Cab Sim vs. Room Reverb - What's the Difference?

[sampleaccurate]

Moderator's note: this exchange has been separated from the original "Redwirez tone recipe" thread

This topic is of intense interest to me. I've owned the Ultra almost 4 weeks and I'm now seriously getting into the user cab sims. I've used IR reverb extensively in my DAW, but never before have I used IRs for cab sims.

What's still missing?

Scott, you convinced me to purchase the BigBox IRs from Red Wire. I believe you when you say you're getting good results mixing IRs. I've been having my own fun creating cab IRs with GR3 and IRs I already have using a VST host in my computer to run parallel IRs. I disable the cab sim on my AXE before processing the output with a mix of cab sims from GR3 and "real" IRs with some interesting results. It's fun to experiment with the infinite possibilities, and I'm also having a blast exploring the capabilities.

I think mixing IRs will put the finishing touch on what I see as currently the weakest yet vitally important link in the signal chain. The main reason this will improve the realism, I believe, is that by using multiple IRs you are "filling in" the gaps or notches in the frequency response of a single IR by itself. Look at the response of most IRs and you'll see a pronounced comb filtering effect from the microphone placement. The fact that the sound from different parts of the speaker arrive at the microphone at different times is what causes the phase notching. Sound travels about 1 foot in 1 millisecond - that's an easy to remember approximation. So if you mike a 12" speaker near the edge, the sound from the far edge of the speaker reaches the mic about 1 ms later than the sound from the close edge (although it's much lower in amplitude) canceling frequencies near 500 HZ. The interaction is far more complex than that due to the fact the sound doesn't come from a point but from a large diaphragm that's 12 inches more or less in diameter, but the result is that some frequencies are notched out, that is, the EQ curve of the IR has deep, narrow depressions where a "slice" has been removed from the signals spectrum. I said in another post that most of the IRs to me sound like "caricatures" of cabs with exaggerated EQ and phase notching compared to a real guitar amp. Mixing the IRs should eliminate this problem since the average response obtained from doing so will be free of deep notches at any one frequency.

If you take a look at the EQ curves of the IRs you're creating by mixing them together I would suspect that the curves aren't as "jagged" and have fewer, much less pronounced notches in the response than single IRs. That's my experience with the GR3 cabs I've been playing with. Tonight I'll have the BigBox IRs to play with. Maybe I should have waited until I could experiment with the BigBox before writing this, but I'll have a report tomorrow.

I'd like to add another opinion to my post.

There are two serious problems as I see it with the IR method of cab simulation in the AXE. The first is being addressed by mixing IRs to get rid of those pesky notches in the frequency response. That one is or will be licked.

The second problem will have to wait for the AXE-FX II with gigabytes instead of megabytes of memory, and slots for hundreds of user cabs. The important thing is that it's designed such that the IRs are permitted to be much longer, perhaps as long as 1 second. This will require more processing power - a lot more. I believe the CPU load from an IR is proportional to its length, so 1 second is a pretty big leap (I could be wrong on that but I seem to remember reading it). But computers can process 1 second stereo IR convolutions with low CPU usage, so I see no reason the AXE-FX II can't be designed to do the same.

The reason this is necessary is that a 20ms IR isn't long enough to capture a room sound. 20ms of room reflection doesn't do much for the sound other than mold the frequency response through phase cancellation and reinforcement. The 20ms IR is effectively an EQ - a very complex one, but still just an EQ. IMO being able to capture the sound of a room and the full decay of the sound captured by the distant mikes would add the final touch of realism that is needed. As a substitute, the reverb sim module could be used to get a small room sound for a decent simulation, but it's not the same as having an IR recreate the complex pattern of delayed sound of distant mikes and real room reflections. It's OK, but a good IR would trump the reverb sim in the simulation of miking techniques that result in an IR that is longer than 20ms. I see this as the only "gap" in the effects and sims in the AXE-FX: not being able to capture the room reflections in the cabinet IR.

Are the Red Wirez IRs only 20ms long, and if so, how can they claim to capture the room reflections? If they are longer than 20ms, I guess the AXE just uses the first 20ms and truncates the decay. That's rhetorical since I'm buying the BigBox tonight and will find out for myself.

I quote the Red Wirez website: "voila, out comes the sound of your guitar played through a well-mice’s speaker cabinet, in an acoustically treated live room, run through a Neve 1073" (bold added by me for emphasis).

All sound that travels about 20 feet or more, including reflections, will be truncated by the AXE IR. That's not what I call the sound of an acoustically treated live room. Even the tiniest room will have reflections longer than 20ms. The reflection from a wall 11 feet away would be completely truncated since it would take 22 ms for the sound to make the round trip back to the mic.

I'm not here to rag on the AXE - it's the best deal I've ever gotten on a piece of gear - period. I'm just salivating at the next generation AXE that will accommodate (I hope) long stereo IRs for both cabinet simulation and room ambience and reflections and endorsing the concept of mixing IRs. Long IRs will add the only thing I can think of that my computer can do and my AXE can't. The alternative in the meantime is to run longer IRs in my DAW.

Stephen Cole

 

[Jay Mitchell]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

The main reason this will improve the realism, I believe, is that by using multiple IRs you are "filling in" the gaps or notches in the frequency response of a single IR by itself.

No. I have been constructing mixed IRs for use in the Axe-Fx two years - and looking very carefully at the spectra of both the original IRs and the mixed result - and the result you describe is most definitely not the result you will get from that process.

Look at the response of most IRs and you'll see a pronounced comb filtering effect from the microphone placement. The fact that the sound from different parts of the speaker arrive at the microphone at different times is what causes the phase notching.

What you are ignoring is that this is precisely the same phenomenon you hear when you listen to a guitar speaker live. Neither of your ears is ever equidistant from every point on the speaker cone - nor from every point on the cabinet, a huge contributor (due to diffracted energy) to the response that you hear.

So if you mike a 12" speaker near the edge, the sound from the far edge of the speaker reaches the mic about 1 ms later than the sound from the close edge (although it's much lower in amplitude) canceling frequencies near 500 HZ.

In an effort to educate, here are spectra of the same speaker due to three different mic placements:

The above are nearfield responses taken at the center of the dust cap (green), halfway between the dust cap and the outer edge of the cone (violet), and the edge of the cone (red). The differences are generally counterintuitive (unless you can intuitively solve surface integrals :lol: ).

These are response spectra of the same transducer (actually two of them in a 2x12 cab) taken at a distance of 2 meters on axis (green) and 30 degrees off axis of the median plane separating the transducers (red).

Note that the greatest differences in the NF spectra lie almost entirely above ~1200 Hz, whereas there are substantial differences in the FF spectra extending downward to 700 Hz. Note also the downward tilt of the 30 degree farfield response. This is representative of the response you would hear in a typical playing position.

I said in another post that most of the IRs to me sound like "caricatures" of cabs with exaggerated EQ and phase notching compared to a real guitar amp.

That is most definitely not due to the presence of response notches. Deep comb filtering is an essential part of the character of multitransducer cabs, particularly 4x12 cabs. If you eliminate that, you eliminate one of the most important aspects of the sonic signature of that type of cab.

Mixing the IRs should eliminate this problem

Incorrect on two counts: it is not a "problem," and mixing the spectra does not eliminate it in general.

Here is the spectrum of a mixed IR:

This is an equal mix of the two farfield IRs shown above.

Note that the notches present in the individual IRs are altered but not eliminated.

There are two serious problems as I see it with the IR method of cab simulation in the AXE. The first is being addressed by mixing IRs to get rid of those pesky notches in the frequency response.

