Sonar Timestretching Quality

Author
Xspringe
Max Output Level: -89 dBFS
  • Total Posts : 66
  • Joined: 2005/04/01 18:05:23
  • Status: offline
2005/05/26 14:21:32 (permalink)

Sonar Timestretching Quality

I noticed that changing the length of a wav file by +/- 10% (and sometimes as little as 5%) while maintaining the same pitch (timestretching) causes very significant artifacts, making this feature basically useless. The source files are 44/24. Is this normal behaviour or a problem with my setup?

I can post audio clips if required!

My hardware:
Athlon64
Asus A8V
FF800
Sonar 4.02
Windows XP
#1

21 Replies Related Threads

    daverich
    Max Output Level: -41 dBFS
    • Total Posts : 3418
    • Joined: 2003/11/06 05:59:00
    • Location: south west uk
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/26 14:38:01 (permalink)
    you need to use the correct setting - i.e. mono/polyphonic source and always use best quality.

    I assume you're using sonar pro and have the Mpex timestretching.

    Kind regards

    Dave Rich.

    For Sale - 10.5x7ft Whisperroom recording booth.

    http://www.daverichband.com
    http://www.soundclick.com/daverich
    #2
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/27 18:51:34 (permalink)
    ORIGINAL: daverich

    you need to use the correct setting - i.e. mono/polyphonic source and always use best quality.

    I assume you're using sonar pro and have the Mpex timestretching.

    Kind regards

    Dave Rich.


    I tried but that didn't make a whole lot of a difference quality wise. I will post up some examples later on.
    #3
    instantdan
    Max Output Level: -82 dBFS
    • Total Posts : 432
    • Joined: 2005/02/21 20:23:26
    • Location: Chicago
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/27 20:04:20 (permalink)
    I noticed the same thing sometimes. It is really bad when working with a stereo sample. When it does work I've had better luck using groove clip looping but then you need the original bpm and all that stuff (also uses RAM). Also I sometimes get a silent sample when I use timestretch. Feature could use some rehashing.
    #4
    name1432
    Max Output Level: -85 dBFS
    • Total Posts : 295
    • Joined: 2005/02/15 19:02:47
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/27 21:17:18 (permalink)
    Sounds bad to me too; but time-stretching is an effect that can be provided by a third-party plugin. So that Sonar can remain affordable, I actually hope Cakewalk won't spend their resources on competing with plug-in providers.

    edited to add: mine isn't pro, it's studio, without Mpex as far as i know
    post edited by name1432 - 2005/05/28 07:14:36
    #5
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/28 07:05:58 (permalink)
    ORIGINAL: name1432

    Sounds bad to me too; but time-stretching is an effect that can be provided by a third-party plugin. So that Sonar can remain affordable, I actually hope Cakewalk won't spend their resources on competing with plug-in providers.


    Is there any consensus on what the best time stretching application is?
    #6
    Master Chief [Cakewalk]
    Max Output Level: -69 dBFS
    • Total Posts : 1053
    • Joined: 2003/11/03 19:20:44
    • Location: Boston, MA, USA
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/28 09:25:38 (permalink)
    Is there any consensus on what the best time stretching application is?
    When we were developing SONAR 4 the consensus was MPEX, that's why we licensed it. It's the same time stretcher licensed by Steinberg and I think Samplitude.
    #7
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/28 13:18:52 (permalink)
    Here's two examples:

    Sonar at 90%: http://www.dikkechill.com/xspringe/sonar90.mp3
    Sonar at 110%: http://www.dikkechill.com/xspringe/sonar110.mp3

    I've got a few more examples, some are a lot worse but I haven't had the opportunity yet to transfer them from the studio.
    post edited by Xspringe - 2005/05/28 13:20:43
    #8
    Richard Brian
    Max Output Level: -40 dBFS
    • Total Posts : 3519
    • Joined: 2004/02/09 11:16:38
    • Status: offline
    RE: Sonar Timestretching Quality 2005/05/28 15:13:47 (permalink)
    Here are two pristine examples and the original using Best-Single Instrument:

