MIDI Weirdness...

Author
Troykington
Max Output Level: -90 dBFS
  • Total Posts : 15
  • Joined: 2010/12/05 11:38:58
  • Status: offline
2011/07/27 06:51:15 (permalink)

MIDI Weirdness...

Hells Bells I love this forum....you go into it in search of a solution and you find one for a different problem by mistake!
Anyway, any help on this one would be appreciated - I have just imported a lot of audio into a new project and since doing it, for some reason, when I input midi from my keyboard, the input data appears ( and plays back) late? anyone seen this before?? It was fine until the audio was imported.
All the best
Troy
#1

26 Replies Related Threads

    jackn2mpu
    Max Output Level: -47.5 dBFS
    • Total Posts : 2765
    • Joined: 2003/11/08 17:38:43
    • Location: Soprano State
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 07:22:20 (permalink)
    You've got a couple of things going on:
    latency in the audio stream from both your imported audio and audio generated by the MIDI input

    and

    it takes longer to generate the audio from a softsynth driven by MIDI than it takes for already existing audio to playback; when you imported the audio you now have a reference to compare to that you didn't have before you imported the audio

    Jack
    Qapla!
    #2
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 08:01:49 (permalink)
    Cheers Jack....so am I now doomed to Quantizing my midi tracks after audio importing?
    #3
    jackn2mpu
    Max Output Level: -47.5 dBFS
    • Total Posts : 2765
    • Joined: 2003/11/08 17:38:43
    • Location: Soprano State
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 09:05:36 (permalink)
    Troykington


    Cheers Jack....so am I now doomed to Quantizing my midi tracks after audio importing?


    Quantizing MIDI tracks has nothing to do with what you are experiencing. You'll just have to slide your imported audio or rendered audio from the MIDI-driven softsynth to align with each other. Move either one but not both. Let me guess - you're new to audio recording on a computer, right? There's lots to learn.

    Jack
    Qapla!
    #4
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 09:16:33 (permalink)
    Hiya Jack...I may not have explained it properly but I am not using a soft synth and it is purely the midi data that is the problem. When in the piano roll pane I can actually see the midi data going in at a semi-quaver late to the tempo/bar. I have just opened up a new project and have discovered that it is nothing to do with Audio - it is Midi and doing it again...the stuff is not going down in the right place.....bloody thing....!
    #5
    jackn2mpu
    Max Output Level: -47.5 dBFS
    • Total Posts : 2765
    • Joined: 2003/11/08 17:38:43
    • Location: Soprano State
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 10:21:32 (permalink)
    Troykington


    Hiya Jack...I may not have explained it properly but I am not using a soft synth and it is purely the midi data that is the problem. When in the piano roll pane I can actually see the midi data going in at a semi-quaver late to the tempo/bar. I have just opened up a new project and have discovered that it is nothing to do with Audio - it is Midi and doing it again...the stuff is not going down in the right place.....bloody thing....!


    If all your MIDI is slightly late then what's the problem?
    Moreso - what are you basing your timing on for playing MIDI? A previous audio track, audio coming from whatever softsynth you are using when playing in your MIDI or exactly what? You have to have some audio going from a softsynth or hardware synth when playing in your MIDI, elsewise you're just working in the dark. I'm assuming you're actually playing in via musical keyboard and not computer keyboard. If you are indeed playing in from a musical keyboard and using a softsynth then your problem may well be audio latency - if that's the case you need to lower your latency. Are you using ASIO drivers for your soundcard?

    Jack
    Qapla!
    #6
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 11:05:23 (permalink)
    What are you using as metronome?
    Midi or audio metronome?

    And all midi is external hardware as I read your post?

    Could it be roundtrip delay through system for midi, then for listening to the metronome audio - this would mean what you listen too is late and recording as well.

    Just a view...
    #7
    Bristol_Jonesey
    Max Output Level: 0 dBFS
    • Total Posts : 16775
    • Joined: 2007/10/08 15:41:17
    • Location: Bristol, UK
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 11:11:27 (permalink)
    I think it's time for the OP to provide some more information.