No. "Those pesky notches in the frequency response" are integral to the sound character of guitar cabs. Whenever you play through a physical cab, you are hearing "those pesky notches in the frequency response." If a smooth, notch-free response were sufficient to simulate the effect of a guitar cab, all the cheap modelers would be just great in this department, and the Axe-Fx cab sims would offer no advantage over the others. IR-based cab sims and realtime convolution are necessary specifically because of the highly detailed (i.e., peaky/notchy) response of a physical cab.

The important thing is that it's designed such that the IRs are permitted to be much longer, perhaps as long as 1 second.

No. That will not produce an audible improvement in the authenticity of cab IRs. In fact, the ~21-ms length of the present 1024-point IRs in the Axe-Fx cab sim (note the emphasis: it's very important) is more than enough to completely contain all the direct energy from a guitar transducer.

There are reasons that close-mic'ed IRs (i.e., all but three in the factory cab sims) do not do a good job of representing the sound of a speaker that you actually hear, but your guess as to the nature of those differences is incorrect.

The reason this is necessary is that a 20ms IR isn't long enough to capture a room sound.

First, it is most definitely not necessary. Second, "room sound" is part of the definition of reverb, not cabinet simulation.

20ms of room reflection doesn't do much for the sound other than mold the frequency response through phase cancellation and reinforcement.

If the cab sim IR is worth a damn, it contains zero seconds of "room reflection." That is certainly true of all my farfield IRs.

The remainder of your discussion has nothing to do with cabinet simulation. If enough cpu power and memory become available to actually implement IR-based room reverb algorithms in real time - not very likely IMO, but not altogether out of the question - then the reverb block in the Axe-Fx will undergo a huge improvement. Improvements to the cab sim can come generally from two different areas and will be extremely subtle, to the point that their justification in terms of R&D and cpu resources may not be justified. The present 1024-point IR is extremely powerful, as many users are discovering of late.

I'm not here to rag on the AXE -

It really doesn't matter, because your critique is incorrect.

 

[sampleaccurate]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

The main reason this will improve the realism, I believe, is that by using multiple IRs you are "filling in" the gaps or notches in the frequency response of a single IR by itself.

No. I have been constructing mixed IRs for use in the Axe-Fx two years - and looking very carefully at the spectra of both the original IRs and the mixed result - and the result you describe is most definitely not the result you will get from that process.

Look at the response of most IRs and you'll see a pronounced comb filtering effect from the microphone placement. The fact that the sound from different parts of the speaker arrive at the microphone at different times is what causes the phase notching.

What you are ignoring is that this is precisely the same phenomenon you hear when you listen to a guitar speaker live. Neither of your ears is ever equidistant from every point on the speaker cone - nor from every point on the cabinet, a huge contributor (due to diffracted energy) to the response that you hear.

[quote:17tt70vw]So if you mike a 12" speaker near the edge, the sound from the far edge of the speaker reaches the mic about 1 ms later than the sound from the close edge (although it's much lower in amplitude) canceling frequencies near 500 HZ.

In an effort to educate, here are spectra of the same speaker due to three different mic placements:

The above are nearfield responses taken at the center of the dust cap (green), halfway between the dust cap and the outer edge of the cone (violet), and the edge of the cone (red). The differences are generally counterintuitive (unless you can intuitively solve surface integrals :lol: ).

These are response spectra of the same transducer (actually two of them in a 2x12 cab) taken at a distance of 2 meters on axis (green) and 30 degrees off axis of the median plane separating the transducers (red).

Note that the greatest differences in the NF spectra lie almost entirely above ~1200 Hz, whereas there are substantial differences in the FF spectra extending downward to 700 Hz. Note also the downward tilt of the 30 degree farfield response. This is representative of the response you would hear in a typical playing position.

I said in another post that most of the IRs to me sound like "caricatures" of cabs with exaggerated EQ and phase notching compared to a real guitar amp.

That is most definitely not due to the presence of response notches. Deep comb filtering is an essential part of the character of multitransducer cabs, particularly 4x12 cabs. If you eliminate that, you eliminate one of the most important aspects of the sonic signature of that type of cab.

Mixing the IRs should eliminate this problem

Incorrect on two counts: it is not a "problem," and mixing the spectra does not eliminate it in general.

Here is the spectrum of a mixed IR:

This is an equal mix of the two farfield IRs shown above.

Note that the notches present in the individual IRs are altered but not eliminated.

There are two serious problems as I see it with the IR method of cab simulation in the AXE. The first is being addressed by mixing IRs to get rid of those pesky notches in the frequency response.

No. "Those pesky notches in the frequency response" are integral to the sound character of guitar cabs. Whenever you play through a physical cab, you are hearing "those pesky notches in the frequency response." If a smooth, notch-free response were sufficient to simulate the effect of a guitar cab, all the cheap modelers would be just great in this department, and the Axe-Fx cab sims would offer no advantage over the others. IR-based cab sims and realtime convolution are necessary specifically because of the highly detailed (i.e., peaky/notchy) response of a physical cab.

The important thing is that it's designed such that the IRs are permitted to be much longer, perhaps as long as 1 second.

No. That will not produce an audible improvement in the authenticity of cab IRs. In fact, the ~21-ms length of the present 1024-point IRs in the Axe-Fx cab sim (note the emphasis: it's very important) is more than enough to completely contain all the direct energy from a guitar transducer.

There are reasons that close-mic'ed IRs (i.e., all but three in the factory cab sims) do not do a good job of representing the sound of a speaker that you actually hear, but your guess as to the nature of those differences is incorrect.

The reason this is necessary is that a 20ms IR isn't long enough to capture a room sound.

First, it is most definitely not necessary. Second, "room sound" is part of the definition of reverb, not cabinet simulation.

20ms of room reflection doesn't do much for the sound other than mold the frequency response through phase cancellation and reinforcement.

If the cab sim IR is worth a damn, it contains zero seconds of "room reflection." That is certainly true of all my farfield IRs.

The remainder of your discussion has nothing to do with cabinet simulation. If enough cpu power and memory become available to actually implement IR-based room reverb algorithms in real time - not very likely IMO, but not altogether out of the question - then the reverb block in the Axe-Fx will undergo a huge improvement. Improvements to the cab sim can come generally from two different areas and will be extremely subtle, to the point that their justification in terms of R&D and cpu resources may not be justified. The present 1024-point IR is extremely powerful, as many users are discovering of late.

I'm not here to rag on the AXE -

It really doesn't matter, because your critique is incorrect.[/quote:17tt70vw]

Where to start? Your data backs up my assertions.

Looking at your composite response curve, the first or lowest notch is at about 4 kHz, but it's FREE OF ANY DEEP PHASE NOTCHING UP TO 4 kHz. You can't say the same about the mic curves you show in your first chart. Can't you see the difference? The graph you provide showing the responses from three mic placements proves my point - the lowest notch is near 1.5 kHz, NOT 7 kHz, and it's clear from looking at the second graph that the mics are complimentary and where one has a deep notch another doesn't. "there are substantial differences in the FF spectra extending downward to 700 Hz."

You also used two IRs that are almost identical in shape to sum together to "prove" that my point about phase notching being smoothed by combining IRs was wrong. If you want to see the effect I'm referring to, use the three IRs in figure 1 and combine them together. Don't cherry pick two IRs to combine that have their peaks and valleys in the same place.

On to the IR length. Apparently Red Wire disagrees with you on your assertion that "No. That will not produce an audible improvement in the authenticity of cab IRs. In fact, the ~21-ms length of the present 1024-point IRs in the Axe-Fx cab sim (note the emphasis: it's very important) is more than enough to completely contain all the direct energy from a guitar transducer.”