    Edited

    Are you sure you're using the Time/Pitch stretch 2? There is a version 1 without the MPEX code, but it's not installed by default, but could be hanging around from a previous version install.
    post edited by Richard Brian - 2005/06/08 20:37:45
    #9
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/01 22:06:15 (permalink)
    I'm one hundred percent sure. You don't even want to hear the quality at >25% stretching, it's just godawful. Could there be a problem with my setup that is causing this?
    #10
    Qwerty69
    Max Output Level: -62 dBFS
    • Total Posts : 1435
    • Joined: 2004/02/19 17:44:10
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/01 22:11:59 (permalink)
    Are you correctly setting the option for a single instrument/voice vs. an ensemble?

    I am continually amazed at the sound quality of the MPEX v3 alogrithm as delivered in v4.0.2...

    Something isn't working correctly.

    Q.
    #11
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/02 07:04:36 (permalink)
    ORIGINAL: Qwerty69

    Are you correctly setting the option for a single instrument/voice vs. an ensemble?

    I am continually amazed at the sound quality of the MPEX v3 alogrithm as delivered in v4.0.2...

    Something isn't working correctly.

    Q.


    I will post additional samples with more details on the settings tonight after I get back from the studio. Personally I think something strange is going on, it almost sounds like the clock/speed at which the sound is being stretched is unstable.

    Thanks a lot for your support so far!
    post edited by Xspringe - 2005/06/02 07:06:44
    #12
    The Scar
    Max Output Level: -75 dBFS
    • Total Posts : 760
    • Joined: 2005/02/18 11:19:18
    • Location: Hackney Biatches
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/02 10:43:17 (permalink)
    Are you correctly setting the option for a single instrument/voice vs. an ensemble?

    I am continually amazed at the sound quality of the MPEX v3 alogrithm as delivered in v4.0.2...

    Something isn't working correctly.
    Ditto. I've been very impressed by the MPEX, but I do get weird artifacts when I get the settings wrong... I'd play around with the options (just try 'em all) 'cause stretching 10% isn't usually a problem.

    Electro Punk 'n' Roll at www.myspace.com/thescar
    #13
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/02 20:04:54 (permalink)
    I've created a recording for demonstrating the problem I am encountering with the sound quality of the time stretching in Sonar. The mp3 contains the following sounds: 1. Original 2. 125%, polyphonic 3. 125%, single instrument 4. 75%, polyphonic 5. 75%, single instrument. Quality was always set to "best" and no other settings were changed.

    http://www.dikkechill.com/xspringe/stretchtest.mp3


    On a side note, has anyone ever experienced Sonar all of a sudden ignoring certain automation envelopes? I just spent over two hours on automating various tracks and all of a sudden a significant part of the automation envelopes began to be ignored by Sonar. This has happened to me before but I thought I did something wrong back then, but now I did not change any settings as far as I can tell. Could this be a bug? If it is, it's a pretty evil one!

    #14
    Al
    Max Output Level: -35 dBFS
    • Total Posts : 4047
    • Joined: 2003/11/07 01:03:27
    • Location: NYC
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/02 20:47:11 (permalink)
    This has happened to me before but I thought I did something wrong back then, but now I did not change any settings as far as I can tell. Could this be a bug? If it is, it's a pretty evil one!


    I see some complaints lately about this one although the envelopes bug/bugs were reported as FIXED by Cakewalk in this version ( 4.02) ... hmmm ... maybe under some conditions the envelopes still stop doing what their supposed to do ( ? )

    ok, going to listen to your mp3 examples now ...

    Much fun with ours to natural resound! <G>
    #15
    Jake68
    Max Output Level: -86 dBFS
    • Total Posts : 224
    • Joined: 2003/11/23 06:23:04
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/02 21:52:01 (permalink)
    You are asking too much of it to stretch like that. You need to develop the "technique".

