Axe-FX XL+ Upgrade, Improved Audio Quality

[AtomicBlaze]

I just upgraded to the Axe-FX XL+ today and I immediately noticed that the audio quality seems much better than the Mark II. Is the difference because of the lower noise floor? Everything sounds so much clearer. I wasn't really expecting the XL+ to sound better. I wonder if maybe I had some global setting that is different between the 2 that could account for the difference?

 

[unix-guy]

I just upgraded to the Axe-FX XL+ today and I immediately noticed that the audio quality seems much better than the Mark II. Is the difference because of the lower noise floor? Everything sounds so much clearer. I wasn't really expecting the XL+ to sound better. I wonder if maybe I had some global setting that is different between the 2 that could account for the difference?

You should not notice any difference. Are you using the same firmware version? The same presets, cabs, etc?

I just made the same transition a few weeks back... No difference.

 

[NY Guitarist]

Yeah everything is the same. I have only tried through headphones on the XL+ But I am very familiar with the sound of my presets since I only use a couple. Does lower noise floor mean operating noise or audio noise?

I'm not positive, but I thought Cliff mentioned the headphone amp was improved on the XL.

 

[unix-guy]

I'm not positive, but I thought Cliff mentioned the headphone amp was improved on the XL.

That could be... I rarely use headphones and would not have noted that difference

 

[TG3K]

You should not notice any difference. Are you using the same firmware version? The same presets, cabs, etc?

I just made the same transition a few weeks back... No difference.

I made the move up a year or so ago, and I also noticed no difference. But like you, I very seldom use headphones with it.

 

[DLC86]

I moved from axe mk1 to ax8 a couple weeks ago and noticed a bit of a difference too, I think it could be caused by the input secret sauce that's been upgraded in the xl and ax8.
Here's a sample I recorded just to test this difference in the short time I had both side by side.

-First part is guitar into ax8 input recorded through spdif
-Second part is the dry track of the first part reamped through the axe fx (identical preset)
-Third part is guitar into axe fx input recorded through spdif

What I hear is that the first two parts sound pretty much identical while the third has something different. Not a huge difference but I could clearly hear it especially while playing.
What do you think about?

 

[Bakerman]

What I hear is that the first two parts sound pretty much identical while the third has something different. Not a huge difference but I could clearly hear it especially while playing.
What do you think about?

Something's giving the third clip more gain/volume.

 

[DLC86]

Something's giving the third clip more gain/volume.

Exactly, but I checked several times all the possible settings and couldn't find a difference between the two units, and considering that the two first clips sound almost identical the only difference is the input stage (well, and the playing).
Anyway I heard a little difference even in terms of frequency response of the guitar, it seemed that the axe fx had a deeper bass response but lacked something in the upper mids.
While playing I always preferred the ax8 though, it seemed to impart a more "3D" quality to the sound.

 

[oldstar1975]

I hear the difference between Mark II and XL. I thought it is a mind trick till this post. I"m not the only one.

 

[Bakerman]

Exactly, but I checked several times all the possible settings and couldn't find a difference between the two units, and considering that the two first clips sound almost identical the only difference is the input stage (well, and the playing).
Anyway I heard a little difference even in terms of frequency response of the guitar, it seemed that the axe fx had a deeper bass response but lacked something in the upper mids.
While playing I always preferred the ax8, it seemed to impart a more "3D" quality to the sound.

A better test would be to first adjust for the level difference at Input/gate block with a test tone, while checking level at the grid. (Watch output or a compressor block's gain meter.) Then you've removed the obvious gain difference, whatever the cause might be.

 

[DLC86]

A better test would be to first adjust for the level difference at Input/gate block with a test tone, while checking level at the grid. (Watch output or a compressor block's gain meter.) Then you've removed the obvious gain difference, whatever the cause might be.

Sure, my test is to be taken with a grain of salt and I originally made it just for my curiosity without the intention to be scientifically flawless, I posted it now only because I felt it was interesting for this topic.
Anyway, before recording the dry track, I ensured that the ax8 was outputting a unity gain signal by sending a test tone at its input with my daw and re-recording it through spdif. Actually I had to lower the output level by 3 or 4 dB to match it.
Once done that basically I was sure the two units were fed by the same signal (first two parts) and there the gain difference was removed so I assume it's being introduced before the AD conversion since it appears when the input stage of the axe fx comes into play.
Another possible cause could be the instrument level setting in the i/o menu that maybe causes a slight increase of gain at extreme settings. I had it set at 100% cuz I play with low output single coils and I always had the impression that it changed the sound a bit when pushing it past 90%.

 

[Bakerman]

Anyway, before recording the dry track, I ensured that the ax8 was outputting a unity gain signal by sending a test tone at its input with my daw and re-recording it through spdif. Actually I had to lower the output level by 3 or 4 dB to match it.

I think this is the error in your method. You're assuming the AX8 instrument input sensitivity should match the output level of whatever you're sending the test tone from.

