Page 1 of 1
Combi possible bug
Posted: Mon Dec 16, 2013 12:49 am
by Redrain
Hi
when I use a kronos to play external modules something strange happen :
if I play only an expander setting EX2 as status, the expander work normally.
if I enable another channel in order to play 2 expander (so I set another channel status to EX2), I have problems using the lower joystick as mod weel (pushing it forward) .
the situation is this :
timbre n. 6 in EX2 mode midi channel 6 playing Novation supernova II rack
timbre n. 7 in EX2 mode midi channel 7 playing Roland JD990
if I set timbre 6 and 7 to EX2 mode when I use the joystick as mod weel the note played are delayed (this happen expecially if i move very quinck the joystick forward and backward)....and it do not depend on the expander because if I set only a tinbre to EX2 and the others to OFF the joustick/mod weel work great !
it seems that the Kronos can't be able to menage too many midi data ...
I use MIDI-OX to see the midi data coming out from my kronos and, if I use 2 track set EX2 and move the joystick 5 times forward quinckly I see the data come out from the kronos also 1 second after I stop to move the joystick (this happen only when I move it on the Y axis and not on the X axis used normally for pith bend)
if I use only one channel set to EX2 the problem is not present...
thank you
roberto
Posted: Fri Dec 20, 2013 10:56 pm
by Redrain
Hey
this is a heavy BUG
Posted: Sat Dec 21, 2013 12:38 am
by ronnfigg
Redrain wrote:Hey
this is a heavy BUG
I don't think it's a bug. Probably something to do with the receiving unit's MIDI implementation. I believe you are talking about the Pitchbend lever- NOT the joystick. There is a joystick on the Kronos. The description of your problem is a bit difficult to understand especially when you are not using the right terminology. If you are daisy chaining the 2 external modules be sure you are using the "Thru" connector to connect module 2 to module 1.
Posted: Sat Dec 21, 2013 2:04 am
by X-Trade
I have actually perceived some latency usually in note off messages on even internal sounds if I have been playing heavily with the pitch/mod joystick.
However this is nothing compared to the MS2000 and MicroKorg's arpeggiator skipping notes if you use the pitch or mod wheel at the same time.
Just seems to be a data bandwidth limitation. Pitch Bend messages can be special high-resolution messages, I'm not sure if the Kronos implements this MIDI feature though, but they would take up twice as much bandwidth as a normal control change.
It is possible though that the Kronos is sampling the joystick position too often and triggering more messages than it could get away with, causing a bottleneck which prevents other messages from being sent. This is purely a performance optimisation. The real problem there is the relatively limited bandwidth of the traditional 5-pin DIN socket MIDI standard. It has been around for 30 years now.
Posted: Sat Dec 21, 2013 2:28 am
by Redrain
Hi
ok, it's a pitch bend lever ("using the lower joystick", and there is only one lower joystick)
...but...
the problem is the Kronos because :
the problem is present in the midi monitor output of MIDI-OX, so without no one expander connected with the Kronos ;
the Kronos is connected to the PC via USB;
My expander work good played with Prophet 08 and Roland JP8000.
The supernova is connected via midi out of the kronos..end...
the JD990 is connected to the Soundcard that is connected via firewire to the pc, MIDI-OX connect the usb of the Kronos with the soundcard..
...but...
the problem is present as seen in the midi monitor
roberto
Posted: Sat Dec 21, 2013 3:08 am
by geoelectro
I wonder if this is USB related...can you replicate this except using the MIDI port?
Geo
Posted: Sat Dec 21, 2013 4:45 am
by ronnfigg
I don't understand why people want to go with the "bug" thing right out of the gate.
Posted: Sat Dec 21, 2013 5:47 am
by ronnfigg
Good point Geo. I was going to recommend the same thing. Sounds like his current set up has a complicated MIDI stream.
Posted: Sat Dec 21, 2013 9:20 am
by Redrain
ronnfigg wrote:Good point Geo. I was going to recommend the same thing. Sounds like his current set up has a complicated MIDI stream.
Korg Kronos USB to PC USB is a complicated stream ??
Posted: Sat Dec 21, 2013 9:33 am
by ronnfigg
Redrain wrote:ronnfigg wrote:Good point Geo. I was going to recommend the same thing. Sounds like his current set up has a complicated MIDI stream.
Korg Kronos USB to PC USB is a complicated stream ??
You also claim you are using the Kronos MIDI out and a Firewire connection with a "soundcard". that sounds like something beyond the standard fare. And all of this goes to a PC? Sounds like you are taking it to the limits, but then I don't do much with my Kronos hooked up to a computer, so maybe I don't know what I am talking about. I just don't think you should call this a Kronos "heavy bug". Since you are using a variety of different technologies, different manufactures hardware and 3rd party software, you problem cannot necessarily be attributed to a Kronos "bug".
Posted: Sat Dec 21, 2013 9:36 am
by Redrain
Ok I did the test :
very simple stream :
Kronos (midi out) Supernova (midi in ) then Supernova (midi thru) JD990 (mid in )
if I more the pitch bend forward the Supernova and the JD 990 sound retarded....
I repeat, U must move the pitch bend forward very quickly like 5 time in one second !
the BUG don't happen if I set to OFF one of the two timbre so..
with only a timbre set to EX2 all work fine
with 2 or more timbre set to EX2 the BUG is present here !
all this don't happen if I change the midi master keyboard witn my Prophet 08.....the 2 expander sound without delay on the note if I move the mod weel.....
now NO PC was used
Posted: Sat Dec 21, 2013 8:00 pm
by ronnfigg
if you are so sure this is a bug then I recommend you send a report to Korg support. I hope you find your solution.
Posted: Sat Dec 21, 2013 11:16 pm
by X-Trade
It does definitely sound familiar. As I said, I have experienced similar things although I have not tested it extensively.
It seems to me that the joystick is sampled for controller data far too often and floods the MIDI stream when used too enthusiastically.
Posted: Fri Jan 03, 2014 11:57 pm
by danatkorg
I tested Combi cc output from four simultaneous MIDI channels, using four KARMA modules to generate arpeggios and CC#10 data while I manually moved the joystick to generate CC#1. Everything stayed in sync, and when I stopped playing everything stopped immediately. Below is a screen-shot of the output as recorded into my DAW via 5-pin MIDI (the joystick is the yellower line, as hilighted with black dots in the second channel). The CC output resolution seems sufficient and not overly dense. My conclusion is that the KRONOS appears to be working properly here; if you continue to hear laggy response on your externally controlled synths, I would look at the other parts of your MIDI system. Hope this helps!

Posted: Sat Jan 04, 2014 12:08 am
by danatkorg
As a corollary, here's the same Combi with me wildly moving the joystick back and forth on the pitch-bend axis. The results seem similar to the CC example above: everything stays in sync, and stops instantly when I stop playing.
