Don't remember seeing anything like that when patching between DACs. Outputs went to the Crown power amps spec'ed by JBL for the M2s.
There is various stuff going on inside the Crown boxes, of course. M2s while there are here with me are running on different amps with analog crossover. They sound good, although the bass is a little looser than it might otherwise be given the design approach's ported cabinet and sorta-large-ish internal volume.
Not the 4x10HD I have that def uses 2068.
They have a lot of products you possibly have to define which one you think is midfi and under what operational mode.
The level of the signal. If you hit a bell for example, the tone rises very fast to full amplitude, then slowly decays to zero. This is an envelope with a steep front slope. That fast envelope rise causes high frequency components.
If it is too fast a rise, the filter will be unhappy and will stop some of it.
There is a trade off between the frequency created and how fast it's amplitude will rise.
Scott, in his plot, showed the difference between filter in and out. That plot, I called the Gibbs envelope, is a direct measure of the filter removing stuff that exceeds the filter cutoff. I propose using it to look at specific music or instruments to determine is the frequency/envelope combined to make frequency side bands that exceeded the filter. If it does, raising the filter break to find out where it stops cutting information will show you the sample rate needed to insure good fidelity.
The only caveat is if the instrument by design creates lots of ultrasonic tones, that would also be rejected.
I can see looking for Gibbs envelope, then examining the waveform to see if it's just ultrasonics, or if the signal envelope caused the action.
Jn
Last edited:
My simple conceptually oriented minded approach is just to solve the matter, as far as what is heard, by up the sample rate and move any complex issues further away from audibility. Try it and listen. If it makes things more realistic/accurate, keep iit.
The gory details are important to how we go about engineering the solution. But, that is up to others'. Maybe the DSP/ADC/DAC designers. Thats definitely, not me.
New views other than Nyquist etc is refreshing. It always helps to know more by filling in missing information.
I wish more engineers would listen to what people, even if not the majority, hear and try to figure out how that can be in engineering terms. Instead of thinking its good enough for some is good enough for all.
THx-RNMarsh
Last edited:
Nope. I was not using the miniDSP. yes it uses opamps. I replaced them. i saw no HF filters on input.
THx-RNMarsh
It take time to accept new technology.
From AM radio to FM radio.
From tube to solid state.
From VCD to DVD to Blueray.
From mp3 to losseless compression.
Etc.
Not only from people who use the technology, but also from the industries.
Engineer who made the implementation learn to make better solution.
New technology is not always better in all aspect in the beginning, but the implementation improved time by time.
But still, there are people who love their memories and still keep the old technology no matter what.
From AM radio to FM radio.
From tube to solid state.
From VCD to DVD to Blueray.
From mp3 to losseless compression.
Etc.
Not only from people who use the technology, but also from the industries.
Engineer who made the implementation learn to make better solution.
New technology is not always better in all aspect in the beginning, but the implementation improved time by time.
But still, there are people who love their memories and still keep the old technology no matter what.
I know the relationship holds for any arguments to the sin() function. I am questioning what you are relating. As someone already pointed out, there are an infinite number of, umm... numbers, that can be plugged into that formula and produce a correct result.
You are the one who keeps saying that people here do not understand fundamentals. Like this: The sine function - math word definition - Math Open Reference
So sin() is a function which takes a number as an argument and returns a number. The result is unitless, and in order to produce the results seen in sine tables (I remember those) it needs to know the units of the arument, because the only valid arguments are angles. The sine of 45 degrees is not the same as the sine of 45 radians. When I ask my calculator to take a sine it needs to know the units I am using, and Hz is not a choice. So exactly what does your "sine identity" actually show?
I think my question actually is interesting, and your answer is unsatisfying. I think the trigonometric identity equation you dug up is, by definition, exactly correct, and unrelated to your example.
With what? You just said you couldn't use it and now you replaced the "741's" not noticing they were dual op-amps. Do you make this up as you go along?
It's highly likely it was a datasheet style circuit, too. Which is usually at least a 2nd order LPF.
I will say this again, if you use a math algorithm to create a signal and sample it at 44.1kHz it will contain images and if >22050Hz aliases. End of story you can not use this data to prove anything.
- Status
- Not open for further replies.