AMP Resolution

G

Gizmo

Inspired
From the manual

AMP block Resolution
Presets which use a single amp block run in high-resolution mode, utilizing an entire processor. This provides greater fidelity and resistance to aliasing. This mode is automatic and is selected whenever there is only one amp block in the layout grid. Adding a second amp block will revert both to normal resolution. Note that switching between presets with differing number of amp blocks may introduce an additional delay as a “soft reset” of the amp blocks must be done whenever changing the resolution.

Getting ready to redo my patches this weekend to try out scenes. I like the idea of using 2 AMP blocks with X/Y switching, but was wondering about the difference in Hi and Lo Res for the amp. I remember trying this about 6 months ago and I thought I could tell a difference, but since I didn't do a "blind" test, I might have been biased. Any thoughts on this?
 
I seem to recall even Cliff commenting that the difference is essentially inaudible. I'm like you, did a quick test and thought I could tell the difference, but I think it could have been the placebo effect. I really don't think anyone could tell the difference in a live setting, band mix and all of that.

Scenes really make this thing come to life as far as control goes. I think you'll have fun this weekend. (A lot of work, but fun work.)
 
I was about to post the same thing. Whats alliasing. Would it occure more on high gain patches.
I did patches with 2 amps and i was also not sure if i hear a difference. And where would be the difference
More on high fq or low or both?
 
What is aliasing? | EarLevel Engineering

Basically it means that above a certain point in frequency, there aren't enough sample points to accurately represent the waveform. This will lead to a misrepresentation of the frequencies above that particular point.
I'm in no way a specialist on these things, but the way I understand it, this would probably mostly relate to the upper harmonic content of the sounds we have in the Axe :)
 
Last edited:
I believe all it is doing is a higher rate of oversampling.

Any aliasing in normal res was inaudible as far as I could tell. Cliff tests before hi-res was around seemed to support that. Hi res, just improves on that. I would say if any one hear were to do a proper blind A/B test of hi res and normal res, no one would be able to reliably pick between the two (at least without measurement tools).
 
  • Like
Reactions: yek
The difference is easy to hear for me.

Fr example; the aliasing of the kemper makes it totally unplayable to me, I'm sensitive to these things.

I never use two amps in one patch because for me it's easy to hear the difference. It's the same as the difference between 44.1 and 96k sample rates in a daw for me, in this case obviously 48k.
 
^ What he said.

I'm always skeptical of claims that higher sampling rates sound *vastly* better than standard 44.1 and 48 sampling. The hi-res in the Axe makes certain that artifacts created in signal processing won't show up as aliasing in the post-signal, but when you're resampling, the computer should have plenty of time to apply a decent algorithm (including a filter with bandpass range beyond 20kHz and minimal ringing.)
 
Skeptical about easily hearing the difference also.

In my experience there are other influences contributing to the difference. The audio interface being used, the presets being compared etc.

Not saying there is no difference, but it is not obvious, to my ears.

Richard
 
Aliasing is well within the hearing range. It's not in supersonic ranges.
For instance, I have Ampkit for my iphone. Cute practice tool completely useless for real gigging. Not because the models suck (which they do - a little) or the effects are terrible (which they are somewhat) but because of the aliasing noises which are very noticeable.
Now I can't really hear a difference between hi and low res in the amps, but L6, Digitech and other modelers are quite obvious with aliasing noises. Just gotta know what you're listening for.
Take a high note and bend it up. Listen for a harmonic of the note coming down. A bit like heterodyne. It's there, buried in the sound but audible and it grates my teeth. But not in the "low res" amps. Can't hear it.
 
Cliff should've never mentioned about the difference in resolution. Now that the cat is out of the bag, there's always going to be someone who claims he can hear a difference.
 
  • Like
Reactions: yek
I think I remember cliff saying something like "This new firmware sounds great!" when announcing the new resolution saying something like he had doubled the resolution in all the essential filters.
 
this was the thread

http://forum.fractalaudio.com/axe-fx-ii-discussion/44345-new-firmware-sounds-great.html

this was the post regarding the changes:

None of the models have changed.

The new firmware is just some bug fixes plus automatic "high resolution" mode for the amp block. If you only have one amp block in the patch then the amp block runs at twice sampling frequency. Also, the internal bit depth has been increased for many of the filters. All the critical filters now use an extra 8 bits of precision in the mantissa. There is also a minor change to the speaker resonance modeling that increases the dynamic range.

It just happens to sound really good to me.
Click to expand...

http://forum.fractalaudio.com/axe-fx-ii-discussion/44345-new-firmware-sounds-great-6.html#post583511
 
Gizmo said:
From the manual

Getting ready to redo my patches this weekend to try out scenes. I like the idea of using 2 AMP blocks with X/Y switching, but was wondering about the difference in Hi and Lo Res for the amp. I remember trying this about 6 months ago and I thought I could tell a difference, but since I didn't do a "blind" test, I might have been biased. Any thoughts on this?
Click to expand...

