Need an "listening from another romm" impulse

[006]

I don't give a fuck how mathematic things are, that to me is meaningless. It is all about the endresult and since I've proven to myself that impulses do not sound the same like a mic'ed up cab my point remains valid for me.

Exactly. They are close, and can sound good, but it's not the same at all.

Mathematically I should be president of the United States. In practice, I'm not.

Something "mathematically" being equivalent doesn't mean shit if sonically it is different man. Numbers don't mean anything when time and time again impulses can get picked out of a blind test amongst the real cab(s) very easily. That there is proof that the math doesn't mean shit. The SOUND is what matters, not the numbers. If somebody made a trashcan cabinet that sounded just like a Mesa 4x12, nobody would care about the math, they care that it sounds like the fucking Mesa 4x12 sonically.

See what we are saying?

~006

 

[Metaltastic]

Yeah, the main thing is that people on here have done exactly this:

Put the mic at whatever position. Record a swept sine being played from the speaker and perform deconvolution to extract the impulse response. If you record audio being played through the speakers with the same mic setup, the result will sound exactly the same as convolving the dry signal with that impulse response.

And there have been noticeable differences, mainly on the palm mutes where the speakers DO respond differently, so the tone pumps and breathes more than the impulse, which is like a "static" eq-curve (yes I know you don't like that word, but I think it's accurate). And also, abyssofdreams and 006 know their shit when it comes to getting good sounds, trust me on this

 

[schnarf2]

Exactly. They are close, and can sound good, but it's not the same at all.

Mathematically I should be president of the United States. In practice, I'm not.

Something "mathematically" being equivalent doesn't mean shit if sonically it is different man. Numbers don't mean anything when time and time again impulses can get picked out of a blind test amongst the real cab(s) very easily. That there is proof that the math doesn't mean shit. The SOUND is what matters, not the numbers. If somebody made a trashcan cabinet that sounded just like a Mesa 4x12, nobody would care about the math, they care that it sounds like the fucking Mesa 4x12 sonically.

See what we are saying?

~006

I think your problem is you don't understand the math in the first place, or aren't taking the effort to try to.

When these tests are being done, is it that someone positions a mic, captures an impulse, records a dry signal, and plays that through both the entire miced system, and also just through the amp and then applies the impulse response? Is the convolution plugin working right? I remember somebody posted something around here recently, and a bunch of convolution plugins aren't doing what they should be. Is the impulse being captured properly?

If something is different to your ears than the numbers are different too. If it sounds different, then its different, and that's either because of the reasons above, or because the system is not LTI. In either of those cases the math doesn't say that an impulse will work.

All I'm saying is, don't be so quick to reject the math. I'm not telling you your ears are wrong, I'm saying, if you hear a difference in a blind test, then there's more to it, and in particular there is a cause. If you did those tests and found you liked something better, there's a reason for it, and the thing to take away shouldn't just be a blanket "impulses always sound bad."

If you're trying to say that for a LTI system, you can hear a difference between the actual response and the stimulus convolved with an impulse (i.e. difference between a miced signal and a signal with an impulse applied, where the speakers do not distort), then that's totally unfounded. It's like saying that perpetual motion machines work or that gravity doesn't exist.

But if you're just saying that you can hear a difference, all I'm saying is maybe you should be aware that it's due to either a fault in methodology or the system is nonlinear, which it probably is. If it's nonlinear you shouldn't be expecting to get good results with an impulse. My point is, there's no divorce between the numbers and math, and what you hear, but rather that it's important to understand what that connection is and how to take advantage of it.

In your hypothetical trashcan sounds like a mesa 4x12 case, the math would agree! The impulse response of the trashcan would be the same as the 4x12!

 

[26]

i didnt read that carefully, but i have a feeling that there´s some misunderstanding going on.
the time the impulse lasts, might be caught mathematically correct and is maybe identical.
but that means nothing because a miced cab produces different colours of tone with ongoing time.

like i said maybe i misunderstood something....

