Is it possible to use Bomes to 'thin' midi data on the fly?

[djmoonie]

Bento,

Would you be on a mac by any chance ?

As you indulged Jack, you'll know this issue only affects windows users. You will also know that I posted up a TINY midi mapping that can crash traktor (if you are affected by the midi out bug) in under a minute, 100% repeatable.

The issue may not be so clearly defined on this board, but trust me when I say that its been extensively tested to death (by me and others), and there are threads on the NI board (referenced by posted on this board by me), and on the VDJ boards about this issue.

[BentoSan]

Bento,

Would you be on a mac by any chance ?

I am using Windows of course lol

I am trying to get the VCI-100 to crash purely by spamming the device without touching it to get to the cause of the issue.

Your mapping doesn't clearly define what exactly the issue is as you have midi going into Traktor at the same time as midi is going out of Traktor - thats not showing whether if its a midi input issue or a midi output issue. Though it could be a combination of them both, thats what i am trying to uncover at the moment.

[djmoonie]

Sounds like your one of the happy windows users. If you send me your spam script for bomes I'll run it here.

[BentoSan]

Sounds like your one of the happy windows users. If you send me your spam script for bomes I'll run it here.

I am actually using Traktor to spam the device, i have multiple phase monitors mapped to midi outputs going into the VCI-100 - you can do obscene amounts of LED changes without crashing the VCI-100 this way which clearly demonstrates its not just a midi output issue.

I have a second midi output going to Bomes so i can monitor the amount of midi messages that are being sent by Traktor to the device using timers and a global variable that counts the midi messages that the VCI-100 would be receiving while the timer is activated.

I tried your mapping earlier and it does indeed skrew the VCI up, but like i said in my last post its still does not 100% clearly demonstrate as to exactly what is causing the issue. The tests i am doing at the moment are trying to get 100% clear picture of exactly what causes the issue to see if there is anything i can do about it.

[djmoonie]

I tried your mapping earlier and it does indeed skrew the VCI up, but like i said in my last post its still does not 100% clearly demonstrate as to exactly what is causing the issue. The tests i am doing at the moment are trying to get 100% clear picture of exactly what causes the issue to see if there is anything i can do about it.

k.

So basically you're running beat phase and just letting a track play and mapping it to loads of LEDs ?

With my mini mapping, try turning off midi-out and then try it again. Also, try taking out the output commands and try it again also. I bet with either adjustment you can't get it to crash.

Curiously tho, if you change the mapping from this:-

www.djmoonie.co.uk/traktor/crash.jpg

to this:-

www.djmoonie.co.uk/traktor/nocrash.jpg

It DOESNT crash, even tho the LED's flashing is the same. (although the outputs are both mapped to deck a).

[djmoonie]

Oh, and as for a new firmware update.. I thought that in one of Eans recent posts, he mentioned that vestax wasn't going to be making a 1.4 ?

[Jo3]

I think they are making 1.4 just with no 14 bit jogs etc.

[Jack Bastard]

Id still like to hear from more people who have tested disabling the midi output for an extended period of time to hear that they haven't had any dropouts - theres a few people who have claimed this but not enough people that i am 100% sure that this is the issue.

I'm sure that Moonie has already covered this, but let me confirm that this is the case. I'd describe myself as a semi-pro dj, I gig at least a couple of times each month in clubs in the UK and have a residency at a successful, long running night in my home city so a lot of my experience could be considered to be 'in the field'. Every other possible fix for his issue has been tried by me (barring buying a PSU, after I'd seen reports that it didn't work I didn't want to waste my money). The only fix for this issue is to disable midi out.

No promises though as i am still not entirely convinced that this is the issue - there just hasn't been enough users report that disabling the midi out fixes the issue.

I just think you haven't read enough around the issue, or you've been confused by people posting incorrectly in threads (hence my asking for people not to post unless their issue matches this exactly in the thread Moonie attempted to make on here). Anyone who has the issue will confirm that it is resolved by disabling the midi out. Anyone who says that this doesn't fix the issue didn't have it in the first place.

I'm sure you've already confirmed all of this by now, please feedback any findings you make.

[djmoonie]

Interesting that Bento said he can get his VCI to crash using my mini mapping.
I'm beginning to think this issue affects all windows users, its just that under normal conditions, its rare enough that many aren't sure they have it.

[BentoSan]

I have done a fair amount of testing and it seems that the crash only happens when trying to send a midi at the same time as receiving midi.

Mapping the phase meters to the leds and spamming a crazy huge amount of midi doesn't crash the unit and from what you guys are saying only sending midi doesn't crash the unit.

So i guess the only user fix that would be possible(short of disabling midi out completely) is to buffer the LED messages being sent to the unit and only send the messages when none of the jog wheels, knobs and faders are being touched - which definitely isn't as easy as it sounds, but it is possible.