Fundamental problem across all hosts and all platforms AFAIK. Part and parcel of the TI - sorry, but that is my experienced assessment.
My take is the USB1 spec will NEVER handle mutliple audio streams reliably. End of story.
Posts by Dusk
-
-
OS4 a charm here.
No bugs, glitches or slowness when changing parameters or patches.
XP32, FL9 fyi.
-
-
1.04 - 1.30 - lots of noise!
but all of it really
-
we sense a certain unrest in here ... and yes, it stopped snowing in the rhine area ... but be aware, this winter could be even longer than it already is. and colder. so cold that frozen bits could hold us back from releasing the update. nevertheless, don't you worry. we are save and sane. as a proof, i've attached a new demo focusing on step-sequencer-esque blippy melodies inspired by snowmen and snow cones. don't you think the new comb filter sounds somehow ... cold ... ?
reporting from the headquarters
m
F*CKING HELL!
Any chance of a video how you made this?!? Or some basic ideas??
is it all virus?!
-
Display More
That had me running to the Virus to see if I hadn't been missing the simplest solution ever for over a decade!Unfortunately, it is unlikely to be sufficient, unless you want a severe, 'Zombie Nation' type detune.
There is a workaround:
Set LFO1 to a square wave.
Set LFO1 Contour to +63
Use LFO1>Osc2 Pitch in negative amounts to detune downwards.Maybe an even better result would be to set (LFO1>) Osc1Pitch and Osc2 Pitch amounts to opposite values, i.e. Osc1 = -12.5% and Osc2 = +12.5%.
Be sure not to use Env Mode and remember to use the hard-wired LFO1>Pitch assignments in the MODULATORS section - using the mod matrix for this won't work properly.
Of course, you tie up LFO1 by doing this.
Inside the mind of sound design master. Thanks Ben.
-
marc, if you're able to, can you confirm if this OS will be for Ti mk1 as well?
Personally, I'm not one of those that feels endlessly entitled - as I know you eventually have to encourage your users to buy more hardware - but I just hope us Ti1 users have another couple of updates left in the tank!
cheers,
Andy
-
This has been going on from the very outset with Ti using VC, and tbh I don't beleive it will ever be cured completely.
The best OS was the release version before the current one (not the beta's) - I think it was 3.01.15 (IIRC?) That was the only OS that gave me arps that could withstand tweaking of parameters, or playing other parts with heavy patches, without losing sync every single time. I'm going to try the new beta later, as this 3.1 release seems as bad as any previous OS on the timing front.
There was an article on the old site (troubleshooting section) pretty much saying the problem was unavoidable with certain computer architectures, certain combinations of processor and chipset, etc. It wasnt really detailed. My take is that once again the limitations of the USB1 protocol are the chief factor - time synchronisation is apparently one of the hardest thing to maintain along the signal path, and hence one of the biggest technical challenges of the Ti concept. I doubt whether standard, non specialised components, on board USB, and non calibrated OS' are capable of guaranteeing the sort of free, robust flow of data that would benecessary for rock solid timing.