 

[schnarf2]

i didnt read that carefully, but i have a feeling that there´s some misunderstanding going on.
the time the impulse lasts, might be caught mathematically correct and is maybe identical.
but that means nothing because a miced cab produces different colours of tone with ongoing time.

like i said maybe i misunderstood something....

You just don't understand how impulse responses work. The impulse response does capture that.

 

[26]

The impulse response does capture that.

what? the future?
sorry, i`m tired of that.

 

[006]

I think your problem is you don't understand the math in the first place, or aren't taking the effort to try to.

Yep. I hate math, but I don't do math when I am working audio. I use my ears...because it's audio.Thanks.

When these tests are being done, is it that someone positions a mic, captures an impulse, records a dry signal, and plays that through both the entire miced system, and also just through the amp and then applies the impulse response? Is the convolution plugin working right? I remember somebody posted something around here recently, and a bunch of convolution plugins aren't doing what they should be. Is the impulse being captured properly?

If something is different to your ears than the numbers are different too. If it sounds different, then its different, and that's either because of the reasons above, or because the system is not LTI. In either of those cases the math doesn't say that an impulse will work.

All I'm saying is, don't be so quick to reject the math. I'm not telling you your ears are wrong, I'm saying, if you hear a difference in a blind test, then there's more to it, and in particular there is a cause. If you did those tests and found you liked something better, there's a reason for it, and the thing to take away shouldn't just be a blanket "impulses always sound bad."

If you're trying to say that for a LTI system, you can hear a difference between the actual response and the stimulus convolved with an impulse (i.e. difference between a miced signal and a signal with an impulse applied, where the speakers do not distort), then that's totally unfounded. It's like saying that perpetual motion machines work or that gravity doesn't exist.

But if you're just saying that you can hear a difference, all I'm saying is maybe you should be aware that it's due to either a fault in methodology or the system is nonlinear, which it probably is. If it's nonlinear you shouldn't be expecting to get good results with an impulse. My point is, there's no divorce between the numbers and math, and what you hear, but rather that it's important to understand what that connection is and how to take advantage of it.

In your hypothetical trashcan sounds like a mesa 4x12 case, the math would agree! The impulse response of the trashcan would be the same as the 4x12!

Like I said, I'm pretty much a big advocate for using impulses. I use them all the time. I want them to be good enough to where there is no reason to argue against them and to where they are a widely accepted tool from amateurs to pros. But everybody that has a problem with impulses is right - they don't sound 100% identical in any way. They may sound similar. They may sound good/great with tweaking. But you can always pick them out of a blind test.

If impulses were identical, then why hasn't everyone sold off their cabinets just yet?

~006

 

[schnarf2]

what? the future?
sorry, i`m tired of that.

It has nothing to do with the future. Any system that actually exists, linear or nonlinear, has the property that future values depend on past values. You can't have a value depend on future values or that's an acausal system. Acausal systems don't exist.

With convolution of an impulse response, each output value is expressed as a linear combination of input values.

I think you have a bit of a misunderstanding about how this all works, and if there's something you would like me to clarify for you, I'm happy to, but I'm not going to argue this.

 

[006]

Ok, lets try another way of getting you to understand....if impulses were identical, then I guess alllllllllll the people who either don't use them because they don't think it sounds as good or the people that use them anyway but still acknowledge the fact that they aren't 100% (yet ), like me, are all crazy and you are right simply because "the math is correct." I guess they are all wrong when it comes to how they are perceiving the sound and need to go back to math class eh?

Give me a fucking break...

~006

 

[schnarf2]

Ok, lets try another way of getting you to understand....if impulses were identical, then I guess alllllllllll the people who either don't use them because they don't think it sounds as good or the people that use them anyway but still acknowledge the fact that they aren't 100% (yet ), like me, are all crazy and you are right simply because "the math is correct." I guess they are all wrong when it comes to how they are perceiving the sound and need to go back to math class eh?

Give me a fucking break...

~006

I don't think you're listening. If it sounds different, obviously the numbers are different too. All I'm saying is that means the system isn't linear, which isn't surprising. So an impulse won't sound exactly the same.

However, in the case of wanting the sound of listening to something from another room, that is linear, and an impulse will perfectly reproduce that.