    SYSTEM spec please - all of it
    Signal chain - start to finish



    CbB, Platinum, 64 bit throughout
    Custom built i7 3930, 32Gb RAM, 2 x 1Tb Internal HDD, 1 x 1TB system SSD (Win 7), 1 x 500Gb system SSD (Win 10), 2 x 1Tb External HDD's, Dual boot Win 7 & Win 10 64 Bit, Saffire Pro 26, ISA One, Adam P11A,
    #8
    Cactus Music
    Max Output Level: 0 dBFS
    • Total Posts : 8424
    • Joined: 2004/02/09 21:34:04
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 11:19:21 (permalink)
    Do you "hear" the MIDI delay when striking a key?

    Johnny V  
    Cakelab  
    Focusrite 6i61st - Tascam us1641. 
    3 Desktops and 3 Laptops W7 and W10
     http://www.cactusmusic.ca/
     
     
    #9
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 11:37:00 (permalink)
    Thanks everybody....Ok its Sonar 8.5.3, XP , I5 Processor, 4 gb RAM, RME Fireface 400, and the mother keyboard is a Roland XP80. I am using audio metronome. There is no audible latency delay at all when recording midi, all is perfectly in time to the grid but on playback, it is both visually ( in piano roll view) and audibly late.
    #10
    Chappel
    Max Output Level: -52.5 dBFS
    • Total Posts : 2300
    • Joined: 2009/07/11 14:55:32
    • Location: California
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 11:43:52 (permalink)
    What device are you using to play back the Midi?

    Are there any tempo changes in your project? The audio metronome can be very offbeat if the tempo map becomes complicated.
    #11
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 12:01:57 (permalink)
    There is only one tempo change and I am playing back with Gigastudio and XP80 internals through the Fireface 400. As I say, I have never had this problem before...and there are, superficially at least, no latency issues evident anywhere.
    #12
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/07/27 14:05:43 (permalink)
    Troykington


    Thanks everybody....Ok its Sonar 8.5.3, XP , I5 Processor, 4 gb RAM, RME Fireface 400, and the mother keyboard is a Roland XP80. I am using audio metronome. There is no audible latency delay at all when recording midi, all is perfectly in time to the grid but on playback, it is both visually ( in piano roll view) and audibly late.


    If it's late visually in PRV it's going to be played back late.

    Why this is becoming placed late as midi, is to be narrowed down.

    Gigastudio, is that a standalone that has a midi port that Sonar plays towards then?

    Do you echo midi through Sonar when recording, or going directly towards GigaStudio?

    My idea is if it's sounding alright while recording, because you go directly towards Gigastudio.
    But there is a difference in the sound you listen to from Sonar, and playing on your keyboard.

    So Sonar's latency might not be audible because you play directly towards Gigastudio.
    But it becomes apparent when midi notes are placed in time.

    I have a very vague memory of some setting regarding Gigastudio.

    If Gigastudio is standalone then both Sonar and Gigastudio needs to handle audio at the same time. So setting in Sonar should be to share audio device with other.

    Could something like this matter?

    Just a couple of ideas...
    #13
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 03:27:56 (permalink)
    Yes, gigastudio is standalone and has seperate ports. The real mystery for me is that nothing in my setup has changed. There is however one more point - the problem started since importing the audio and those files were ten years old from an old ProTools project and had to be converted before I could transfer....could this have anything to do with it??
    #14
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 04:48:51 (permalink)
    Really odd.

    Converting this audio, is there some kind of stretch involved to make it line up?

    I'm thinking if it's done with a tool that writes certain data into wav-files telling about an operation that is to be done, but Sonar does not look for this.

    Maybe an offset from start or something - not visible but audible.

    In this way Sonar might play back imported audio with samples in a different spot timewise than what is intended.

    I would try freeze these audio tracks and see if they come up differently after this operation.

    I don't know if you need to pop something into fx bin to make Sonar do something, if so just an EQ doing nothing would do.

    Anyway it would make Sonar interpret and read waves and then save in it's own way.

