Properties of Reflection Free IRs

AlbertA

Member
Messages
66
What I'm asking is, what would an informed user do to make proper use of far-field IRs, that's different from what they should do with near-field ones?
It depends on your goal. If your goal is to accurately reproduce the sound of a cab captured by a reflection-free far-field IR, then you need a playback system of sufficient fidelity to do so.

Jay may be alluding (but I don't want to put words in his mouth) to the fact that in the past people having tweaked an amp model while using a close mic'ed IR, then switching to using the far-field IR and expecting magic to happen without changing any amp settings. Instead, the user should just tweak from scratch just as if you were using the real amp.
 

Chocol8

Member
Messages
318
There's a difference between a recording that used close mic'd amps and an impulse response with partial reflections; in the recording, the reflections are not chopped up. What you get in a typical commercial IR, is the response of the cabinet at that close mic'ed position + partial reflections from whatever room was in. Note I'm not imparting judgment on the sound quality (that's a subjective assessment to be done by the listener)

Note that it is still possible to capture close mic'ed responses reflection free, it's just that it's a bit harder than far-field IMHO - you still need a big room, but to eliminate ground reflections you will have to elevate the cab at least around 12ft (which is what Jay mentioned earlier in the thread); because the most direct reflection will travel 2*elevation_height - Assuming the speed of sound is 343m/s, and that you want the reflection to arrive after 20ms, then the reflection should travel roughly at least 343m/s*.020s = 6.86m. That means the 2*eleveation_height > 6.86m or elevation_height > 3.43m


Even if you had the precise room dimensions, the position of the mic, etc - the main problem with reflections is that they cause null's in the response (i.e. because reflections will introduce a comb like response) which is not possible to undo.
The 70 meters was to get 35 meters from any wall so 200 ms travel time. Would 20 ms be enough?
 

dlc86

Member
Messages
306
Just tried the IRs Jay sent me today.
Even though it's not specified which one is the far-field IR, it's pretty easy to tell once they're auditioned, it's one of the most natural sounding IRs I've tried to date (and I've heard literally thousands of them), it has no phasiness at all and I don't need to push the mids on the EQ as I usually do to balance close-mic IRs.
The close-mic IR sounds good too but has a bit of phasiness and the highs are not as smooth as on the far-field one.
The "fake" far-field instead is pretty much unusable, I guess who captured it had absolutely no clue about what he was doing.

Thank you very much for having shared them, looking forward to try the far-field at higher volume tomorrow with my atomic CLRs, I guess it will sound even better thru those speakers.

Now I just hope more reflection-free far-field IRs reach the market, I've tried a few of them over the last few years but most (if not all) of them clearly had some comb filtering caused by reflections.

I have one question for you Jay if it's not too long/hard to explain:
I had the impression that if I apply the minimum phase transform to your far-field IR the sound changes noticeably and becomes a bit "phasey" while with most close-mic IRs the difference is almost impossible to hear (unless they contain huge room reflections).
Why is that happening? Maybe something to do with the ground-plane measurement or is it just that particular speaker not being close enough to a minimum phase system?
 
Last edited:

yeky83

Member
Messages
2,634
I have one question for you Jay if it's not too long/hard to explain:
I had the impression that if I apply the minimum phase transform to your far-field IR the sound changes noticeably and becomes a bit "phasey" while with most close-mic IRs the difference is almost impossible to hear (unless they contain huge room reflections).
Why is that happening? Maybe something to do with the ground-plane measurement or is it just that particular speaker not being close enough to a minimum phase system?
That's interesting. Perhaps it's in part due to the greater presence of the "construction of the cab" part of the far-field sound that Jay talks about in post #102? No idea.
 

Jay Mitchell

Member
Messages
5,648
I had the impression that if I apply the minimum phase transform to your far-field IR the sound changes noticeably and becomes a bit "phasey" while with most close-mic IRs the difference is almost impossible to hear (unless they contain huge room reflections).
Why is that happening? Maybe something to do with the ground-plane measurement or is it just that particular speaker not being close enough to a minimum phase system?
I've said this many times: no speaker in an enclosure is even approximately a minimum phase system. There is never a reason to apply a minimum phase transform to an IR of a guitar speaker.
 

