Ooooh... Charts and Graphs Redux

Only, if you have an input that is on par with the AxeFX and can get similar latency. I run a Roland Studio Capture, and it might not be the best in terms of latency, but when I tried (bought) the recent Pete Thorn amp plugin, I just felt disconnected from the sound in a way that never happens with the axe.
I never tried a plugin that did not make me want to plug into my AxeFX.
I seem to be able to get great sounds out of all devices. A Fractal vst would be one more option. I'm not sure what was discussed previously but the Helix native package is great as are the neural options with next to no latency on my audient. There is definitely room in the quadrant for Fractal and would be most welcome.
 
Last edited by a moderator:
Once I got my hardware down to the point where I could run the buffers at 64 or even 128, the feel became viable - I don't notice loss of feel with plugins any more, but, I just have not found any plugin modelling that's as good as Axefx, and, though the plugins have improved a lot over the last couple of years, they all seem to suffer from difficult input levelling effort as there does not seem to be any plugins that have the compensated input like Axefx.
I was running at 32 samples buffer, and still the experience was far from just running through my Axe
 
Once I got my hardware down to the point where I could run the buffers at 64 or even 128, the feel became viable - I don't notice loss of feel with plugins any more, but, I just have not found any plugin modelling that's as good as Axefx, and, though the plugins have improved a lot over the last couple of years, they all seem to suffer from difficult input levelling effort as there does not seem to be any plugins that have the compensated input like Axefx.
A plugin has no way of knowing anything about actual input levels and how they end up translating to dBFS, so developers have to assume something, and they all have different assumptions, it seems.

Some, like Helix Native, display the area of optimal level though - it's an easy enough workaround, and works quite well.

Most audio interfaces, even expensive ones, however, are plain bad for guitar. They are way too noisy, don't have enough headroom, and some distort way too much (hello Audient). There are only a handful decent ones.
 
I was running at 32 samples buffer, and still the experience was far from just running through my Axe
Out of curiosity, how do you use a computer with such a small buffer, what do you do with it? Is it just for playing/recording or are you able to also do editing, mixing etc.?
 
I don't, this was just to try it with as low latency as possible. My normal setting for mixing (which is what I mostly do) is 128 or 256 sample buffer
 
I am able to get ~5 ms latency on my daw at 64 samples which is roughly like standing 5 feet from an amp. The speed of sound is roughly 1 foot per ms, so standing from an amp 10 feet away would be ~10ms. I find 5ms unnoticeable. To minimize your own distance from your studio monitors, you could track with headphones.
 
Only, if you have an input that is on par with the AxeFX and can get similar latency. I run a Roland Studio Capture, and it might not be the best in terms of latency, but when I tried (bought) the recent Pete Thorn amp plugin, I just felt disconnected from the sound in a way that never happens with the axe.
I never tried a plugin that did not make me want to plug into my AxeFX.
It's extremely useful for re-amping. You record both your DI and modeler output, then if you want to tweak your tone later or add effects, you mute the original and put a VST on the DI track.

You could run the whole track through the unit again, but then you're waiting several minutes to record the new output. The VST is easier/faster.
 
It's extremely useful for re-amping. You record both your DI and modeler output, then if you want to tweak your tone later or add effects, you mute the original and put a VST on the DI track.

You could run the whole track through the unit again, but then you're waiting several minutes to record the new output. The VST is easier/faster.

That, and also it's much more convenient to practice a lot of times just using a laptop.
 
Anyone to share aliasing measurements on the FM3?
How is it compared to AFXIII?
Do you thing the turbo will have any positive effects? (Higher frequency…?)
How is the oversampling on AxeFxIII?
Thanks :)
 
A Fractal vst would be one more option. I'm not sure what was discussed previously
This was discussed quite a lot a few years ago. I don't recall any definitive official answer from Fractal Audio, but a reason given against releasing software was piracy, as far as I remember.

I would also think that maintaining desktop OS versions of Axe-FX requires significant resources in addition to what FA have. With uncertain benefits.
 
Anyone to share aliasing measurements on the FM3?
How is it compared to AFXIII?
Do you thing the turbo will have any positive effects? (Higher frequency…?)
How is the oversampling on AxeFxIII?
Thanks :)
Well it’s the same models, so aliasing should be the same. And turbo sounds like it just allows larger presets. So the aliasing and sound is the same across the Fractal units.
 
No so sure. Axe FX II had not the same aliasing while having the same code, but the TigerSHARC was not as fast and powerful as the TI dsp. despite being a beast. Even when using 2 blocks of amps instead of one they turned over sampling from 16x to 8x if I remember
 
Only, if you have an input that is on par with the AxeFX and can get similar latency. I run a Roland Studio Capture, and it might not be the best in terms of latency, but when I tried (bought) the recent Pete Thorn amp plugin, I just felt disconnected from the sound in a way that never happens with the axe.
I never tried a plugin that did not make me want to plug into my AxeFX.
Truth!
 
Higher oversampling is the most common approach. There is a technique known as anti-derivative anti-aliasing. The problem with that is it really only works with waveshaper approaches. You use the anti-derivative of your waveshaper function and then take the derivative of the result.
I remember you said the AxefxIII had 16x oversampling. Someone asked why not going to 32x oversampling. You said it was hard for the DSP AND that would cause some other issues (filter instability?)
Now that you have the new TI DSP, is that possibly an option to turn 32x oversampling? Is that something directly correlated to DSP frequencies?
(Kemper said his DSP was 700MHz and so he could have 16x oversampling for his profile gain algorithm… even if I can clearly reproduce aliasing with the kemper and sold it because of the issue of layering guitars)
BTW I assume FM3 has a lower oversampling rate but frankly I don’t really catch the difference with my AxeFxIII mkII. So is that still and issue or you considering that solved for your products even with very high gain presets?
Thanks
 
I usually don't make any difference between 8X and 16X
I can barely spot 2x and 8x difference sometimes ;)
Cliff is a cool guy letting us know how to spot the good from the bad in digital world. But for some guys like me it can be source of unnecessary worries. I spent years having fun with AxeFx II and now the III. I'll get the Turbo of course because I can't stop my GAS, and that's enough for me :)
 
Back
Top Bottom