horizon is a fantastic synth but seems to have some problems. I have had similar problems with other complex SE vsts. Using cubase, sometimes it wont bounce, or wont load properly when a project is opened.The only way to fix it is to reload the plugin. Sometimes cubase says "a serious problem has occured". This is with multicore processing switched off
any ideas?
oli
horizon problems
-
- KVRian
- Topic Starter
- 876 posts since 24 Jun, 2002 from Berlin
-
- Banned
- 12368 posts since 30 Apr, 2002 from i might peeramid
horizon has one issue beyond any that may be caused by the edition of se it was compiled with (the multicore thingy is probably not a factor for most hosts now??) -
the BLITs are shared between voices.. i had (have) no idea how to implement this in shared memory, so i basically coded it from pointers jeff gave me. it was either that or 45 seconds per instance load time, every time..
there are two solutions:
this first, which won't happen soon, but perhaps in the next year.. write the BLIT data and include it in the .sem, fixed for 44.1k host rate.
the second is for someone else (?
) to be buggered with the original code.. i think there's only half a dozen pertinent lines if that, though the files are verbose otherwise.
i think the filters in isnonu (chris' biquads instead of standad rbj) were a better choice as well.
the BLITs are shared between voices.. i had (have) no idea how to implement this in shared memory, so i basically coded it from pointers jeff gave me. it was either that or 45 seconds per instance load time, every time..
there are two solutions:
this first, which won't happen soon, but perhaps in the next year.. write the BLIT data and include it in the .sem, fixed for 44.1k host rate.
the second is for someone else (?
) to be buggered with the original code.. i think there's only half a dozen pertinent lines if that, though the files are verbose otherwise.i think the filters in isnonu (chris' biquads instead of standad rbj) were a better choice as well.
you come and go, you come and go. amitabha neither a follower nor a leader be tagore "where roads are made i lose my way" where there is certainty, consideration is absent.
-
- KVRian
- Topic Starter
- 876 posts since 24 Jun, 2002 from Berlin
Well I'd love to see the code because I think this is one excellent synth and it would be interesting to find out why, however I don't think I'm able to fix it at the moment due to the ridiculous number of projects I have half finished.
Ps how is the sitar coming along?
Ps how is the sitar coming along?
-
- Banned
- 12368 posts since 30 Apr, 2002 from i might peeramid
the sitar is alright, "as good as it's going to get for now". similarly, i decided to do a few other things (phaser, flanger, reverb.. exciting things..) before finishing it. the reverb exploraton will probably help for body resonances, so it was a good idea to wait.
the horizon osc code is fairly straightforward, w/o checking, i expect the detune stretching is faded to cubic nonlinearity (not all the way). oscs are standard BLIT with gentle reduction towards nyquist (i like to leave more highs in than the gibbs effect reduction in the native osc).
i've got a ~wicked thought about making my 'girl' oriented synth the horizon update..
the horizon osc code is fairly straightforward, w/o checking, i expect the detune stretching is faded to cubic nonlinearity (not all the way). oscs are standard BLIT with gentle reduction towards nyquist (i like to leave more highs in than the gibbs effect reduction in the native osc).
i've got a ~wicked thought about making my 'girl' oriented synth the horizon update..
you come and go, you come and go. amitabha neither a follower nor a leader be tagore "where roads are made i lose my way" where there is certainty, consideration is absent.
-
- KVRian
- Topic Starter
- 876 posts since 24 Jun, 2002 from Berlin
So are some waveshapes blit based and some phase distortion? Are you referingto the unison detune-the way the pitches are spread around the main pitch?
Oli
Oli
-
- Banned
- 12368 posts since 30 Apr, 2002 from i might peeramid
yes and yes - there's a piccie in the manual of the spectral effects of linearity and nonlinearity in unison mode. equidistant pitch produces overt regular beating.
i've gotten a fair amount of use out of it too. perhaps a built-in center-side process would complement the panning as well
i've gotten a fair amount of use out of it too. perhaps a built-in center-side process would complement the panning as well
you come and go, you come and go. amitabha neither a follower nor a leader be tagore "where roads are made i lose my way" where there is certainty, consideration is absent.