    Which is basically a combination of micro edits cross fades AND timestrectch chosen for its suitability over each factor of the sound.
    MPEX is the best on the right settings, however certain material will require more clarity than any timestretch alone can deliver.

    You might like to try a demo of melodyne, unfortunetely it has an interface design by a blind man and its as intuitive as a self build house, but it is the best at these jobs. I just wish they would develop version 5 all at once! so that it was at least as operable as Sonar Cubase etc.
    #16
    DroopyPawn
    Max Output Level: -77 dBFS
    • Total Posts : 665
    • Joined: 2004/04/19 23:55:04
    • Location: Fox, OK
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/02 22:25:12 (permalink)
    Doesn't anyone use TimeFactory anymore?
    #17
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/03 06:15:35 (permalink)
    ORIGINAL: Jake68

    You are asking too much of it to stretch like that. You need to develop the "technique".

    Which is basically a combination of micro edits cross fades AND timestrectch chosen for its suitability over each factor of the sound.
    MPEX is the best on the right settings, however certain material will require more clarity than any timestretch alone can deliver.

    You might like to try a demo of melodyne, unfortunetely it has an interface design by a blind man and its as intuitive as a self build house, but it is the best at these jobs. I just wish they would develop version 5 all at once! so that it was at least as operable as Sonar Cubase etc.


    Interesting. I'd like to know more about the "technique" :)

    I still find it rather curious though that even a 5% stretch introduces quite significant quality degradation issues with the soundfiles I'm testing on.
    post edited by Xspringe - 2005/06/03 06:21:29
    #18
    shea
    Max Output Level: -84 dBFS
    • Total Posts : 343
    • Joined: 2003/11/06 16:21:31
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/03 11:38:37 (permalink)
    When we were developing SONAR 4 the consensus was MPEX, that's why we licensed it. It's the same time stretcher licensed by Steinberg and I think Samplitude.


    But not protools ,its like apples and oranges.



    shea
    post edited by shea - 2005/06/03 11:41:27
    #19
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/03 11:40:41 (permalink)
    ORIGINAL: Xspringe

    On a side note, has anyone ever experienced Sonar all of a sudden ignoring certain automation envelopes? I just spent over two hours on automating various tracks and all of a sudden a significant part of the automation envelopes began to be ignored by Sonar. This has happened to me before but I thought I did something wrong back then, but now I did not change any settings as far as I can tell. Could this be a bug? If it is, it's a pretty evil one!


    I did some additional testing. It seems like the problem was limited to the volume envelopes this time. The strange thing is when I deleted the envelope and then undeleted them they worked fine again. Another possible clue for the cause of the problem is the fact that the mouseover information of all the envelopes that stopped working was put between parentheses and no longer displayed any value.

    User error or bug? And if it's a bug, where can I submit it?
    #20
    crash_3210
    Max Output Level: -72 dBFS
    • Total Posts : 916
    • Joined: 2003/11/23 06:13:26
    • Location: Suez, Egypt
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/05 09:08:20 (permalink)
    i had the same STOP-FUNCIONING envelopes .... t happened on a fade out of an ending of a song.... the 1st track (drums) (all audio by the way) refused to fade out and kept playing....i had to put a destructive editing method and faded it out inside SONY Sound Forge 8.0
    #21
    Xspringe
    Max Output Level: -89 dBFS
    • Total Posts : 66
    • Joined: 2005/04/01 18:05:23
    • Status: offline
    RE: Sonar Timestretching Quality 2005/06/05 10:05:47 (permalink)
    ORIGINAL: crash_3210

    i had the same STOP-FUNCIONING envelopes .... t happened on a fade out of an ending of a song.... the 1st track (drums) (all audio by the way) refused to fade out and kept playing....i had to put a destructive editing method and faded it out inside SONY Sound Forge 8.0


    Have you tried deleting and undeleting (CTRL-Z) the envelope? That fixed it for me. Also, does the malfunctioning envelope display it's paramater between parentheses and without a value on mouseover?
    #22
    Jump to:
    © 2024 APG vNext Commercial Version 5.1