the difference between 170ms and 500ms IR lenghth

S

Soultrash

Guest
maybe i am picky but there is quite a lot of "information" missing in the 170ms IR...

https://www.dropbox.com/s/hiuwicwx76cgxpt/170vs500.wav?dl=0





tenor.gif
 
Last edited by a moderator:
There will also be missing information in an IR that is 500 ms compared to a 5 second one. The length required to obtain all possible information is determined by the decay time of the recording space.

All this information isn't always necessary though. When I capture IRs, I crop them short on purpose in order to remove the room reflections contained in longer IRs.

Any information past 170ms will be well into reverb tail territory anyway (I crop my IRs to be much shorter than this). Personally I'd prefer to have manual control over this using reverb.
 
Last edited:
Any information past 170ms will be well into reverb tail territory anyway (I crop my IRs to be much shorter than this). Personally I'd prefer to have manual control over this using reverb.

i do not hear any "reverb" in the soundfile i posted but i hear frequencies that are simply missing in the 170ms IR,
i think that's quite a big difference any may be the key why short IR's sound so flat, lifeless and "two dimensional"
compared to a micd signal.
 

"Truncating an IR destroys information by definition. We don't care where the information comes from, be it the speaker or the room or the mics or the preamps. We want all the information. If a plot of the frequency response of a truncated IR differs considerably from the non-truncated version then we have lost information and concomitant accuracy."

maybe i do not understand this correctly but if i do this actually confirms my point that 170ms IR's are OK but far from being an accurate representation of a real cab
 
The sound of a cab is adequately captured at 170 ms or less. Additional length doesn't capture the cab itself but the reflections of the sound of the cab in the room.
 
There will also be missing information in an IR that is 500 ms compared to a 5 second one. The length required to obtain all possible information is determined by the decay time of the recording space.

All this information isn't always necessary though. When I capture IRs, I crop them short on purpose in order to remove the room reflections contained in longer IRs.

Any information past 170ms will be well into reverb tail territory anyway (I crop my IRs to be much shorter than this). Personally I'd prefer to have manual control over this using reverb.
Kevin at Ownhammer would disagree with you.

I tried his experiment and that is how I decided to follow Kevin's example of also offering 200 ms & 500 ms at 96Khz.

Don't forget, when you mic a cab, even in the deadest of rooms, you are not only capturing the speaker's on axis and off axis sounds, you are also capturing to a lesser degree the resonance of the cabinet which has it's own harmonics and resonances.

When you elongate the duration of the sweep, you are sampling more times per second each waveform through the travel of that sweep.

20 Hz alone takes 50 ms to develop one complete wavelength. 30 Hz alone takes 33.34 ms to develop one complete wavelength.

https://www.unitjuggler.com/convert-frequency-from-Hz-to-ms(p).html?val=30
 
Last edited:
The sound of a cab is adequately captured at 170 ms or less. Additional length doesn't capture the cab itself but the reflections of the sound of the cab in the room.

and i think that the reflections are significant to how the cab sounds at the end.
i also would not necesarily call these reflections "reverb"
 
Kevin at Ownhammer would disagree with you.

I tried his experiment and that is how I decided to follow Kevin's example of also offering 200 ms & 500 ms at 96Khz.

Don't forget, when you mic a cab, even in the deadest of rooms, you are not only capturing the speaker's on axis and off axis sounds, you are also capturing to a lesser degree the resonance of the cabinet which has it's own harmonics and resonances.

When you elongate the duration of the sweep, you are sampling more times per second each waveform through the travel of that sweep.

20 Hz alone takes 50 ms to develop one complete wavelength. 33 Hz alone takes 33.34 ms to develop one complete wavelength.

https://www.unitjuggler.com/convert-frequency-from-Hz-to-ms(p).html?val=30

sounds logical :)

my main point about all these IR discussions is that FAS is extremely concentrating on amps,
parameters and refining them over and over again getting it to sound closer to the real thing but
a great recorded guitar tone is mostly defined by the sound of the cab so why is there no more development
in this sector which seems quite more important to me than anything else.
 
There shouldn't be that noticeable a difference. 50 ms is enough to capture most cabs. There is a marked difference between the two sounds which indicates something was not done correctly. I would need the original .wav files to investigate further.
 