Mark Al

Member
Messages
851
Just tried the IRs Jay sent me today.
Even though it's not specified which one is the far-field IR, it's pretty easy to tell once they're auditioned, it's one of the most natural sounding IRs I've tried to date (and I've heard literally thousands of them), it has no phasiness at all and I don't need to push the mids on the EQ as I usually do to balance close-mic IRs.
The close-mic IR sounds good too but has a bit of phasiness and the highs are not as smooth as on the far-field one.
The "fake" far-field instead is pretty much unusable, I guess who captured it had absolutely no clue about what he was doing.

Thank you very much for having shared them, looking forward to try the far-field at higher volume tomorrow with my atomic CLRs, I guess it will sound even better thru those speakers.

Now I just hope more reflection-free far-field IRs reach the market, I've tried a few of them over the last few years but most (if not all) of them clearly had some comb filtering caused by reflections.
I also tested out three IRs from Jay today, which I asked him via PM and Jay kindly shared with me. I was instantly wow-ed by one of the IRs, it's easily one of the most natural sounding, balanced IR I've heard in a long time (I am no stranger to IRs, owning tons of IR from OH, RedWires, Celestion). It is almost entirely void of the typical boomy lows and hyped highs from near field IRs. It's balanced and rounded, it worked really well for Fender tones, and there is minimum amount of reflection I can hear from that IR as well. It's just great!

I have no problem identifying one of the other IR as the near field IR, as I am quite familiar with the typical NF sound signature. And the last one is a far field IR with TONS of reflections, rending it almost useless.

Immediately, I asked Jay, where do I find more IRs like this!! I know it's very limited experience with one particular far field IR, but I am definitely hooked, this far field IR sounds so much more natural, balanced and closer to the AITM feel. Man, we need more of those, please :)
 

Jay Mitchell

Member
Messages
5,648
What I'm seeing so far is a direct correlation between the graphic data I have presented and the audible characteristics of the IRs as identified by the auditioners. I've worked with this kind of data for decades, so the correlations are well-known to me. It would appear that they are also sufficiently obvious that the "average person" can detect them and connect them to the data I published, ergo there is a purpose in presenting "theory" (actually hard data based on proven principles) in discussions about sound.
 

Chocol8

Member
Messages
318
Is one of these 3 IR’s the Atomic Fender one posted earlier? I downloaded that one to test, but if there is something better I will send a PM.
 

hippietim

Silver Supporting Member
Messages
6,472
Jay sent me the three IRs this evening. I have limited time to test. I loaded them on my AxeFX III and used the Deluxe Reverb factory. I turned off the reverb and loaded all three IRs in the cab block with default settings. I tested with my PRS PS Brent Mason which has a great mix of humbucker and single coil sounds. I only had time to listen with my Genelec monitors.

There is no question which IR was not Jay's. That thing sounded terrible. There was a truly bizarre resonance thing happening like a cheap acoustic guitar simulator with a very brittle top end. This IR was harsh with any amount of gain from the amp. Using a drive pedal was completely unusable IMO.

Ok, now the good. Jay's IRs are freakin' great. Really. Either one was usable with no adjustments to the preset. The edge of breakup sounded just like a DR should sound - a little angry but still keeping it together. The clean tones were really nice when I switched to singles or rolled back the volume. The high end was thick like a Fender should be with the volume opened up to the point of breakup when you dig in. The low end held together really nice. The two combined sounded very good together - if I had more time I would have experimented with the mix. I tried the drive pedal in the preset (sorry I don't recall which is used). I got exactly what I would expect from a DR - good top end crunch, the mids more pronounced but smooth, and the low end stayed tight.

Bottom line for me is that Jay's IRs capture the essence of a DR cab and have a natural resonance. They really want to make me just play the guitar - unfortunately, I need to go do grown up stuff for a bit before I crash.
 

Dave Merrill

Member
Messages
2,451
So this sounds promising. People are swapping these into their regular modeling rigs and noticing a real difference. That's really great.

Jay, you might have a side business there if you weren't opposed to it. Seems like musicians would appreciate it too.
 

Jay Mitchell

Member
Messages
5,648
Is one of these 3 IR’s the Atomic Fender one posted earlier?
No. I have not shared that one in wav format.

I downloaded that one to test, but if there is something better I will send a PM.
It's not a matter of "better." Of the three IRs I'm making available, two are of the same cab, taken with the same mic. One is close-mic'ed, the other was taken with the mic at 2 meters. It's an interesting comparison with no uncontrolled variables that is not available elsewhere.
 