I had no idea the amp res changed if you used 2 amps, but it makes perfect sense, as the amps are handled by one processor. I changed my effects chain in mid-december after v9.02 came out, to 2 amps and 2 cabs in parallel, with a modifier that mutes the amp if bypassed. With the advent of scenes, I really don't need the modifier, but I am still triggering it that way with MIDI, and a scene change command, and then saving my Axe Preset, after I've cycled through all the scenes for that preset, and had my controller trigger all the on/off's and x/y's. So, essentially, I could simply trigger scenes, and everything would come on and off the way I need it to.

The thing I did give up when I went to 2 amps and 2 cabs, was that I had to run my cabs in LowRes because I kept exceeding the CPU usage otherwise. Is there a difference? Perhaps, but the Fractal sounds so freakin good, who could tell? In a live room, with a full band, and other guitars, at over 100 db, no one is. It sounds amazing, and 2 amps gives you seamless switching. I usually set up 4 presets per song: verse, chorus, bridge, solo, if you will. I set up my amp switching like this: verse: amp1x/cab1x, chorus amp2x/cab2x, bridge amp1y/cab1y, solo amp2y/cab2y. Yes, sometimes, it means I'm only switching between amp1x and amp1y which gives a slight loss of sound, but most of the time I'm switching between amp 1 and amp 2, and in those situations, there is no loss of sound - no hiccup. If I set them up identically (which would be a waste of time) there would be no way of telling I'm switching between the two...it's THAT good. I highly recommend it.

Of course, there are songs, where the same amp and cab can remain on, and the scene master volume can be used to provide boost for solos or changes to the volume for louder or softer song passages, and in those cases, I could change to a single amp/single cab signal chain, for maximum quality. If I was recording, and did not need seamless switching between song passages, I would use a single amp/single cab pathway for highest quality.

Look up the thread Adam Cook posted about 2 amps with absolutely no drop outs....for a detailed explanation.
 
javajunkie said:
I believe all it is doing is a higher rate of oversampling.

Any aliasing in normal res was inaudible as far as I could tell. Cliff tests before hi-res was around seemed to support that. Hi res, just improves on that. I would say if any one hear were to do a proper blind A/B test of hi res and normal res, no one would be able to reliably pick between the two (at least without measurement tools).
Click to expand...

If it isn't because of better audio quality why did he do it?

It must be because he can only model certain things on higher precision... I guess
 
I can't and never have been able to hear or feel the difference....

and I so like the sound of two similar amps panned hard that I use two amp blocks live...
sounds amazing...

studio though I reamp with presets config'd with just a single amp and hires cab
although I can't actually hear the difference I guess this is just a belt'n'braces thing I do..
just cos I can't hear it, don't mean it's not there...

so
I record with the Axe performing as best as possible
I play live with the Axe rocking as hard as possible
 
It's easy to try the low res. Just take a preset that you know and add another ampblock somewhere in the nowhere in the grid away from the soundrow and let the new block unconnected, so nothing can go wrong because of a wrong cabling or wrong settings. Now listen if you can hear a difference.

I still got to try that when I'm playing with the band, that's the ultimate test for sound qualitiy, because every weak signal is drowning there and only the strong survive.
 
Patzag said:
Aliasing is well within the hearing range. It's not in supersonic ranges.
For instance, I have Ampkit for my iphone. Cute practice tool completely useless for real gigging. Not because the models suck (which they do - a little) or the effects are terrible (which they are somewhat) but because of the aliasing noises which are very noticeable.
Now I can't really hear a difference between hi and low res in the amps, but L6, Digitech and other modelers are quite obvious with aliasing noises. Just gotta know what you're listening for.
Take a high note and bend it up. Listen for a harmonic of the note coming down. A bit like heterodyne. It's there, buried in the sound but audible and it grates my teeth. But not in the "low res" amps. Can't hear it.
Click to expand...

I took this thread to be about being able to hear defects in the quality of a preset that uses 2 Amp blocks, vs. a preset that uses only 1 Amp block.

I can't hear that. But I haven't tried extreme parameter settings either. And don't see the point.

Richard
 
Well, whether you'd be able to hear the difference or not would depend on the frequency of the signal components that exceed the Nyquist frequency (1/2 the sample rate). If the Axe runs with a sample rate of 48Khz, the Nyquist frequency would be at 24Khz. Any frequency above this point that's created in the digital modelling, would fold back around this mirror point. So if for instance there was a strong frequency component at, let's say 30Khz, this would fold back and occur as an alias at 18Khz, which is within theoretical hearing range. The higher the frequency that exceeds the 24Khz point, the lower in frequency the alias would occur.

All that being said, I still haven't heard any of this occur in the Axe, highres or not :)
 
so maybe your dog could complain about the aliasing then?
but only if it could speak English I guess...

and given that dogs never buy my CD's or come to my gigs I don't care about them anyhow...
 
You must log in or register to reply here.
Back
Top Bottom