AFIII Axe-Fx III sounds JUST LIKE MY MESA BOOGIE MARK V!

ML SOUND LAB

Cab Pack Wizard
Vendor
First it's the Axe-Fx III and second is my real Mark V. (the ground noise should give it away)



Not 100% sure if you know me but a quick recap is that I love real amps, I love the Axe-Fx, I even like a couple of other modelers. The amp that I believe most modelers struggle with is Mesa Boogie and to be even more specific the Mesa Mark series. Throughout all these years I've given people a hard time about Mesa amps, and for that reason I've also made this a quest of mine to really nail my real amp sounds with Fractal gear and obviously share my findings with you guys.

I'll be honest, I've been struggling with the Mesa low end. I just can't get those palm mutes to sound right (=not just tight but MESA TIGHT!) and I'm one of the tweakers of this forum with 6 years of Fractal experience so it's not a case of me being new to all this. So today I decided that I'm doing this properly. A huge shoutout to a plugin that I LOVE that's called GlissEQ by Voxengo. It allows you to get multiple sources in the same frequency analyzer so I could compare the real amp and the Axe-Fx III "on top of each other" while a loop was playing. It's really easy to see where the differences in the spectrum are.

First comparison: Mark V FX loop with GEQ off VS Axe-Fx III without power amp modeling

This was quite a weird one for me. Although I'm using the Mark IV setting on the Mark V the only amp sim in the Axe-Fx III that has the same EQ balance to my Mark V is the IIC++ amp sim. All the other Mesa Mark series amp sims have way too much low end even when I have the bass knob at zero. The IIC++ EQ settings actually match my exact amp settings and I get the same EQ balance in GlissEQ as well so we're off to a good start. I have my gain on the amp at around 7. To match this amount of gain I needed to set both gains on the IIC++ amp sim to 10. Also to get the low end just a little bit tighter I had to raise the input EQ low cut to 250hz. Okay, these sounds are not identical, off to the...

Second comparison: Mark V FX loop with GEQ on vs Axe-Fx III without power amp modeling and GEQ tweakage

This was actually surprisingly realistic. The only difference was that I had to set the 80hz EQ knob to -1,5dB while in real life I have it probably around +1,5dB. My GEQ on the real amp is not as drastic with the 80hz and 6600hz sliders as you'll see with most people. These sounds are now identical again so...

Third comparison: Mark V + real cab miked vs Axe-Fx III with power amp modeling and an IR of that mic up

This is where things got weird. Just so people don't blame the IR, I compared the Mark V slave out and DI box out signals with the IR to the miked signal and those sounds are identical. No difference anywhere. So... this leads me to believe that there are big differences in how the power amp modeling is by default when compared to my Mesa Mark V that's connected to a Mesa 4x12. At this point I'm completely aware that these differences could come from just about anywhere but I fixed it with just the speaker page. These settings are a bit crazy, BUT if you listened to the clip above I'm sure you'll want to try these. Very simple: Low res Q: 0.300 and LF resonance: 5.00. That's it. Now that's a ridiculously wide Q setting. The thing is, it sounds just like my Mark V.

Small tweaks here and there as is to be expected but here you go. I made this preset that's using stock IR's of mine that are from the same exact cabinet that I was using so it should sounds very realistic.

WARNING: I'm using as much gain as I can handle. It does require you to palm mute quite heavily to keep things clean. If you feel like the preset is clammy and too gainy, lower the input drive to around 8. :)

EDIT: I added Fractool conversions for Axe-Fx II and AX8. I don't know if these work or not but give them a try if you're interested. At least the Cab Block will not work so try the stock CK TRAD IR's that you have or this free IR that's "the best IR in the world": bit.ly/MLSoundLab

EDIT EDIT: The Fractool conversion doesn't infact work but Fremen was kind enough to convert the preset and you can find it later in this thread. Thank you @fremen
 

Attachments

  • Mikko's Mark V.syx
    48.2 KB · Views: 346
Last edited:
I think the III sounds better. It's tighter in the lows.

I did a similar test and had a similar result, although the differences were far more drastic. I did a shootout comparing the Axe-Fx II XL+ to the TriAxis/2:90 into a Captor (Two Torpedoes) load box. In my comparison, the II was tighter and cleaner than the TriAxis/2:90/Captor. That said - and this could be because of the 2:90 (I was using the Modern mode) - the TriAxis and it's corresponding model in the Axe-Fx II didn't sound all that similar...and yes, I used the same IR, I turned off the power amp simulation when using the TriAxis/2:90 in the FX loop, etc.

More, the whole experience turned out to be an expensive and tumultuous exploration as the TriAxis showed up damaged (USPS's fault, not the seller's). I sent it in for repair. M/B sent it back. It still didn't work, so I had to send it back to them again. After the second repair it worked properly. Then after doing the shootout and preferring the Axe, I sold the TriAxis and 2:90 at a loss.

The moral of my story - the Axe is all I need. It's an improvement on the real thing and exponentially more convenient to use.
 
Interesting. I still hear a difference. Notice how the higher end fizz decays with the low end on the real amp. On the Axe, the fizz decays much more quickly.

I wonder what this would sound like if you had Fractal's actual Mark V? I wonder if it's modeling inaccuracy, or just the difference between amps. I mean, my old Mesa didn't even sound the same from day to day.
 