Yes, it's long enough to "completely contain all the direct energy". So what? Are you saying that room reflections aren’t part of miking a cab? And if that’s true, how do you recreate them? With the reverb module? Good luck. Red Wire claims they used specially treated great sounding rooms, and that that's part of the sound. Why are some of their IRs over TEN TIMES as long as the ones the AXE can handle? One is over 200 ms, and MANY are well over 20ms. I guess the Red Wires guys don't know what they're doing, what with wasting all that good space on useless information concerning the room reflections that make up a part of the sound of ANY cab not miked in an anechoic chamber.

I constructed several IRs last night and it doesn't take two years for me to look at the curves and tell you that the notches are indeed being filled in. You're wrong in your assertion based on my tests. And you used two almost identical IRs to reach the erroneous conclusion that the notches don't cancel when using multiple mics. Try again with the 3 IRs in the first graph.

You go on to say in reference to phase notching caused by speaker miking: "What you are ignoring is that this is precisely the same phenomenon you hear when you listen to a guitar speaker live. Neither of your ears is ever equidistant from every point on the speaker cone - nor from every point on the cabinet, a huge contributor (due to diffracted energy) to the response that you hear."

Your response indicates a fundamental misunderstanding you have. I was ignoring nothing. You are overlooking that it is precisely the fact that each ear is not equal distant from the amp that is the reason you should not "force" each ear to hear the same notches recorded by one mic. Regardless of whether you're listening to a guitar cabinet or an FRFR, your ears will each hear a comb filtering effect due to the fact you mentioned yourself, each ear is not the same distance from or oriented the same with respect to the speaker, but the effect will be DIFFERENT IN EACH EAR. Why you insist that putting permanent notches on the signal that will be heard by both ears should sound natural is beyond me. Why you ignore the fact you will ultimately be listening to your sound through a speaker which will give you similar phasing effects, DIFFERENT in each ear is also beyond me.

What you are doing is the same thing I was wrongly doing with pickups. You are "stacking" phase cancellations. The first from the cab IR and the second from the interaction of your ears with WHATEVER speaker you end up using, whether it be a cabinet or an FRFR system. Are you saying that an FRFR speaker won't create phase notching or different responses in each ear as you move around the room, but a guitar cab will, therefore if we use an FRFR we need to add some artificial phase notches to make it sound like a real cab?

Then you say "If the cab sim IR is worth a damn, it contains zero seconds of "room reflection." That is certainly true of all my farfield IRs."

Why? NOBODY mics cabs in anechoic chambers. Real world miked cabs have room reflections as part of their sound. So how are YOU going to simulate that? Again, with the reverb module? HA!

I tried some of the Red Wirez IRs last night, and I hosted them in my DAW, not the AXE. Some of the IRs sounded great through the DAW, and it's PRECISELY because they weren't truncated. When I truncated the files to 20ms, or, used them in the AXE_FX the result was the same - the realism was compromised. If you think that everything beyond 20ms is irrelevant to reproducing the sound of a cab then I'd suggest you (or anyone else) do the same. Host the Red Wirez IRs, especially with some ambient mics used, in your DAW, then host them in the AXE. Sorry, but the 20ms truncation causes a significant loss of realism. I heard it for myself last night, and I believe my ears before your flawed reasoning and irrelevant graphs that don't address the issue.

The bottom line is that there is no way for the AXE-FX to accurately or convincingly create the short room reflections that are part of miking an amp. I don't care if you call it part of the cab sim or reverb, the fact is sir that it is created at the time the amp is miked, not later, but it's irrelevant anyway. It's a part of the guitar sound the AXE can't duplicate or emulate with nearly the realism of an IR. If it is reverb as you say, then the AXE-FX has a deficient reverb algorithm. I really don't care is you call it 6 or half a dozen, the result is the same.

I love my AXE, it was the best purchase I ever made, but last night after several hours of experience hosting the Red Wire CABINET IRs it became CLEARLY evident the box doesn't do it all. So what? Why can't you accept that without acting like the thing is absolutely flawless. It isn't. It has shortcomings. Get over it and enjoy.

Stephen Cole

 

[Jay Mitchell]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

Looking at your composite response curve,

First things first. You fail to mention the NF IRs, and how your speculations are completely at odds with those spectra.

the first or lowest notch is at about 4 kHz, but it's FREE OF ANY DEEP PHASE NOTCHING UP TO 4 kHz.

Look again. The lowest notch in the mixed IR is below 1kHz.

You can't say the same about the mic curves you show in your first chart.

You guessed that there would be cancellation at the "1/2 wavelength" frequency, ignoring the fact that a 12" speaker's moving diaphragm is closer to 9" than to 12". Your guess is completely wrong.

Can't you see the difference?

It is you who fails to see the differences. See above. There is no interference in any of the plots until well above 1kHz. This is completely at odds with your guesses.

The graph you provide showing the responses from three mic placements proves my point - the lowest notch is near 1.5 kHz, NOT 7 kHz,

Perhaps you should revisit the metric system. 7kHz is seven thousand hertz. My description reference seven hundred hertz.

and it's clear from looking at the second graph that the mics are complimentary and where one has a deep notch another doesn't. "there are substantial differences in the FF spectra extending downward to 700 Hz."

Which is in stark contrast to the close-mic'ed IRs, which are near-identical to above 1kHz. This specifically refutes your speculation about the notch frequencies that appear when you relocate a mic that is very close to the speaker cone.

You also used two IRs that are almost identical in shape

Uhh, no. There are substantial differences in the "shapes" of the two responses - take a look at the 800-1.6kHz range, which is audibly very important - and the combined IR retains some aspects of both shapes, but it does not eliminate the peaks and notches entirely or even substantially.

On to the IR length. Apparently Red Wire disagrees with you on your assertion that "No. That will not produce an audible improvement in the authenticity of cab IRs.

I'll rely on rediwirez to state their agreement or disagreement, thank you. FYI, I have far more technical background and technical experience in loudspeaker design, testing and measurement, and room acoustics than they do. If indeed they disagree, they are mistaken. It's a simple matter - if you understand the relevant mathematics - of the relationship between the time and frequency domains.

Yes, it's long enough to "completely contain all the direct energy".

Read carefully. Think about this, and the light may come on: Direct energy is the only energy you want in a cab sim. Why? Because you are playing your guitar in an acoustic environment. That environment adds its own reflected energy, and therefore its own set of colorations to your sound. Including the colorations due to early reflections from another acoustic environment only makes things worse. To demonstrate this highly noncontroversial fact, all you need do is listen to a youtube video that was recorded in a bedroom. It's got all the acoustic cues from the room in it. Do you really want to add that kind of crap to your guitar sound? If you do, then I can show you how to do it with the reverb block, which is the only logical place to create room sound.

Are you saying that room reflections aren’t part of miking a cab?

If you're generating an IR for a cab sim,They'd better not be. See above. Close-mic'ed IRs contain no audible room cues, because the room refections that arrive at the mic are greatly reduced in level (typically by more than 40dB) compared to direct.

And if that’s true, how do you recreate them? With the reverb module?

How about with the room you're playing in? That is always there.

Red Wire claims they used specially treated great sounding rooms, and that that's part of the sound.

Redwirez were not targeting Axe-Fx users when they decided on their acquisition methodology. In a software postprocessing environment, it may - and I stress that word - be acceptable to lump the cab sim and room cues together, but that is never the best approach. It is far more powerful and flexible to separate the two and process them independently of each other. I've been a participant in a very large project dedicated to capturing room IRs, primarly for design-level purposes. A colleague (and customer) of mine works exclusively with the creation of room reverberation. In case you're not getting this, I have a real, professional-level background in this subject.

