KVR Audio

DevonB · Post by **DevonB** » Mon Oct 15, 2007 12:14 pm

stefancrs wrote:And yeah, quite advanced synths can be very small if you don't need all that "GUI junk". The one we used back in the days in 64kb demos used about the same size, including song player.

Heck, don't forget the few of the 4k demos that had music as well. Very impressive. Granted, they were cheezy chip tunes, but still.

Devon

D-Fusion · Post by **D-Fusion** » Mon Oct 15, 2007 12:18 pm

So is there any fix for my Sonar issue so that i can start using this wonderfull synth?

Chris Walton · Post by **Chris Walton** » Mon Oct 15, 2007 12:21 pm

AdmiralQuality wrote:Genuine attempt to help: Why not implement the MIDI all notes off message? I see you have a function for it already ADNotelogic::allNotesOff(), but aren't filtering for the msssage in ADNotelogic::midi()

Not sure and don't have time to prove it, but I'll bet that's your issue, not sorting or lack thereof. (Which again, Poly-Ana doesn't do but AFAIK works fine in Sonar anyway.)

hate to do it, but ugh

I have to agree with AQ, that's the problem. Sonar is lazy about sending note-offs when it can send allnotesoff instead. It does return sorted events.

stefancrs · Post by **stefancrs** » Mon Oct 15, 2007 12:27 pm

DevonB wrote:
stefancrs wrote:And yeah, quite advanced synths can be very small if you don't need all that "GUI junk". The one we used back in the days in 64kb demos used about the same size, including song player.
Heck, don't forget the few of the 4k demos that had music as well. Very impressive. Granted, they were cheezy chip tunes, but still.

Devon

I've made/coded music for a 4k too

Wouldn't call it chip tunes when it's real-time softsynths with filters etc. I guess it's a matter of definitions though. Granted, my 4k tune sucked.

I _really_ like this 4k:
http://pouet.net/prod.php?which=30222

aciddose · Post by **aciddose** » Mon Oct 15, 2007 12:53 pm

"Sonar is lazy about sending note-offs when it can send allnotesoff instead."

well, that goes against the midi spec and the authors should definitely be informed of that one. i'm sure they wouldnt like their host being known as "non-conforming to the midi specification". the midi-specs explicitly state that you may not use all-notes-off in that way.

anyway yes i've already fixed it and i've also already added event sorting. i'm just in the middle of another edit which puts the code into a state where i can not release it. the next release will include event routing, loading event tables via text files, all-notes-off fix for sonar including a gui switch for ignore and event sorting, plus some additional tweaks to the code as usual.

next weekish? i'm in the process of assembling and testing my first hardware synth prototype here so i've no time for the software. i have a schedule to keep up to on this hardware project and there is a lot of work to be done.

to above: interestingly xhip was originally intended for 64k intros. i also wrote a 100% software 3d engine and renderer however that never got used for anything. i'll be releasing the source for that too although it isnt very useful.

check that out --> http://xhip.cjb.net/temp/public/something.exe
(no xhip included though

)

Chris Walton · Post by **Chris Walton** » Mon Oct 15, 2007 1:01 pm

aciddose wrote:"Sonar is lazy about sending note-offs when it can send allnotesoff instead."

well, that goes against the midi spec and the authors should definitely be informed of that one. i'm sure they wouldnt like their host being known as "non-conforming to the midi specification". the midi-specs explicitly state that you may not use all-notes-off in that way.

Actually, that's perfectly valid according to midi, as far as I know

oh, and SUPER DUPER AWESOME YAY

D-Fusion · Post by **D-Fusion** » Mon Oct 15, 2007 1:01 pm

aciddose wrote:"Sonar is lazy about sending note-offs when it can send allnotesoff instead."

well, that goes against the midi spec and the authors should definitely be informed of that one. i'm sure they wouldnt like their host being known as "non-conforming to the midi specification". the midi-specs explicitly state that you may not use all-notes-off in that way.

anyway yes i've already fixed it and i've also already added event sorting. i'm just in the middle of another edit which puts the code into a state where i can not release it. the next release will include event routing, loading event tables via text files, all-notes-off fix for sonar including a gui switch for ignore and event sorting, plus some additional tweaks to the code as usual.

next weekish? i'm in the process of assembling and testing my first hardware synth prototype here so i've no time for the software. i have a schedule to keep up to on this hardware project and there is a lot of work to be done.

to above: interestingly xhip was originally intended for 64k intros.