Call the test tone clip A and the AX8 re-recording B, and the Axe-FX re-recording (with same test tone) C. It's not important that A & B match. It's important that B & C match. It doesn't sound like you found C in order to compare to B.

 

[DLC86]

I think this is the error in your method. You're assuming the AX8 instrument input sensitivity should match the output level of whatever you're sending the test tone from.

Sorry, forgot to mention that I also tried to reamp the recorded dry track through the ax8 and I got the same exact result of the first recording once I lowered by 3-4 dB the dry track for unity gain, so output level of my interface and ax8 input were matched. If they weren't I would have got a sample with more or less gain. I know it's not the most precise way to test it but there surely wasn't the same clear difference we hear above.
Furthermore the difference between the ax8 and the axe was evident even by alternately plugging the guitar in both, that's why I made this test.

Call the test tone clip A and the AX8 re-recording B, and the Axe-FX re-recording (with same test tone) C. It's not important that A & B match. It's important that B & C match. It doesn't sound like you found C in order to compare to B.

Given what I said above, basically I have:

-a dry track recorded through the ax8 (let's call it A1)
-the same dry track reamped through the ax8 (B1)
-the same dry track reamped through the axe fx (C1)

Now from my sample I think we can tell that B1=C1 (parts 1 and 2).
This means that, if the guitar signal is acquired by the same input stage and AD converter, the processing in the digital domain is the same and the resulting recordings are pretty much identical.
It changes as soon as you change the input stage (aka guitar in the axe fx input - part 3).

 

[mr_fender]

Don't underestimate the power of the placebo effect. You need a double blind test to eliminate bias.

 

[Jason Scott]

Pure placebo. The lower noise floor won't improve clarity or dimensionality.

 

[Jason Scott]

I am certain I hear a difference through headphones, no way it is placebo. I am actually having to revisit all my OH IRs because they sound less muddy and have more clarity now. My 3rd party patches from Fremen also sound a lot better.

If there is a difference then it's possible there was a setting in your previous unit which you didn't account for. They both use the same hardware and firmware to process audio. There's nothing in the XL that would provide more clarity or dimensionality. You can bet if it did, Fractal Audio would advertise it.

 

[yek]

Yes, something must have been wrong with your previous unit. No way there's that much difference between the hardware models.

 

[DLC86]

If there is a difference then it's possible there was a setting in your previous unit which you didn't account for. They both use the same hardware and firmware to process audio. There's nothing in the XL that would provide more clarity or dimensionality. You can bet if it did, Fractal Audio would advertise it.

They don't use exactly the same hardware since the XL/XL+ have a newer "secret sauce" version, that means the input stage design has changed and that, for how well and precisely it can be designed, can definitely lead to substantial differences in the audio spectrum, especially when it is interfaced with a high output impedance device like a passive pick-up.

I think who marks this as a placebo effect (even if it could be too) has no idea of how an analog electronic circuit works and how even a different resistor value can affect the impedances and the gain of that circuit.
Furthermore components' construction tolerances can come into play and I'm sure that we'd find slight (but probably non-audible) differences even if we analized two identical units with an oscilloscope and a rta.

And then I found this from Cliff himself that makes me believe the difference is expected:
https://forum.fractalaudio.com/thre...etween-axe-ii-mk1-and-mk2.97766/#post-1178592

 

[unix-guy]

They don't use exactly the same hardware since the XL/XL+ have a newer "secret sauce" version, that means the input stage design has changed and that, for how well and precisely it can be designed, can definitely lead to substantial differences in the audio spectrum, especially when it is interfaced with a high output impedance device like a passive pick-up.

I think who marks this as a placebo effect has no idea of how an analog electronic circuit works and how even a different resistor value can affect the impedances and the gain of that circuit.
Furthermore components' construction tolerances can come into play and I'm sure that we'd find slight (but probably non-audible) differences even if we analized two identical units with an oscilloscope and a rta.

And then I found this from Cliff himself that makes me believe the difference is expected:
https://forum.fractalaudio.com/thre...etween-axe-ii-mk1-and-mk2.97766/#post-1178592

And did you see this post a few replies below the one you referenced?

https://forum.fractalaudio.com/thre...etween-axe-ii-mk1-and-mk2.97766/#post-1178604

My understanding of the "secret sauce" is that it lowers the signal to noise ratio. As far as I know that should not affect the "tone"...

 

[DLC86]

And did you see this post a few replies below the one you referenced?

https://forum.fractalaudio.com/thre...etween-axe-ii-mk1-and-mk2.97766/#post-1178604

My understanding of the "secret sauce" is that it lowers the signal to noise ratio. As far as I know that should not affect the "tone"...

Yes but that's a comment of a user that judges by hear, Cliff is the one who designed both machines and I think he knows better what he's talking about

Ps: to lower the noise of the circuit you have to tweak it (aka: change it) and that can bring wanted or unwanted differences