I guess the Red Wires guys don't know what they're doing,

Perhaps you are not aware of who I am or what my background is. That's fine, but I will say that it ill-suits your purposes - whatever they may be - to attempt to start a pissing match between redwirez and myself by proxy.

Regardless of whether you're listening to a guitar cabinet or an FRFR, your ears will each hear a comb filtering effect due to the fact you mentioned yourself, each ear is not the same distance from or oriented the same with respect to the speaker, but the effect will be DIFFERENT IN EACH EAR.

As it will be when you play an authentic IR through a looudspeaker. The notches in the farfield IRs I posted will be heard almost identically by both ears up to ca. 1200 Hz. Above that frequency, the HRTF (Head Related Transfer Function) and pinnae cues will cause greater differences than the time-of-flight differences between the ears. I've done a bit of work with binaural hearing mechanisms and was a leader on an engineering team that implemented a very sophisticated binaural recording/playback system. I have some experience-based insight into what is and is not significant in aural terms.

The bottom line is that there is no way for the AXE-FX to accurately or convincingly create the short room reflections

Wrong. You're looking in entirely the wrong place for that, so it's no surprise that you aren't finding what you want. It's in there, however.

that are part of miking an amp.

Wrong. Most amps are close-mic'ed, and there are no room reflections in that signal.

I don't care if you call it part of the cab sim or reverb,

Obviously not. There are compelling reasons to keep them completely separate, however. Your failure to recognize those reasons is not an argument against their existence.

This is not a defense of the Axe-Fx. I have never said it is "flawless." You are attempting to put words in my mouth. However, if you're going to discuss flaws, my suggestion is to first identify real ones as opposed to attempting to fabricate them where none exist.

 

[MikeyB59]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

I always learn something when someone gets into it with Jay ;-)

Part of the conversation seems like apples and oranges as Jay pointed out. There are IRs for cab simulation and IRs for room sounds and IRs for etc etc etc. The IRs for the Axe are simply trying to capture the cabinet and NOT the room. Once you're extending the time or pulling the mic back, then you're into what's often done with reverb as Jay says. If you're using far mic'd IRs, you're combining the cabinet and reverb portions into one thing. There may be reasons to do it that way, but it totally makes sense to me that you don't need a long IR to capture the direct impact that a cabinet has on sound. The rest is the acoustic space and that seems like what people trying to capture cabinet IRs DON'T want to capture.

Thanks for all the back and forth. It really helped me clarify aspects of the topic in my head. I love the Redwirez IRs I have. I know that much.

Mike

 

[Jay Mitchell]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

If you're using far mic'd IRs, you're combining the cabinet and reverb portions into one thing.

This has been cussed and discussed many times over the past two years, yet many folks still miss the key point. If you properly acquire a farfield cab IR - as I have done with all of mine - there are no room reflections present in it. This is extremely important to grasp. Every IR I use is a farfield IR, but none of those IRs have any room reflections in them.

There may be reasons to do it that way,

There is one compelling reason to use farfield IRs: that is the only way that the IR can possibly contain the response that you would hear when playing that cab. Close-mic'ed IRs will always sound very different from the cabs from which they were taken.

FWIW, this sequence of posts is very much on topic: it is extremely important for people to understand the purposes for cab sims vs. other effects blocks. A 1024-point IR is more than enough to fully capture the sound of the cab. Capturing the sound of even a small room in one would require an IR that is more than a second (i.e., 48k+ points) long and the ability to convolve that IR in real time. IMO it is extremely unlikely that an upcoming Axe-Fx will have this capacity within the near term (say, less than five years.)

 

[mortega76]

Damn... so much to read... so much information... thanks for the schooling fellas!

 

[sampleaccurate]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

Looking at your composite response curve,

First things first. You fail to mention the NF IRs, and how your speculations are completely at odds with those spectra.

the first or lowest notch is at about 4 kHz, but it's FREE OF ANY DEEP PHASE NOTCHING UP TO 4 kHz.

Look again. The lowest notch in the mixed IR is below 1kHz.

[quote:3p3s5pwq]You can't say the same about the mic curves you show in your first chart.

You guessed that there would be cancellation at the "1/2 wavelength" frequency, ignoring the fact that a 12" speaker's moving diaphragm is closer to 9" than to 12". Your guess is completely wrong.

Can't you see the difference?

It is you who fails to see the differences. See above. There is no interference in any of the plots until well above 1kHz. This is completely at odds with your guesses.

The graph you provide showing the responses from three mic placements proves my point - the lowest notch is near 1.5 kHz, NOT 7 kHz,

Perhaps you should revisit the metric system. 7kHz is seven thousand hertz. My description reference seven hundred hertz.

and it's clear from looking at the second graph that the mics are complimentary and where one has a deep notch another doesn't. "there are substantial differences in the FF spectra extending downward to 700 Hz."

Which is in stark contrast to the close-mic'ed IRs, which are near-identical to above 1kHz. This specifically refutes your speculation about the notch frequencies that appear when you relocate a mic that is very close to the speaker cone.

You also used two IRs that are almost identical in shape

Uhh, no. There are substantial differences in the "shapes" of the two responses - take a look at the 800-1.6kHz range, which is audibly very important - and the combined IR retains some aspects of both shapes, but it does not eliminate the peaks and notches entirely or even substantially.

On to the IR length. Apparently Red Wire disagrees with you on your assertion that "No. That will not produce an audible improvement in the authenticity of cab IRs.

I'll rely on rediwirez to state their agreement or disagreement, thank you. FYI, I have far more technical background and technical experience in loudspeaker design, testing and measurement, and room acoustics than they do. If indeed they disagree, they are mistaken. It's a simple matter - if you understand the relevant mathematics - of the relationship between the time and frequency domains.

Yes, it's long enough to "completely contain all the direct energy".

Read carefully. Think about this, and the light may come on: Direct energy is the only energy you want in a cab sim. Why? Because you are playing your guitar in an acoustic environment. That environment adds its own reflected energy, and therefore its own set of colorations to your sound. Including the colorations due to early reflections from another acoustic environment only makes things worse. To demonstrate this highly noncontroversial fact, all you need do is listen to a youtube video that was recorded in a bedroom. It's got all the acoustic cues from the room in it. Do you really want to add that kind of crap to your guitar sound? If you do, then I can show you how to do it with the reverb block, which is the only logical place to create room sound.

Are you saying that room reflections aren’t part of miking a cab?

If you're generating an IR for a cab sim,They'd better not be. See above. Close-mic'ed IRs contain no audible room cues, because the room refections that arrive at the mic are greatly reduced in level (typically by more than 40dB) compared to direct.

And if that’s true, how do you recreate them? With the reverb module?

How about with the room you're playing in? That is always there.

Red Wire claims they used specially treated great sounding rooms, and that that's part of the sound.

Redwirez were not targeting Axe-Fx users when they decided on their acquisition methodology. In a software postprocessing environment, it may - and I stress that word - be acceptable to lump the cab sim and room cues together, but that is never the best approach. It is far more powerful and flexible to separate the two and process them independently of each other. I've been a participant in a very large project dedicated to capturing room IRs, primarly for design-level purposes. A colleague (and customer) of mine works exclusively with the creation of room reverberation. In case you're not getting this, I have a real, professional-level background in this subject.

I guess the Red Wires guys don't know what they're doing,

Perhaps you are not aware of who I am or what my background is. That's fine, but I will say that it ill-suits your purposes - whatever they may be - to attempt to start a pissing match between redwirez and myself by proxy.

Regardless of whether you're listening to a guitar cabinet or an FRFR, your ears will each hear a comb filtering effect due to the fact you mentioned yourself, each ear is not the same distance from or oriented the same with respect to the speaker, but the effect will be DIFFERENT IN EACH EAR.

