KVR Audio

caulixtla · Post by **caulixtla** » Fri Jan 20, 2017 4:57 am

spencerlee wrote:Hey caulixtla
I have a heavily chorused reverb that I really like at 7 secs. But if you bring it down under 2.5 secs it turns to shit.
I have posted some of it here before.
Would you be able to have a look at it and maybe turn it into a thing of beauty. You are much smarter than me.
Spencer

Without even the topology posted here, I can only offer general advice:

To make something sound good with a shorter decay, decrease the size. If that doesn’t solve the problem, add l(-1) or l(-2) between parts of the topology to make the sound decay more quickly while it’s circulating through the tank.

spencerlee · Post by **spencerlee** » Fri Jan 20, 2017 7:51 am

Sorry here is Late Reflection.

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

XComposer · Post by **XComposer** » Fri Jan 20, 2017 11:59 pm

Chandlerhimself wrote: @XComposer - (…) I don't have B2 so I can't really say, but the halls I made all seem to work at long decay times and of course you can add modulation, although in my examples I usually don't use any. The default algorithm actually works really well for that type of thing. Can you show me some examples of what you're talking about?

Yes, I will try to make some examples as soon as possible. It's not easy for me to do it very soon, it's a very busy period, but I'll do my best whenever possible.

caulixtla · Post by **caulixtla** » Sat Jan 21, 2017 8:29 pm

spencerlee wrote:Sorry here is Late Reflection.

The full preset was not correctly pasted — I could not load it. Instead of pasting the entire preset (which is problematic), please just cut and paste the “Algorithm” part of the late reflection for your preset. If you use multiple late reflections, paste all of the “Algorithm”s for late reflections separately.

onerob · Post by **onerob** » Sat Jan 21, 2017 9:56 pm

Is there still time to submit new active presets? I've got a nice, natural sounding room verb here.

https://dl.dropboxusercontent.com/s/3kg ... 20Room.mp3

spencerlee · Post by **spencerlee** » Sat Jan 21, 2017 10:15 pm

Here is LR Preset you can import.
And LR Algorithm.

an[8a;f(5000);[swap;[20a;xfdn4d];st[8a;b[v(8;1)];[8b[a]]]];4a]cc[30l;16v]f(3000)

Concert Wave Sortta.mpresets.zip

Chandlerhimself · Post by **Chandlerhimself** » Mon Jan 23, 2017 9:28 am

onerob wrote:Is there still time to submit new active presets? I've got a nice, natural sounding room verb here.

https://dl.dropboxusercontent.com/s/3kg ... 20Room.mp3

Sounds good. You should still have time to submit.

Chandlerhimself · Post by **Chandlerhimself** » Mon Jan 23, 2017 9:38 am

I thought I'd post something to try to help those out there making pitch shifting plugins. When I first tried it I noticed it had a dirty sound and I thought maybe there was nothing I could do, but I found out there is a way to make it work a bit better. I recommend using

Code: Select all

pitch(4098;c;0)

The first number is the samples or resolution. I forget which, but increasing the number makes it more accurate, but causes latency, you can experiment with this to suit your needs. The "c" lets you adjust the pitch according to the pitch param. The most important part is the "0". By default it is set at 1 which means it follows the formant, I believe and IMO causes noise with some material. By setting it at 0 it sounds much better.

In my algo, I set things up like this

Code: Select all

pw($1;$2)[l;[pitch(4098;c;0);l(3)]]

This allows you to control the pitchshifted to non-shifted ratio. Set a multiparm to move user parm 1 between 1 and 0 and user parm 2 between 0 and 1. That should allow the user to choose the amount of effect they want. Hopefully that helps.

MeldaProduction · Post by **MeldaProduction** » Mon Jan 23, 2017 10:40 am

onerob: Sure there's time, as you probably know now, I added one more month

.

spencer: Huh

, don't tell me you wrote the algorithm

. It uses like half of the modules

.

chandler: Good point. This basically disables the formant compensation, I suppose that may be useful indeed.

onerob · Post by **onerob** » Mon Jan 23, 2017 3:52 pm

onerob: Sure there's time, as you probably know now, I added one more month .

Indeed. Thanks.

DarkStar · Post by **DarkStar** » Mon Jan 23, 2017 5:18 pm

Quick Q:

Info page for Algorithm wrote:Instead of writing the weights themselves, you can just enter "c", which stands for coefficients. In that case the module uses the input coefficients for weights.

What are 'input coefficients' and where do they come from?

Also does "b(0.1)[5a]"
mean that there is 10% chance that all 5 allpasses will be bypassed and a 90% chance that all 5 allpasses will be active? Or does the probability apply to each allpass individually (so I could end up with 0, 1, 2, 3, 4 or 5 active allpasses)?

MeldaProduction · Post by **MeldaProduction** » Mon Jan 23, 2017 6:47 pm

Input coefficients are some sort of pseudorandom numbers generated from the seed along with delays and output coefficients. Their usage depends on the algorithms. They are mainly relevant for R, RD, CC and CA modules. The advantage is that you can control the types of these values in the Designer panel.

As for b(0.1)[5a] - this one will create a bypass module and inside it will be 5 allpass diffusers in series and with 10% chance it will bypass the whole thing. If you want 5 allpasses, each with some random chance of being active, write 5b(0.1)[a], hence 5 bypass modules, each with one allpass.

DarkStar · Post by **DarkStar** » Tue Jan 24, 2017 10:02 am

Got it, thank you.

PS The [Seed] button does not have an F1 Info window at the moment.

spencerlee · Post by **spencerlee** » Fri Jan 27, 2017 10:45 am

I have finished my first preset. Can I get some opinions please.
http://spencerlee.audio/RandomHallShort ... resets.zip
Spencer

onerob · Post by **onerob** » Fri Jan 27, 2017 8:31 pm

spencerlee wrote:I have finished my first preset. Can I get some opinions please.
http://spencerlee.audio/RandomHallShort ... resets.zip
Spencer

Whoa! That is an amazing amount of ER presets. Must have taken ages to do?

I did notice a bit of high end fizz coming through if using a synth sound with a lot of presence, but it's easy to adjust High Cut or Damp High, so no problem.

Sounds very good, IMO.

MTurboReverb preset making action, let's make the ultimate reverb! ;)