Cher Geoff,

Thank you for answer and your time. To confess you all, what you explain me is quite technical (a little too much for me) My idea, is to be able to record in live my step, that they are played with the keyboard, or with my sequencer!
The sequencer was really to build loops with my Redrum or Tb 303 Vst (Reason) but it's not in sync with the rest (and notes played randomly + or - 2 rms) How I could show you a concrete example ? I'll try again with Ableton to see if it's not Reason that's the problem. Thanks a lot for your help

Yes, my aploogies if I get too technical. But, it's the technical things that really interest me. Especially the data inside a midi file.

If you can create an example of what's going wrong, get this into a midi file, and note which parts of what's there are supposed to be wrong, and which parts are not, that that will be a big help. So far, I have not really been able to tie together the data you've sent, and what you suggest is wrong (or not as you think it ought to be).

By the way, you can attach midi files to your messages, BUT. yes, this is a MIDI forum, but the file types (i.e. .txt, .zip etc) do NOT include .mid - yes, honestly. The way around this is to put any .mid files into a ZIP archive file, as .zip is OK.

Geoff

Hi,
I'm so pissed off... I don't want to be bothered for 30 years... I'm going to buy an external midi clock... and if it's still not totally in sync... I'll burn everything 😀

I'll share a Zip file with all the proofs of my problem tomorrow ! Thanks for your help and patience

One last question because I am lost ... I am looking for an external midi clock, but my 2 sequencers are usb. how can I find the right hardware for my configuration?

Didider

I look forward to seeing a complete midi file, this may shed light on the problem.

And apologies, but I must add - IF there IS a problem. No trace of anything in the previous midi file, but maybe something was hiding?

Hence, please supply notes with .mid, saying - NB at bar xx:xx:xx there is xxx happening but this SHOULD be yyyy.

Also, still need some info - how are units connected. A --> B --> C or

A --> [
B --> [ C i.e. A and B are in parallel into C, rather than in series
[

Geoff

Thanks for your answer.

Here is a Zip with 2 .mid files and a .png

Mid_Kick : kick 1/4 played from my DAW on Redrum drum machine ( midi master rhythm ). the 2nd sound which arrives is the one played from my Beatstep sequencer in Midi slave via USB and which drives the Kong of Reason.

Midi_HH : Kick 1/4 played from my DAW and its rhythm box Redrum ( midi master rhythm ), the 2nd sound is a HH on 3/4 played from my sequencer in Midi Slace via USB and which drives Kong of Reason

The image is 3 of the notes played (the HH) from my sequencer. You can see that each note is not quite at the same time (not to mention the huge delay not configurable from the Reason preferences (nor from my ESI sound card)

I understand that the problem is Windows but I'm still amazed that it's so out of sync and delayed.

According to my research, there is no Midi clock that would be able to fix the problem ( no Midi IN ports ) A usb HUB doesn't seem to be the solution either. Could a MAC virtual machine on my windows pc get around the problem ? Or the only solution would be to set fire to my BEATSTEP as well as to my NEKTAR AURA ?

I'm super angry and upset with the midi 😀

Thanks for your help

Didier,

OK, I see the .png file, but I don't know how this relates to the midi data files. The .png shows that the notes are all delayed from the beat as suggested by the grid. I do NOT know if the image refers to the start of the midi file, or some other point within the midi file(s).

You could explain these points?

I'm still working through the midi files. So far, I can see that the data in both midi files starts at tick 0, which is NOT what the .png file shows. BUT the .PNG might refer to a point later in the midi data? My number-crunching has not got there yet.

Also, as I follow the data through the midi files, I see a succession of notes of 1920 ticks long, each starts at a multiple of 15360 ticks, so all consistent/regular.

This data is one one track.

Then there is data on another track, this might not fit though. More number crunching. The fact that your midi files are working with a resolution of 15360 ticks per quarter note (rather than a more usual 480 or 960) makes for a lot of extra number crunching.

I'll get back to it.

Geoff

Didier,

When I'm doing this sort of thing, I use a utility prog which converts a midi file into readable text.

I attach the TXT files for the two midi files you sent me, so you can see what I'm looking at.

Can you see where things are OK, and where things are going wrong? I'll keep on with my study, but if you can see things, you can edit the file to draw attention to things, and send it back to me?

