KVR Audio

hauke · Post by **hauke** » Thu Mar 03, 2016 8:27 am

Good morning

!

Hmmm.. this sounds interesting. I will see what I can do here, but next Thursday is the earliest date where I can do that.
Some thoughts and insights:

Apart from the double tempo issue: As long as Sundog is running as slave, the timing should be very solid (at least if S1 is sendig out the clock ticks constantly). Sundog responds to the S1 clock tick directly. As soon as the tick comes in, Sundog calculates the resulting notes. I would say that Musical Gym listened correctly here ("it sounds like it's keeping time with S1 metronome")

The BPM indicator is different: Sundog collects the ticks and then does a BPM calculation from time to time. A range from 179 to 181 BPM is a little bit strange, though. When a host is sending at 179.5 BPM, Sundog might jump between 179 and 180. But 181 is too much.

So, do I understand you correctly: Sundog responds / syncs to S1 in the right tempo (e.g. a quarter note is still a quarter note, not an eighth), but it displays the wrong tempo (double tempo)?

dsan: A big thank you from my side as well for helping here. I saw the thread, but I am not familiar enough with S1 (to be precise: not at all^^). So I was glad that you could jump in here.

Cheers,
Hauke

Musical Gym · Post by **Musical Gym** » Thu Mar 03, 2016 8:39 am

hauke wrote: So, do I understand you correctly: Sundog responds / syncs to S1 in the right tempo (e.g. a quarter note is still a quarter note, not an eighth), but it displays the wrong tempo (double tempo)?

I will need to analyze whether a quarter is a quarter, etc. and get back on that.
Yes, Sundog displays roughly double tempo.
Timing in S1 seemed consistent tonight but seemed to slow down/speed up earlier today. Also, I think Sundog got up to at least 106 from 90 when S1 was running at 45.
I deselected the Timecode checkbox in S1 to see what would happen. All seemed ok, but I don't know if that had an effect.
Thanks much for your response.

hauke · Post by **hauke** » Thu Mar 03, 2016 8:50 am

Musical Gym wrote: I will need to analyze whether a quarter is a quarter, etc. and get back on that.

That would be great, thanks! The BPM display routine and the note calculation routine are independent parts of Sundog. If a quarter is a quarter and the display is double tempo, then something is calculated erroneously on my side. However, if a quarter is played back in double tempo (the same temo as in the Sundog display), then I would conclude that S1 is sending clock ticks twice as fast as it should do for some reason.

Maybe it's also a good idea to test the whole thing with a MIDI debugging utility like MIDI-OX (Windows) or snoize MIDI Monitor (Mac OS X). Use one of these utilities as clock slave instead of Sundog, and it should give us a second opinion regarding the "real" tempo.

All the best,
Hauke

Musical Gym · Post by **Musical Gym** » Thu Mar 03, 2016 9:08 am

I'll see what I can do. You can also download S1 version 3 demo or even the free version of S1 and perform tests on your end. It might be easier that way.

One thing to add: Since dsan isn't having the problem in S1 version 2, perhaps it's either a version 3 related problem or something else entirely.

dsan@mail.com · Post by **dsan@mail.com** » Thu Mar 03, 2016 1:16 pm

Well, this is odd. TBH I've not paid attention to notice this previously so may just have missed it was happening, but,

I'm using Bitwig Studio at the moment, v1.3.2. Sundog Studio is v2.2.0 (I have the update 2.3.0 sitting on my desktop and one of these days I'll get around to installing it !

)

I have the tempo at 110 BPM. Sundog Studio is registering between 109 and 111 BPM during playback and sits at 110 BPM when stopped, but it seems to be in time no matter which pattern I switch to.

So I opened a new instance of S1 and hit the play button. Sundog Studio sits steady at 110 BPM during playback.

I'm using rptMIDI (over LAN) as my clock source.

