Just discovered that nasty MIDI bug. Ouch! :-(

General questions about the fusion belong here.

Moderators: Jesse, parametric, jeebustrain

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

Don't rest too easily. :-)

Agreed, having both settings available is nonsensical. However, just turning off MIDI Clock RX while using MIDI Clock TX does not completely avoid the problem. I've reproduced it quite a number of times as noted above with MIDI Clock RX off. I even turned off both MIDI sequence control options and removed the MIDI IN cable just to limit to a single variable - scientific method. In such cases, it can still be produced with a very dense MIDI stream like the Bartok file. No two ways about it, there is a bug in the MIDI Out code probably something like I detailed above. Once the queue gets too long in response to dense MIDI data, something breaks and it never returns to normal until power cycled. That's a very reasonable explanation for the behaviour.

Give it a shot. Import that Bartok midi file, set it to loop the whole song and press play. Walk away for 30 minutes and likely when you come back, strange delayed and/or stuck notes will be playing that sound even stranger than the Bartok itself. You will also likely hear stuck notes on whatever module the Fusion is controlling until you press Stop a second time.

I think another reason this bug was not frequently reported is that most people are just playing the Fusion as opposed to controlling external modules. The Fusion doesn't really need external modules for most things. The internal note timing on the voice engines is never affected by the bug so many people who probably got the Fusion into that condition never noticed it because they were only listening to the Fusion. I imagine Alesis also had inadequate testing in regard to tests with external modules.
Last edited by 2ManyKeys on Mon May 28, 2018 10:22 pm, edited 5 times in total.

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

As most of us aren't Bartok attempting to jam as many notes as possible into our measures, this bug is not seen too frequently. However, it can show up even with normal playing combined with the Mod wheel or other dense controller data. I did it once that way.

Overall it's not a show stopper. If it happened all the time Alesis would have caught it in their QA testing.

User avatar
EXer
General
General
Posts: 103
Joined: Sat May 02, 2015 5:42 pm
Do you own a fusion?: Yes
Location: France

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by EXer »

Never trust Bartók to test MIDI timing :twisted:


I played Bartok's Mikrokosmos piece 149 on my 8HD when something utterly strange happened: the timing seemed completely AWOL.

Yet MIDI TX was OFF.


Suddenly I realized that I and my Fusion were playing correctly :idea:

I was just playing one of the 6 dances in Bulgarian rhythm and the strangeness was all in the 7:8 time signature of that piece :mrgreen:


Here it's a 5:8 time signature, but you get the idea:

https://www.youtube.com/watch?v=RZAeSFqCTAw
Fusion 8HD Klaus Schulze + 2 x Sector101 EXM-E3 128MB DRAM modules ; original 80 GB HD replaced with a SSD

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

:-)

LOL

Agreed. I'm a lover of many classical categories including impressionists, but Bartok is too far out there for me. I was amazed how many tempo changes were in that MIDI file!

Supposedly he was influenced by Debussy.
I'll take Debussy. :-)

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

If anybody wants to try that Bartok + MIDI Clock Tx test it would be good to see if this can be just as easily reproduced on anybody's Fusion. It certainly should be.

User avatar
Jesse
Site Admin
Site Admin
Posts: 1433
Joined: Thu Jul 07, 2005 5:49 pm
Do you own a fusion?: Yes
Location: Wilmington, NC USA

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by Jesse »

The Fusion is the only keyboard I have, one working, one not working.

It might be interesting to see if the problem persists using two Fusions :)
Tempus & Ludwig drums, Roto Toms-Djembe, Bongo's, Ibanez Bass, Fusion8HD, SUMO300 Amps, Alesis16 Firewire mixer, AKG D112-Behringer, B2-Shure, Beta 58a mics, Win10, Audigy soundcard, Audition3 & East West Goliath sound bank

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

Well then it's not a problem for you! :-)

I thought you got that busted Fusion fixed.

User avatar
Jesse
Site Admin
Site Admin
Posts: 1433
Joined: Thu Jul 07, 2005 5:49 pm
Do you own a fusion?: Yes
Location: Wilmington, NC USA

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by Jesse »

No, it needs a new mainboard unless I can get it repaired :(
Tempus & Ludwig drums, Roto Toms-Djembe, Bongo's, Ibanez Bass, Fusion8HD, SUMO300 Amps, Alesis16 Firewire mixer, AKG D112-Behringer, B2-Shure, Beta 58a mics, Win10, Audigy soundcard, Audition3 & East West Goliath sound bank

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

What does it do / not do?

User avatar
Jesse
Site Admin
Site Admin
Posts: 1433
Joined: Thu Jul 07, 2005 5:49 pm
Do you own a fusion?: Yes
Location: Wilmington, NC USA

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by Jesse »

I have made repairs to the top half and connected it to the bottom half of my working 8HD and the top half passes all diagnostic tests.

When I try to operate my non working 8HD there is no video, the main board controls the video output to the display so my guess there is a problem with a processor chip or circuit on the main board :(

I guess I need to test the 40 pin connector on both the working and non working Main Boards and compare the results.
Tempus & Ludwig drums, Roto Toms-Djembe, Bongo's, Ibanez Bass, Fusion8HD, SUMO300 Amps, Alesis16 Firewire mixer, AKG D112-Behringer, B2-Shure, Beta 58a mics, Win10, Audigy soundcard, Audition3 & East West Goliath sound bank

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

The cable is just a standard ribbon cable/connector. Unlikely failure point, but possible.

User avatar
Jesse
Site Admin
Site Admin
Posts: 1433
Joined: Thu Jul 07, 2005 5:49 pm
Do you own a fusion?: Yes
Location: Wilmington, NC USA

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by Jesse »

I already checked that, I actually made a new cable and checked the solder joints on the connectors, no visible cracks or bad solder joints :(
Tempus & Ludwig drums, Roto Toms-Djembe, Bongo's, Ibanez Bass, Fusion8HD, SUMO300 Amps, Alesis16 Firewire mixer, AKG D112-Behringer, B2-Shure, Beta 58a mics, Win10, Audigy soundcard, Audition3 & East West Goliath sound bank

2ManyKeys
Advanced
Advanced
Posts: 223
Joined: Mon Feb 12, 2018 3:55 am
Do you own a fusion?: Yes

Re: Just discovered that nasty MIDI bug. Ouch! :-(

Post by 2ManyKeys »

Interestingly enough, I recently picked up a Korg Radias which is also an awesome machine. I could not dump SYSEX out of that thing reliably until I shut off MIDI CLOCK transmit! So it appears the Fusion is not the only machine to have trouble sending high density MIDI data interleaved with MIDI clock.

Post Reply
  • Similar Topics
    Replies
    Views
    Last post