This new firmware sounds great...

Status
Not open for further replies.
viabcroce said:
While I understand your point, you seem to forget that this is just a forum. From a scientific point of view a generalist forum, which doesn't require a graduation in some specific fields in order to be joined. Users here express free ideas, which may at times not correspond to the state of the art of human knowledge.
Some naive - to say the least - concepts have been written on this board in the years (from myself as well), each of them probably touching the sensitivity of those who were more skilled in the corresponding field.
There have probably been bigger misconceptions expressed here than the ones you're complaining about: it's just that you felt not touched by those, maybe because you're not competent to judge them. I'm referring for example to some economical or psychological opinions I've come across since I'm a member. And never I felt allowed to tell people to shut up but, were I felt I could, I tried to help.

While I can accept that a person writes "you'd be not competent enough to understand such explanation" (even tho I totally dislike the manner), I can't accept instead someone stating I'm not allowed an opinion because my specific competences in the field are low. I want to feel free to be stupid, and I'll always assume the responsibility of what I write, at the cost of being considered an ignorant.

There's nothing personal in what I'm writing, but it seems to me there's a principle to defend here.
Hope my English has made me clear.

Peace,
Click to expand...
Preface: If I could provide a hover caption that provides a reason for my (frequent at times) miscommunications, I would. I wish I could. For this I apologize.

In my post, I qualified "opine" with "prosthelytizingly." Opinions are absolutely fine, of course. Say someone were to state, "48k is better for recording digital audio because it is more samples.. sort of like how more pixels is better with a picture." This is patently false, and to spread this.. especially in a prosthelytizing manner (check some DAWquencer manuals, for example).. is a horrible disservice to those who may happen to read it. If someone is to state that they prefer the sound of recordings made with XYZ Converter at 24/96k, then there is nothing to argue there (IMO). I would not get involved in that, unless I saw a response to that opinion that was providing false information in an attempt to qualify it, or disqualify it in some manner. If I were to state something as objectively true when it is, in fact, false, I would hope to have it pointed out. Citations/material that I could then consult for further study would be a very qelcome bonus.. for me :D

Further, in my post I did not intend to state that ANYONE is above learning something. I provided a link to Nika's book for that very reason. If that is what anyone infers from what I wrote, I apologize for not wording/phrasing well enough. I attempt to avoid technical discussions that involve these types of opinions for just that reason, but this one caught me off guard I suppose. I will refrain from doing so out of respect for those here.
 
Okay, my better half was reading this over my shoulder and said I was being an ass, and to simply apologize. To anyone I offended, I am truly sorry. It was not my intention.
 
FractalAudio said:
You guys are not understanding the intricacies of sample rate as it applies to non-linear systems.
Click to expand...

AAhhh, of course... thanks for that. I thought about this today, it's clearer now..

(What other product can you buy where you actually learn something in the forums from the designer himself? Talk about value for the buck!)


My reference to the 192 vs 96 listening seminar describes a linear process dealing only with the creation of a digital 'snapshot' of audio.

Completely different than using the digital engine to create a NEW output signal based on a zillion and a half different subtle and not-so-subtle properties of the input, static and more importantly dynamic.

Fascinating.

I guess the main thing is that most of us don't need to understand these technical behind-the-scenes in any great detail.. but lucky for all of us, Cliff does!

Really looking forward to hearing this upgrade.

Fractal is the gift that just keeps on giving and giving...

Peace.
Lou
 
Just a simple explanation. Say you got a 3KHz sine wave with an 8khz sample rate. Its all fine because you can represent as high as a 4Khz signal (Nyquist limit).
Now suppose you square that sine wave. Now you got a signal that's supposed to be 6KHz, but since you are only using an 8kHz sample rate it ends up aliasing into a 2KHz. Now if you resampled this wave into a 12khz sample rate, then apply the square operation, there's no aliasing now. Note that this has nothing to do with A/D or D/A. This is all in the digital domain. Non linear processing in the digital domain can cause aliasing if not properly dealt with.
The usual approach is a multi rate system, where before doing nonlinear processing , you up sample internally to some higher rate that is adequate to represent the additional higher frequencies that will be generated during the non linear process, the its low pass filtered and down sampled to the base rate again. For the Axe-FX, the base rate is 48 KHz and I believe the amp block runs at 384KHz. With this update the amp block would run at 768KHz.

So what does this mean? It means you'll get less audible aliasing. This is especially useful on high gain.
 
Last edited:
@Albert: You're proposing two ideas, both of which are incredibly difficult for me to wrap my head around.

1) That, with a bandwidth of 192k under a sample rate of 384k, an antialiasing filter cannot or has not been enacted within the amp block which adequately shifts all signals generated above 192k below the threshhold of audibility.
2) That enough components of the signal would be generated above 192k for these to fold down into the audible range, which would take a pretty large number of unfiltered signal components above 364k to achieve.

This can't just be for antialiasing, methinks...listen to the amps as they are now. Do you hear an abundance of annoying ringing high frequency components in the signal, as they stand now?

To be clear, I am not going to ask for more details on the reasons for this transition. The only reason I persist is that what I've learned about the sampling theorem seems to be at odds with the idea that an audible improvement could be achieved this way. However, I am by no means familiar with programming nonlinearity in a DSP environment, so my ignorance of that is very much in play here.
 
Last edited:
VegaBaby said:
no matter how much this is technically being discussed, I can tell you one thing... I will not be going back to dual amps until the next gen Axe runs both amps on hi-res. THAT's how good it sounds !
Click to expand...
You could get second II and run dual :)
 
Im aware its not out yet, but even still after trying to revert back to v3.0 there are still problems with my unit.
 
Status
Not open for further replies.
Back
Top Bottom