As it will be when you play an authentic IR through a looudspeaker. The notches in the farfield IRs I posted will be heard almost identically by both ears up to ca. 1200 Hz. Above that frequency, the HRTF (Head Related Transfer Function) and pinnae cues will cause greater differences than the time-of-flight differences between the ears. I've done a bit of work with binaural hearing mechanisms and was a leader on an engineering team that implemented a very sophisticated binaural recording/playback system. I have some experience-based insight into what is and is not significant in aural terms.

The bottom line is that there is no way for the AXE-FX to accurately or convincingly create the short room reflections

Wrong. You're looking in entirely the wrong place for that, so it's no surprise that you aren't finding what you want. It's in there, however.

that are part of miking an amp.

Wrong. Most amps are close-mic'ed, and there are no room reflections in that signal.

I don't care if you call it part of the cab sim or reverb,

Obviously not. There are compelling reasons to keep them completely separate, however. Your failure to recognize those reasons is not an argument against their existence.

This is not a defense of the Axe-Fx. I have never said it is "flawless." You are attempting to put words in my mouth. However, if you're going to discuss flaws, my suggestion is to first identify real ones as opposed to attempting to fabricate them where none exist.[/quote:3p3s5pwq]

Sir, if anyone is putting words in anyone's mouth it's you. You misrepresented most of what I said, and you ignored the glaring errors I pointed out in your arguments or made factually incorrect statements to cover your mistakes.

Case in point: You say "You guessed that there would be cancellation at the "1/2 wavelength" frequency, ignoring the fact that a 12" speaker's moving diaphragm is closer to 9" than to 12". Your guess is completely wrong."

That was a gross oversimplification on my part on purpose, not a guess, to get people who are not experts like you to try to visualize how sound from the speaker doesn't reach the mic all at exactly the same time, you know it, and the very next sentence makes this clear. The reference to the 9" cone size is a Red Herring, and again you know it.

When anyone starts talking how smart and educated they are instead of refuting arguments with facts it throws up a red flag, and in your case it's a big one. "Perhaps you are not aware of who I am or what my background is." No sir, I don't know who you are, I don't know what your background is, and I don't care in the least. I judge you not by your experience or education but by the flawed arguments you make, your condescending attitude, and your false claim that the AXE-FX can accurately simulate room reflections without using IRs.

One of your worst answers was how you're going to simulate the room reflections. Your answer was the reverb block.

Here's where we really diverge. The reverb block sounds NOTHING like a real space. I consider the early reflection to be an extremely important part of the sound. I've set the sim on the smallest room and experimented with all of the parameters. Sorry my friend, but if you think the reverb sim sounds even close to as natural and convincing as some of the cabinet IRs that Red Wire sells then your ears are shot.

If you want to keep the cab IR a simple EQ and handle what you call reverb separately, no problem. But I have yet to hear a reverb processor, the AXE included, do anywhere as near a good of a job at creating very small spaces than an IR, and the axe won't handle IRs that incorporate the time domain response - it's truncated.

As far as the guys at Red Wire, like I expected, "you have more experience than they do and they don't know what I know" to paraphrase.

When asked how you're going to create (I should have said record) the sound of the room that the cab is in you said: "How about with the room you're playing in? That is always there." Then later you say "use the reverb module"

OK, so how are you going to capture the room sound? If the answer is the reverb module forget it. Not even close.

Some other BS statements from your post:

-"The lowest notch in the mixed IR is below 1kHz.". Sure, if you call a 3 dB dip a notch. A notch is a very high Q bandstop filter that's deep, like at least 10db deep, not 3 dB. At about 4k there is an 18dB drop over from about 3800 HZ to about 4300 HZ, which then rises 8dB at about 4600 Hz.

-"there are substantial differences in the FF spectra extending downward to 700 Hz." (concerning the two mic plots in the second figure). What planet are you on? The SHAPES of the curves match almost perfectly. Look at the locations of the peaks and dips - almost every single one is identical in position (not magnitude, but position - if the curve on the bottom were increased in level by about 12 dB it would sit almost right on top the first one. IGNORE THAT like you did all the other things I said, or distort it, whatever.

-"Which is in stark contrast to the close-miked IRs, which are near-identical to above 1kHz. This specifically refutes your speculation about the notch frequencies that appear when you relocate a mic that is very close to the speaker cone." More BS. The higher the frequency the more sensitive to mic placement it will be. 1 kHz isn't that low. Much of the tone is between 1K and 6K. So from 1 to 6K they don't resemble each other in the least. OK, I agree, your bass won't phase notch if you keep all your mics close. You know that wasn't the point.

-"The notches in the far field IRs I posted will be heard almost identically by both ears up to ca. 1200 Hz. Above that frequency, the HRTF (Head Related Transfer Function) and pinnae cues will cause greater differences than the time-of-flight differences between the ears." That's true, and I did say in one post it's the position of your ears that affect your perception, not just the time displacement, but again, it doesn't matter. How are your ears going to regenerate portions of the spectrum that have been almost completely obliterated and discarded by your IR? No answer?

-"There are substantial differences in the "shapes" of the two responses - take a look at the 800-1.6kHz range, which is audibly very important - and the combined IR retains some aspects of both shapes, but it does not eliminate the peaks and notches entirely or even substantially." I specifically exempted the region around 1 k (800-1600HZ works for me). No mention of that.

-This one is the worst and demonstrates a fundamental lack of understanding on your part. Sorry to be so blunt, but you turned this into a pissing match, not me. You say "Because you are playing your guitar in an acoustic environment. That environment adds its own reflected energy, and therefore its own set of colorations to your sound.[/i] Including the colorations due to early reflections from another acoustic environment only makes things worse. To demonstrate this highly non-controversial fact, all you need do is listen to a youtube video that was recorded in a bedroom. It's got all the acoustic cues from the room in it. Do you really want to add that kind of crap to your guitar sound? If you do, then I can show you how to do it with the reverb block, which is the only logical place to create room sound." So I don't want to add any early reflections to my sound because it will "make things worse" by combining with the reflections in the room I'm playing and listening in??? You've got to be kidding!!! And if I want to listen to my recordings through headphones what do I do, listen dry or use the poor substitution of the reverb block? Is that a joke? And YES, I do want some of the carefully tuned room reflections that Red Wire uses on their IRs as part of my sound. Not the sound from one camcorder mic (again you get ridiculous) in a bedroom, but the sound of several carefully positioned mics with some of them capturing ambient reflections lasting more than 20ms. I want this because the AXE-FX isn't realistic enough for reverb with very short decays.

I have no time to go through the rest - I wish I did. I merely speculated on the behavior of mixing IRs, and what it should and shouldn't do for your sound. I provided numerous specific examples. You ignored the most important points, misrepresented others and concentrated on those to try to deflect the incorrect assertions that you have made.

So how are you going to put that early reflection reverb on your guitar? The reverb module? :roll:

You simply can't replicate the sound of guitar amp recorded in a particular small room with ambient mics using the AXE. If you think you can do it with the reverb module then feel free to delude yourself. It's simply not that good and there are not enough controllable parameters to do so. I'll even send you an IR with a room sound on it. You won't come close to the room with the AXE-FX reverb. You can get the sound of small rooms, but the parameters necessary to customize them to sound like a particular room don't exist. An IR is far superior for duplicating very short reverb or early reflections than the module in the AXE, or any other reverb unit I've ever tried.

If you are satisfied with your cab sim being an EQ and nothing more, and want to separate it from the reverb, and use an artificial algorithm to simulate a room sound, be my guest. I'll use IRs for my rooms with superior results, but unfortunately I'll have to wait until I record the tracks to introduce room ambience on my DAW.

