well, with those 10k programs, especially the ones with the 'UFE' controls, i'm really going into uncharted territory. that program has 5,600 samples in it. i'm not going to say i'm the only dev to go over 1,000 in a program, but i think i might be. not that i'm saying raw sample count is a record i'm striving for, but it's needed to cram more sampled controls into a program. the ufe progs have 3 controls which were sampled at various positions so it comes to lots of samples and creates a very complex program.lipa wrote: Seriously, has anyone had any problems loading 10 kernels programs? My reaper crashes almost every time..even with totally new session.. but 1k and 5k programs work great!
i tested them myself on a few hosts, and a few beta testers did also, but i couldn't be 100% sure those programs would work for everyone in every host. i think it is safe to say nobody else has made a program with close to 5600 samples so it's hard to predict what happens at that point. but the 10k versions aren't always as necessary as you might think. i would really strongly recommend only using the 10k versions if you are going to use heavy compression. like maybe over 8-10dB. because, having seen the structure of the samples up close, i can tell you that those upper harmonics (6-10) will not even exist unless you are using lots of compression, or very hot inputs close to 0dBfs. so you'll just be spending more time loading the program, and you'll still be using more cpu and ram, but won't be getting anything for it. this is because of how insanely clean these units are. further, the 5k version will even be a little more (slightly) accurate with the harmonic structure of the harmonics it does generate. so honestly i kind of think of the 5k versions as the main part of the set.
that said, as lipa asked, is anyone else having that issue?