Ooooh... Charts and Graphs

Status
Not open for further replies.
FractalAudio said:
I'm assuming the green trace is the DUT in which case it's only 30 dB. At 5 kHz there's a small spike that touches -30 dB.

Oh, and nice title on your project...
Click to expand...
Green is Helix SLO at factory settings and Blue is the the same model in Helix with the gain lowered to produce the best possible Peak signal to aliasing result. We both know that this test Is garbage when you change the gain settings the aliasing changes.
 
GeekeryGeek said:
Green is Helix SLO at factory settings and Blue is the the same model in Helix with the gain lowered to produce the best possible Peak signal to aliasing result. We both know that this test Is garbage when you change the gain settings the aliasing changes.
Click to expand...

He specifically mentions high gain and clarity in the op.
 
GeekeryGeek said:
Green is Helix SLO at factory settings and Blue is the the same model in Helix with the gain lowered to produce the best possible Peak signal to aliasing result. We both know that this test Is garbage when you change the gain settings the aliasing changes.
Click to expand...
Now get your Helix blue trace's second harmonic at 20 kHz to be equivalent or higher than green, get it up to -25 dB. That's what the Axe-Fx III is doing.

Level of harmonic corresponds with gain level, which corresponds with aliasing performance. Cliff's test matched the gain level, yours shows itself to obviously not do so. If you're trying to make a point, it's not working too well, and it just goes to show your lack of knowledge in the subject matter...

You've studied this topic for maybe a week, a little more humility is in order before you keep making arguments with your false data.
 
Last edited:
GeekeryGeek said:
We both know that this test Is garbage when you change the gain settings the aliasing changes.
Click to expand...

I don't want to fan the flames here too much, but I think this argument is quite poor. I've spend my last 5 years in scientific research, and while I have no claim to expertise on aliasing measurements, I do lay claim to being quite familiar with scientific process and hypothesis testing. I think there are two things to consider here.

Firstly, before doing any measurements or experiments, you need to be absolutely clear what your hypothesis is. You also need to ensure that you set up you experiments so that you can accurately test your hypothesis, and control for any other variables which may influence your interpretation. From my understanding, the hypothesis behind Cliff's original testing may have been "The Axe FX III has better aliasing performance than Product X". This is a reasonable hypothesis, but in designing an experiment to test this you need to account for other variables that might confound your results. What might these be? Amplifier choice. Gain settings. Input frequencies. The primary concern when designing your experiment is to ensure that these conditions are the same between both products/samples. In the case of Cliff's original data, I believe he did this well.

With regards to your observation that aliasing changes when you change the gain settings, I don't think this is a terrible surprise. Cliff accounted for this in his testing by ensuring that all the relevant settings were the same between the two products. Now, if your hypothesis was "aliasing is always audible (<60dB) at all settings for Product X" then perhaps the test you performed could help support/reject this hypothesis. As it stands, doing tests on the Helix alone does not add any evidence one way or the other towards Cliff's original hypothesis, except to show that there is aliasing present in the Helix.

Secondly, in any scientific field, there is always a level of trust with reported data, based on the credentials of the person responsible for the experiments. Even during peer review, I have to trust to some degree that the authors have performed the experiment according to their description, that they haven't falsified or misrepresented data and have some level of expertise in their field. In academia, previous publications go a long way towards building this trust. In the case of Cliff's testing, we need to trust that he knows what he's doing with regards to testing aliasing. My trust in this is heavily supported by the fact that he designed the Axe FX. Those are his credentials, and anyone trying to argue against his conclusions has to bring enough evidence to outweigh his credentials. I don't think we're anywhere near that point yet.
 
GeekeryGeek said:
Green is Helix SLO at factory settings and Blue is the the same model in Helix with the gain lowered to produce the best possible Peak signal to aliasing result. We both know that this test Is garbage when you change the gain settings the aliasing changes.
Click to expand...

What exactly is your point? That the OP test results are manipulated? Why would he do that when as soon as the Axe 3 starts shipping people can test it for themselves and call fraud if it's not repeatable?
 
stereotactic said:
What exactly is your point? That the OP test results are manipulated? Why would he do that when as soon as the Axe 3 starts shipping people can test it for themselves and call fraud if it's not repeatable?
Click to expand...
It could be a misunderstanding - there are two graphs to consider - I can see how someone might interpret what's been said the wrong way. I still have no idea what the confused party is on about, it's not been clearly stated well enough for me to follow along.
 
BillyZeppa said:
It could be a misunderstanding - there are two graphs to consider - I can see how someone might interpret what's been said the wrong way. I still have no idea what the confused party is on about, it's not been clearly stated well enough for me to follow along.
Click to expand...
Half a dozen posts ago, it could have been a misunderstanding. At this point, he's made his agenda clear. I think all his baiting is just an attempt to drive traffic to his blog.
 
Status
Not open for further replies.
Back
Top Bottom