    Because it's related to imported audio, my gues is that it's not a midi issue really - but audio is delayed at playback by Sonar for some reason.

    Just another bunch of thoughts...
    #15
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 06:01:09 (permalink)
    You are dead right. It is not a Midi issue as I have just opened an old project and the midi input/output is perfectly fine and working as it should!...something has happened since the import into this particular project. There was no stretching for line up.
    Weird.... 
    #16
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 07:28:30 (permalink)
    How about start-offset as I mentioned?

    Waves may contain empty space at start. So clips may appear to start in one location but is actually played back as another.

    Try freeze as suggested to eliminate that.

    Any misalignment will show up on freezed tracks.

    I don't know how you import audio from other project without having to align stuff.

    How did you do that?

    Just selecting same tempo from protools might not give the same result in Sonar.

    I sense some information is missing here.

    And you mention in your a post that in realtime it aligned " all is perfectly in time to the grid ".

    What grid are you looking at in realtime?

    When you mention PRV you say it is late(about a 16th).

    If snap to grid is on for recording and set to 16th and a bit late i might be like that, it snaps to the next 16th note.

    My bet is on imported audio tracks with some starting offset.

    When exporting audio for clips not at immediate start of project different software do it different ways. Som export just what is audio, other create silence at start for resulting clips to line up later.

    If you solve it - would be great to hear what you did.

    Good luck.
    #17
    jackn2mpu
    Max Output Level: -47.5 dBFS
    • Total Posts : 2765
    • Joined: 2003/11/08 17:38:43
    • Location: Soprano State
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 08:22:07 (permalink)
    This thread is why we need to know EXACTLY what the op did to get the 'problem' he is having. How did he export the files from ProTools? BWF (broadcast wave) would be the best way to go as it gives time stamps. Even then you may well have to slide the audio around in Sonar to get things to line up.

    In post 10 the op states there was no audible latency when recording, only on playback. There wouldn't be any audible latency on recording depending on how the op is monitoring the audio generated by the MIDI triggered synth (we don't specifically know if he's using Gigastudio or the XP80 for audio while he's recording) during the record process. He has to have audio generated while recording - what's doing it?

    What I don't understand is why we can't get complete info on what the op has done or is doing to create this scenario in one post? Why is he just coming up with dribs & drabs (breadcrumbs)?

    Jack
    Qapla!
    #18
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 11:18:44 (permalink)
    I agree - a little more knowledge about details would help.

    Maybe he can chop up 15-30s or so from one of these audiofiles and we can try it out if we can reproduce it.

    Or a full mini project with just 15s or so.

    Blank project, import a 15s file, save and done.

    Then we can rule out Sonar's handling of files, and whether Gigastudio external stuff has anything to do with anything.
    #19
    brundlefly
    Max Output Level: 0 dBFS
    • Total Posts : 14250
    • Joined: 2007/09/14 14:57:59
    • Location: Manitou Spgs, Colorado
    • Status: offline
    Re:MIDI Weirdness... 2011/07/28 12:19:54 (permalink)

    It is not a Midi issue as I have just opened an old project and the midi input/output is perfectly fine and working as it should!...something has happened since the import into this particular project.

    I'm guessing that the audio import is a red herring, and a plug-in that uses Plug-in Delay Compensation was inserted somewhere at about the same time, causing all existing tracks (and the audio metronome) to play back with added latency.

    But the audio for the new MIDI track is being direct-monitored, or is on a track with PDC bypass enabled such that it sounds in time when played live, and has no audible latency, but the MIDI lays down late in the timeline, because it is in fact being played late relative to the M:B:T grid. 

    SONAR should automatically be compensating the MIDI recording position for the PDC, but I seem to recall there is a problem with this.

    Bottom line, if this is indeed the causes, is that the use of Plug-ins needing PDC should be avoided until tracking is complete.