We've been at least 5 years past the need for tube amps. I still love owning real amps and sometimes turning real knobs and being limited is exactly what I feel like.

That being said there's no sound the Fractal can't do. This is the hardest thing to model - high gain but tight with an EQ inbetween changing the feel. The minor detail difference here is still the low end accuracy but we're talking under half a dB.

You can barely hear a difference raw and clear like this - imagine it smeared in a mix context. It's crazy. :)
 
I added Fractool conversions for Axe-Fx II and AX8 in the first post. I don't know if these work or not but give them a try if you're interested. At least the Cab Block will not work so try the stock CK TRAD IR's that you have or this free IR that's "the best IR in the world": bit.ly/MLSoundLab
 
Have you tried using the multibandcompression block like in the recent discussion?
I believe the Axe-Fx modeling is at such a high level that we don't need tricks and gimmicks to make it sound right. In this case the biggest problem was the power amp modeling so if I was trying to solve this problem with pre-EQ (which is what the multiband is essentially doing) I would end up with a very different sound and feel since it's an EQ difference before the preamp (read: before drive) while the difference in tone was mainly after the preamp in the poweramp section.

These are the reasons why match EQ/tone matching, profiling etc. will not give you the exact feel or dynamic character of your amp. You will get the final EQ balance but the EQ balance that hits your preamp distortion and power amp distortion IMHO affects the way the amp reacts. This is a very realistic simulation in the Axe-Fx III. Still I feel like I need to take care of getting the EQ balances just like my real amp like I've done here, if I really want to play them side by side and feel like there's no difference between the Axe-Fx amp sim and my real amp. :)
 
I agree that it’s best to match the amp and amp model settings as close as possible!
The multicomp is great to tighten the palm mutes and / or low end. This can be before or after the amp model where the latter is more like in a DAW.
 
As promised on Facebook, I converted the file for Axe-Fx II :



As the Axe II doesn't have an input EQ, I put a PEQ in front of the amp block. And also a tonematch block (not really needed), as the II doesn't have individual SM57/R121/E906/MD 421 IRs like in the III. Sounds not identical but more than close enough to me ;)
 

Attachments

  • Mikko's MK V.syx
    12.6 KB · Views: 91
As promised on Facebook, I converted the file for Axe-Fx II :



As the Axe II doesn't have an input EQ, I put a PEQ in front of the amp block. And also a tonematch block (not really needed), as the II doesn't have individual SM57/R121/E906/MD 421 IRs like in the III. Sounds not identical but more than close enough to me ;)

Wow this is much appreciated! Thank you so much buddy. :)
 
sorry but the distorted overtones have a totally different character than on your real MKV, they are way harsher, raspier and squirelly. On the real deal it sounds like the overtones compared to the normal note has a slight high roll-off. the base note you here is pretty spot on with the Axe. For me this overtonebehaviour is the only thing that makes the Axe Fx really different to real amps.
 
sorry but the distorted overtones have a totally different character than on your real MKV, they are way harsher, raspier and squirelly. On the real deal it sounds like the overtones compared to the normal note has a slight high roll-off. the base note you here is pretty spot on with the Axe. For me this overtonebehaviour is the only thing that makes the Axe Fx really different to real amps.
So is this the moment when I reveal that the clips are the other way around? ;)
 
To me the second one (whichever it really is) sounds a little more compressed, like maybe it has more gain. It's like the mids compress more, which leaves the bass and fizz a little more prominent. Have you tried tweaking this difference out?
 
I like the second one better. I was like - hmmm, well I'm not sure he did nail it because the second clip sounds a bit thicker and warmer. Then you drop the hammer and reveal it's the Axe III - great work man!
 
You always need a safety net! The thing with these comparisons is that once I say which is which, you will listen to the clips in a different way. Sure they're not 100%, nor would two Mark Vs compared to each other either. I'm not trying to make that argument either. The Mesa low end seems to be impossible to model since no modeler gets it 100%. Just saying that I made these types of comparisons blindfolded in the Axe-Fx Standard days with match EQ and the majority always voted for the Axe-Fx to be the real amp.

I'm not sure my point got across here. The reason why this is very impressive is because this is not tone matching. This is not EQ matching and this is not profiling. This is changing a few settings on an amp sim to make it sound pretty much identical to the Mark V in its Mark IV mode which we don't have as an amp sim. I even started with the IIC++ sim and not the Mark IV sim because it was closer to the Mark V in it's IV mode. How close is it? Well let me just make this very scientific for now. Here's a match EQ curve that I got with a Kemper Profiler that's sitting on my table trying to profile this same sound. I refined it very precisely and everything:



That's over 1,5dB peaks here and there. Now... here's the EQ curve of my Axe-Fx III patch compared to the real Mark V:



That's a flat line in comparison. Sure there's a wide mid scoop and the low end after 150hz or so is different between both of these attempts but the main point here is that by turning a few knobs (when you know which settings to turn) you can get so much versatility with the Axe-Fx III. The Mesa Mark series has always been a problem in all modelers if you ask me, this is why I "gave up and got the real amp" but I must say that's I've never gotten results this close before. I'm sure some more fine tweaking with the input low cut vs speaker resonances could give me more accurate results even but I felt like this was pretty sick already. :)
 
Back
Top Bottom