BTW, the discussion on pickup response was great. I learned a lot both from reading as well as my questions that were answered here on the forum. I knew very little about the subject, acknowledged so, but put forth my theories so they could be shot down (which they were) in an effort to understand the subject better. Last time it ended up boiling down to me finally understanding that the pickup EQ changes for each string depending on the open note. That one part threw me off completely, and it was one of the moderators who knocked that into my head. We still couldn't agree that the main difference in the sound of humbuckers vs single coils is the brightness of the single coils, which in my opinion but not yours overshadows the phase notching effect of the pickup's aperture.

It's unfortunate that you now seem to feel this is more of a forum for condescension and talking about your extensive expertise ("I have a real, professional-level background in this subject"; "FYI, I have far more technical background and technical experience in loudspeaker design, testing and measurement, and room acoustics than they do"; "Perhaps you are not aware of who I am".)

You gotta love that last one. Nope. I don't know who you are.

Stephen Cole

 

[sampleaccurate]

I'd like to revise my characterization of the AXE-FX reverb.

Most reverb units sound extremely "metallic" when asked to create very short reverb decay times. Only the highest quality very expensive units sound good (until IRs) when producing decay times under a quarter second or so.

The AXE-FX actually does a remarkably good job with very short reverb times and has an adjustable "early reflection" component that many don't. I would rate the short decay reverb as among the best I've heard excluding IR reverb.

Stephen Cole

 

[Jay Mitchell]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

-This one is the worst and demonstrates a fundamental lack of understanding on your part. Sorry to be so blunt, but you turned this into a pissing match, not me. You say "Because you are playing your guitar in an acoustic environment. That environment adds its own reflected energy, and therefore its own set of colorations to your sound.[/i] Including the colorations due to early reflections from another acoustic environment only makes things worse. To demonstrate this highly non-controversial fact, all you need do is listen to a youtube video that was recorded in a bedroom. It's got all the acoustic cues from the room in it. Do you really want to add that kind of crap to your guitar sound? If you do, then I can show you how to do it with the reverb block, which is the only logical place to create room sound." So I don't want to add any early reflections to my sound because it will "make things worse" by combining with the reflections in the room I'm playing and listening in???

If you actually take a look at the impulse response - log impulse squared is easier to understand intuitively - of reverbs that will work effectively when you are playing in a small room, you will see an element common to all of them: an initial time delay ("ITD" is a technical term used by acousticians, BTW), that exceeds the 20-ms length of the cab IR by several times. To make this as clear as possible, the early reflections will be provided by the local acoustic environment - the one in which you are playing - and are therefore not duplicated in the reverb setting.

And if I want to listen to my recordings through headphones what do I do, listen dry or use the poor substitution of the reverb block?

That is what the reverb block does. Spend some more time working with the advanced parameters. I'll even share some of my settings. It works a lot better than you presently give it credit for.

And YES, I do want some of the carefully tuned room reflections that Red Wire uses on their IRs as part of my sound.....I want this because the AXE-FX isn't realistic enough for reverb with very short decays.

Why not post a request for others' reverb settings, then? It is clear that you have not yet fully utilized it. You might get a pleasant surprise.

So how are you going to put that early reflection reverb on your guitar? The reverb module? :roll:

Yes.

You simply can't replicate the sound of guitar amp recorded in a particular small room with ambient mics using the AXE.

Exactly what part of "cab sim" is not perfectly clear to you? The block is not called "cab and small room sim," and the function you say you want - an IR-based reverb - is not available in any hardware device that operates in real time, including the most sophisticated (and expensive) Lexicon reverb processors ever developed.

As one who has routine contact with the technological side of this issue, I say with complete certainty that you're asking for something that has never been available. If it becomes available in the near future, my customer - whose partner is a pioneer in the field of digital delay lines and reverberation algorithms - is far more likely to develop it than anyone else. If/when that happens, I'll be sure to let you know. Unfortunately, their hardware is not sold separately. It is part of a package that involves detailed acoustic analysis of a space (and testing if the space already exists), the design of a complex sound system dedicated entirely to producing reverberation, and onsite setup by the company principal. Here's a link: http://lares-lexicon.com/welcome.html .

I'll use IRs for my rooms with superior results,

Not on a real-time hardware-based processor. No such animal exists.

We still couldn't agree that the main difference in the sound of humbuckers vs single coils is the brightness of the single coils,

Actually, we agreed completely on that issue.

which in my opinion but not yours overshadows the phase notching effect of the pickup's aperture.

Still missing an important piece. The brightness of SCs vs. HBs is caused by the smaller aperture of the SC pickup.

 

[Stratman68]

Should have known better than to expect an answer
:-(

 

[redwire]

Hey fellas,

For the record, I don't disagree with Jay Mitchell (can I call you Jay?) about the appropriateness of the Axe-Fx's IR length. I don't recall disagreeing with him about anything, actually, nor do I question his expertise.

The fact that we include IRs long enough to capture room sound was a decision based on our original goal of replicating an entire top-notch studio signal chain (including the room) in a convenient, easy to use package, not because we think more than 20ms is required to reproduce a speaker cab.

We were targeting home recordists, but were embraced by Axe-Fx users for an entirely different purpose. While our IRs weren't designed for this purpose they apparently can sound pretty good in the Axe-Fx. As Jay mentioned the close-mic'ed IRs will not have audible room reflections when truncated. The floor is most certainly a factor the further away you place the mic, however and we made no special effort to eliminate room reflections from our FF mics. It wasn't our goal.

To more closely align our sampling techniques with the goals of live players, we are in the process of doing far-field samples in a quiet outdoor space using an Earthworks TC30 reference mic, a clean Earthworks mic pre, and ground planing techniques to eliminate reflections. Based on what I've heard so far, they can stand on their own, but I think they will be most useful with a few close mic'ed IRs blended for a little push in one direction or another.

I tend to view mics as effects and delight in discovering and exploiting their unique characteristics and sonic flaws. That's where my inner audio nerd gets down and dirty with his freaky, cosplayin' girlfriend and her Princess Leia slave girl outfit. I realize that other folks have different approaches and different goals and we are trying our best to accommodate all-comers.

For what it's worth I'm glad you guys discovered our work. You're a passionate bunch and I hope to keep making improvements based on your feedback. Your input has been invaluable. Speaking of which. Why don't y'all take a listen to the Mesa Rectifier IRs we just did with a TC30. I think they might sound really killer with a little Royer thrown in for thump. There's some close mic'ed and some far-field stuff in there. We did a few more FF positions that may or may not make it into the package.

Muchas gracias. Murky buckets.

Now... who wants pie?



PS> The attachment didn't make it in the original post, so you can download it here:

http://www.redwirez.com/ir/MesaRectifierV30s-TC30.zip

 

[Scott Peterson]

Working Princess Leia in THE slave girl outfit and pie into a post? ****ing brilliant. :lol:

Keep on rocking Red Wirez, keep on rocking!

 

[Jay Mitchell]

For the record, I don't disagree with Jay Mitchell

Thanks for weighing in. I had a feeling that was how you felt.

(can I call you Jay?)

Coitainly. I've been called worse, uhh, red. Or is it Mr. wirez?

As Jay mentioned the close-mic'ed IRs will not have audible room reflections when truncated.

If you place the mic, say 2" from the speaker cone, and the nearest wall (or ceiling) is 10 feet away, the first reflection from a room surface other than the floor will be 41.6 dB lower in level than the direct sound from the speaker. It will potentially be further attenuated by the mic's directivity. If you have the ability to display your IRs in log-impulse-sqared format, you can easily see the level relationship between direct sound and room reflections. At -40dB or lower relative to direct, if you listen to the guitar track by itself, you'd be able to hear the tail, but it would have to be virtually inaudible in almost any mix.

