RE: jammidimystery

[Phil Diem <pdiem@edcen.ehhs.cmich.edu>]

Bob,

In reply to my ongoing problems trying to get my two Jamthings to buddy 
up via midi connection you wrote:

>I'm not familiar with the term "brothersync". Are both systems sending
>sync information to eachother?

As I am now begining to more fully understand the term, which Kim Flint 
credits Matthias Grob with coining, yes - no slave, no master, brothers. 
Apparently the Echoplex has the capability to interact in this manner.

>Regardless, the JamMen can't be configured to cleanly work in stereo.
>Unfortunately, it was not considered when the product was spec'd and
>designed, wasn't tested in-house and apparently doesn't work.

Why does page 21 of the Jammanual contradict this statement? (This is a 
rhetorical question. I do not disbelieve your statement, nor do I believe 
that you are in any way responsible for what is printed in the manual.)

>When asked about this before, the only suggestion I could come up with 
>was to 
>feed the MIDI commands to the two independent systems in parallel and run
>them open loop. The lock in this configuration is coarse at best however
>due to the lack of a synce between systems and the latency of the
>system's response to incoming MIDI commands.

Could both be paralleled to one external clock for sync lock and then run 
on open loop independantly?

>It's been a few years since
>I looked at it but I suspect that using MIDI commands could decrease the
>responsiveness/resolution by as much as 10ms (it may very well be less)
>compared to standard JamMan resolution of 0.5ms. With 10ms of slop, it
>won't take long to hear the delta. In the best possible case, the
>systems will never be locked at a sample level so there will always be
>drift and you will not get proper stereo imaging (even briefly I 
>suspect). 

I think that few of us loopers would argue that the drift which results 
from looping unsynced 
can be extraordinarily beautiful at times but I, at least, have found 
that it can be equally annoying at other times. As for increased time lag 
using midi control, I seem to recall several people on the list reporting 
improvement when they switched from the Lexicon footswitches to 
midipedal. Personally, I haven't noticed any difficulty with the midi 
signals arriving late. My problem seems to be that the signals are not 
always interpreted the same way/correctly.

>Regarding the odd behavior; I suspect that, as Greg Hogan suggested, the
>second system is attempting to slave to incoming MIDI clock so it
>ignores or misinterprets the second tap. During product development
>there were precious few systems available so not a whole lot of
>multi-unit testing was done. Apparently systems are not supposed to
>"soft" thru MIDI data but we added it in fairly late in development
>anyway so that 2 JamMen could be used together without a MIDI
>merger/splitter (albeit not in stereo).

This answers my last statement, and my original post.

>Part of the problems with trying to cascade the JamMen (brothersync?) 
>is that the CPU in JamMan that handles MIDI (a measly 4MHz Z80) also 
>has to manage the realtime looping without missing a beat. 
>The audio and MIDI clock are the highest priorities so MIDI thru 
>operations can be delayed (the aformentioned 10ms).

Is the CPU upgradeable? 

>I wish I had better news but thats the way it is. Sorry.

Has Lexicon responded to your proposal for licensing the code? If this 
happens (the upgrade) can/will any of these problems be addressed?

Both Greg Hogan, Lexicon Customer Service and Bob Sellon, Stech/Lexicon 
have been extremely responsive and helpful? ;-) in my time of need. In 
this respect Lexicon is way ahead of many other companies, who still 
"just don't get it." It's really great that Greg and Bob are on the list! 
Thanks guys. I'll just keep noodling around until I come up with 
something that works for me (or until plexs finally arrive).

Phil