KVR Audio

bpblog · Post by **bpblog** » Sat May 22, 2010 7:13 pm

edit.

guitarzan · Post by **guitarzan** » Sat May 22, 2010 7:53 pm

bpblog wrote:a logical answer for me is that a reverb impulse is naturally longer, so the reverb impulse hosts are probably optimized for such use.

I think that's pretty much it, so the ones for cab impulses are more likely to be capable of running in real time with lower added latency, lower cpu use, etc. Kefir has one of the cooler options for running cab impulses, where you can decide on the fly how much of the legnth of the impulse you really need - a lot of the IRs only really have any useful info during the first half second or less.

I'd like to see more specialization in the cab convolution plug ins....like I think it would be useful to have power amp-like compression before the IR (so sims w/o enough power amp compression could be enhanced, as well as hardware preamps, pedals, etc.) and also a different compressor/limiter after the IR to represent speaker cab physical compression. Ideally, it would be cool if you could load and mix multiple IRs, then maybe use multi-band compression afterwards to dial in the speaker cab dynamics. Maybe one channel could be loaded with an especially "cabby" sounding IR followed by an exciter or something like that - to get a little speaker/cab buzz tone going to mix in subtly - or not so subtly if you're going into Neil Young territory or whatever.

IRs alone get the tone pretty much right, but they don't mimic a cab's non-linear behavior at all - there's a lot to be done in this area still! It's a big item that still hasn't really been seriously dealt with in amp sims yet, and it is probably the thing that keeps amp sims from having great amp "feel" (IMO).

bpblog · Post by **bpblog** » Sun May 23, 2010 11:09 am

thanks for the extremely informative reply guitarzan! i've been struggling with making some DI recordings of electric guitar parts and have failed at making them sound good thus far. but some points you made there like adding compression before a cab impulse have certainly given me some food for thought. :]

Ken McLaren · Post by **Ken McLaren** » Sun May 23, 2010 10:27 pm

Hey Guitarzan...Hows things? Good to see you...

IRs alone get the tone pretty much right, but they don't mimic a cab's non-linear behavior at all - there's a lot to be done in this area still! It's a big item that still hasn't really been seriously dealt with in amp sims yet, and it is probably the thing that keeps amp sims from having great amp "feel" (IMO).

You and I had that very same conversation way back..we're both on the same page. We tried and got a little closer with the freeware ABG Cabinet Enhancer, although it helped its still not quite there...

ANYWAYS Agreed, I think Nebula still remains the best option to date...
KM

guitarzan · Post by **guitarzan** » Sun May 23, 2010 10:52 pm

JohnnyMcFly wrote:ANYWAYS Agreed, I think Nebula still remains the best option to date...

Yeah...I've been thinking about cab sims lately and I think it's the last big frontier for amp simulation

. Nebula is cool but it still doesn't do the natural compression/limiting/distortion thing a cab does and it has a lot of system resource overhead, so I'm starting to think more along the lines of parallel IRs combined with various dynamic processes like multi-band compression, exciter/wave-shaper...really somewhat like the stuff you were trying on your acoustic sim, because it shares a lot with that type of thing... the speaker and cab is like the body of an acoustic guitar, or like an acoustic drum, so a lot of the techniques used to model those would be useful in a cab sim.

Well, if you've read any of my posts over the last week you can tell I'm trying to rally support and interest in cab sim research and development...let me know when you'll be revisiting your cab sim development - I'll be there!

strangedogs · Post by **strangedogs** » Mon May 24, 2010 12:01 am

JohnnyMcFly has some great news for you then - I'll let him tell ya'

p.s. this is the largest banner they would allow - 300 pixels wide and 1 line of text is max for a sig

Ken McLaren · Post by **Ken McLaren** » Tue May 25, 2010 4:26 am

Yeah we're going to be releasing the Cabs as a separate VST to Shred 1.5x. Now thats not it either, we have plans on something else new called Cab Lab..Which allows you to create your cabs, and then of course, we can't forget Cab Enhancer 2.0 which is very dynamic to itself. So, by no means have we stopped with Cab research, we just took a detour to get Shred out, and happy to say we're really close to releasing version 1.5x of that.

Now, on the topic of cabs, I have found that they are nothing without a decent room simulation as well. The one released in Shred 1.06 was the first foray into this, but the new one is a little closer. Just like anything else, you learn more as you go. So, I'm not talking about reverb either, I mean something that works with reflections and dampening properly, a proper room simulation. So far the best results we have obtained has been through convolution. I was always of the school that the mic in front of a cab would pick up all the reflections and dampening properties that were required, I no longer believe that..You can no more define a full reactive room in 2 ms than you can a cab, so, room sims IMO have to be included. I think we're closer to this, but just like everything else, we're still learning too...
KM

PurpleCatfishBettie · Post by **PurpleCatfishBettie** » Tue May 25, 2010 6:29 am

I'm not much of a tekkie, but have also been revisiting ideas about the 'different compression stages' of a 'signal' the past couple of days.

