Another Midi-sync issue

[msascha]

Hi,

I don't know if I'm the only one using Tracktion to sync his external sequencers, but I discovered another - in my eyes pretty nasty - behaviour. When I have Tracktion running every now and then there is a pause in the playback of my hardware sequenzer. It stops for a second or so and than continues.

I don't have this behaviour with other Midi-sync masters (Ableton Live! or Cubase). Before somebody asks - no, I don't simple what to use Live! or Cubase instead of Tracktion

[chico.co.uk]

Do you don't simple what to use Live! or Cubase instead?

Oh, sorry, you already addressed that ...

Is it the Tracktion 2 demo? You've not purchased it yet? Because the demo sends an all-notes-off message to external hardware as a demo restriction. This information shows as a popup every time you open tracktion. So if it's the demo, this is completely normal behaviour, which the full version doesn't do.

[msascha]

I upgraded from T1 in the minute the anouncement mail from Mackie hit my inbox. So yes, I purchased T2.

[IIRs]

Are you using MTC or clock? If MTC check you have the same frame-rate set for all devices..

[msascha]

I'm using MIDIClock not MTC.

Btw. I'm not aware of any reasonably priced external MIDI gear, that supports MTC. MTC is something for external HDRs and alike, but I've never seen a step-sequencer supporting MTC (if there is any let me know, I'm willing to learn).

[valley]

OT: I honestly can't see any good reason for devices to use MIDI clock for critical timing duties. It sucks.

[msascha]

MIDI Clock doesn't suck, it works reasonably well. MTC is nothing else than SMPTE translated to a MIDI specification. MTC and MIDIClock are two totally different things.

MIDIClock is the choice when you want nothing more than (relative) information about the tempo, MTC is the choice when you want absolute timing information (for instance for controlling an external HDR, tape-machine or something alike). For the synchronization of any step-sequencer like device MIDIClock is still the choice to make.

[customdigi]

Msacha,

Are you using a Mac? I had exactly the same problem. Was using T1 with the Mackie Spike, and had horrible midi problems when using T as my sequencer to control my Triton.

I dl'd the T2 demo to see if it was cleared up, and it wasn't. I thought maybe the reason was that I had to mucnh going on with the USB bus and it was affecting the midi timing, so I returned the spike for an m-audio firewire 410...SAME problem. So unfortunately, I returned to trusty pro-tools (I REALLY wanted to make switch to T) "m-powered" and the midi is rock solid.

I noticed the midi timing mostly when resizing windows or zooming in/out...don't know if this is the same for you, but it didn't feel reliable, ever.

So far, it seems as if only MAC users are experiencing this problem..we are not alone.

Anyway, good luck with T2, everyone. I really love the interface and work flow, it just wasn't the right choice for me based on my experience.

--Matt

[valley]

MIDI Clock doesn't suck, it works reasonably well.

For critical timing duties, it sucks.

MIDIClock is the choice when you want nothing more than (relative) information about the tempo,

Indeed. It works great for syncing hardware delays, LFOs, or even none position locked, arps.

That's all rather different from any kind of position dependent device such as another sequencer, or say, an exteral step sequencer. Without any form of error correction, how can you ever be sure you're playing to the correct tick, and once more than 24 quarter notes has passed, even the right quarter.

For the synchronization of any step-sequencer like device MIDIClock is still the choice to make.

No. It's the cheap alternative. The right choice to make is a position locked format that cannot possibly drift out of time. IE, MTC.

[diverdee]

Is the point of this thread a MTC vs MIDIclock comparison? from reading the initial post it is my understanding that msascha has no problems using MIDIclock to synchronise his external devices with his other hosts live & cubase, only with tracktion.
Therefore comments to the effect of 'MIDIclock sucks' have no relevance, as it obviously doesn't 'suck' when used with these other hosts.
If one wants to use suck metaphors here it would appear that tracktion 'sucks' in this particular instance, as it is the only host where this 'sucky' behaviour manifests itself.
msascha didn't state that tracktion sucks though, he stated that he had a problem with tracktion which he doesn't have with other hosts when using this protocol - so telling him 'well use a different protocol cos' the one your using sucks' is the kind of answer that would befit a troll or a juvenile fanboy more than a moderator, & i say that with the utmost respect for valley's past contributions etc.

[valley]

Is the point of this thread a MTC vs MIDIclock comparison?

No. Which would be why I prefixed my post with an OT, now wouldn't it?

Therefore comments to the effect of 'MIDIclock sucks' have no relevance, as it obviously doesn't 'suck' when used with these other hosts.

No. It does suck. It sucks for the simple reason that it is a poor synchronisation method for doing position dendent duties.

If one wants to use suck metaphors here it would appear that tracktion 'sucks' in this particular instance, as it is the only host where this 'sucky' behaviour manifests itself.

Give it a rest. Try reading the post before getting on your high horse.

msascha didn't state that tracktion sucks though, he stated that he had a problem with tracktion which he doesn't have with other hosts when using this protocol

I didn't tell him there wasn't a problem with MIDI clock. I said, as an off topic, that MIDI clock is not a good choice of protocol for this kind of thing.

MSasha chose to challenge that which is unfortunate because his technical understanding of what MIDI clock does is clearly wrong.

I don't care what the issue at hand is, if someone posts as fact a claim that I know to be false, I will generally respond, and for the simple reason that leaving misconceptions unchallenged breeds other misconceptions.

[diverdee]

fair point on the OT - but whatever one's views on MIDIclock maybe, the fact is that his other hosts DO work ok using this protocol, tracktion DOESN'T - so he does raise a valid point, a problem with tracktion's implementation would seem the most likely explanation given the circumstances.
It appears that customdigi had a similar problem, so now it seems it isn't an isolated incident, & may perhaps be mac related.
This is why i think a bug report sticky & a clearly defined method for reporting & verifying bugs would be useful here.

[diverdee]

actually no not a fair point on the OT - as it wasn't off topic at all - as the topic is about MIDIclock & tracktions implementation of it compared to other hosts.
By saying 'OT' that MIDIclock 'sucks' in this thread you are implying that the fault is with 'sucky' MIDIclock & not tracktion, which seems unlikely (given that other hosts work fine).
That reads more like a blanket defence of tracktion, ignores the real issue & still seems like inappropriate behaviour (for a moderator) meant to obscure msascha's issue.

[msascha]

@customdigi
Yes, I'm using a Mac... A G5/2x2GHz to be exact, but even my trusty old Atari ST with Cubase had better MIDIClock timing than Tracktion.

@valley
MIDIClock has its uses, for instance for me. To be honest even simple Roland-style SYNC would be enough for me. I just want it to be as tight as possible and that's something Tracktion can not deliver at this time.

[chico.co.uk]

By saying 'OT' that MIDIclock 'sucks' in this thread you are implying that the fault is with 'sucky' MIDIclock & not tracktion, which seems unlikely

No he isn't. As he's already said. Learn to read.