Sounds like some Saturday evenin' fun.todd_r wrote:Urs wrote:...I only need to render a new background image for Uhbik-G...
Feasture Brainstorming for Uhbiq
-
- KVRAF
- 13133 posts since 7 May, 2006 from Southern California
At the risk of making you life even more difficult... a delay w/ feedback would be great. I'd also really like much larger maximum grain size and an option to adjust grain window by tempo synced divisions (that is 1/4 notes at base pitch become 1/8 notes at an octave above and 1/2 notes at an octave below).Urs wrote:and I might have to add a delay at some point, not sure... could do that while/after public testing...
Cheers,
Urs
Pie in the sky requests... an adjustable window envelope, a "freeze buffer" button (keep the manual reset active to introduce new audio into the buffer without unfreezing!), the ability to adjust the position of play back within the buffer.
The last one would probably be a toughie, the way you currently buffer audio may not allow for this. I imagine a start and an end knob with values from -1 thru +1 and a button to link them. The start knob would control the position that buffered audio begins to play. Moving the start knob into the negative would introduce samples from previously buffered audio (modulating this value at the same rate as the size of the buffer should result in a reversing effect) and moving the start knob into the positive should introduce a gap of silence (obviously, you can't playback audio that hasn't been recorded yet). The end knob would control the position that the loop recycles (making it so the main grain window size does not change when adjusting the pitch). Negative values of the end knob would decrease the amount of buffered audio that is played back before looping, positive values would introduce a gap of silence (again, you can't play back what hasn't been recorded.
Pretty please!
-
- u-he
- 30215 posts since 8 Aug, 2002 from Berlin
Arrrgh.justin3am wrote:Pretty please!
The feedback delay is of course possible with Uhbik-X then.
The principle behind G might be different from other grain tools. The duration in time of the grain does not change with transpose. Instead, the number of samples is being adjusted dynamically, along with the start/end points. This makes it as low latency as possible, and as realtime-tweakable as it gets. IMHO.
I'm reluctant to allow for longer delays, it would kill memory resources in a surround scenario.
There are however chances that one day there'll be a more sophisticated grain thing, which would complement G in the way that MFM2 complements D.
Urs-
- KVRAF
- 2911 posts since 3 Mar, 2006
So no comment on the intelligent pitch shifting then
-
- KVRAF
- 13133 posts since 7 May, 2006 from Southern California
Don't sweat it. I've said it before, the features I request would rarely be useful to anyone but me. I've already built my dream granular machine in Reaktor (8x minute long buffers) and it eats ram like pommes frites. Right now, I'm learning to program the same functionality into dedicated hardware (probably FPGA+TI DSP and a grip of ram). At this rate I think I'll have a prototype in 1.5yrs, a production model in 2yrs.Urs wrote:Arrrgh.justin3am wrote:Pretty please!
There is something about your synths and processors though... I can't quite put my finger on it. It takes a lot to get me really excited these days but every time I use a U-He product it's like the first time all over.
-
- u-he
- 30215 posts since 8 Aug, 2002 from Berlin
Why FPGA?!? - You'll spend a year just to teach it addition and subtraction...justin3am wrote:probably FPGA+TI DSP and a grip of ram
I've seen a phantastic combo recently where you get a tiny little proto board with a 32bit microcontroller, a reasonably powerful dsp and 2in - 4out audio dacs ready to go... I was contemplating to get that myself, but I can't justify another time eater... I think it was an Atmel thing.
Urs-
- KVRAF
- 13133 posts since 7 May, 2006 from Southern California
Well, my employer is sponsoring my education so I can program embedded systems for them. Their primary focus is FPGA based going forward. I picked up a copy of Musical Applications for Micro Processors and I figure that I'll do some experimentation with those 32bit AVRs but I was under the impression that they couldn't handle the kind of high resolution granular stuff I want to do. I'm still in my research phase so I'd like to hear your thoughts on the subject. Send me an e-mail if you like.Urs wrote:Why FPGA?!? - You'll spend a year just to teach it addition and subtraction...justin3am wrote:probably FPGA+TI DSP and a grip of ram
I've seen a phantastic combo recently where you get a tiny little proto board with a 32bit microcontroller, a reasonably powerful dsp and 2in - 4out audio dacs ready to go... I was contemplating to get that myself, but I can't justify another time eater... I think it was an Atmel thing.
-
- KVRAF
- 13133 posts since 7 May, 2006 from Southern California
-
- KVRist
- 36 posts since 3 Jun, 2009 from Wales
You need a faster rendering machine Urs!
Aaaargh!!! Can't wait!
Aaaargh!!! Can't wait!
-
- u-he
- 30215 posts since 8 Aug, 2002 from Berlin
Sorry... still sitting over a couple of bugs (see Zebra thread, these bugs affect everything) before starting to render anything...robspaull wrote:You need a faster rendering machine Urs!
Aaaargh!!! Can't wait!
The actual render process is done in, uhm, 3 minutes (Cinema4D), after editing a bit on the model. Then a bit of Photoshop, then placing files in directories, then firing up Uhbik-G in edit mode, then moving knobs, then saving that, then distributing the config files to all directories and to the FAT32 partition, then create installer for Mac, then upload, then reboot into Win, then do everything again on Win side from after Photoshop or so until everything is on the server. Then have it tested first before posting here.
Actually the layout part (Photoshop, ui editor) is a bit more elaborate because the new knob overlaps margins of a tabbed pane... hmmm... involves a bit extra redesign.
It'll be done soon though, I'm sure. Most bugs seem to be under control by as of today...
Urs