It's going to be great to see/hear what this new Shred is about!

guitarzan · Post by **guitarzan** » Tue May 25, 2010 8:04 am

JohnnyMcFly wrote:Now, on the topic of cabs, I have found that they are nothing without a decent room simulation as well. The one released in Shred 1.06 was the first foray into this, but the new one is a little closer. Just like anything else, you learn more as you go. So, I'm not talking about reverb either, I mean something that works with reflections and dampening properly, a proper room simulation. So far the best results we have obtained has been through convolution. I was always of the school that the mic in front of a cab would pick up all the reflections and dampening properties that were required, I no longer believe that..You can no more define a full reactive room in 2 ms than you can a cab, so, room sims IMO have to be included. I think we're closer to this, but just like everything else, we're still learning too...

Yeah Ken...I agree the room definitely needs to be represented...the ultimate would be to include acoustic feedback simulation too (maybe like Softube's) - then the entire environment would be represented.

I haven't even tried Softube' Acoustic Feedback Sim (I have a Fernandes Sustainor equiped guitar, though) so I have no idea how it works, but there's a lot of good word out on it around here.

A small article at www.musicdsp.org about Guitar Feedback says:

It is fairly simple to simulate guitar feedback with a simple Karplus-Strong algorithm (this was described in a CMJ article in the early 90's):

Code :
Run the output of the Karplus-Strong delay lines into a nonlinear shaping function for distortion (i.e. 6 parallel delay lines for 6 strings, going into 1 nonlinear shaping function that simulates an overdriven amplifier, fuzzbox, etc.);

Run part of the output into a delay line, to simulate the distance from the amplifier to the "strings";

The delay line feeds back into the Karplus-Strong delay lines. By controlling the amount of the output fed into the delay line, and the length of the delay line, you can control the intensity and pitch of the feedback note.

I don't know if that's the same type of effect as Softube's or not...just seems like feedback should be represented somehow in a virtual guitar environment!

I also read a post by the late guitar speaker manufacturer Ted Weber where it was his opinion that emulating motional impedance was key to realistic speaker cab emulation, and that it really hadn't been done yet.

Ken McLaren · Post by **Ken McLaren** » Tue May 25, 2010 8:26 am

Yeah agreed, Softube makes great stuff, and pretty interesting concept...but if you have your guitar up in volume and you are close enough to your speakers,Shred will feedback, really close to what you would get from a real amp...I mean you have to have enough gain, but it will do that...I am not sure that under every condition it will, but I want to try and isolate the best conditions for it and release it as a preset in Shred 1.5x

As much as I love their feedback device, and I am not trying to take away from it, its really quite an excellent piece of work...I still like the idea of it being real feedback. Now I guess for playing on head phones that would be really handy, but someone told me the other day that they got feedback when they placed their head close to the guitar(obviously with head phones on)...As much as I've tried I was not able to get that to happen at all with Shred...But if we could get that to happen...Man thats exactly where I want to be in terms of feedback...Now if I could only get my hearing back as a result of trying that!

MrBlitz, we are in a development stoppage right now as a result of our development system deciding it was time to die. But once we get another one we'll be really close to Public Beta. Also this time we're holding the Public beta at our own forum, thats where it will be released too when done. Its open to all members. I hope to have the first set of trials in another week after we get the new system.

We have held public betas everywhere, here, GAM, Reaper, Ultimate Metal, Ultimate Guitar..etc etc. So we just figured it was time to host one ourselves at our own home for a change..

KM

bpblog · Post by **bpblog** » Tue May 25, 2010 1:33 pm

get the hell out of my thread!

guitarzan · Post by **guitarzan** » Tue May 25, 2010 4:55 pm

bpblog wrote:get the hell out of my thread!

Resistance is futile!! Everything ends up being about AcmeBarGig......EVERYTHING!!! Bwahahaha!!!

Ken McLaren · Post by **Ken McLaren** » Thu May 27, 2010 1:58 am

HAHA....I didn't even actually start that, at least I don't think I did, see this is what happens when you get old kiddies...you forget...

I just got carried away as I have enjoyed all my conversations with guitarzan and have not spoke for a while...Or...It could have been shameless self promotion and outright thread hijacking from a serial hijacker and shameless self promoter...

sorry bpblog...I'll send you the beta of the new cabs when they're done too if you want em, which is what I probably was thinking of doing anyways before I lost my train of thought again...Damn the 80's were good to me...

KM

guitarzan · Post by **guitarzan** » Thu May 27, 2010 2:11 am

Yeah, it's every bit as much my doing, that's why I was kidding that it was like a plot or something....I actually was actively trying to bait developers into a cab sim development conversation.

Speaking thereof, Ken, I have avoided hijacking this thread further - Best amp sim for midrange "bluesy" long sustain w/o distortion? - because it is already way OT...but it is now all about cab sim development, too....maybe someone should start a Cab Sim development thread instead?

bpblog · Post by **bpblog** » Thu May 27, 2010 6:26 am

nooo you guys, continue the conversation here. there, i've edited the topic title.

and sure, those betas would be nifty, i've be planning on experimenting with different software amps and impulses in order to improve the sound of the DI recorded guitars on my tracks anyway, so there you go. ;]