EMI Timing Issues and MIDI Instability

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

18 Jun 2018

Just did a quick test, multiple buffer settings, 16ths from Dual ARP to EMI via USB to OP1 (teenage engineering), back to Reason via Focusrite Clarett 2Pre.

Worst case, @1024 buffer, was 2ms late (one note out of 64). Average less than ±1ms, certainly not enough to hear/feel.

Worst case at 64 samples was higher (3-4ms, 3 out of 64) for ±2ms total (almost all notes were late after correcting for latency).

Doesn’t seem to correlate with other experiences, not sure what to take away from this?
[emoji3]


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
chimp_spanner
Posts: 2908
Joined: 06 Mar 2015

19 Jun 2018

I sometimes do work with outboard stuff and I can't say I've had any problems (beyond latency inherent to certain devices, like the Boutiques). Otherwise it seems pretty stable to me. The only thing I'd really kill for is per track negative delay, and multiple MIDI clock destinations. The lack of both these things makes working with hardware a little more difficult than I'd like.

User avatar
Carly(Poohbear)
Competition Winner
Posts: 2871
Joined: 25 Jan 2015
Location: UK

19 Jun 2018

I decided to look at the EMI as others are looking at external devices.

I was a little surprise that there was a difference between a low buffer size and a higher one as I was using a loopback midi so my midi would not go near any other hardware.

Using a Low buffer @88 the avg 3 ticks off (lowest being -4 and highest 11 ticks out of 96 notes)

Using a buffer of @225 there were no negative ticks, avg 8.19 ticks off (lowest 0 and highest 16 ticks out of 96 notes)

Using a buffer of @440 there were no negative ticks, avg 16.75 ticks off (lowest 1 and highest 30 ticks out of 96 notes)

Note notes were 1/16th and tested at BPM 180

ticks (ms rounded down)
240=1/16th=83ms
120=1/32th=41ms
60=1/64th=20ms
30=1/128th=10ms
15=1/256th=5ms
7.5=1/512th=2.5ms

Also did a quick test in Reason 7, was seeing the same results.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

19 Jun 2018

Carly(Poohbear) wrote:I decided to look at the EMI as others are looking at external devices.

I was a little surprise that there was a difference between a low buffer size and a higher one as I was using a loopback midi so my midi would not go near any other hardware.

Using a Low buffer @88 the avg 3 ticks off (lowest being -4 and highest 11 ticks out of 96 notes)

Using a buffer of @225 there were no negative ticks, avg 8.19 ticks off (lowest 0 and highest 16 ticks out of 96 notes)

Using a buffer of @440 there were no negative ticks, avg 16.75 ticks off (lowest 1 and highest 30 ticks out of 96 notes)

Note notes were 1/16th and tested at BPM 180

ticks (ms rounded down)
240=1/16th=83ms
120=1/32th=41ms
60=1/64th=20ms
30=1/128th=10ms
15=1/256th=5ms
7.5=1/512th=2.5ms

Also did a quick test in Reason 7, was seeing the same results.
Did you compensate for latency? After compensating (higher buffers, higher latency) I saw no real difference in jitter - if anything it was worse at higher buffer settings.

I notice the jitter was 16 ticks for both 88 and 225 buffer settings, which is why I wonder about latency.

BTW, what is the first value in your table above (240 in the top row)?

Could the differences I’m seeing have to do with the device used, or the audio interface, or what?


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
Carly(Poohbear)
Competition Winner
Posts: 2871
Joined: 25 Jan 2015
Location: UK

19 Jun 2018

I did not compensate as what is there to compensate when at the midi level going from software to software..

Being at the midi level I did not expect buffer size effecting this...

There are 240 ticks per 1/16th notes (visible, the scale goes down more but you only see a * on the position marker in Reason).


Working this stuff out is going to be hard\fun as everything the midi signal goes through can effect the timing..

jimmyklane
Posts: 740
Joined: 16 Apr 2018

19 Jun 2018

My experience has been that at extremely low buffer settings my jitter problems all but disappear. High buffers like 1024samples make for an unlistenable experience for me
DAW: Reason 12

SAMPLERS: Akai MPC 2000, E-mu SP1200, E-Mu e5000Ultra, Ensoniq EPS 16+, Akai S950, Maschine

SYNTHS: Mostly classic Polysynths and more modern Monosynths. All are mostly food for my samplers!

www.soundcloud.com/jimmyklane

User avatar
Ahornberg
Posts: 1904
Joined: 15 Jan 2016
Location: Vienna, Austria
Contact:

20 Jun 2018

On smaller buffer size MIDI-out will have more jitter because audio-out comes first.
Audio dropouts are more audible than MIDI timing issues.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

20 Jun 2018

Interesting - seems the audio buffer size is tied to MIDI. Maybe because of what Ahornberg says (audio priority), which would mean that the driver/device used affects the MIDI response, and would explain the timing differences we are seeing when comparing low buffer settings on different hardware. Maybe? Further tests are warranted…


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
Carly(Poohbear)
Competition Winner
Posts: 2871
Joined: 25 Jan 2015
Location: UK

20 Jun 2018

Ahornberg wrote:
20 Jun 2018
On smaller buffer size MIDI-out will have more jitter because audio-out comes first.
Audio dropouts are more audible than MIDI timing issues.
But that is backwards from what Jimmy and I have seen...
selig wrote:
20 Jun 2018
Interesting - seems the audio buffer size is tied to MIDI. Maybe because of what Ahornberg says (audio priority), which would mean that the driver/device used affects the MIDI response, and would explain the timing differences we are seeing when comparing low buffer settings on different hardware. Maybe? Further tests are warranted…
I still can't fully see how buffer size would be tied to midi, could it have an effect on the priority of signals? I could only see Reason effecting (controlling) that part, yes hardware could at the micro level, in my case my midi is not going to hardware (except for the computer it's running on) so Reason is the only real thing that could effect it.
If I was to look my full midi setup, I have the computer midi interface, a midi merge\spliter box, Keyboard using Midi (soft)thru to another keyboard, all of them points can and will effect the signal timing, we are all use to that sort of latency.. Jitter is down to a timer not timing correctly.... Maybe it's time to re-test using external timers....

