I've seen you mention Bristol 0.50 in a couple of forum posts, but the newest tarball I see for download from Sourceforge is still 0.40.7. Have you released 0.50 yet, or will you soon?
Long time no speak.
The original plan was to release 0.40.8 and 0.50.0 at the same time at the end of december, the 0.40 version being a maintenance of fixes for issues found whilst working on new features in the 0.50 stream. That plan failed for a few reasons, firstly sf.net stats got damaged and I do actually track the downloads but more importantly a few bugs came up that I could not really ignore, they related to how midi events were being handled at a pretty basic level (now finally resolved).
The features for 0.50 include:
LADI level 1 compliance
MIDI Event redistribution (caused plenty of problems)
SYSEX redefinition (solution to a problem that has not happened yet)
QA testing routines (and lots of memory corruption fixes)
More debuging options
Rakes of fixes.
There only remains one issue that I would like to fix however I am seriously considering pushing that out to 0.50.2:
For correct LADI integration bristol must only register once with Jack. At the moment I do this twice, once in the audio thread and again in the MIDI thread, each with a different symbolic name. This confuses LADI as the process ID is the same for both registrations making it pretty unique in the Jack world.
The reason I might push it out is that bristol will work with Jack and ALSA MIDI, should work with the lightly tested Jack MIDI interface but then not with LADI, and there are actually two or three more related issues surrounding Jack, Jack MIDI, LADI and Event Redistribution so if I have a crack at a fix now it will probably delay the release another month.
You can get the current beta (not gone through the QA process yet) at the site below and I would be very happy for you to use it, it is a release candidate - if it passes the QA process of a couple of thousand GUI restarts I am likely to post it here.
Do you use LADI? I would be very interested in having that code tested. If you don't then any feedback is naturally still going to be useful.
Kind regards, nick
Thanks for the link to 0.50.1d. I hope I'll have some time to have a play with it over the weekend, and of course I'll post any feedback here.
I don't use LADI at the moment: I may get round to trying it out eventually, but I haven't really needed it (or anything like it) for any serious use yet.
OK, don't worry about the LADI stuff - I am working with the author, Nedko, on the implementation. He is doing tests of different code I put together which gives me at least some testing from some other than just my system.
If you get into issues with the code let me know. I should try and release it next week pending some testing here.
By the way, how will the version numbering go?
0.40.8 -> Released at the same time as 0.50.1
0.50.1 -> Released at the same time as 0.40.8
The target was to release 0.50.0 and 0.40.8 at the same time, 0.50 with some enhancements and 0.40.8 with fixes coded whilst writing the enhancements. Version 0.50.0 will not not be released, it will roll up to 0.50.1 and they will go out at the same time.
Version 0.50.1 was finally uploaded with a few changes compared to the code you downloaded. Let me know how you get on and what issues you run in to.
I downloaded and compiled 0.50.1 with no problems.
I couldn't seem to get Jack MIDI working from my little USB MIDI keyboard (using 'jackd -X seq' to try to link ALSA to Jack MIDI), but I don't think that's anything to do with Bristol: I've never tried to use Jack MIDI from an external device for anything before, so I probably need to tinker some more to make sure I understand how to make that work.
'startBristol -jack -midi seq …' works nicely, and I can get sound out of at least the mini, b3 and arp2600. I'll let you know if I find anything funny as I carry on playing around. I just wish I had more time to spend making strange noises!
Log in to post a comment.
Sign up for the SourceForge newsletter:
You seem to have CSS turned off.
Please don't fill out this field.