The floor is most certainly a factor the further away you place the mic,

Since the floor is much closer than the other room surfaces it can cause colorations in close-mic'ing. That effect is not necessarily undesirable, but it is definitely there.

 

[sampleaccurate]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

-This one is the worst and demonstrates a fundamental lack of understanding on your part. Sorry to be so blunt, but you turned this into a pissing match, not me. You say "Because you are playing your guitar in an acoustic environment. That environment adds its own reflected energy, and therefore its own set of colorations to your sound.[/i] Including the colorations due to early reflections from another acoustic environment only makes things worse. To demonstrate this highly non-controversial fact, all you need do is listen to a youtube video that was recorded in a bedroom. It's got all the acoustic cues from the room in it. Do you really want to add that kind of crap to your guitar sound? If you do, then I can show you how to do it with the reverb block, which is the only logical place to create room sound." So I don't want to add any early reflections to my sound because it will "make things worse" by combining with the reflections in the room I'm playing and listening in???

If you actually take a look at the impulse response - log impulse squared is easier to understand intuitively - of reverbs that will work effectively when you are playing in a small room, you will see an element common to all of them: an initial time delay ("ITD" is a technical term used by acousticians, BTW), that exceeds the 20-ms length of the cab IR by several times. To make this as clear as possible, the early reflections will be provided by the local acoustic environment - the one in which you are playing - and are therefore not duplicated in the reverb setting.

And if I want to listen to my recordings through headphones what do I do, listen dry or use the poor substitution of the reverb block?

That is what the reverb block does. Spend some more time working with the advanced parameters. I'll even share some of my settings. It works a lot better than you presently give it credit for.

[quote:2pju8efh]And YES, I do want some of the carefully tuned room reflections that Red Wire uses on their IRs as part of my sound.....I want this because the AXE-FX isn't realistic enough for reverb with very short decays.

Why not post a request for others' reverb settings, then? It is clear that you have not yet fully utilized it. You might get a pleasant surprise.

So how are you going to put that early reflection reverb on your guitar? The reverb module? :roll:

Yes.

You simply can't replicate the sound of guitar amp recorded in a particular small room with ambient mics using the AXE.

Exactly what part of "cab sim" is not perfectly clear to you? The block is not called "cab and small room sim," and the function you say you want - an IR-based reverb - is not available in any hardware device that operates in real time, including the most sophisticated (and expensive) Lexicon reverb processors ever developed.

As one who has routine contact with the technological side of this issue, I say with complete certainty that you're asking for something that has never been available. If it becomes available in the near future, my customer - whose partner is a pioneer in the field of digital delay lines and reverberation algorithms - is far more likely to develop it than anyone else. If/when that happens, I'll be sure to let you know. Unfortunately, their hardware is not sold separately. It is part of a package that involves detailed acoustic analysis of a space (and testing if the space already exists), the design of a complex sound system dedicated entirely to producing reverberation, and onsite setup by the company principal. Here's a link: http://lares-lexicon.com/welcome.html .

I'll use IRs for my rooms with superior results,

Not on a real-time hardware-based processor. No such animal exists.

We still couldn't agree that the main difference in the sound of humbuckers vs single coils is the brightness of the single coils,

Actually, we agreed completely on that issue.

which in my opinion but not yours overshadows the phase notching effect of the pickup's aperture.

Still missing an important piece. The brightness of SCs vs. HBs is caused by the smaller aperture of the SC pickup.[/quote:2pju8efh]

I'll agree with most of that, and I'll concede a couple of other points. Didn't realize we agreed on the single coil issue - we just mis-communicated. I do appreciate your help with the Tillman articles. I skimmed them, thought I "got it", but missed the fact the damn response is dependent on the frequency of the open string. I still won't claim to have a full intuitive understanding for how that effect is created (intuitive meaning other than mathematically).

I also agree it's "logical" to separate the instantaneous response (or simple EQ IR) from the time delays beyond those of the direct sound that the mics pick up.

I don't agree that what I'm asking for doesn't exist in hardware. A computer will do it in real time quite nicely, and a computer will fit in 2 rack spaces (I built 4 of them for live audio processing). The caveat is that you will have about a 10ms or so predelay in the IR from the latency of the computer. It's inaudible to my ears as long as the dry signal is not delayed.

What got me started on this were the claims on the Red Wire site about their room acoustics, knowing that the room reflections were being discarded in the AXE files, then actually hearing how great an ambient sound created by an IR is (using my computer and a VST host). But rethinking it I agree it makes sense to keep the cab EQ direct mic sound separate from time domain effects, even if it's room acoustics. It also allows more flexibility in the creation of patches.

I'll need to experiment with the parameters more. I use one module for just ER. I use the other for long reverb. Even committing an entire reverb module to just the early reflection hasn't allowed me to create what I've heard in some short room IRs. The ideal design would be a module for cab IRs and another for early reflection or room reverb IRs. I'm satisfied with the long reverbs; they sound awesome. But rooms or spaces aren't a steady decay that smoothly fades away like long reverb, which is the typical shape of the reverb tail of hardware reverb units. Maybe I just need to do a lot more experimenting, and I will.

Stephen Cole

 

[sampleaccurate]

Hey fellas,

For the record, I don't disagree with Jay Mitchell (can I call you Jay?) about the appropriateness of the Axe-Fx's IR length. I don't recall disagreeing with him about anything, actually, nor do I question his expertise.

The fact that we include IRs long enough to capture room sound was a decision based on our original goal of replicating an entire top-notch studio signal chain (including the room) in a convenient, easy to use package, not because we think more than 20ms is required to reproduce a speaker cab.

We were targeting home recordists, but were embraced by Axe-Fx users for an entirely different purpose. While our IRs weren't designed for this purpose they apparently can sound pretty good in the Axe-Fx. As Jay mentioned the close-mic'ed IRs will not have audible room reflections when truncated. The floor is most certainly a factor the further away you place the mic, however and we made no special effort to eliminate room reflections from our FF mics. It wasn't our goal.

To more closely align our sampling techniques with the goals of live players, we are in the process of doing far-field samples in a quiet outdoor space using an Earthworks TC30 reference mic, a clean Earthworks mic pre, and ground planing techniques to eliminate reflections. Based on what I've heard so far, they can stand on their own, but I think they will be most useful with a few close mic'ed IRs blended for a little push in one direction or another.

I tend to view mics as effects and delight in discovering and exploiting their unique characteristics and sonic flaws. That's where my inner audio nerd gets down and dirty with his freaky, cosplayin' girlfriend and her Princess Leia slave girl outfit. I realize that other folks have different approaches and different goals and we are trying our best to accommodate all-comers.

For what it's worth I'm glad you guys discovered our work. You're a passionate bunch and I hope to keep making improvements based on your feedback. Your input has been invaluable. Speaking of which. Why don't y'all take a listen to the Mesa Rectifier IRs we just did with a TC30. I think they might sound really killer with a little Royer thrown in for thump. There's some close mic'ed and some far-field stuff in there. We did a few more FF positions that may or may not make it into the package.

Muchas gracias. Murky buckets.

Now... who wants pie?



PS> The attachment didn't make it in the original post, so you can download it here:

http://www.redwirez.com/ir/MesaRectifierV30s-TC30.zip

For the record I wrote"

"I guess the Red Wires guys don't know what they're doing, what with wasting all that good space on useless information concerning the room reflections that make up a part of the sound of ANY cab not miked in an anechoic chamber."

