800 #34 is a Noisy Beast!

Can you post a couple DI tracks taken with the Axe-FX and your other interface? I'd be interested in seeing the difference in noise levels.

I can't check right now, but this should be Apogee One, with a guitar plugged in, via a 3 meter cable. Guitar volume turned all the way down, so there's no sound from pickups.

I don't have anything recorded with Axe handy, and it's at my rehearsal place, quite far away, don't use it at home.

It’s easy to do though, just plug your guitar into your Axe FX, turn guitar volume down, make sure you turn off the input gate (otherwise you’ll get better than 120 dB), and record a few seconds. Than loop the samples and A/B them with some high gain preset using some digital connection. The difference is quite noticeable. With an interface using real top notch converters, it’s drastic. I don’t own one though.
 

Attachments

  • Apogee One Noise.zip
    333.7 KB · Views: 8
Last edited:
It’s easy to do though, just plug your guitar into your Axe FX, turn guitar volume down, make sure you turn off the input gate (otherwise you’ll get better than 120 dB), and record a few seconds.

I did this and the Axe-FX II (Mark II) DI is far quieter than your clip. Your noise peaks around -101.3 dB. Axe DI (Front Instr. input, 54.5%) is -116.8 dB.

I'll mention I removed some DC variance with a 20 Hz highpass on both clips to get those values. Without doing this, the highest peaks in each clip were around -100 dB (Apogee) and -107 dB (Axe-FX). Still lower noise on the Axe-FX, but misleading about how much quieter it really is.

Here's your clip then my Axe-FX DI, sent through the same preset with a ton of gain.

https://dl.dropboxusercontent.com/s/pes81xj7toxu5i8/apogee-axe.wav
 
I did this and the Axe-FX II (Mark II) DI is far quieter than your clip. Your noise peaks around -101.3 dB. Axe DI (Front Instr. input, 54.5%) is -116.8 dB.

I'll mention I removed some DC variance with a 20 Hz highpass on both clips to get those values. Without doing this, the highest peaks in each clip were around -100 dB (Apogee) and -107 dB (Axe-FX). Still lower noise on the Axe-FX, but misleading about how much quieter it really is.

Here's your clip then my Axe-FX DI, sent through the same preset with a ton of gain.

https://dl.dropboxusercontent.com/s/pes81xj7toxu5i8/apogee-axe.wav

The XL and XL+ have even lower noise specs than the MK II as well.
 
-6 dB is roughly half the signal level. -3 dB would be only around 30% less.
Depends on what you call "half" — half the voltage or half the power.

In either case, 3 dB is a noticeable but small difference. If you turn up your volume by 3 dB, it won't be twice as loud.
 
Twice by amplitude I meant and yes, that’s wrong.

But there’s something off here. Is the file I uploaded 44 or 48 kHz? I only have my phone now, so it’s difficult to check. Resampling on the fly may mess up the measurement.

Also, what do you use to measure the level?

Peaks aren’t important, they may be high in my clip because it was recorded on a phone with whatever recording app, so there are no fades in the beginning and the end. It’s RMS level that matters.

Also, unless you have some golden special edition Axe FX, there’s no way to get -116 dB. So you either have input gate on or use a measurement tool that’s somehow deceived by the “secret sauce” thingie. It’s 110 dB by specs and that’s what I’ve always been getting with mine. 116 is weird.
 
Last edited:
Depends on what you call "half" — half the voltage or half the power.

In either case, 3 dB is a noticeable but small difference. If you turn up your volume by 3 dB, it won't be twice as loud.

This is true, 3 dB is noticeable but not drastic, I did say that. 6-8 dB is very large, though. I don’t have a card with those specs, however.
 
...unless you have some golden special edition Axe FX, there’s no way to get -116 dB...It’s 110 dB by specs...
It's not 110 dB by specs, it's >110 dB. And that's for the Mk I. The XL and XL+ have less noise than the original Mk and Mk II.


The "Special Sauce III" uses a combination of things to get a lower noise floor. One of these things is new, premium Burr-Brown op-amps in the signal path which have extremely low noise and distortion (and are very expensive). As always I don't design stuff to be cheap, I design it to be good.
 
It's not 110 dB by specs, it's >110 dB. And that's for the Mk I. The XL and XL+ have less noise than the original Mk and Mk II.

Well, my XL+ manual still quotes 110 dB. When I recorded and measured, I was getting about 110dB on both inputs. The “secret sauce”, as far as I understand, doesn’t actually lower noise, it just shapes it in such a way that it sounds less noticeable. Since you can apply filters to any signal, that’s irrelevant really.

That “>” sign - I don’t know what that means. But I doubt it’s a 6 dB hidden gift.
 
Well, my XL+ manual still quotes 110 dB.
The manual quotes >110 dB. ">" means "greater than." In other words, the specified dynamic range is more than 110 dB.


