@baconpaul Would you consider creating a sampler VCO similar to the wavetable VCO for single waveforms? This would be perfect for creating a sample-based Surge Drum Machine. 2.1 update is awesome!
We have plans to revive the shortcircuit sampler vst this year. Depending on how that goes maybe we will rackify it too. But no plans to expand sampler in surge right now
1 Like
Ah neat, heard about that. Looking forward to trying that too.
Is there a way to automatically convert non looping Serum wavetables to be Surge compatible? I have a bunch of tables which work fine in Serum, Vital, and Pigments but they load as one shots in the Wavetable VCO.
What is a “non looping Serum wavetable”? You mean a wav file with no markers?
If you choose the menu item “load untagged wavetable as / 2048 sample wavetable” does it work?
If so then that’s how you would do it. I don’t think we added that as an auto setting but we could in a future version.
Thanks @baconpaul. Yeah, loading through the menu works for those types of wavetables. But I have hundreds or thousands of those, so an auto-option would be amazing. Tangentially, do you know of a way to make them have markers?
Well we read lots and lots of different wav chunks. Depending on your OS those are easy or hard to add in batch.
We look for the chunk clm followed by 2048 as data, which is what serum uses when you author. We also read uhWT as a chunk type, which contains no data and implies a 2048 wide hive table. Finally we have a custom chunk type srge followed by the integer of the chunk size.
The more standard cue point writing in a wav file also works.
We provide some python scripts which are community documented to help you add some of these tags.
2 Likes
Thanks! Will review.
I found (annoyingly) a bug with how I was making json in 2.1.0 which means we have a small leak. I’m going to get a 2.1.1 together next few days. So if you have found any small bugs in 2.1.0 which aren’t on our GitHub, would love to know.
also: json_object_set_new every time folks. Sigh.
3 Likes
Hi Paul,
I just have a nasty bug with the filter in V. 2.1.3
when I turn the frequency all the way up I get a nanV on the output.
see this little patch:
2023_02_23_Surge XT_filter_bug_01.vcv (26.2 KB)
1 Like
Thanks. What’s your sample rate and os? I can look for 214
Is it possible to have all mono inputs on Surge modules to be normalled to stereo? This way, if a mono signal is input on any Left input, it will be duplicated to the Right one for processing and stereo outputs.
The fx all do that now For the filters and waveshapers it doesn’t make sense since they are really dual mono not stereo. For the egxvca it uses different pan laws already
So I’m guessing I missed normalization on some other module? Which one? Easy to add
I’m on Win11 and sample rate is 48kHz
1 Like
Cool, thanks @baconpaul. Cool, re fx, that works. Filter makes sense though would be nice to have. I believe the Mixer module is not normalled currently.
1 Like
I suppose in filter if l in and lr out are connected I could just dup l to r at the output
I don’t want to dup at input tho since that will potentially double load in poly cases
And I’ll check the mixer! Thanks!
1 Like
Cool, thanks for taking a look.
I logged both of these as issues for our 2.1.4 release. I’m a bit heads down rebuilding shortcircuit right now but won’t forget em! Thanks for the reports
2 Likes
So either (1) can you reproduce this with modules just in the library or (2) how do I get a version of rack windows that you are running?
Without the rack windows in the way I didn’t get a nan with the cutoff all the way up once i wired galactic to the filter.
@baconpaul : Two small ones:
Delay shows tooltip ‘Clock/BPM Input input’ on Clock in
String VCO shows ‘Audio Input input’ on Audio in.
There are others I have seen, seems any input you add Input to, or output you add Output to end up with the redundant already added (input/output respectively in lowercase) tooltip text. Too lazy to trawl them all but should be easy to find in source…
1 Like