• We would like to remind our members that this is a privately owned, run and supported forum. You are here at the invitation and discretion of the owners. As such, rules and standards of conduct will be applied that help keep this forum functioning as the owners desire. These include, but are not limited to, removing content and even access to the forum.

    Please give yourself a refresher on the forum rules you agreed to follow when you signed up.

Non-Minimum Phase IRs

api4u

Experienced
So.... and forgive my ignorance here.... do "non-MPT" and "RAW" mean the same thing?

I'm considering re-loading my cabs, to try things out, and will likely add a suffix of "MPT" or "RAW", if that's the case, so I can keep track of what I'm comparing. It would be awesome if there was an optional auto color coding, italics, bolding, or something along those lines, within Axe-Mange.
 
So, I usually use Ownhammer IRs for the most part, as they sound good and it takes the thinking away from searching by figuring out what cab I want and then using a quick mix. That being said, Ownhammer's recommendations are:

Recommended content if mixing with stock cabinets: 200ms MPT\48000 Hz
Recommended content if using only with OwnHammer IR's: 200ms RAW\48000 Hz

Does that hold up pretty good? I suppose the idea of mixing Ownhammer with Fractal cabs is very cool, but sounds daunting admittedly as the options are endless! haha That being said, would the safe bet be just to use 200ms RAW/48000 Hz? That's what this conversation is about, no? NON MPT IRs meaning RAW, no?
 

DLC86

Fractal Fanatic
So, I usually use Ownhammer IRs for the most part, as they sound good and it takes the thinking away from searching by figuring out what cab I want and then using a quick mix. That being said, Ownhammer's recommendations are:

Recommended content if mixing with stock cabinets: 200ms MPT\48000 Hz
Recommended content if using only with OwnHammer IR's: 200ms RAW\48000 Hz

Does that hold up pretty good? I suppose the idea of mixing Ownhammer with Fractal cabs is very cool, but sounds daunting admittedly as the options are endless! haha That being said, would the safe bet be just to use 200ms RAW/48000 Hz? That's what this conversation is about, no? NON MPT IRs meaning RAW, no?
Yes, non-MPT=raw.
In the Axe fx III and FM3 (or in Cab-Lab) you can easily align and mix them anyway even if they're from different producers.
 

pitta

Inspired
Yes, non-MPT=raw.
In the Axe fx III and FM3 (or in Cab-Lab) you can easily align and mix them anyway even if they're from different producers.
I had issues with some Ownhammer IRs: When they are imported as RAW, the resulting delay in the IR is so large it cannot be aligned with the MPT factory cabs in the Axe-FX III. I had to manually shorten the wav files.

PS: There is a thread on this board on this topic...
 

DLC86

Fractal Fanatic
I had issues with some Ownhammer IRs: When they are imported as RAW, the resulting delay in the IR is so large it cannot be aligned with the MPT factory cabs in the Axe-FX III. I had to manually shorten the wav files.

PS: There is a thread on this board on this topic...
Did you try importing them with the auto-trim processing enabled?
That's usually all you need to do but sometimes there's some noise in the leading silence of some IRs that makes the auto-trim trigger too early.
 

pitta

Inspired
Did you try importing them with the auto-trim processing enabled?
That's usually all you need to do but sometimes there's some noise in the leading silence of some IRs that makes the auto-trim trigger too early.
Yes, I did. There is a tiny bit of noise at the beginning of many OH wav files, which messes up the auto trim.
 

DLC86

Fractal Fanatic
Yes, I did. There is a tiny bit of noise at the beginning of many OH wav files, which messes up the auto trim.
Yeah, then there's not much you can do apart from manually cut them.
I suggested some time ago to add a threshold parameter for the auto-trim, or a manual-trim option where you can specify how many samples/ms you want to trim, but none of those suggestions has been considered yet.
 

greiswig

Power User
I had issues with some Ownhammer IRs: When they are imported as RAW, the resulting delay in the IR is so large it cannot be aligned with the MPT factory cabs in the Axe-FX III. I had to manually shorten the wav files.

PS: There is a thread on this board on this topic...

Haven't seen that other thread, but I ran into this issue with the Ownhammer IR's recently, too. I can't remember what I turned on or off in Cablab, but the next time I tried, that initial delay was gone. So you might try a couple of configurations of Cablab...or just a second try exporting.
 
Wait, confused now. I mostly use Ownhammer IRs. I don’t plan to mix them with Axe FX cabinets. What does that mean for me? What’s the delay and noise all about? Does that mean Ownhammer doesn’t work in axe fx for some reason? Sorry; this stuff goes over my head fast, and concise answers help the best... It’s the polack in me...
 