The “secret sauce”, as far as I understand, doesn’t actually lower noise, it just shapes it in such a way that it sounds less noticeable. Since you can apply filters to any signal, that’s irrelevant really.
The "secret sauce" uses preemphasis/deemphasis to reduce noise at the output of the converter. It's the same tech that's used in FM broadcasting to reduce noise.
 
"Greater than" is used by all kinds of manufacturers, and it's usually a meaningless difference. Like 110.1 dB. Nobody ever quotes 116 dB as >110.

When I measured my Axe FX noise a few years ago, both front and rear inputs yielded same results. Spectrum was different. But RMS level was the same.

BTW, here's my measurement of (I hope) the file I uploaded here.

There's something way off in your results.

apogee noise screenshot.png
 
Last edited:
But there’s something off here. Is the file I uploaded 44 or 48 kHz? I only have my phone now, so it’s difficult to check. Resampling on the fly may mess up the measurement.

Also, what do you use to measure the level?

Peaks aren’t important, they may be high in my clip because it was recorded on a phone with whatever recording app, so there are no fades in the beginning and the end. It’s RMS level that matters.

Your file was 44.1 kHz, but resampling doesn't have any noticeable effect on the measurement or noise level.

I was using Reaper's normalize function to check peak level. Did you listen to the clip I made? Forget about the peak values if that helps. The Axe-FX DI was obviously quieter.

Axe input level setting has an effect on the noise level being measured here. At lower values like 20% I'm seeing about -112 dB peak. Around 90% reduces it to -120 dB. Being able to set it as high as possible for a given guitar is an advantage over other devices, but it's still quieter than your clip near minimum.

I don't know exactly what might have happened in your earlier test. Maybe you should try it again sometime.
 
Your file was 44.1 kHz, but resampling doesn't have any noticeable effect on the measurement or noise level.

I was using Reaper's normalize function to check peak level. Did you listen to the clip I made? Forget about the peak values if that helps. The Axe-FX DI was obviously quieter.

Axe input level setting has an effect on the noise level being measured here. At lower values like 20% I'm seeing about -112 dB peak. Around 90% reduces it to -120 dB. Being able to set it as high as possible for a given guitar is an advantage over other devices, but it's still quieter than your clip near minimum.

I don't know exactly what might have happened in your earlier test. Maybe you should try it again sometime.

Well, you obviously did something wrong with the way you do it because Axe FX isn’t capable of 120 dynamic range. And if you get -120 dB peaks, what’s the RMS? I only get such values with input gate enabled, these are simply insane.

As for my sample, if it’s 44 kHz, it may be a wrong one. Like, with gain not at zero or something, not sure and cannot check now, don’t have access to either Axe or PC or that card. That’s what I had on my phone.

As for Axe FX is quieter - well, it may be if the file is wrong. But that’s certainly not what I get when using it compared to better interfaces. Audibly, front input trickery may make it appear quieter sometimes, but that’s what it is - trickery, and I don’t use front input anyway - it has lower headroom and tends to distort. Some may like it, I don’t.
 

Also, let’s just forget this Apogee card for a second and assume, for the sake of argument, that it's much worse, and everything is correct with the way you did it - after all, I'm not arguing that everybody should abandon their Axe FXs and switch to Apogee Ones with plugins or something.

Still my initial point was that converter and preamp noise matters a lot with high gain presets, and their contribution may be very noticeable, and sometimes drastic. You can hear the difference from your own example. Neither of these interfaces are high end or top notch. With really good ones, the results are way better. That's what really matters.
 
Guys, while the nerd in me understands the fun about discussing these matters, the engineer in me thinks:

- you are highjacking another person's thread

- that level of noise is nothing compared to the noise coming from the guitar

As a side note, it's nothing compared to the thermal noise of the real components of real guitar amps.
 
Well, you obviously did something wrong with the way you do it because Axe FX isn’t capable of 120 dynamic range.

I didn't say 120 dB dynamic range; I said the noise level drops to around -120 dB. With the input level that high, a signal that clips the input has peaks around -17 dB.

- that level of noise is nothing compared to the noise coming from the guitar

This is true. Once you turn up the guitar volume knob, the noise level might be around -80 dB or higher. Having an ultra-low-noise input doesn't matter so much then.
 
Guys, while the nerd in me understands the fun about discussing these matters, the engineer in me thinks:

- you are highjacking another person's thread

- that level of noise is nothing compared to the noise coming from the guitar

As a side note, it's nothing compared to the thermal noise of the real components of real guitar amps.

I used to think something like that, too. Until I discovered, in practice, that I was wrong, and it’s not nothing at all. It was weird to find out.

But you’re certainly right about hijacking.
 
Back
Top Bottom