Geoff

I read your file well. it's a bit confusing for me.

About the png image ( it's in the middle of the track ( so you can see the time difference of the recorded notes.

Here is a test with midiview. on 1/4. Maybe you can see better?

I don't understand, I have the impression to be the only one to have this problem... I'm looking at installing macos virtually on my Windows (you see how far it goes) what a waste of time... I saw external midi host, at Kenton or Doremi, but it doesn't seem to me that it would suit me. And there is also Midronome which can be a solution... but I'm a bit tired of it 😀
Merci en tout cas de votre patience que je n'ai plus

Didier,

Well, the tick numbers are WAY too large and make things too difficult to see easily, so I decided to cheat.

I use the SynthFont program on my PC to do virtual synth messing, this is linked to a virtual synth (rather than the usual boxes I use for normal analogue midi).

So, I loaded in turn your two midi files, and looked at each using the Piano Roll display.

I see ALL the notes. Everything looks totally normally. Everything is dead on the beat. I see no problem. No hint of any problem.

I press the play button, and off they go. I don't get the right sounds, I just get a sort of piano sound, as I don't have the correct boxes. That's to be expected.

But I hear all the sounds, boom, boom, etc. Everything SOUNDS totally as I'd think it would, should, might - allowing for it being a piano rather than HH or kick. or whatever.

So, I do not SEE or HEAR anything wrong. I DO NOT!

Why do you?

We are looking at different things?

Something plays the notes, sends midi data to the computer, which saves the midi data which you send to me and this seems to be correct. As I SEE and HEAR.

BUT when you play the notes and send the midi data to something - what exactly??? - this does something different, and what is played, and YOU hear, is NOT correct?

What is different between what you do to create the data that you've just sent to me, and what you do when you just play and listen? This difference is where something is going wrong??

Geoff

The photo is still speaking, right?

here is another picture .

The 2 tracks were played with the Beatstep at the same rate. the top track was afterwards, quantified to stick perfectly. but impossible to quantify in LIVE. so when I play with my sequencer.. the result is desynchro so useless for what I wanted to do

Didier,

Yes, I agree, the image shows a non-sync situation. Do you have a midi recording of this??

So what are you showing in the midi files you've recently sent? If you're showing two clearly different things, then this is not helping explain anything.

If some of the midi data has already been quantized, then this is what the police would call 'tampering with the evidence' so it becomes useless. You agree?

What happens if you get a midi file containing the 'recorded' track, NOT Quantized, with the live track, as live?

If these are then both 'wrong', but in sync, this might be a step towards getting an answer re 'What is happening?' If they are both 'wrong', and NOT in sync, then ....???

Geoff

The .mid files are the recordings made in 2 different versions. The image comes from a new realization.

OK?

But I don't see anything wrong with the midi files. SHOULD I see something wrong there?

What is different with what you did to get the later image (which DOES show a problem) compared to what you did to get the midi files that seem to be OK?

Geoff

I send you 2 .mid files (made by exporting from reason)

The first one is the result of the live recording with the Beatstep Sequencer. The second one is the same one once you have quantized.

Unfortunately I can't be more precise. But I'll try with ableton to see if the problem doesn't come from Reason

Edit :

Here is an mp3 file of a recording made with FL Studio.

Only 1/4 with the sequencer ... you can notice that the rhythm changes tempo or am I crazy ?

Regarding the .mp3 file, I've just listened to it, twice.

No, I don't notice that it changes tempo. Just seems to me to be 12 or so regular beats. If it does change tempo, it's so slight that I cannot tell.

Just to pin it down, at what point does the change happen. i.e. at about beat nn?

I accept that you're having a problem of some sort, the question is what, what might be causing it, and how might it be fixed. I cannot do much (or even anything) from an audio file, whereas a midi file could offer some big clues.

The .csv file contains very little actual midi data, not at all sure what the point of this is?

Edit:

Listened again. No, no change of tempo.

I hear 17 beats, the file len is 12 secs. At what beat do you hear a change of tempo. Does it speed up, or slow down?

Maybe your computer is doing something funny, so you hear a change of tempo, but I do not. If this is so, it could explain a LOT

Geoff