Why is it necessary to apply high/lo pass filters to cabs?

henryrobinett said:
I do it especially for live FRFR cabs. The full range is a bit unnatural for guitar cabs. The low end on my Xitone cabs takes up too much frequency and drives the bass player and FOH a little crazy. So I cut.
Click to expand...
This is where I like the Dyna-Cabs' ability to move the mic away from the speaker which reduces the proximity effect of a mic against the cone so the lows are reduced. Then, by moving the mic across the cone away from the dust cap the highs are reduced.

In the middle of the floor of the stage my PXM-12MP and/or ELIS.8 sound different than when they're against a back and/or side wall. I kind of like how the lows are supplemented by the wall for the ELIS.8 but it can get out of hand with the PXM-12MPs so I tend to roll off some lows on the modeler to deal with that situation and let FOH have their own unadjusted feed.

I also noticed that the frequency response of guitar speakers and cabs and the associated IRs naturally roll off the highs and lows pretty aggressively. I think that's why Cliff says it's not necessary to do anything, and why Justin York tends to leave the high end alone. EQing for the mix can sit on top of the natural EQ curve but I haven't found a particular setting I want in different rooms and with different speakers, so I adjust to fit depending on my mood basically. :)

I think it's another "ears not eyes" situation, and is somewhat black magic and art colliding with the location and number of people expected to be in the room and the allowed volume.

PS - And then there's the whole acoustic-coupling of the speaker and the guitar "thang" that can make the sound get fatter as the guitar's volume is rolled up and the speaker pushes out more sound. It's a balancing act.
 
I use them because I want a studio produced album tone when I play live 🤷‍♂️ I dont need a bunch of bass or air in my guitar tone...
 
  • Like
Reactions: BTU
Depends on your playback system. With my studio monitors they play from about 40 hz on up to 20k. I like to find an impulse that has a good amount of high end (not too bright or dark) so I don’t have to cut there. Then I will usually cut low end because there may be excess rumbling.

Also I found my speaker low end changed a lot after using room correction. Before there was a huge spike around 150 hz and a huge dip around 100 so nothing made sense. With room correction the boost is flattened out a lot and the lows extend really deep. Cuts work a lot better.
 
It depends on what you want to hear that’s all. It isn’t a mandatory rule some IR’s don’t need it at all and some others sure could use some cleaning up. It will vary from every IR how much highs or lows you may want to cut and also the application for example for a lead guitar sound you don’t really want a lot of lows it really helps sit in the mix while for rhythmic guitars you want to let the low end “breathe” a bit but even then that’s all subjective. I find it’s best application specially when playing at gig volumes through a monitor wedge. With the proper IR I personally don’t apply any cuts it sounds more alive to me more old school with all that top end fizz that some people either hate or love and if anything I trim the lows a bit if playing with super low tunings. Not a big fan of the ultra modern overproduced guitar sound although I agree it sounds huge
 
It is because of the microphone. Most IRs are “close mic’d,” which means the bass and low mids will be fattened by the proximity effect. On top of that, many are also much more centered or “on-axis” to the speaker cap than you would put your ears with a real cab, which makes high cuts necessary.

In theory, a mic far enough away and at the right angle off axis would need neither low nor high cuts.

I have gotten reasonably close to that using dynacab packs (max distance goes up to 12 inches) and positioning as far away from the center of the speaker as possible. Using a R121, it still needs some adjustment for proximity effect though. Based on the product manual, I’ve been using 6dB slope at 180.
 
I mostly used my ears, but as far as modeled amp+cab sounds (or real preamp into modeled cab via 4CM), I wanted to see the theory aspect. So I looked carefully at the signal levels frequency via the SSL Realtime Frequency Analysis meter plugin in my DAW and there is quite a bit (meaning -60/70db but not -120db) of "residual" signal from the dynacabs between 20hz and 80hz and likewise above 10kHz. My high quality Genelec monitors do reproduce those. I assume it is ambient and proximity frequencies that are picked up (or even generated by) by the high quality microphones used to create the IRs and dynacab files, even if such frequencies did not come from the speaker cab itself. So indeed a high pass at 80Hz with a 12db or even18db removes that. On the high end, I found that I can come really close to my actual MarkV + 1x12 Mesa cab by using the 4x12 M90 AxeFX cab (I think I read someone that it's considered the "Cliff Boogie" cab) with a blend of off-axis SM57+R121. For high gain MkV in IIC+ red mode, cutting highs more aggressively to 6kHz and 12db slope retains the "fuzz" that the actual amp+cab have. But not removing any highs at all has too much such (hifi sounding) fuzz.
 