That is great news

Looking forward to it

That note off thing in Sonar has been there atleast since V3 and has been pointed out to Cakewalk many times and still no fix in 6.2.1

It is a new option in Sonar 7 which is suposed to fix this but i will wait until a demo of V7 is out before i upgrade.

aciddose · Post by **aciddose** » Mon Oct 15, 2007 1:14 pm

"Actually, that's perfectly valid according to midi, as far as I know"

well, it isnt. you can use "all-sound-off" in that way as it's intended to be like a panic command which instantly cuts all sound and all notes. that is cc #120. all-notes-off #123 however is intended as a stop command for a sequencer. the idea is that you might be playing the local keyboard while a sequencer plays notes to the midi-in. all-notes-off is supposed to turn off the notes triggered by the sequencer while not touching the local-keyboard notes - although the spec says sustained notes should be sustained even if they came from the sequencer, until sustain is released. maybe it means a sustain sent from the sequencer as well.

in any case, it again explicitly states that if a device doesnt have a local keyboard or otherwise cant distinguish between it's local keyboard and midi-in port, all-notes-off shouldnt even be implemented at all.

what it says here... http://www.midi.org/about-midi/table1.shtml
...makes it seem like this is not the case, but the full documentation fills in all the details.

spuddle · Post by **spuddle** » Mon Nov 19, 2007 2:07 am

Hi friends,

What is the current status of the lovely xhip synth? I see that you are now working on a hardware project, good for you I say! I hope that's a success. Is the soft synth "safe" to use in projects yet is all I'm asking for at the moment?

Thanks.

aciddose · Post by **aciddose** » Mon Nov 19, 2007 3:18 am

it always has been if you intend to continue using the same dll for that particular project. it is still not forward compatible though, no.

download the latest version from http://xhip.cjb.net/xhip/releases/ and use that (adxhip.0.6.14.8d.dll ?) - do not remove the version number attached to the file name.

disturb · Post by **disturb** » Tue Nov 20, 2007 7:54 am

bug report (flaky one though)
i've had recuring crashes happening when trying to get a particular patch going.
not sure exactly what's causing it, there is some audio rate pwm and filter mod going on, it always happens when trying out different combinations of fm and sync modes over that.
you might wanna look into this ad, i'll try to narrow it down though if that's too vague...

latest version btw, but i've had this happen with previous ones too.

aciddose · Post by **aciddose** » Tue Nov 20, 2007 9:57 am

there is some potential for a combination of very strong modulation (xmod in combination with others) while using sync to cause a out-of-bounds access to occur which will crash. it is a strange corner case and while i'm aware of it i havent been able to figure it out or fix it yet. a quick and dirty solution might be to bound the index for this memory access, but that wouldnt solve the root logical problem. i'd rather have the osc code 'pure' and free of logical glitches or corner cases. since this occurs so rarely i havent really spent much time on it.

if you can provide a patch and event sequence that can reliably reproduce a crash i can definitely fix that. the reason i havent is i've only been able to encounter the bug while not looking for it - when i start looking it suddenly refuses to occur.

MaliceX · Post by **MaliceX** » Fri Nov 30, 2007 9:04 am

Hey, I really want to make and distribute patches for this thing, but would one say the patch format's safe to work with yet, or is it likely it'll break on the next versions?

sinkmusic · Post by **sinkmusic** » Fri Nov 30, 2007 10:23 am

MaliceX wrote:Hey, I really want to make and distribute patches for this thing, but would one say the patch format's safe to work with yet, or is it likely it'll break on the next versions?

+1

aciddose · Post by **aciddose** » Fri Nov 30, 2007 6:56 pm

it definitely may not work in the next versions, although as for the synth itself it is pretty well locked down at this point so it's likely that any patches or banks will be able to load.. i just would need to leave the code in for the old formats.

one major problem at the moment is i'm willing to bet you'll find yourself very frustrated not being able to patch stuff like velocity and the modulation wheel with things the way they are. i dont really know when i'll be able to get the next version going, i havent touched the code in nearly a month now. i recently posted something on the site about this: http://xhip.cjb.net/

pretty much, you can use the current version so long as you're happy with the features it offers since i'm not planning to change anything else in the gui. at least not anything that wouldnt be perfectly compatible.

XHip--Please finish your synth!!