Did CAB IR`s (Impulse Response technology) really represent the real thing?

Morphosis

Morphosis

Official G66 Support
Yesterday i did an A/B comparison between a miced Speaker Cabinet and the counterpart Impulse Response, i captured with the Axe-Fx II. Recorded OUT2 (miced Cab w/o CAB IR) and OUT2 (CAB IR from that cab/mic/position) the same time. Here is the result:

Test 1.: JM45 + Strat -> Eminence Wizard -> SM57 Capedge on grill

2013-01-15%2018.06.25%20%28Klein%29.jpg


MICED CAB


IR


Test 2.: Roomsound vs. Direct-Signal: Two Room Mic`s, Sound is the same like above!
Note: I switch beetween the two room mics and seperate them later into stereo panorama. So you`ll can get the idea what`s all about Amp-in-the-room vs. DI / Nearfield recordings ... sorry for different playing in the next sample, but my interface has just 2 Inputs ...

2013-01-15%2018.48.09%20%28Klein%29.jpg


ROOM MICS (SM57 + ECM8000)



Double checked the Signal routings to be save not recording the same signal twice with another example:

Test 3.: Friedman BE + ES335 -> Eminence Governor -> Beta87A Capedge on grill

2013-01-15%2019.27.17%20%28Klein%29.jpg


MICED CAB


IR



For those who won`t believe: Here are the original 6 .WAV files in zipped folder. RAW as recorded. Here you can check by inverting the phase, that the files are NOT identical. If so, the Sound would cancel completeley out ... it does not!

https://dl.dropbox.com/u/47593922/BOX vs IR + NEARFIELD vs ROOM.zip
[48KHz / 24bit / .WAV format - RAW as recorded]


CONCLUSION:
1.) Impulse Response technology is superior to represent real Speaker-cabinets. Real miced cab vs IR sounds almost identical.

2.) DO NOT compare acoustic recordings (mics in a room) with DI-recordings. This is senseless and illegitim -> compare Test1 with Test2 and think about "3D" and depth tonewise ...
 
Last edited:
Awesome!That's some great comparison,and anwers questions for me! Very accurate! What power amp are you using ?
 
RDH said:
Awesome!That's some great comparison,and anwers questions for me! Very accurate! What power amp are you using ?
Click to expand...
Matrix GT1000

I hope that this open eyes for some people ... This comparison is apple to apples IMO! Because the whole signal chain is identical. One time i capture the IR with the chain, next time i recorded the miced cab (and simult. the IR routing). But the chain is identical incl. Box, mic. ,position and recorded guitar thru the same Amp-Block.

The only difference is the added 3m Audio cable while capturing the IR. For total accuracy i recommend to check out the RAW recorded files i posted under the dropbox link ....

We can discuss, if there is an audible difference, because of room reflections hitting the nearfield mic, which are longer than the 42ms or if 3m cable does matter ... For me personally, questions ended latest after this comparison session I´ve made yesterday.
 
Last edited:
Thanks Man! Awesome! After this demonstration it makes me mention again how much I'd like to see the actual ir's from the amps cliff models be available!Maybe they're no better than what is available,but I can't help believing they would be accurate to the amps!! When Cliff posts clips they just sound right!
 
Yeah it's amazing the number of people that think you can't capture "the magic" of a "speaker moving air" with an impulse response.

I usually request them to make an mic recording (while at the same time recording the raw output of the axe-fx. Then I have them do a sweep without them moving the mic so I can create an IR.

I post process the axe-fx raw track and compare it against the mic recording. You can even phase invert after time alignment and just hear the difference. In one user session, you could hear the only difference was the mic picking up the user hitting the strings (acoustically) and some really low level room reflections depending on how far the mic was from the cabinet. In short a true impulse response of just the cabinet.
 
hah, i've already been waiting for this, great test! :)

on the clean track i can't hear ANY difference, on the other there is a slight difference.
i am still not 100% convinced though. it seems that the IR gets more and more "innacurate"
the more distortion you add.
 
check the original files in your DAW! Loop some seconds which represent differences, you`ll think you hear. A/B it solo/mute automatically with one click to switch fast the files directly in your DAW (original 48khz / 24bit). Take Care of different levels ...

Now: If you hear differences any longer, is it dynamically or tonewise all over? When you still think there is a gap: think about the 3m more cable i needed to take the IR and think about room reflections longer 42ms, the IR cannor represent. No? Other difference you think you hear? A difference bigger than change you ears in the rehearsal room 30cm from your real cab? A difference which make it really different?
 
Morphosis said:
We can discuss, if there is an audible difference, because of room reflections hitting the nearfield mic, which are longer than the 42ms
Click to expand...

You could always emulate these by shooting a reverb impulse response of the room with your computer and FRFR system, then add it back in your DAW.
 
antcarrier said:
You could always emulate these by shooting a reverb impulse response of the room with your computer and FRFR system, then add it back in your DAW.
Click to expand...

Shure, but this is not my intension. For recording i can place convolution reverbs like Altiverb in the chain or create longer IRs using in external VST CAB Player.

My intension is perhaps using my guitar cabinet as Monitor for myself and feed DI-Signal to my in-ear and FOH with IRs of my cab for live playing.