Holyoli

Experienced
No. If you mix non MPT OH IR's with other manufacturer's ones (or the stock one which are all MPT processed), you 'll have the delay problem at the biginning when you try to align them in Axe Edit. OH IR's mixed between them, no problem
 
No. If you mix non MPT OH IR's with other manufacturer's ones (or the stock one which are all MPT processed), you 'll have the delay problem at the biginning when you try to align them in Axe Edit. OH IR's mixed between them, no problem

Got ya. So no weirdness with simply importing and using Ownhammer IRs on their own?
 

brianv4

Fractal Fanatic
I know there's a lot on your plate right now Cliff, just wondering... oh mighty crystal ball, is a set of non-mp factory IR's still in our future. I'm addicted, my ears are craving that next fix.
 

Frank.F

Member
That's just delay. Same thing as moving further away from the speaker (or using a delay block with 100% mix). Minimum vs. non-minimum phase changes the "delay" of the individual frequency components of the waveform. In a minimum phase system the individual sine waves have the least phase possible which concentrates the energy near the start of the waveform.

For example consider a sine wave with an isosceles triangle envelope. The energy is concentrated at the center of the waveform (at the apex). The Fourier transform of that is mostly the primary frequency with a bunch of other sine waves at various amplitudes added. We can phase shift the component sine waves and the magnitude (frequency response) will not change but the waveform will. If we make it minimum phase the sine waves will add up so that the energy is concentrated at the beginning of the waveform and the waveform will then look something like a sine wave with a right triangle (ramp down) envelope.

Remember the old Behringer V-Amp. It's IRs were "maximum phase" because they forgot to reverse the data when processing (since convolution is x(t)*h(t-tau)). If you played one the IRs sounded okay but the feel was a little strange. Reversing an IR gives the exact same frequency response but adds latency equivalent to the IR length.

I am just working on my own IRs and finished my own app for shooting IRs. All is working for my purposes but last step is getting them minimum phase. Is there a "easy" way to get an IR (I have amplitude and phase spectra) out of an non minimum IR?.
All I was reading is about Hilbert transforms and zero/pole reflections, and this all looks pretty much like there should be lots of real handcraft and finet uning in choosing the details for each individual IR to really get a good result. So I am little wodering how an automatic "one button" transfer could word like this? Is there an "easy" way ?
 

AlbertA

Fractal Fanatic
So I am little wodering how an automatic "one button" transfer could word like this? Is there an "easy" way ?

Yeah, by definition a minimum phase system is one where the phase can be derived from the negative of the Hilbert transform of the log-magnitude response.

One way to compute the MPT is the following:

The complex cepstrum of a minimum phase signal is causal - i.e. 0 for negative time.
You can derive the complex cepstrum from the real cepstrum, given their relationship: i.e.

Rceps(n) = (Cceps(n) + Cceps(-n))/2; // i.e. the even part of the complex spectrum
CCeps(n) = 0, for n < 0; //i.e causal for minimum phase
CCeps(0) = Rceps(0);
CCeps(n) = 2*Rceps(n);

To estimate the real cepstrum you can do IFFT(ln(abs(FFT(original_ir))))

The forward transform of that complex cepstrum would then give you the minimum phase version of the original IR.

To forward transform the estimate of the complex cepstrum: IFFT(e^(FFT(CCeps)));

Just be aware of aliasing, since we are estimating the complex cepstrum with a DFT (in other words use a big enough FFT size)
 

DLC86

Fractal Fanatic
After the recent discussion on the subject I've spent another bit of time doing comparisons between raw and mpt IRs, and this time I did it with reflection free ones to be sure the difference I heard wasn't imputable to an alteration of the reflections contained in the IR.

When listening to recorded samples there's no hearable difference between the two, but when I play the mpt ones give a different feel.
I previously thought that was due to a different interaction between the guitar and the speakers' feedback, and partially it can be it since the mpt alters the phase response.

But the difference that strikes me the most is in the attack of the notes, it feels "dryer" with a minimum phase IR while it has a some kind of "bloom" when using a raw IR.
Sort of like knocking on a concrete wall vs a wooden door.

After thinking a bit about it and looking at the differences in the waveforms of the two impulses, I think the MPT alters the transient response, and that's what's causing that different feel.
Does it make sense?
 

FractalAudio

Administrator
Fractal Audio Systems
Moderator
After the recent discussion on the subject I've spent another bit of time doing comparisons between raw and mpt IRs, and this time I did it with reflection free ones to be sure the difference I heard wasn't imputable to an alteration of the reflections contained in the IR.

When listening to recorded samples there's no hearable difference between the two, but when I play the mpt ones give a different feel.
I previously thought that was due to a different interaction between the guitar and the speakers' feedback, and partially it can be it since the mpt alters the phase response.

But the difference that strikes me the most is in the attack of the notes, it feels "dryer" with a minimum phase IR while it has a some kind of "bloom" when using a raw IR.
Sort of like knocking on a concrete wall vs a wooden door.

After thinking a bit about it and looking at the differences in the waveforms of the two impulses, I think the MPT alters the transient response, and that's what's causing that different feel.
Does it make sense?
I doubt it. The transient is usually almost identical. It's most likely placebo.
 
Top Bottom