My point is, there's no separation between what you hear and the math, and it is, in my view, a lot more productive to be able to say, "I hear this difference and know it is for this reason," rather than "I hear a difference and fuck you." It's worthwhile to know when one technique is appropriate, because, for example, you can say, "I know an impulse will work perfectly for capturing the sound of this room", and some other time say, "This amp is really cranked so an impulse won't work so well."

There's no special subjective voodoo. You're saying fuck the math, I hear this, but I'm not saying, fuck your ears, I'm just saying, there's math that explains what you're hearing, and it's productive to have some sort of idea how things work.

 

[schnarf2]

Look, I don't really want to argue this point anymore. I don't think we're really disagreeing.

I think the only fundamental difference here is that some people are saying that as a blanket statement, impulses don't sound as good as the real thing. I am saying that there are situations in which they sound (and are) exactly the same, and situations in which they don't, and there is a way to characterize those situations and it's useful to know that.

 

[abyssofdreams]

Where are Jeff and Ermin when we need them.

 

[abyssofdreams]

Look, I don't really want to argue this point anymore. I don't think we're really disagreeing.

I think the only fundamental difference here is that some people are saying that as a blanket statement, impulses don't sound as good as the real thing. I am saying that there are situations in which they sound (and are) exactly the same, and situations in which they don't, and there is a way to characterize those situations and it's useful to know that.

First thing I mostly agree with you. But the statements weren't blanket just because you label them so. I just don't need to justify my claims each and every time I post a common FACT on this board. Thanks for getting your point across and now it's time you be a little bit more peaceful and enjoy this space

 

[Metaltastic]

Where are Jeff and Ermin when we need them.

Haha, SERIOUSLY

 

[26]

It has nothing to do with the future. Any system that actually exists, linear or nonlinear, has the property that future values depend on past values. You can't have a value depend on future values or that's an acausal system. Acausal systems don't exist.

i know, i`m a determinist too ;-)
i think it`s all in what you said about linearity/non-linearity (the disagreement imo stems from us being guitar cab impulse people in the first place) .
the "future" i meant, is the changes the non-linearity of my cab produces.

 

[VH100R]

I think your problem is you don't understand the math in the first place, or aren't taking the effort to try to.

He doesn't need to. He only cares about the end result. (like the rest of us)

The math is simple in this case:

1+1=2

IRs are great and are one of the most exciting things to hit the recording community lately, but for most people in this forum (myself included) using a real cab+mic combo is better that using an impulse period.

Sure we can talk about signal convolution and Fourier but... our ears are the ones that matter.

 

[Morgan C]

I'm gonna have to take Schnarf's "side" here, perhaps to clarify because I don't think you guys understand him totally. He's saying that while a cab sound is non-linear (it changes when you do palm-mutes, or whatever, and over time, etc. which you are all arguing yet agreeing on), 'listening from another room' is linear (temperature differences, etc. are so minute they can pretty much be discounted). So while impulses are not 100% on non-linear sources such as guitar cabs, they are 99.9% on sources such as "listening from another room".

edit: to clarify further, he's NOT saying that impulses are perfect for guitars. Cabs are obviously a better choice. But for a linear source impulses can do the job 99.9% as well as the real thing. This whole argument of "math" vs "sound" is ridiculous. If they don't match up, one (or both) is wrong, usually from oversimplification. The math DOES work, just not for a non-linear source, which is why there's a sound difference.

 

[Erkan]

I don't give a fuck how mathematic things are, that to me is meaningless.

Not to be an ass but you do realize all of your entire audio, project, effects etc are math, unless you're recording straight to tape which I doubt you do? To not give a fuck about the math when you're working on a digital system is like not giving a single fuck about the wood in your guitar, drums, cabs, or the material in the speakers etc etc. Don't be silly.

I understand what he is trying to explain to everyone, and I "support" him on this because... well, I just get what he's trying to explain. Impulses are not the way to capture a 99.9% accurate sound of a guitar cab, it just isn't. It's not that impulses are bad or "worse" than the real stuff, it's just that they're not made for guitar cabs. I'm not that into Nebula and all that but from what I've heard, that would be more appropriate for guitars.

