|
|
|
11-01-2016, 05:51 PM
|
#1
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
MIDI Timing Since 5.27
I noticed the following in the changelog for 5.27:
Quote:
+ MIDI: reduce Windows MIDI hardware output jitter
|
There was a lengthy discussion prior to all this about MIDI jitter which went on over the course of numerous threads and we ended up learning that setting the MIDI hardware outputs to TGT (low precision mode) made a significant improvement. However, MIDI timing has still been an issue at higher latency settings (it is ok at the lowest settings).
I'm curious to know what the changes mentioned above entail. Should we still be using TGT mode? Do the changes improve timing at higher latencies?
Any specific details of what this update did and what it changes would be great to know.
I use REAPER with both hardware synths and VSTi plugins. The MIDI jitter I'm most concerned about is with hardware synths.
|
|
|
11-01-2016, 06:21 PM
|
#2
|
Administrator
Join Date: Mar 2007
Location: NY
Posts: 15,821
|
The changes should improve MIDI hardware output jitter at all latencies, with any timestamping preference. The most important change was switching to a finer resolution timer internally (on both Windows and MacOS) when looking at the timestamped MIDI events and deciding exactly when to send them.
|
|
|
11-01-2016, 06:33 PM
|
#3
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Thanks heaps for the quick response Schwa. Much appreciated.
I guess leaving my output settings at TGT (low precision mode) would still give the best timing then?
|
|
|
11-01-2016, 08:33 PM
|
#4
|
Human being with feelings
Join Date: Jan 2010
Location: Kalispell
Posts: 14,759
|
Quote:
Originally Posted by ReaDave
Thanks heaps for the quick response Schwa. Much appreciated.
I guess leaving my output settings at TGT (low precision mode) would still give the best timing then?
|
Hi Dave, is this for your outboard midi modules, keyboards, etc..?
|
|
|
11-02-2016, 02:00 AM
|
#5
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by Tod
Hi Dave, is this for your outboard midi modules, keyboards, etc..?
|
Yep. Sure is.
After the lengthy discussion some of us had a while back, one of our resident REAPERites (might've been ELP), suggested changing the output to low precision (TGT) mode and that made a huge difference to the output jitter. It went from almost unusable to quite good just with that setting alone. There are still issues at higher buffer settings though.
|
|
|
03-21-2017, 05:02 PM
|
#6
|
Human being with feelings
Join Date: Jan 2011
Posts: 445
|
Quote:
Originally Posted by ReaDave
Yep. Sure is.
After the lengthy discussion some of us had a while back, one of our resident REAPERites (might've been ELP), suggested changing the output to low precision (TGT) mode and that made a huge difference to the output jitter. It went from almost unusable to quite good just with that setting alone. There are still issues at higher buffer settings though.
|
Is this really true? I would say you should disable the low-latency mode, at least after the update:
Before the update low-latency mode actually seemed like it had less jitter than normal latency mode. Keep in mind though that there is a bit of variance in the figures, but I did test and it's not that bad, but I would say there is no change in low latency mode. I used an ESI M4U and random HW synth to test (and the synth for sure contributed to the figures). The average can be calculated by taking the square root of the combined jitter figures. Seems like the update improved the average jitter by about .10ms, not bad.
All tips to minimze jitter are sure welcomed!
Last edited by Aesis; 03-21-2017 at 05:09 PM.
|
|
|
03-26-2017, 09:01 AM
|
#7
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Aesis, indeed the latest versions of REAPER work better allowing REAPER to do timestamping rather than using TGT mode. TGT is now the least accurate which is what I would typically expect when things are working correctly.
|
|
|
01-29-2018, 04:35 PM
|
#8
|
Human being with feelings
Join Date: Dec 2011
Location: Finland
Posts: 792
|
I would like to know what software/methodology Aesis used for those graphs!
|
|
|
01-29-2018, 05:20 PM
|
#9
|
Human being with feelings
Join Date: Apr 2008
Posts: 1,910
|
Quote:
Originally Posted by ReaDave
I noticed the following in the changelog for 5.27:
There was a lengthy discussion prior to all this about MIDI jitter which went on over the course of numerous threads and we ended up learning that setting the MIDI hardware outputs to TGT (low precision mode) made a significant improvement.
|
Say, I know nothing about this. It seems rather important.
Can you post any significant URLS?
|
|
|
01-30-2018, 06:13 PM
|
#10
|
Human being with feelings
Join Date: Dec 2011
Location: Finland
Posts: 792
|
I've found the software used: MLA - MIDI Latency Analyser V2
Unfortunately it has completely vanished from the internet since the page has been down for a year or so. ( Wayback Machine link)
I tracked down the developer on a few forums and I'll try to contact him regarding this.
If anyone has the software on their computer, now would be a time to upload it somewhere so it's available again till the developer gets their page together again.
|
|
|
Thread Tools |
|
Display Modes |
Linear Mode
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -7. The time now is 04:23 PM.
|