Last edited:
Randalljax said:
When you close mic a cabinet the microphone is going to pick up much more high end information than you would be used to standing in a room with a cab , we never put our ears directly on the speaker so by nature what our ears hear is a much narrower band of frequency than a mic would
Due to this many players or artists will try to get the cab IR as close to the listening experience they are used to
To do this they are cutting off the frequencies that the player standing 2 ft away from a cab would not hear

There are different schools of thought, Justin York for instance does no high cuts
If he has fizz he just turns down the treble or presence
But his theory is there is still energy and value in the higher freq getting through and it sounds more dull by removing them

Some bands like Aerosmith I think I read use hi cuts down to 4500 to 5000. Likely because they are more comfortable with that response
Click to expand...
the reason for aerosmith to cut the hi freq so low is: they don’t hear high frequencies anymore and they crank up the highs insanly on their amps. Says Warren Huart, who produced them.

As simple as possible: with the low cut you get out of the way of bass player, and the high cut to let that freq range for cymbals etc….no rules but almost always low and hicut is applied regardless live or studio.
 
Thank you. The topic was to compare a real guitar amp+cab on a standalone basis versus modeled and I think we have achieved that. The topic of how to make a guitar sit in the mix is way more complex and subjective as there is no comparison tone to try and match. But having a starting point that matches the real standalone tone seems useful to me prior to delving into mix considerations. This was quite helpful so thank you all.
 
I’ll echo mostly the same. I never use or make those adjustments.

For the stage/room, I use guitar cabs and speakers without a microphone. Makes me and people up-front happier. I never, ever use FRFR speakers, for the same reason I never used HiFi cabs and speakers when I was a kid, lol.

For FOH, I use the IR (now dynamic cab) that sounds best in the venue. I found that the sound guy likes to fiddle with the dynamic cab feature on the Axe FX.
 
The only thing that matters is whether you like the sound. If it's too bassy, add a high pass. If it's too fizzy, add a low pass. If you're mixing a record you're probably gonna be chopping those frequencies off anyway, and if you're playing live with a sound guy he'll probably do the same thing in the PA, so there's not much point in being precious about it.
 
Thanks all. I think the purpose of my initial question got diluted, but I still think there is value in knowing how to set the modeled amp
+cab to match what I hear in my studio from the real amp+cab. I also agree that the real goal, which becomes 100% subjective and song dependent, is to ultimately do whatever sounds good, but I wanted a matching reference base line.
To put a final end to this chapter, I compared the CabClone from my MarkV to both real amp+cab and to the accurately simulated all AxeFX preset. What I found is that CabClone played through my studio monitors (same chain as the AxeFX) requires absolutely NO EQ to sound very close to the real amp. This makes 100% sense to me, and further is what I was expecting from the modeled amp+cab with NO EQ, which is the only reason why I started this thread to begin with. We have debated all the reasons why the modeled sound doesn't match and that's fine, and maybe (most likely) the CabClone ultimately may have built-in low/high pass filters although there is no way to know and it is sort of moot.
I figured it was interesting to mention that CabClone is good to go right from the XLR output, although surprisingly, there is some sort of conventional wisdom that CabClone is no good. I tend to disagree based on the above although I didn't play it in a live and loud environment.
 
j20056 said:
the CabClone ultimately may have built-in low/high pass filters although there is no way to know
Click to expand...
Yes there is, and you have the tool: Tone Match.
Ideally, you'd want to compare it to the speaker out into an X-Load LB-2, but you can still see what it's doing by comparison of the Axe-Fx Model (no cab) to the Mesa amp + Cab Clone.
 
I tried Tone Match (CabClone versus amp only AxeFX) but unsure I did it right, as it seems unclear what input it's using as the volume adjustments thereof don't seem to impact the levels measured. So the actual levels not matching measn nothing, the sahpe matters more I suppose. In addition, ToneMatch doesn't seem to display anything below 100Hz. But there is useful info on the high frequencies as it appears that CabClone more aggressively cuts high, reflecting a high slope low pass filter. I suppose the presumption is that there is likely a high pass as well below 100Hz.
Lastly, out of curiosity, I ran ToneMatch with the AxeFX cab I tweaked and the low/high pass filters as I had set them up to match by listening, and interestingly, the two curves are much more similar, same slope of low pass cuts for the high frequencies. So this actually proves that my overall modeling is accurate. So indeed yeah, go ToneMatch!!! The AxeFXIII is a remarkable unit.

No AxeFXIII - Reference is CabClone
No Sim Cab.png

AxeFXIII Cab - Reference is CabClone

Sim Cab.png
 
You must log in or register to reply here.
Back
Top Bottom