    EDIT: I just did a quick test, using a hardware synth with PDC bypass enabled on the monitored track, and the MIDI did indeed lay down late for the new track, though simultaneous audio was compensated correctly for PDC.
    post edited by brundlefly - 2011/07/28 12:32:20

    SONAR Platinum x64, 2x MOTU 2408/PCIe-424  (24-bit, 48kHz)
    Win10, I7-6700K @ 4.0GHz, 24GB DDR4, 2TB HDD, 32GB SSD Cache, GeForce GTX 750Ti, 2x 24" 16:10 IPS Monitors
    #20
    Troykington
    Max Output Level: -90 dBFS
    • Total Posts : 15
    • Joined: 2010/12/05 11:38:58
    • Status: offline
    Re:MIDI Weirdness... 2011/07/29 05:45:57 (permalink)
    Sorry for the drip feeding of information Lads! I am a bit new to this forum and didn't understand the need for detail on what I presumed was something really simple. How wrong could I be.....
    Anyways, it appears to be that the problem HAS been caused by PDC!! something I was blissfully unaware of. There have been loads of soft synths winging in and out of this studio without my knowledge but now the override is in place, All seems to be working and in sync . Cheers Brundlefly and everyone else. Have a top weekend... 
    #21
    brundlefly
    Max Output Level: 0 dBFS
    • Total Posts : 14250
    • Joined: 2007/09/14 14:57:59
    • Location: Manitou Spgs, Colorado
    • Status: offline
    Re:MIDI Weirdness... 2011/07/29 12:52:07 (permalink)
    I feel like one of those healers who works by transferring other people's illnesses to himself. 

    PDC Override just keeps the audio on input monitored track from having latency when you rehearse/record with live input. It shouldn't affect MIDI recording one way or the other, and I am able to reproduce the issue with MIDI being laid down late whether PDC Override is enabled or not. So I'm curious how PDC Override fixed your MIDI timing issue, and also curious how you were not experiencing audio latency from the synth without it - unless you are direct-monitoring the synth...?

    Maybe better to just let sleeping dogs lie if you feel you have a solution, but it's a little confusing. 


    SONAR Platinum x64, 2x MOTU 2408/PCIe-424  (24-bit, 48kHz)
    Win10, I7-6700K @ 4.0GHz, 24GB DDR4, 2TB HDD, 32GB SSD Cache, GeForce GTX 750Ti, 2x 24" 16:10 IPS Monitors
    #22
    jackn2mpu
    Max Output Level: -47.5 dBFS
    • Total Posts : 2765
    • Joined: 2003/11/08 17:38:43
    • Location: Soprano State
    • Status: offline
    Re:MIDI Weirdness... 2011/07/29 13:25:07 (permalink)
    brundlefly


    I feel like one of those healers who works by transferring other people's illnesses to himself. 

    PDC Override just keeps the audio on input monitored track from having latency when you rehearse/record with live input. It shouldn't affect MIDI recording one way or the other, and I am able to reproduce the issue with MIDI being laid down late whether PDC Override is enabled or not. So I'm curious how PDC Override fixed your MIDI timing issue, and also curious how you were not experiencing audio latency from the synth without it - unless you are direct-monitoring the synth...?

    Maybe better to just let sleeping dogs lie if you feel you have a solution, but it's a little confusing. 


    You're not the only one confused. I don't think the op knows exactly what he's talking about or even what his problem is. He keeps adding bits and pieces of info, never giving us the whole story of exactly what he's doing. I mean - he's unaware of PDC in ProTools (and really - what does that have to do with importing the tracks into Sonar - why doesn't he just slide the tracks around to line them up in Sonar?)?

    I'm wondering if we're just dealing with a troll? I mean - he doesn't think we need complete info on what's going on to help him? It's not like we're mind readers here. Hate to sound mean, but I have limits to what I'll put up with in helping people.

    Jack
    Qapla!
    #23
    brundlefly
    Max Output Level: 0 dBFS
    • Total Posts : 14250
    • Joined: 2007/09/14 14:57:59
    • Location: Manitou Spgs, Colorado
    • Status: offline
    Re:MIDI Weirdness... 2011/07/29 19:21:55 (permalink)
    jackn2mpu


    brundlefly


    I feel like one of those healers who works by transferring other people's illnesses to himself. 