The advantage doing this is to have three cabinet sounds in real (2x12" stereo cab with Wizard + governor = Wizard (similiar to the G12H30) for clean and classic rock, Governor (similiar to V30) for hard and heavy stuff and both, when this will fit the bill better ...

I can reach these three variations of CAB sounds by using the Balance Parameter of the FX LOOP to OUT2 ->Matrix -> physical stereo Cab (LEFT->Wizard, RIGHT->Governor, NOON-> both for Monitor an match this with correspondending IRs with stereo cab block panned hard left/right as DI-Signal to OUT1 for FOH.

I like the idea to get the feel from a real Cab again, tweak the Amp-Block the same way as a tube head + cabinet not hearing thru a "transmitter technology" like FRFR with third party IRs. You`ll never be shure, if the FRFR System is representative, or the third party CAB IR is lacking, or you tweak Amp Parameters to compensate unknown lacks in the cab/monitor section.

Tweaking the Amp-Block by monitoring thru the real cab feels for me easier, because no "medium like FRFR to hear anything" is in the chain. Also i get the guitar cabinet feeling and response.

Playing with a real cab is no compromise in possible lacking FRFR. IRs from THAT cabinet give me all the advantages, DI-Signals can have ... best of both worlds (for me)!
 
Great work! I wish I had some cabs but since I sold my Mesa Mark V to finance my Axe II I dont have any real amps or cabs anymore. Well I dont miss it!
It would be great to have some cabs to make IRs from since I have almost finished my masters degree in Acoustics and have access to all the equipment needed to make flat IRs. Anechoic rooms, flat mics and so forth.
Regards/ Jesper
 
Morphosis said:
check the original files in your DAW! Loop some seconds which represent differences, you`ll think you hear. A/B it solo/mute automatically with one click to switch fast the files directly in your DAW (original 48khz / 24bit). Take Care of different levels ...

Now: If you hear differences any longer, is it dynamically or tonewise all over? When you still think there is a gap: think about the 3m more cable i needed to take the IR and think about room reflections longer 42ms, the IR cannor represent. No? Other difference you think you hear? A difference bigger than change you ears in the rehearsal room 30cm from your real cab? A difference which make it really different?
Click to expand...

it's more dynamically, which at the end affects the tone. it's hard to describe but it's dedfinitively there. well, the strange thing is why it seems to affect distorted tones only...
 
Good job. +1 to what AlbertA said.

What an IR won't capture is any speaker distortion and speaker/guitar interaction differences. Speaker distortion is mostly irrelevant though since it is typically much less than the amp distortion. Furthermore, we simulate it anyways. Speaker/guitar interaction causes reinforcement of bass and low mids so playing through a cab at moderate to high volumes will sound slightly different than playing through the IR and listening through monitors or headphones. This becomes more prevalent at higher gains because there is more reinforcement at higher gains (this is why you can get controlled feedback easier at higher gain). The difference is so minute though that it doesn't really matter. It cracks me up that people nit-pick about minute differences when those differences are minuscule in comparison to the difference between the response of two monitors (even two of the same brand and model).

You can compensate for the difference in interaction by putting EQ before the amp block and boosting the bass and low mids slightly. Or use big monitors and crank 'em.
 
German guy lacking on his english skills ...
Morphosis said:
A difference bigger than change you ears in the rehearsal room 30cm from your real cab? A difference which make it really different?
Click to expand...

The "Master" himself...
FractalAudio said:
It cracks me up that people nit-pick about minute differences when those differences are minuscule in comparison to the difference between the response of two monitors (even two of the same brand and model).
Click to expand...


... puhh ... :)
 
AlbertA said:
Yeah it's amazing the number of people that think you can't capture "the magic" of a "speaker moving air" with an impulse response.

I usually request them to make an mic recording (while at the same time recording the raw output of the axe-fx. Then I have them do a sweep without them moving the mic so I can create an IR.

I post process the axe-fx raw track and compare it against the mic recording. You can even phase invert after time alignment and just hear the difference. In one user session, you could hear the only difference was the mic picking up the user hitting the strings (acoustically) and some really low level room reflections depending on how far the mic was from the cabinet. In short a true impulse response of just the cabinet.
Click to expand...

I still don't quite buy this yet and am one of the said knit pickers :D. Countless tests over the years have generally shown the IRs to lack dimension when compared to the mic'd speaker cab and many producers and engineers steer well clear if possible because of it. This is the case in the OP to my ears (although I do note the mic'd clip is also louder). That's not to say I don't use IR's though, they have their place of course. To me they still slightly feel rigid, like a picture rather than a flowing video most of the time. What cliff mentions about the low mid frequencies though is interesting as that is generally the frequency area that is the troublesome one...

Interesting idea on the phase flip btw. I might give that a go when making some IR's and see if it'll help put my demons to rest!
 
Last edited:
themaidenmaniac said:
(although I do note the mic'd clip is also louder)
Click to expand...

For correct A/Bing all my files are always normalized to -0.3db. If the miced clip is louder (PEAK?), it must be more compressed (RMS), no? - can`t believe this ... or my DAW works incorrect ... don`t believe this also ...

anyhow ...
 
You must log in or register to reply here.
Back
Top Bottom