TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
Switch = on is discrete. Switch = off is discrete.
Answer this: On a vibrating bass string (kinetic energy engaged), what switch-like states are on or off? (I am not referring to nodes of constructive/destructive wave interference either; tone persists)
Then how can an amp with discrete states of on and off binary switches faithfully reproduce the continuous waveform of a bass string? Especially compared to class A amps?
So if I have a benchtop power supply that is fixed voltage and variable current (say 5V at 0-1 amp) - it's a digital?
It's either on (5V) or off (0V)...
On the speaker terminals of a class-D amp, what switch-like states are on or off?
The same way a class A amp does.
What you're missing is that Class-D (generally) doesn't involve a conversion of an analog signal to digital. It involves the conversion of one type of analog signal to another type of analog signal. A different encoding scheme effectively.
Kind of like AM and FM are two different ways of modulating a signal for transmission - but both are analog.
randysmojo said:That's been my point to. Is the actual audio from the instrument being converted from analog to digital at any point in the power supply? If not, no argument.
I appreciate the spirited debate on this thread!
I think class D being a relatively new tech requires some public observation and comparison.
Perhaps the class D doesn't quantize the original analog input signal, but perhaps it does somehow quantize dynamic changes in volume perception and/or wave amplitude? I am just asking!
In other words how does class D "decide" what switches are on or off and how does it interpret/manage amplitude thresholds for these on and off states?
The biggest question to me is: does class D or SMPS interpret analog waveform (with respect to the original analog input waveform) more or less accurately than older technology or amp classes?
Like with all new tech and their advent (CD's, cell phones, and more), the improvement of this new tech usually goes through multiple improvements and upgrades. Perhaps our dialogue here might facilitate some new ideas and improvements?
The way I understand class D technology (and maybe the reason why you might hear a difference between class D and class A, for example) is, that there is a certain "dead zone" in a transistor, where the switch is neither on or off, but somewhere in-between. This effect seems to be more apparent in class D amps than in class A or B.
That "dead zone" causes a small unwanted voltage drop which in turn results in harmonic distortion. The voltage drop can be countered or buffered by a capacitor to an extent, so that the signal is effectively interpolated between subsequent transistor switches.
I might be totally off on electronics theory here, but that's what I understand from various literature. So yes, you could argue that if all of the above is correct, a class D amp will be potentially slightly less accurate in reproduction of the source signal because of the introduction of low harmonics distortion. I do still fail to see where the digital vs. analog argument comes into play then, though![]()
Thanks Alban!
This seems like a highly possible and reasonable observation to me!
I am not sure there is an absolute explanation for what I am hearing, but I enjoying asking questions to find out more!
My digital/analog comparison is more philosophical regarding the minimal qualifications for something being digital. When I started workin on my masters in computer science, my professor always had us pay attention to where we thought something "becomes" digital. The definition can be difficult to interpret (below just being one).
Digital - Wikipedia, the free encyclopedia
If computers are digital, is an abacus (primitive computer) digital? Is a microprocessor a glorified abacus on steroids?![]()
Even the performance frequency of a "digital" microprocessor can be plotted as a sine wave.....which of course does not preclude it's digital nature, but does show the interplay of analog and digital interpretation.
Micro Bass Heads, why doesn't everyone use them?
waiting for the micro cabs.
Micro Bass Heads, why doesn't everyone use them?
waiting for the micro cabs.
Yeah, a one or a zero!You can pick up a micro head with one digit
I highly doubt you could identify class-D vs. other amplifier topologies in a blind test. I suspect what you heard is something particular to the voicings of the class-D amps you used.
For me - it's quantization. If there is no data quantization (outside that imposed by physics) it's analog, if there is data quantization it's digital.
if you've ever seen LVP (laser voltage probe) of an on-die clock signal, they look frighteningly sine wave-ish in the highest speed CPUs.