Kevin at Ownhammer would disagree with you.

I tried his experiment and that is how I decided to follow Kevin's example of also offering 200 ms & 500 ms at 96Khz.

Don't forget, when you mic a cab, even in the deadest of rooms, you are not only capturing the speaker's on axis and off axis sounds, you are also capturing to a lesser degree the resonance of the cabinet which has it's own harmonics and resonances.

When you elongate the duration of the sweep, you are sampling more times per second each waveform through the travel of that sweep.

20 Hz alone takes 50 ms to develop one complete wavelength. 30 Hz alone takes 33.34 ms to develop one complete wavelength.

https://www.unitjuggler.com/convert-frequency-from-Hz-to-ms(p).html?val=30

The length of the sweep only determines the signal-to-noise ratio. If the room is completely silent the sweep can be infinitely short (an impulse). To overcome ambient noise you need more energy in the applied stimulus. With an impulse you can only increase the power so much before the amplifier or the speaker or the mic or the preamp, etc. distort. However, if you spread that power out over a longer period of time you can increase the energy and therefore increase the SNR. Think of it this way: a 1 ms pulse at 1000W has the same energy as a 1 second pulse at 1W.

Now you can't just put a 1 second pulse into a system because the pulse has little frequency content. A 1 second sweep over the band of interest allows the transfer function (IR) of the system to be obtained via deconvolution. There are other signals you can use like pseudo-noise and MLS sequences but a "chirp" has the best characteristics.

In the early days of room IR capturing they used impulses generated by popping a ballon, firing a starting pistol or clapping two boards together. The results were poor due to low SNR. This lead to the development of signals that have higher energy.

To get the IR of a room long sweeps are typically used because there is a lot of ambient noise and the "returned signal" is weak (the reverb portion of the response is very low compared to the direct signal). When close mic'ing a speaker the ambient noise is low and the signal strength is very high so a short sweep is adequate. In fact you could probably get away with 100 ms or less in a studio environment.
 
There shouldn't be that noticeable a difference. 50 ms is enough to capture most cabs. There is a marked difference between the two sounds which indicates something was not done correctly. I would need the original .wav files to investigate further.

what's the correct way to do such a test?
 
what's the correct way to do such a test?

Post the .wav files and I'll do an FFT on them.

I've never seen a cabinet IR (and I've examined thousands) that has any significant content beyond 150 ms or so. Most cab IRs are under 40 ms.

The exception to this would be a "room IR" where the mic is very far from the speaker and the room is significantly reverberant. But one wouldn't normally use that as the primary tone, instead to add a little ambience to the tone and the loss of information would be imperceptible in context.

Modeling products typically use IR lengths of 1K samples as this covers 90% of IRs ever captured. We support 2K and Ultra-Res (which is equivalent to 8K) which covers 99% of IRs.

The amount of CPU power required to process an IR is proportional to the length of the IR. To support a 500 ms IR (24,000 samples) would require over ten times the CPU power of a 2K sample IR. It also requires over ten times the memory for storage. Given that that vast majority of IRs do not have any information beyond 40 ms it is wasteful of CPU and memory resources to support IRs longer than 2K.
 
i do not have the files anymore but if you find the time you can post an example, maybe i did something wrong!

i recorded the short riff, copied it, added the two different lenghth IR's and then flipped the phase on one signal,
i can't think of any other way how to set up sch a test.

regarding the needed processing power/memory it sure makes no sense using longer IR's
i fully understand that, but in general when speaking about the most accurate representation
of a mic'd cab i still think that longer IR's are "better"

then again, i am not really chasing for a "perfect representation",
i am mostly trying to find out why IR's sound like they sound compared to a mic'd cab.

i've heard quite a lot of samples, even where the mic position that was used to record the orignal take
was not changed when the IR was shot.

people claim that it sounds exactly the same and that they can not hear a difference
but i do clearly hear a difference especially the mid frequencies sound completely different...
 
i do not have the files anymore but if you find the time you can post an example, maybe i did something wrong!

i recorded the short riff, copied it, added the two different lenghth IR's and then flipped the phase on one signal,
i can't think of any other way how to set up sch a test.

