Tonex and Compressors

[dean.winter]

Did something interesting today - I know that the Tonex manual says that it cannot model time based effects like compressors, but something has always seemed a bit off to me about that. After all, guitar amps are compressors (among other things), and in order to capture one correctly Tonex must model some information about time interactions, otherwise the transient and sustain behavior would be exactly the same on a super clean Fender model and a super hi-gain Mesa.

So, I figured I'd test it. I had the vintage Fender out and mic'd up, so I figured I'd try a chickin pickin, country-style sound since it (often) relies heavily on compression. I dropped a Cali 76 pedal in front of it and did the capture, and I think it worked great! I can still hear the compression going on when I play the model, and when I A/B on the amp it feels and sounds pretty much the same. The model is here if anyone is interested: https://www.tone.net/tonex/tonemodels/11988

I'm kinda curious to hear from IK on this as well - what's the story with capturing time based stuff? How does Tonex handle time based information?

Thanks,
Dean

[molonlabe23]

I had some good results with the Philosophers Rock pedal. With and without the grit switch on.

[RiF_2205]

I captured a compressor pdeal in front of my Champion 600 as well with quite some success. Although I am not sure, maybe TONEX 'thinks' that it's part of the amp's compression and might not do the compression 100% accurate and that's why IK says that it cannot capture it to be on the safe side. But still, if it sounds good, it is good .

[dean.winter]

I captured a compressor pdeal in front of my Champion 600 as well with quite some success. Although I am not sure, maybe TONEX 'thinks' that it's part of the amp's compression and might not do the compression 100% accurate and that's why IK says that it cannot capture it to be on the safe side. But still, if it sounds good, it is good .

Hi MutantLabs - that's great. Did you share the model? I'd love to hear it.

Dean

[RiF_2205]

I captured a compressor pdeal in front of my Champion 600 as well with quite some success. Although I am not sure, maybe TONEX 'thinks' that it's part of the amp's compression and might not do the compression 100% accurate and that's why IK says that it cannot capture it to be on the safe side. But still, if it sounds good, it is good .

Hi MutantLabs - that's great. Did you share the model? I'd love to hear it.

Dean

I thought I did, but actually didn't. Will upload it tomorrow and report back here.

[RiF_2205]

So, I figured I'd test it. I had the vintage Fender out and mic'd up, so I figured I'd try a chickin pickin, country-style sound since it (often) relies heavily on compression. I dropped a Cali 76 pedal in front of it and did the capture, and I think it worked great! I can still hear the compression going on when I play the model, and when I A/B on the amp it feels and sounds pretty much the same. The model is here if anyone is interested: https://www.tone.net/tonex/tonemodels/11988

I can clearly hear the compressor on your Cali76/Band-Master capture.

I tested it again today with my Champion 600 and I have to withdraw my statement before. While the compressor in front of the amp changed the dynamics of the tone model a little compared to no compressor, it was nowhere near the real thing that was set on extreme settings to really exaggerate the squishy compression.
As I said before, it sounds more as if the algorithm listened to the compressor but then applied what it heard to the amp's qualities, hence resulting in less obvious compression. And yes, I have selected Stomp+Amp.

Need to do more tests, though.

And BTW, there is this IK Premium Tone Model named "Heavy traffic" where the creator used a $3500 API 2500 compressor for the capture.

[MaxTwang]

I believe moderate to extreme attack time and release time settings will likely not be picked up by Tonex as high settings would be outside the window Tonex is likely analyzing for an amp's attack. Compression will affect tone and modify the attack of a model but Tonex, or any other 'profiler', will probably not accurately reproduce a squishy compressor or other compressor with high attack and/or release settings.

[RiF_2205]

I believe moderate to extreme attack time and release time settings will likely not be picked up by Tonex as high settings would be outside the window Tonex is likely analyzing for an amp's attack. Compression will affect tone and modify the attack of a model but Tonex, or any other 'profiler', will probably not accurately reproduce a squishy compressor or other compressor with high attack and/or release settings.

That may well be the case. What I noticed is that more moderate compressor settings are having some effect but that effect hits a ceiling and doesn't get more pronounced when the compressor settings got more extreme and result in a squishy-sound that no amp would normally produce.

[dean.winter]

I believe moderate to extreme attack time and release time settings will likely not be picked up by Tonex as high settings would be outside the window Tonex is likely analyzing for an amp's attack. Compression will affect tone and modify the attack of a model but Tonex, or any other 'profiler', will probably not accurately reproduce a squishy compressor or other compressor with high attack and/or release settings.

Possible, but I still haven't figured out the parameters under which it works and doesn't. For example, the preset I tested with originally (the one in my first post) was using an Origin Effects Cali 76, and the compression is very noticeable and playable. On the other hand, yesterday I figured I'd try something similar but move the compression, so I ran the mics into a couple 1176's I have lying around.
It looked like this:
Input 1: U87 -> API 512 -> BAE EQ -> 1176 -> Tonex
Input 2: 121 -> API 512 -> BAE EQ -> 1176 -> Tonex

I thought that since the Cali 76 is trying to do a similar thing as the 1176's it would get a similar result, but in this test the compression I was getting playing the amp did not translate nearly as much to the model. While this could be due to variations in attack and release, I thought I set the 1176's in a similar way as the Cali 76....and honestly, how much variation in Attack/Release can there be between an 1176 and a Cali 76? I mean, they're 1176's...it's pretty much "Fast" or "Really Fast"

Anyway, I'm going to keep messing with it and if I get useable sounds or come to any realizations I'll post about them.

Dean


EDIT Actually, having just typed the above post, now I'm wondering if part of the problem with the 1176's was the fact that there were two of them. When I used the pedal there was only a single compressor in the signal chain, but when I used the big 1176's there were two of them, likely with small attack/release variations. Wonder if that caused a problem with the learning algos. I could just use a single mic and one 1176 to test. Or I could mult the mics to a single channel and run that to one 1176.

Hmm...guess I know what I'm doing tomorrow.

[molonlabe23]

My tonex capture box crapped out on me after a week, I'm getting my replacement today.

Think I will try a test running into my Ashley compressor. I think the results I was getting from the Philosopher's Rock might be from the circuitry with the germanium transistor, not so much the compression, even though I didn't have the grit engaged.

Unfortunately I can't figure out how to upload a capture to toneNet to share it. I must be blind, I looked in the manual and the web site and the only instructions for sharing on toneNet are for Amplitube.

[ebiggins]

I just did a test with my Philosopher's Rock using it in the grit mode and and Tonex will not pick up the germanium circuit in it even though technically it's really a Compressor/OD with it engaged. I also did a test with my rack mount Asley SC-50 just for giggles and can say no luck with that.

My initial thoughts that Philosopher's Rock was doing something really was just loudness curve fooling my ears so I'm done messing with it in an amp capture but might try a pedal capture to see if it will capture it in grit mode with the germanium engaged.

[qpress]

I compared the sound of my amp and the Tonex. The Tonex transients are much sharper, faster. There is a need to dampen transients.

It may model the sound (saturation) of the compressor, but not its behaviour.
Before the Tonex, the same overdrive is louder when the amp is captured with a compressor. That is, the compressor compression is not captured.

[Peter_IK]

TONEX never claimed to capture compressors... From the TONEX pages:

Model amps, cabinets, combos and pedals (distortion, overdrive, fuzz, EQ or boost) using regular studio gear in as little as 5 minutes per model

So, as guessed above, some characteristics that could be in the realm of "compression" are part of an amp's tone but actual compressors will not give expected results.