**TEST** ARA vs. Drumagog for MIDI conversion

Author
emwhy
Max Output Level: -62 dBFS
  • Total Posts : 1402
  • Joined: 2006/01/03 15:09:02
  • Status: offline
2013/12/07 10:21:55 (permalink)

**TEST** ARA vs. Drumagog for MIDI conversion

Let me start by saying I'm a longtime SONAR user since Pro Audio 8 in '99. This test was done by me on a kick drum track to see which had the more accurate method of converting audio to MIDI for drum augmentation. It's not the most scientific test, but it yielded an interesting result.
 
Specs:
SONAR X3c with Melodyne Win 7 64
Drumagog 5 (32 bit thru jBridge)
M Audio Delta 2496 with a buffer setting of 64 (1.5 ms reported latency roundtrip latency 4.5 ms)
Project settings 44/24
Intel i7 8 GBS of RAM
 
First test was to convert the kick track to MIDI using the methods outlined on Cakewalk's blog (the sticky is at the top of the forum page for how to do this).
 
For the most part the conversion was clean, there were a few barely audible ghost MIDI notes that I easily erased in PV mode. The velocities were relatively high averaging around 105-110. Melodyne also for some reason places my kick track at MIDI note A5 which is quite high on the keyboard, so I had to transpose several octaves and a few notes to get it to trigger a kick in Kontakt. Overall it was right on with the original audio, no hits seemed off at all.
 
 
With Drumagog I had to switch it to Live Trigger mode, not a big deal, and also enable a virtual 3rd party MIDI cable in SONAR (Loop Be MIDI). I then had to load a kick sample and enable Drumagog's MIDI out to send on channel 1. The first difference is that Drumagog lets you determine which MIDI note you want to send. In this case I chose C1 (none of this is a deal breaker, two mouse click at most). I then created a MIDI track below the kick and set its input to Loop Be MIDI and let 'er rip. The main difference in the two tracks was the velocity, but that can be easily adjusted with either Drumagog's dynamic tracking option or SONAR's velocity slider on the MIDI track.
 
After this was done I compared both tracks using SONAR's event viewer looking at where the MIDI notes fell under HMSF and the results were IDENTICAL. Both methods yielded near perfect results. 
 
Bottom line, if you can't afford Drumagog, or Trigger and want to convert audio to MIDI for drum enhancement, SONAR and Melodyne's ARA integration does a very good job IMHO. 
 
Again this was not a huge scientific test, your results may vary, but I liked what I saw. We now return to our regularly scheduled programming.........
 
 
#1

4 Replies Related Threads

    brundlefly
    Max Output Level: 0 dBFS
    • Total Posts : 14250
    • Joined: 2007/09/14 14:57:59
    • Location: Manitou Spgs, Colorado
    • Status: offline
    Re: **TEST** ARA vs. Drumagog for MIDI conversion 2013/12/07 12:27:21 (permalink)
    emwhy
    Bottom line, if you can't afford Drumagog, or Trigger and want to convert audio to MIDI for drum enhancement, SONAR and Melodyne's ARA integration does a very good job IMHO. 
     
    Again this was not a huge scientific test, your results may vary, but I liked what I saw. We now return to our regularly scheduled programming.........

     
    Good info. I'll just add that SONAR's existing Audiosnap Audio-MIDI conversion works better than Melodyne in some ways when you don't need pitch info, and let's you specify what MIDI note number to generate - easily altered after the fact in any case. It also gives you some flexibility on what hits get converted. 
     

    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
    #2
    wmb
    Max Output Level: -77 dBFS
    • Total Posts : 658
    • Joined: 2004/03/18 00:49:16
    • Status: offline
    Re: **TEST** ARA vs. Drumagog for MIDI conversion 2013/12/07 13:03:52 (permalink)
    brundlefly
     
    Good info. I'll just add that SONAR's existing Audiosnap Audio-MIDI conversion works better than Melodyne in some ways when you don't need pitch info, and let's you specify what MIDI note number to generate - easily altered after the fact in any case. It also gives you some flexibility on what hits get converted. 
     


    Maybe it's just me but I've found Audiosnap audio-midi conversion much more time consuming for drum performances. I've not done the tests of the OP regarding this but any time I use AS I have to go in and look at EVERY SINGLE transient marker and confirm it's precisely where it should be. For some reason I will get many markers just slightly out in front of the actual hit. Makes me crazy. Honestly I don't expect Melodyne to do much better.

    X2 & X3c - GA-Z77X-UP5 TH mobo | i7 3770k | 16 gigs RAM | Win7/64 | Audio Drive 2TB RAID0 | Data & Backup 2TB RAID1 | OS drive 256 gig SSD | Apogee AD16x -> Tascam DM3200 -> IF-FWDM-mk2 via Apple FW to Thunderbolt adapter -
    #3
    emwhy
    Max Output Level: -62 dBFS
    • Total Posts : 1402
    • Joined: 2006/01/03 15:09:02
    • Status: offline
    Re: **TEST** ARA vs. Drumagog for MIDI conversion 2013/12/07 17:40:31 (permalink)
    wmb, not to worry Melodyne put all the hits where they should be. I too have tried the audio snap method, which lead me to purchase Drumagog several years back due to similar frustration. I routed the MIDI data to Kontakt and listen to both the MIDI kick and Audio kick and there were no off time notes.
     
     
    #4
    emwhy
    Max Output Level: -62 dBFS
    • Total Posts : 1402
    • Joined: 2006/01/03 15:09:02
    • Status: offline
    Re: **TEST** ARA vs. Drumagog for MIDI conversion 2013/12/10 13:52:11 (permalink)
    *****Bump***** Maybe the mods should move this under the drum month thread.
     
    #5
    Jump to:
    © 2025 APG vNext Commercial Version 5.1