regarding the needed processing power/memory it sure makes no sense using longer IR's
i fully understand that, but in general when speaking about the most accurate representation
of a mic'd cab i still think that longer IR's are "better"

then again, i am not really chasing for a "perfect representation",
i am mostly trying to find out why IR's sound like they sound compared to a mic'd cab.

i've heard quite a lot of samples, even where the mic position that was used to record the orignal take
was not changed when the IR was shot.

people claim that it sounds exactly the same and that they can not hear a difference
but i do clearly hear a difference especially the mid frequencies sound completely different...
Have you tried messing around with motor drive and speaker drive parameters? The differences you hear could be caused by the non-linearities of the speaker
 
i have tried these parameters but on low settings there is no real difference, on higher settings the tone gets
"distorted" in a strange way. haven't checked it for a while though, iirc these parameters were changed/upgraded in the last FW,
gonna give it another try.
 
There will be a difference between the IR of a cab when Re-Amping with a modeler and the mic signal of the cab, when using a "real" poweramp (for example a tube head). Because the interaction between the power amp and the cab is extremely important. You would have to
set up exactly the same behavior virtually, which won't be possible, because even the same models from the same amp can sound different.
You could try to experiment with transformer mismatch.
 
There will be a difference between the IR of a cab when Re-Amping with a modeler and the mic signal of the cab, when using a "real" poweramp (for example a tube head). Because the interaction between the power amp and the cab is extremely important. You would have to
set up exactly the same behavior virtually, which won't be possible, because even the same models from the same amp can sound different.
You could try to experiment with transformer mismatch.

i've taken this into consideration but dismissed it because i am thinking that the AMP section in the
AXE FX II is pretty much spot on incl. power amp modeling etc.

of course, it will never be exactly the same, as you said even two exact same amps can sound different
but i thought that this difference was extremely minimal and insignificant to the sound.

i must admit i haven't tried a real amp with IR's yet, still waiting for a good loadbox to hit the market (;))
but i might be ready to unleash some Diezel VH4 madness when a certain product is available.

EDIT:
well, i just accidentally discovered something. i kind of randomly watched a youtube video
that had an amazing guitar tone.
i asked about the setup and was surprsied by the answer.

the chain starts with an ampsim, gois into a poweramp, to a cab and a mic.
the sound is extremely dynamic and clear, it has this certain "3D sound", so it
can't be the "amp", it must be the IR's.
 
Last edited by a moderator:
The length of the sweep only determines the signal-to-noise ratio. If the room is completely silent the sweep can be infinitely short (an impulse). To overcome ambient noise you need more energy in the applied stimulus. With an impulse you can only increase the power so much before the amplifier or the speaker or the mic or the preamp, etc. distort. However, if you spread that power out over a longer period of time you can increase the energy and therefore increase the SNR. Think of it this way: a 1 ms pulse at 1000W has the same energy as a 1 second pulse at 1W.

Now you can't just put a 1 second pulse into a system because the pulse has little frequency content. A 1 second sweep over the band of interest allows the transfer function (IR) of the system to be obtained via deconvolution. There are other signals you can use like pseudo-noise and MLS sequences but a "chirp" has the best characteristics.

In the early days of room IR capturing they used impulses generated by popping a ballon, firing a starting pistol or clapping two boards together. The results were poor due to low SNR. This lead to the development of signals that have higher energy.

To get the IR of a room long sweeps are typically used because there is a lot of ambient noise and the "returned signal" is weak (the reverb portion of the response is very low compared to the direct signal). When close mic'ing a speaker the ambient noise is low and the signal strength is very high so a short sweep is adequate. In fact you could probably get away with 100 ms or less in a studio environment.
this was the clip that made me think of following Kevin's example:

https://m.soundcloud.com/ownhammer/room-irs-tail-length
 
The cab itself sounds identical. What's different is the ambient room reflections. The shorter sample gives you more options since that baked in room sound is largely missing. You can then add whatever room sound you want using reverb. You can then place the track in the soundscape where you want instead of being stuck with the baked in room sound. That's also not really a fair comparison since that is clearly a room mic mix. Truncating that sample will be very obvious since there's so many ambient reflections to start with. Do the same comparison with a close mic'd cab alone in a less live room and the difference will be far more subtle.
 
Last edited:
Back
Top Bottom