Ooooh... Charts and Graphs

Status
Not open for further replies.

FractalAudio

Administrator
Fractal Audio Systems
Moderator
Since the days of the original Axe-Fx I've been obsessed with the aliasing performance of modelers as I think this is one of the things that separates the good from the not-so-good. I believe that a target goal for a "high-gain" amp should be a minimum of 60 dB of what I call "Signal-to-Peak-Aliasing". Analog amps don't alias so any aliasing above the threshold of audibility (60 dB) reduces clarity and definition. Aliasing is particularly nasty because it's harmonically unrelated to, and can occur at frequencies well away from, the desired frequencies (and, hence, is not masked by our perceptual process).

The high performance of the Axe-Fx III CPU allows aliasing performance to easily exceed that. Below is a graph of the aliasing performance of the III vs. a popular competitor's product. III in blue (obviously).

aliasing_III_vs_hx.JPG
 
Since the days of the original Axe-Fx I've been obsessed with the aliasing performance of modelers as I think this is one of the things that separates the good from the not-so-good. I believe that a target goal for a "high-gain" amp should be a minimum of 60 dB of what I call "Signal-to-Peak-Aliasing". Analog amps don't alias so any aliasing above the threshold of audibility (60 dB) reduces clarity and definition. Aliasing is particularly nasty because it's harmonically unrelated to, and can occur at frequencies well away from, the desired frequencies (and, hence, is not masked by our perceptual process).

The high performance of the Axe-Fx III CPU allows aliasing performance to easily exceed that. Below is a graph of the aliasing performance of the III vs. a popular competitor's product. III in blue (obviously).

aliasing_III_vs_hx.JPG

Nice.
Any comparison between the XL+?
 
for sure he'll not release any comparison with his own devices. this could only be negative.

Well, negative why? more like showing the improvements over the previous generation. IF there is any noticeable aliasing on the now obsolete units, i cant hear them to be honest.
 
well you have the chart with another competitor on top. now adding in his own stuff may show how close the competition is.
 
I particularly like how the Axe trends pretty flat through the whole spectrum where as the competition only trends up, and especially over the most guitary frequencies... Ima nickname the green line "squirrelly" ;)
 
Dumb question: what causes this huge peak at 10k ?

That's the excitation. The excitation is a swept sine wave from 9 - 11 kHz.

For example if you put 10 kHz in you should get 20, 30, 40, etc. kHz out but nothing else. Anything between 10 and 20 kHz and anything below 10 kHz is an alias. The problem with using a single tone is that aliases can completely disappear at one frequency and then spike up just a few Hz away. So by sweeping the input over a range you get the true idea of the aliasing performance.
 
Probably a dumb question; my knowledge of this is quite limited, but if you used a distortion pedal in front of the axe fx, there by increasing the harmonic input, ie upper harmonics, would this increase aliasing? I'm assuming a 48k sampling rate and therefor a nyquist frequency of 24k. I would imaging that any harmonics that hi would be very low in amplitude but then I can't even imaging that a straight guitar frequency would possibly be that hi... which makes me wonder why there would be any aliasing at all....aliasing happens when frequencies above the nyquist frequency are sampled, right?
 
Now I’m curious which competitor was used in the comparison. But this also just pumped me up even more for the release of the III. My credit card is ready and waiting for that email.
 
Its an interesting Graph, at 2 points "the other things" performs better than the Axe Fx II and at 20K the II performs better.
 
Probably a dumb question; my knowledge of this is quite limited, but if you used a distortion pedal in front of the axe fx, there by increasing the harmonic input, ie upper harmonics, would this increase aliasing? I'm assuming a 48k sampling rate and therefor a nyquist frequency of 24k. I would imaging that any harmonics that hi would be very low in amplitude but then I can't even imaging that a straight guitar frequency would possibly be that hi... which makes me wonder why there would be any aliasing at all....aliasing happens when frequencies above the nyquist frequency are sampled, right?

A distortion pedal would create harmonics above Nyquist but those are filtered out by the A/D converter.

A guitar doesn't create any frequencies above Nyquist but an amp does. A lot. Because it distorts.
 
Status
Not open for further replies.
Back
Top Bottom