    PDC Override just keeps the audio on input monitored track from having latency when you rehearse/record with live input. It shouldn't affect MIDI recording one way or the other, and I am able to reproduce the issue with MIDI being laid down late whether PDC Override is enabled or not. So I'm curious how PDC Override fixed your MIDI timing issue, and also curious how you were not experiencing audio latency from the synth without it - unless you are direct-monitoring the synth...?

    Maybe better to just let sleeping dogs lie if you feel you have a solution, but it's a little confusing. 


    You're not the only one confused. I don't think the op knows exactly what he's talking about or even what his problem is. He keeps adding bits and pieces of info, never giving us the whole story of exactly what he's doing. I mean - he's unaware of PDC in ProTools (and really - what does that have to do with importing the tracks into Sonar - why doesn't he just slide the tracks around to line them up in Sonar?)?

    I'm wondering if we're just dealing with a troll? I mean - he doesn't think we need complete info on what's going on to help him? It's not like we're mind readers here. Hate to sound mean, but I have limits to what I'll put up with in helping people.
    I dunno. Seems legit to me. I think he just didn't realize how many factors go into determining how things sync up, and how precisely you need to describe the circumstances for anyone to provide meaningful help.

    My reading was that the import went fine, and was properly aligned, but right after doing that, newly recorded MIDI was landing late in the timeline - by a lot. The only thing I know of that can mess up timing that badly and that suddenly (other than messing with Timing Offset) is PDC, so I have little doubt that's the culprit. But based on what I've seen, the late MIDI recording is a bug, and would not be fixed by enabling PDC Override.



    SONAR Platinum x64, 2x MOTU 2408/PCIe-424  (24-bit, 48kHz)
    Win10, I7-6700K @ 4.0GHz, 24GB DDR4, 2TB HDD, 32GB SSD Cache, GeForce GTX 750Ti, 2x 24" 16:10 IPS Monitors
    #24
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/07/30 02:28:28 (permalink)
    Freezing the tracks before recording midi would have eliminated the PDC delays and bypassed plugins.

    Playback would the run without any PDC and you listen to something with just latency buffer delay.

    That's what I think.

    I put a question to Cake whether X1 is compensated for if PDC is used, and the recorded clips would be offset - since what you listen to is delayed compared where playback clips are located.

    They said this is done - recorded clips are move according to PDC.

    And I did not quite get whether this was a new feature or not - if Sonar 8.5 is having this too. But it sounded like it was not new - just the on/off button.

    Quote from Cake answer:

    "SONAR does have delay compensation built in for all of those scenarios. There is also a new 'PDC' button which will enable you to turn Plug-in Delay Compensation on and off depending on your recording and monitoring situation."

    But maybe they missed the midi clips then - to move midi according to PDC involved.
    #25
    brundlefly
    Max Output Level: 0 dBFS
    • Total Posts : 14250
    • Joined: 2007/09/14 14:57:59
    • Location: Manitou Spgs, Colorado
    • Status: offline
    Re:MIDI Weirdness... 2011/08/01 18:38:35 (permalink)
    But maybe they missed the midi clips then - to move midi according to PDC involved.



    Yes, my testing indicates they did forget to apply it to MIDI clips.

    SONAR Platinum x64, 2x MOTU 2408/PCIe-424  (24-bit, 48kHz)
    Win10, I7-6700K @ 4.0GHz, 24GB DDR4, 2TB HDD, 32GB SSD Cache, GeForce GTX 750Ti, 2x 24" 16:10 IPS Monitors
    #26
    lfm
    Max Output Level: -53 dBFS
    • Total Posts : 2216
    • Joined: 2005/01/24 05:35:33
    • Location: Sweden
    • Status: offline
    Re:MIDI Weirdness... 2011/08/02 04:33:35 (permalink)
    brundlefly



    But maybe they missed the midi clips then - to move midi according to PDC involved.



    Yes, my testing indicates they did forget to apply it to MIDI clips.


    Great, then I add that to Cake's bugtracker. Thanks.
    #27
    Jump to:
    © 2024 APG vNext Commercial Version 5.1