Casio CZ/ VZ/ FZ - Pro Series

The CZ-5000 doesn't have the same Send Request 2 (command = 0x19). It has a Send Request 2 to get sequencer data (command = 0x14, memory location = 0x61). I am not sure if I tested the 0x14 command with the midi-us, but I did with WinXP SP3. (It worked.)

Do you send F0 44 00 00 70 19 00 70 31 F7 when it crashes? I know the MIDI Guide book says the dummy data (7th byte) should be within 00 ~ F7, but that has to be a typo.  At most it should be 00 ~ 7F.

When I do a sequencer data request I can chain at least six 70 31 acknowledgement pairs in the request without any ill effects. Perhaps the CZ-101/1000 can't handle the 70 31 F7 before it send out the data. Since I don't have the 101/1000, unfortunately I cannot test that command with the midi-usb interfaces I have.

Since the 0x19 command is just for getting programmer data (the patch selected) and effect data (vibrato and portamento on/off states). I don't think its a big loss if it doesn't work, though it might be handy for you to have.

The bigger problem is the 0x10 command. If that doesn't work, you cannot request and receive dumps from the CZ. Are you sending F0 44 00 00 70 10 60 70 31 F7 and getting a crash instead of receiving the data dump?  If so, perhaps it is that the 101/1000 cannot make use of the 70 31 F7 being sent with the first part of Send Request 1.  I believe I read that it is supposed to work with the CZ-101/1000, but maybe not.

It occurs to me that maybe the 1x1 (and its driver) are not really full duplex. It might be that the driver (or firmware if it has any) gives control to one channel (in or out) when a F0 is transmitted and that channel keeps control until a corresponding F7 is received. 

I have no way to test that.

With the receive request (command = 0x20 + memory location), it wouldn't matter that the interface doesn't send the F0 44 00 70 30 to the computer, EXCEPT if its presence somehow presence prevents the CZ-101/1000 from receiving all of the Tone Data from the computer. If that were the case, I'd expect a lock up or corrupt data. However, you said you have been able to send tone data to the CZ successfully with the 0x20 command.

With my sequencer (Cakewalk) it wouldn't be possible send the different parts of the handshaking requests (header, ack, F7) individually since each bank (what Cakewalk calls a stored sysex command) requires an F0 F7 pair. I am not sure they can be sent out individually with midi-ox or any other utility.

In sum, it might be a problem with the specific 1x1 usb-midi interface cable, or it might be something unique to the CZ-101/1000, or perhaps as I suggested in an earlier post, perhaps over time components in the CZ age and their abilities start to fail.

While I hope this helps, I am thinking the best way to try to solve the anomaly it is to try other interfaces. I don't know anyone with a CZ-101/1000, otherwise I'd be very interested to test it.

Steve 






 



--- In CZsynth@yahoogroups.com, Gordon JC Pearce <gordon@...> wrote:
>
> On Sun, Sep 01, 2013 at 09:55:22PM -0000, steve_the_composer wrote:
> > Let me ask: Are you saying this is the sysex header for the pach data the CZ1000 sends to the computer?  Or are you saying this is the header you use to send the patch data from the computer to the CZ?
> 
> Command 0x20 tells the CZ to start receiving patch data.  Commands 0x10 and 0x19 tell the CZ to send data.
> 
> If I send command 0x20 to the CZ it does actually dump data that appears to be correct, but if I send 0x10 or 0x19 it crashes.  In my previous email I'd pasted in the string for getting it to send data, which was incorrect.
> 
> -- 
> Gordonjcp MM0YEQ
>