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

20 Jun 2018

Carly(Poohbear) wrote:
20 Jun 2018
I still can't fully see how buffer size would be tied to midi, could it have an effect on the priority of signals? I could only see Reason effecting (controlling) that part, yes hardware could at the micro level, in my case my midi is not going to hardware (except for the computer it's running on) so Reason is the only real thing that could effect it.
Well the thing is Reason has to sync everything in some way too.. 1024 samples from the interface, 64 internally, the MIDI clock is running at a certain speed etc. In Logic the audio timing was tied to the MIDI ticks for a long, long time.

jimmyklane
Posts: 740
Joined: 16 Apr 2018

20 Jun 2018

Wait!!! I just thought of something that might effect all of our results.

Up in the master section, we have “input focus” and “play focus”....I’ve always had BOTH checked. I’m in the middle of a mix right now and can’t really do the testing, but I wonder if this changes things? Opinions?
DAW: Reason 12

SAMPLERS: Akai MPC 2000, E-mu SP1200, E-Mu e5000Ultra, Ensoniq EPS 16+, Akai S950, Maschine

SYNTHS: Mostly classic Polysynths and more modern Monosynths. All are mostly food for my samplers!

www.soundcloud.com/jimmyklane

User avatar
Carly(Poohbear)
Competition Winner
Posts: 2871
Joined: 25 Jan 2015
Location: UK

20 Jun 2018

jimmyklane wrote:
20 Jun 2018
Wait!!! I just thought of something that might effect all of our results.

Up in the master section, we have “input focus” and “play focus”....I’ve always had BOTH checked. I’m in the middle of a mix right now and can’t really do the testing, but I wonder if this changes things? Opinions?
From what I read it won't effect my results as I only have one document open..

From manual
Input Focus and Play Focus
If you activate (external) MIDI Clock sync on the Options|Sync menu, the Transport Panel controls will be disabled,
and Reason will not run unless MIDI Sync data is provided from an external device.
The Input Focus (MIDI + Audio) and Play Focus (MIDI Sync) buttons (located on the Reason Hardware Interface) relate
to how incoming MIDI and MIDI sync should be handled if there are several open Song documents. If you have
two or more Songs opened, and no MIDI sync is used, the currently selected Song (the document “on top”) always
has Input focus.
If (external) MIDI Clock sync is enabled (which is global for all currently open Song documents), this functionality
changes in the following way:
• If both “Play Focus” and “Input Focus” are activated for a Song, incoming MIDI data, MIDI sync (and Audio)
will be sent to this Song, regardless of whether another Song is currently in focus.
• If only “Input Focus” is activated for Song, and another Song has “Play Focus”, incoming MIDI (and Audio) will
be sent to the former and MIDI sync to the latter (i.e this Song will play back), regardless of which Song is currently
in focus.

jimmyklane
Posts: 740
Joined: 16 Apr 2018

20 Jun 2018

Carly(Poohbear) wrote:
20 Jun 2018
jimmyklane wrote:
20 Jun 2018
Wait!!! I just thought of something that might effect all of our results.

Up in the master section, we have “input focus” and “play focus”....I’ve always had BOTH checked. I’m in the middle of a mix right now and can’t really do the testing, but I wonder if this changes things? Opinions?
From what I read it won't effect my results as I only have one document open..

From manual
Input Focus and Play Focus
If you activate (external) MIDI Clock sync on the Options|Sync menu, the Transport Panel controls will be disabled,
and Reason will not run unless MIDI Sync data is provided from an external device.
The Input Focus (MIDI + Audio) and Play Focus (MIDI Sync) buttons (located on the Reason Hardware Interface) relate
to how incoming MIDI and MIDI sync should be handled if there are several open Song documents. If you have
two or more Songs opened, and no MIDI sync is used, the currently selected Song (the document “on top”) always
has Input focus.
If (external) MIDI Clock sync is enabled (which is global for all currently open Song documents), this functionality
changes in the following way:
• If both “Play Focus” and “Input Focus” are activated for a Song, incoming MIDI data, MIDI sync (and Audio)
will be sent to this Song, regardless of whether another Song is currently in focus.
• If only “Input Focus” is activated for Song, and another Song has “Play Focus”, incoming MIDI (and Audio) will
be sent to the former and MIDI sync to the latter (i.e this Song will play back), regardless of which Song is currently
in focus.
Yeah, that was a RTFM moment, my bad :-(
DAW: Reason 12

SAMPLERS: Akai MPC 2000, E-mu SP1200, E-Mu e5000Ultra, Ensoniq EPS 16+, Akai S950, Maschine

SYNTHS: Mostly classic Polysynths and more modern Monosynths. All are mostly food for my samplers!

www.soundcloud.com/jimmyklane

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 18 guests