For example terratec, lexicon have both done this but for my purposes I simply tolerate what latency my system produces or use something else and sometimes the old way is the best way and herin lies the answer.

It has nothing to see with drivers, just let this go. The same controller recorded notes prefectly in time on other DAWs on the same computer. The same controller connected by MIDI and plugged to a MIDI interface with the 5-pin cable recorded early notes as well inside Cubase, so its something soft related and not dependant on drivers.

People who has this issueare having it for quite a long time (in my case, since Cubase 4 and with 2 diferent computers). The strange thing is that apparently, and at the same time unfortunately, not everybody suffers this problem. So it seems Steinberg wont fix it.

That’s right because not all DAW’s are the same, and nor do they use the same technologies.

whatever, but the point is this proves is not the driver.

All threads like this prove Ed, is that it’s the user and not Steinberg that is at fault.

Damn, you are right! I was pluggin the controller to the DVD-R unit! thats why the notes were recorded earlier!

Seriously, your last post is kinda funny :slight_smile:

Funny if you are aligned with the Dark Side of the force maybe but showstopping for those just trying to record their MIDI performances.

The thing about this situation is Steinberg have designed something, and were anyone to read the manual on this subject they would see that it was never made to be some kind of solution for all system ills, in fact it guarantees nothing so how can Steinberg be expected to “fix” anything when the service is a best effort?

Besides, there are way better MIDI interfaces out there than what has been touted thus far, that have addressed these kinds of problems with timing and none of them are USB, except maybe one company who was able to push the envelope a little more than what was previously the case, I’ll give a hint, they are the same people as those making the software, in relation to their personal (not professional) identity.