Guitardave

Supporting Member
Messages
9,719
No. I have not shared that one in wav format.

It's not a matter of "better." Of the three IRs I'm making available, two are of the same cab, taken with the same mic. One is close-mic'ed, the other was taken with the mic at 2 meters. It's an interesting comparison with no uncontrolled variables that is not available elsewhere.
I'm curious to try them. Not sure if you already posted it but what speaker is it?
 

BINGEWOOD

Member
Messages
111
Thanks for these IRs Jay. I think they are great examples to help get the concepts and graphs in the thread across, and two are a lot of fun to play through/sound great!

I prefer the one I think is the far field so thanks again for this information/lesson and the IRs.
 
Messages
1,275
Just tried the IRs Jay sent me today.
Even though it's not specified which one is the far-field IR, it's pretty easy to tell once they're auditioned, it's one of the most natural sounding IRs I've tried to date (and I've heard literally thousands of them), it has no phasiness at all and I don't need to push the mids on the EQ as I usually do to balance close-mic IRs.
The close-mic IR sounds good too but has a bit of phasiness and the highs are not as smooth as on the far-field one.
The "fake" far-field instead is pretty much unusable, I guess who captured it had absolutely no clue about what he was doing.

Thank you very much for having shared them, looking forward to try the far-field at higher volume tomorrow with my atomic CLRs, I guess it will sound even better thru those speakers.
This echoes my sentiments perfectly. Just a bit better articulated. I also got the three IRs from Jay and this was my response to him in an email I sent right after testing them out briefly:

*********************************

Alright,

I couldn't help myself. Late last night after finishing the last episode of Game of Thrones I had to give the IRs a try. Being at home I was limited to using my headphones (which is far from ideal - I know) and give them a go. The headphones are Beyerdynamic DT150s so they are decent without the typical exaggerated bass and treble that many "hi-fi" headphones suffer from (IMO).

Anyway, I suck at these kinds of things so I will probably guess all the IRs wrong. However, I do have definite preferences.

B) Is very loud and sounds messy and brash to me. I did not like that one at all. I think that is the third party IR.
A) Sounds pretty good but I think it is a bit bassy and it's got a bit of hollowness to it. I think this is the close miked one of yours and that the hollowness is caused by the reflections in there.
C) Sounds really good. It sounds like an amp sounds. I think that is the far-field one. With IRs like that there is no need to have a million taken of each speaker with different microphones at different angles, distances, etc. It just works. Now, this may all be self suggestion but I really look forward to trying this IR through my Headrush.

Thanks,
Thomas
 

dlc86

Member
Messages
306
I've said this many times: no speaker in an enclosure is even approximately a minimum phase system. There is never a reason to apply a minimum phase transform to an IR of a guitar speaker.
Ok so maybe the right question is: why most raw close-mic IRs seem to be very close to their minimum phase version?
Is that something caused by the close-micing itself?

I won't say anything about its origin except that it ain't "fake."
I probably chose the wrong word. With "fake" I meant it is an IR that is not what it's advertised to be: a far-field IR (which by definition should be reflection-free)
 

Chocol8

Member
Messages
318
a far-field IR (which by definition should be reflection-free)
I don’t think that is the definition of far field. I would have taken far field to mean the mic was placed far enough back to capture the entire cab/speaker rather than be pointing at a specific spot on a specific driver. More than a few feet as opposed to a few inches or less for close mic’ed.

It would only be reflection free if it is recorded in a large/open space with the mic on the ground. I used to measure home theater subwoofers this way. We would put the sub in the middle of an empty parking lot and lay a calibrated measurement mic on the ground 2 meters away. The sound would still reflect off the ground and back up, but since the mic was laying on the ground it would not pick up the reflections. Measured this way the subs had much less low frequency response but also much smoother curves than you would see in room. Most importantly for us though, the measurements were consistent for apples to apples comparisons and not dependent on the room, placement in the room and mic placement.

I never even thought about doing this for guitar IR captures but it makes a lot of sense. The extra challenge is that we didn’t care about environmental noise above 120 hz but for guitar cabs you are going to want an extra quiet parking lot or a very large room.
 




Trending Topics

Top