It was meant as a sarcastic comment in a discussion with Jay. I didn't mean to seriously imply that there was some "disagreement" between you.

I'm looking forward to exploring your BigBox impulses. A quick combination of some close and ambient mics sounded fantastic. I wish my AXE-FX Ultra could do that.

I would also say that you should put a note on your website concerning the truncation of AXE-FX IRs, but you offer a money back guarantee, and I can't argue with that.

Jay also makes some good points about the amplitude of the reflected room sounds. Far mics would be needed to make the room sound audible.

Thanks for letting me clear that up.

Stephen Cole

 

[sampleaccurate]

Re: Red Wirez "Tone Recipe Formula" Sharing Thread

-This one is the worst and demonstrates a fundamental lack of understanding on your part. Sorry to be so blunt, but you turned this into a pissing match, not me. You say "Because you are playing your guitar in an acoustic environment. That environment adds its own reflected energy, and therefore its own set of colorations to your sound.[/i] Including the colorations due to early reflections from another acoustic environment only makes things worse. To demonstrate this highly non-controversial fact, all you need do is listen to a youtube video that was recorded in a bedroom. It's got all the acoustic cues from the room in it. Do you really want to add that kind of crap to your guitar sound? If you do, then I can show you how to do it with the reverb block, which is the only logical place to create room sound." So I don't want to add any early reflections to my sound because it will "make things worse" by combining with the reflections in the room I'm playing and listening in???

If you actually take a look at the impulse response - log impulse squared is easier to understand intuitively - of reverbs that will work effectively when you are playing in a small room, you will see an element common to all of them: an initial time delay ("ITD" is a technical term used by acousticians, BTW), that exceeds the 20-ms length of the cab IR by several times. To make this as clear as possible, the early reflections will be provided by the local acoustic environment - the one in which you are playing - and are therefore not duplicated in the reverb setting.

And if I want to listen to my recordings through headphones what do I do, listen dry or use the poor substitution of the reverb block?

That is what the reverb block does. Spend some more time working with the advanced parameters. I'll even share some of my settings. It works a lot better than you presently give it credit for.

[quote:20ectg32]And YES, I do want some of the carefully tuned room reflections that Red Wire uses on their IRs as part of my sound.....I want this because the AXE-FX isn't realistic enough for reverb with very short decays.

Why not post a request for others' reverb settings, then? It is clear that you have not yet fully utilized it. You might get a pleasant surprise.

So how are you going to put that early reflection reverb on your guitar? The reverb module? :roll:

Yes.

You simply can't replicate the sound of guitar amp recorded in a particular small room with ambient mics using the AXE.

Exactly what part of "cab sim" is not perfectly clear to you? The block is not called "cab and small room sim," and the function you say you want - an IR-based reverb - is not available in any hardware device that operates in real time, including the most sophisticated (and expensive) Lexicon reverb processors ever developed.

As one who has routine contact with the technological side of this issue, I say with complete certainty that you're asking for something that has never been available. If it becomes available in the near future, my customer - whose partner is a pioneer in the field of digital delay lines and reverberation algorithms - is far more likely to develop it than anyone else. If/when that happens, I'll be sure to let you know. Unfortunately, their hardware is not sold separately. It is part of a package that involves detailed acoustic analysis of a space (and testing if the space already exists), the design of a complex sound system dedicated entirely to producing reverberation, and onsite setup by the company principal. Here's a link: http://lares-lexicon.com/welcome.html .

I'll use IRs for my rooms with superior results,

Not on a real-time hardware-based processor. No such animal exists.

We still couldn't agree that the main difference in the sound of humbuckers vs single coils is the brightness of the single coils,

Actually, we agreed completely on that issue.

which in my opinion but not yours overshadows the phase notching effect of the pickup's aperture.

Still missing an important piece. The brightness of SCs vs. HBs is caused by the smaller aperture of the SC pickup.[/quote:20ectg32]

I just solved 80% of my problem. I took your suggestion and played with all of the parameters some more on the reverb module. I made some definite improvements, but the biggest was one I should have thought of before now.

The problem with my room reverb has mainly been that it was too dull, or not bright or "lively" enough. So I put the reverb on the side, set the mix to 100%, and ran it through an EQ of its own to the output. Now my "room" sounds a lot more convincing.

I'm a little curious as to why my computer can run 10 simultaneous separate instances of a VST convolution reverb plugin with IRs that are 2 or 3 seconds long and yet other hardware can't come close. This is of course with a 10ms latency delay, but that could be tolerated if the dry signal were not delayed (IMO).

The AXE has only about a 1ms latency, certainly less than 2. I measured it, from pickup output to a mic placed on the speaker. The first time I measured almost 2ms, then realized I screwed up and put the mic about 10 inches from the speaker. Putting it close to the speaker brought the measurement down to about 1.4ms (interpolated between 1 and 2 ms on the scale on my DAW software).

My point is that if the AXE processor can't perform the convolution in 1.4ms, why not give it 10ms and use a 100% wet mix so the reverb can be mixed with its un-delayed, un-effected counterpart that was used as the input source to create it? I'm guessing (a big risk in this neck of the woods) the 10ms predelay on the reverb would be inaudible.

Stephen Cole

 

[schnarf]

It could and it would be possible to make a dedicated reverb unit that could do something like that. But convolution is really CPU-intensive. The algorithm that lets it be done in zero latency in the frequency domain, efficiently, is patented and will remain so for a while. Also your home computer probably has at least four cores and many gigabytes of memory, while most embedded DSPs have a core or maybe 2, and on the order of megabytes of memory.

My view is that convolution reverb isn't always so necessary. The early reflections can be done with a reasonably short convolution. After some point the density increases and the impulse response is basically exponentially decaying noise with different EQ and different decay rates for different frequency bands, and there are good-sounding efficient recursive structures for doing that. The only perceptually relevant part about this reverb tail is the frequency response and the decay times for each frequency band. In other words, given an impulse response of a room, a recursive filter that sounds the same can be computed. So basically, there's no point in wasting all that computational power on a 20 second impulse response when a recursive filter can behave the same perceptually at a much lower computational cost. See the work of Jean-Marc Jot.

 

[redwire]

Hey fellas,
It was meant as a sarcastic comment in a discussion with Jay. I didn't mean to seriously imply that there was some "disagreement" between you.

I'm looking forward to exploring your BigBox impulses. A quick combination of some close and ambient mics sounded fantastic. I wish my AXE-FX Ultra could do that.

Stephen Cole

No problem. I added a few disclaimers to the website about the IR truncation for the Axe-Fx to prevent any future confusion.

 

[Jay Mitchell]

My view is that convolution reverb isn't always so necessary. The early reflections can be done with a reasonably short convolution. After some point the density increases and the impulse response is basically exponentially decaying noise with different EQ and different decay rates for different frequency bands,

One scheme for constructing dedicated room reverb for use in "auralizations" in acoustics prediction software involves exactly this method: early reflections are modeled by a ray-tracing algorithm and become part of an IR that is convolved with the signal. The tail is constructed statistically and spliced in. The process is generally not accomplished in real time, but that is not consequential for this use, which invariably involves recorded program material.

So basically, there's no point in wasting all that computational power on a 20 second impulse response when a recursive filter can behave the same perceptually at a much lower computational cost.

Agreed. Even with this hybird approach, the cpu power required is high enough that, at present, it is unlikely that it could be implemented in a MFX unit like the Axe-Fx. Given that Cliff had to dig to find room and cpu power for the second half of the present 1024-point speaker IR and that even an early-reflection IR will need to be on the order of 48k points, a future generation platform will need a lot more horspeower to implement this. I realize it's possible, but my money says it's still a few years down the road.