I'm pretty certain I could bet one of my cymbals on that the impulse of something linear would sound exact enough for the human ear to not be able to hear a difference. Can you hear the difference between something that has the amplitude value of 64389 and 64399?

When we KNOW that 2 sounds are different in some way, though they are supposed to sound the same, we automatically try to force ourselves, to convince ourselves that there is a difference and we can hear it, and we start describing it in all kinds of fucking fuzzy words like "this sound has more shimmer to it", "this sounds a bit more cloudy" ... WTF? However, if some random guy on the forum posted a real miced version of something linear, and then an impulse of it, and if he didn't explain anything, if he just went "Hey guys, lol, here are some sounds from my room to listen to", you wouldn't hear any difference.

Really, I'm not trying to be a smart ass and I'm not trying to say that you don't hear a difference between an impulse cab and a real cab. You just have to understand why it is like that, and what impulses REALLY can reproduce 99.9%. Just because it can't do it on a cab doesn't mean you should trash it, there's plenty of uses for impulses. The guitar cab should be the least use of an impulse... but still it works good enough to get your music across.

 

[Morgan C]

"this sounds has more shimmer to it" ... what the fuck?

LOL. I've always felt so embarrassed because I couldn't hear the shimmer, glad I'm not alone XD


I don't think anyone has done a test of a linear source with impulses vs real thing. Would be interesting, a blind test. Blind tests with cabs have been proven for impulses to be worse, but I honestly don't think anyone here would have more than a 50/50 chance at picking the impulse from the real one in a linear source test.

 

[abyssofdreams]

Not to be an ass but you do realize all of your entire audio, project, effects etc are math, unless you're recording straight to tape which I doubt you do? To not give a fuck about the math when you're working on a digital system is like not giving a single fuck about the wood in your guitar, drums, cabs, or the material in the speakers etc etc. Don't be silly.

I understand what he is trying to explain to everyone, and I "support" him on this because... well, I just get what he's trying to explain. Impulses are not the way to capture a 99.9% accurate sound of a guitar cab, it just isn't. It's not that impulses are bad or "worse" than the real stuff, it's just that they're not made for guitar cabs. I'm not that into Nebula and all that but from what I've heard, that would be more appropriate for guitars.

I'm pretty certain I could bet one of my cymbals on that the impulse of something linear would sound exact enough for the human ear to not be able to hear a difference. Can you hear the difference between something that has the amplitude value of 64389 and 64399?

When we KNOW that 2 sounds are different in some way, though they are supposed to sound the same, we automatically try to force ourselves, to convince ourselves that there is a difference and we can hear it, and we start describing it in all kinds of fucking fuzzy words like "this sound has more shimmer to it", "this sounds a bit more cloudy" ... WTF? However, if some random guy on the forum posted a real miced version of something linear, and then an impulse of it, and if he didn't explain anything, if he just went "Hey guys, lol, here are some sounds from my room to listen to", you wouldn't hear any difference.

Really, I'm not trying to be a smart ass and I'm not trying to say that you don't hear a difference between an impulse cab and a real cab. You just have to understand why it is like that, and what impulses REALLY can reproduce 99.9%. Just because it can't do it on a cab doesn't mean you should trash it, there's plenty of uses for impulses. The guitar cab should be the least use of an impulse... but still it works good enough to get your music across.

I am only defending my own words. I never said anything about that I don't understand his viewport, nor did I jump into the discussion about linear/non-linear issues and/or for "listening from another room"-impulses or for that matter according to a cab.

And I work on trial and error. If you work with a mathematical approach and you take that into account and it works for you then that's fine, nothing against that.

And I don't need to know the mathematical processes that take place as long as I know which knob does what, same goes for the wood used in a guitar. If Wood A works better than Wood B then that's all there is to it, really. I don't need to know what they did to the wood so that it works. That to me, is totally meaningless. I just put a label on it and say to myself: hey, if I combine this guitar with these strings and these pickups it sounds awesome.

Anyhow, I realize we're simply talking about different things here.