@hauke - let me know if there is anything I can do at this end to assist in tracking down a solution. I'm wondering if this is related to the PDC issue that others talk about that I know nothing about

Cyoon · Post by **Cyoon** » Thu Mar 03, 2016 2:27 pm

I do get timing issues (jumping BPM values) in Sundog, when i do have several controller scripts, especially the OSC4Bitwig one running in parallel in Bitwig. This makes the midi clock juggling around, it is a difficult bug to fix i've been told.

As soon as i only send out the midi clock, with no other scripts attached Sundog works fine.

hauke · Post by **hauke** » Thu Mar 03, 2016 2:34 pm

Suloo wrote:I do get timing issues (jumping BPM values) in Sundog, when i do have several controller scripts, especially the OSC4Bitwig one running in parallel in Bitwig. This makes the midi clock juggling around, it is a difficult bug to fix i've been told.

As soon as i only send out the midi clock, with no other scripts attached Sundog works fine.

Is this the PDC issue that dsan is talking about? I didn't hear of PDC problems so far, maybe I missed a thread^^.
Sundog on sync is running as stable as the master can send the clock signal. If the master is sending unstable clock signals, Sundog will respond with wrong timing as well.

All the best,
Hauke

Cyoon · Post by **Cyoon** » Thu Mar 03, 2016 3:17 pm

no, has nothing to do with pdc as far as i know, it is the underlying C++ architecture that gets wobbly as soon as lots of scripts run in parallel. It gets worse with bigger projects, so maybe pdc kicks in additionaly. Just sent a request about this to Bitwig again.

Probably nothing you can do about yea, but sending in a request as well might bring it up to a higher prio.

Musical Gym · Post by **Musical Gym** » Thu Mar 03, 2016 8:31 pm

My results in S1 last night seemed stable, which is what I'd be most concerned with if S1 were jumping around, which it wasn't last night.

Just a thought: Sundog has many arps which are varied lengths. If I have six of these loaded up and they're all different, would that potentially introduce timing errors?

Do measure/bar/loop settings in S1 need to match Sundog? These can also be varied per channel in Sundog, no? Seems like a lot of timing to monitor.

dsan@mail.com · Post by **dsan@mail.com** » Thu Mar 03, 2016 9:15 pm

Since Sundog plays in 4/4 time only, MG, that should have no bearing on tempo.

However, if you have measures in S1 outside of 4/4 then that could cause time issues but should not affect tempo. It would be more like the guitar player ahead of the drummer on the beat. Most often it's the other way around

I think, like Suloo, the problem lies in the host's ability to send proper syncing and it seems the more devices that MIDI clock has to be sent to is causing the timing problems; if they can be called problems.

By that I mean that so far, no matter what tempo is being shown Sundog Studio appears to be staying in time.

But you do raise a really good point about time stretching. It remains to be seen this relevancy.

Musical Gym · Post by **Musical Gym** » Thu Mar 03, 2016 9:39 pm

I was thinking that varied stop/start messages might confuse S1 when sending midi sync messages back to Sundog.

hauke · Post by **hauke** » Fri Mar 04, 2016 8:38 am

Musical Gym wrote:I was thinking that varied stop/start messages might confuse S1 when sending midi sync messages back to Sundog.

Could you slave another program to S1 instead of Sundog? It would be really interesting to know if this program displayed double tempo as well.

Musical Gym · Post by **Musical Gym** » Fri Mar 04, 2016 10:02 am

Tried with RapidComposer, but so far not getting RC transport to move.

Will report back results if I can obtain them.

Musical Gym · Post by **Musical Gym** » Fri Mar 04, 2016 10:51 pm

Still not getting transport on RC to move and Liquid Notes seems to have Java problem.
If I get anything working and testable, will report back, but not sure if problems will be handled soon.

hauke · Post by **hauke** » Sat Mar 05, 2016 10:30 pm

Musical Gym wrote:Still not getting transport on RC to move and Liquid Notes seems to have Java problem.
If I get anything working and testable, will report back, but not sure if problems will be handled soon.

The wonderful world of software

..
Thanks for testing, and good luck!

Sundog Scale Studio 2.2 coming soon: Undo/redo BETA available