Ok, so I just finished another update to Tube FM 1, and believe it or not I actually do think it was justified. I've also decided to use this thread for any updates I may ever release for any of my stuff (if it's for something that has been out for a while), instead of bumping up other old threads or whatever.
First thing I did was improved the 'dry' control the last update added. It bugged me when I released it, and I knew I would have to try to fix it. Now it is a 'wet/dry %' ratio style control and I'm really happy with how it works. Loading these programs into vstanalyser and twisting that knob and watching the frequency responses morph from 'radio filtered' to 100% dry brings a smile to my face. The old 'dry' control I added in the last update back in December just wasn't so hot. This one is waaaaaaay better. Trust me. I think it opens the door to using these things in much more subtle roles. For example you could take the Philco do2 program which is like a fairly sharp band pass filter around 1kHz, put the 'wet %' control at 30% then boost the output by about 3.2db, and you get a mostly dry signal with a 'tubey' boost of 4db at 1kHz. I really recommend checking these things out in vstanalyser or something similar and playing with the new control to see what it can do.
other updates- -the dynamic control envelopes have all been set to EVF 17. they were set to one of the RMS modes and i believe this set was released before EVF 17 was even available. -some controls had inconsistent names across the set, fixed. -timed kernels all have default length of 50ms for all kerns, which is the same length used by the freqd player. the programs still default using freqd, but this way, if you want to switch over to timed before render, it's a lot quicker. -the one with the sampled 'tone' control- Admiral DO3, was 'fixed'. it's more complex in it's structure than the others, and may have actually been the first Nebula program with dynamics and a sampled adjustable control, so it was trickier to make and i don't think it was set up 100% correctly as far as how it handled the dynamics. i think it is now. -a high pass filter is now used on an internal side-chain that controls the dynamics. it's very low at 45hz, and so it barely has any influence. i did this because most of these programs are filtering lots of high and low signal so i thought it would be a bit more realistic to filter some of the low out of the control source. i thought about adjusting the cutoff differently for each program but in the end i decided to go with the same setting for all.
it does still bug me that the 'trim' control directly adjusts the dry level. this means that whatever ratio you set with the new 'wet %' control is thrown off by any adjustments you make afterwards with 'trim'. i spent several hours trying to figure this out and if it's even possible, the math is just over my head. nebula FUNs can be deceptively tricky. this is still an improvement though.
Last edited by Cupwise on Wed Feb 20, 2013 3:56 am, edited 1 time in total.
ok so i just finished up another update to the first part of my Rayphlex set. a lot of it is cosmetic. I renamed all of the extra programs to have clearer descriptions of what they are, and i think the menu/list just looks a LOT better this way. then some controls have been removed, and some added, to many programs. i got rid of 'dist' because these things are pretty clean and it's just pointless, among other reasons. got rid of 'makeup' for programs with feedback detection due to the makeup gain being in the detection path, which is a Nebula flaw and means that makeup control influences compression (it shouldn't). the only way to deal with it on my end is to remove the control from those programs. to make up (get it) for that, i've added look-ahead and dry controls where i could, so now lots more programs have those.
then the 'soft 2' compressor had some bad samples so i resampled it. i think that's it! so, if you have Rayphlex A, you should be getting an email soon with links to the updates.
I've just made a HUUUUGE update to my Casspressors A set. I'm calling it v2 now. I've redone the first audio demo mp3 at the page (http://www.cupwise.com/cup/casspressors-a/) to demonstrate how these things can now sound. Here are most of the main updates:
compression has been drastically improved. i'm not going to say it's perfect because those effects are pretty extreme and also lo-fi, being not just tape but cassette tape. so the tape sat thing still isn't perfect, but they work a lot more naturally now. many of them used various other hardware to intentionally make them more lo-fi/distorted, and the improvements i've made allow for a lot more control over that.
made it so that the look-ahead control can actually go up to 6ms instead of 1.5 ms. this allows for more and in some cases all transients to be caught when combined with the fastest attack setting, which is closer to how tape compression works.
categories that the programs are in has been changed and makes more sense. the whole set is better organized with lots of little improvements.
all programs have been reduced in size with no bad effect on the quality, and this lowers RAM usage and CPU usage by about half.
liquidity control has been removed and replaced with a high pass control for an internal sidechain filter.
made the attack control readout, and the control itself, more accurate.
added an alternate set of programs, which is filed under the 'ALT' sub-category. this set all uses the RMS detection mode and they are labelled as such in their program names.
added 3 other alternate sets which are filed under 'FB1' 'FB2' and 'FB3', which use 33%, 66%, and 100% feedback for the detection path respectively. this means that the 'FB3' progams are fully feedback detection compressors. the main set uses full feedforward detection. there are Peak and RMS versions of the feedback sets. try some of these for slightly different sounding compression.
added info in the description field so that the user can have a better idea about which program they are using at any moment.
switched the pass-through programs from using an RMS mode to using Peak, which i believe makes more sense for that type of program.
i'm going to be emailing the updated set to all customers so if you have the set you should be getting an email with the links.
just finished a pretty big update to my Custom Tremlo set, with a few of the updates being things that people suggested to me, so if you ever have any ideas, don't hesitate to suggest them!
some cosmetic improvements such as adding descriptions that will help the user know which preset they are using (since nebula gui cuts off long program names).
added an additional set of 'manual' presets. these are found in the 'MAN' sub-category. they have no LFO in use. the lfo controls have been replaced with a single control called 'LEVEL', which allows you to manually control what the LFOs were controlling automatically. this way, if Nebula can be automated by your host, and if your host provides its own methods of modulating parameters, you can use them and possibly have a lot more ability to shape the effect. another thing that this might allow is for you to have tempo synched tremolo, all handled by your host program.
liquidity has been renamed to 'smooth'.
changed all programs to use 'peak' level detection mode instead of whatever was used before because i've come to feel that peak makes more sense with this type of thing, and i think it sounds more realistic.
someone suggested to make the control scheme more consistent across the various programs, so that you could audition all of the effects and have the same control setup if using an external midi controller. it was a good idea so now that's how it is (thanks go to the un-named person who had this idea, sorry i forgot who it was). control 2 is always vari, unless there is no vari in which case it is nothing. control 3 is always depth.
renamed 'drive' control to 'dist', to be more consistent with my other releases where it is usually called dist. i think dist is a better name for the control.
added asterisks to the program names of the two programs which do forms of autopanning rather than just tremolo. also added '*panner' to their descriptions.
adjusted the way the 'rate' control works, so that the slower speeds take up more of the control's range. this way it is easier to fine tune within the slower speeds below 1hz.