Axiom Home Page Forums General Discussion What's New At Axiom A1400-8 Digital Amplifier Pics

John, just a little more info relating to what I stated can be found in the "What About Wires Longer Than 50 feet?" section of Mr. Russell's excellent wire explanation . It refers to an article from Audio which apparently isn't available in full online which indicates that a substantial difference is necessary at the highest frequencies before it's audible. Incidentally, in my view the comments from Stereophile would appear to fall into the voodoo audio category.

To quote the Stereophile reviewer:

"...the output above 15kHz dropped significantly, to –3dB at 20kHz. This kind of response is not that audible in itself. However, I am becoming convinced that the better time-domain performance offered by the digital filter that produces this response does sound better, particularly regarding accuracy of stereo imaging."

So he's saying that the (digital) filter that is responsible for this poorer magnitude response offers better phase (time-domain) response.

I would definitely agree with him on that point. If you want to improve phase response (time delay) you can give a little on magnitude response.

ihifi,

This roll off at the top is pretty much standard with high powered amps no matter the class of the design. In a digital amp it is to keep the clock out of the output and in an A/B amp it is to stop ringing. Below I have posted a graph we took of a competitor's high-end class A/B amp and you can see it exhibits exactly the same roll off. The limit of human hearing is around 14 – 16 kHz so you are out of the audible range before you get beyond 0.2 dB of drop. Even if you could hear to 20 kHz then 0.6 dB of a drop would also be inaudible. Hope this answers this one for you.

Ian,

Thank you very much for your thoughtful reply. Just to be sure, my querry was about the response at the high frequency not looking ruler-flat as compared to the rest of the curve, not as compared to other amplifiers' response curves. I am not an audiophile, so I greatly appreciate your explanation regarding the purpose of the slight roll-off. While curves and figures (will see how Mark Johnson will play on the "curves and figures") hint at the performance potential of this amp, I think listening to it will likely provide the ultimate test. I look forward to hearing this amp powering eight of his sisters and brothers in my home in the next few weeks.

On a side note, If the Axiom pre/pro becomes a reality, I may end up with an all-Axiom set-up, save for the DVD player and the pop-corn machine. By the way, do you have any updates that you can share on the planned pre/pro? Thanks again.

John

Quote:

---

Just to be sure, my querry was about the response at the high frequency not looking ruler-flat as compared to the rest of the curve,

---

You have to take into context the scale of the graph.
A drop of 0.2-0.6 dB is essentially inaudible to the human ear. If the Y axis scale were set to values of 2dB increments or more, the curve would not even be visible.

They say you can do anything with statistics but i say the advertising marketers do far more magical wonders with graphs.