If I could have motivated you all to discuss the mass of diaphragms (as you are, sample rates, now), I could have had the aspect covered already lol. Don't worry I will read the books...only a day or two more of work left. =) Dmitrij_S just throw a book at them
I would have had nothing to say on that matter, as I know sweet FA about the subject
I know barely more than that about the intricacies of sampling, so i'll leave it there.
lol, well thats honest. Still I appreciate the convo. I still need to dig deeper on this sample rate topic it appears.
random, or perfectly even doesnt make any difference. Mathematical theories don't know what music sounds like, so as to effectively ignore it AFAIK
And that is the biggest mistake (also the guy in the video) makes.
If you know a thing or two about data acquisition (not just for audio), you know that there is a huge difference between signals that look the same and are the same.
It's not about what it can do, one should ask the questions what it CAN'T do.
With only two samples you can't even distinguish the difference anymore between a sine wave, triangle wave or square wave.
(it all turns into some kind of sine wave, although a square wave will have an higher amplitude)
Let alone sample little blips (transients) that occur somewhere in between.
Or maybe even a big spike exactly on one of the sampling points.
It basically all acts like an integrator (well that is exactly what it is)
Music (definitely) isn't a pure sine wave either.
So pure theoretically speaking, from a data acquisition point of view.
Yes, there is obviously loss of signal information.
The (only) important question is if that is noticeable or not.
Last edited:
ha!!
nah, I just know when i'm surrounded by bigger brains and syn08 is definitely on that list. the gist of that part I was trying to articulate, is that if you use 2 points exactly, you end up with aliasing, because it is too close to (exactly on) the point where it can be mistaken for multiples of that rate. so a 100Hz wave described by 200 points, could allow other frequencies/waves to pass through those same points and only 1 is allowed.
nah, I just know when i'm surrounded by bigger brains and syn08 is definitely on that list. the gist of that part I was trying to articulate, is that if you use 2 points exactly, you end up with aliasing, because it is too close to (exactly on) the point where it can be mistaken for multiples of that rate. so a 100Hz wave described by 200 points, could allow other frequencies/waves to pass through those same points and only 1 is allowed.
It is time for you to start using the question mark (i.e. the "?"). The conversation will be much better if you indicate when you are sure about something and in fact are doing a statement or when you are guessing.
Your "questionstatement" posting style is cumbersome as I see it
//
I thought about whether it should have been a question or not...then a proceeded. We've had this discussion before....
If we actually going debate it, lets be specific. To be fair I'll stick with the same theoretical config that inspired the topic
1 - 15" coaxial and 2 - 18" on the outsides to make a MTM 3 way.
Lets say we can pull off 0,5 inches in between drivers
1/4wl is ~200hz and a ka=2 for an 18" driver is approximately 477hz....
A cross below 200hz or lower not only has optimal CTC, but is well below desired Ka, so dispersion is not colored in the majority of off axis, at least via driver.
A cross at 477hz puts vertical limits the listening window, due to the xo and un-optimal CTC. Ka is still well enough wide to not further exacerbate the banding on the vertical polar...
Theres nothing left to do but linearize phase for the desired listening position with DSP. At almost any distance, this can only be corrected for one spot (some attempt to average over a span of areas) aka as "head in vice" and within the criteria above correcting phase should lean towards good results at 1meter or 3, for the use of the above config. If you disagree, whats the science behind your reason.
I was under the impression that any wave shape can be described by a series of sine waves added upon each other. Square waves do not exist in the real world, cannot be recorded, or reproduced. We describe them with a combination of the required sine waves added on top of one another.
in other words, in the real world, everything is made of sine waves, or at least can be represented by them.
in other words, in the real world, everything is made of sine waves, or at least can be represented by them.
Last edited:
So you dint know the theorem either. 2 samples for 1 period don't produce a triangle, it produces a sine because the theorem uses a sinx/x filter.
I bet 99,9% of the members of this site dont really know how the PCM system really works. But 2/3 behave like the do.
//
But what about when you independently adjust mass of a diaphragm? Yes I get you move sensitivity around. Doesn't this sensitivity come with drawbacks? Seems like everything comes with pros n cons...what are the cons of sensitivity?
Obviously Mr. Fourier, Nyquist and Shannon are not your friends.
Last edited:
In the context of waveform reconstruction, the statement is true, but in the context of correct transmission of spatial information of a stereo recording it is incorrect.
Actually square waves do very well exists in the real world.
Just flick a switch on and off.
They can be represented by sine waves as well, although you will end up with not exactly the same thing unless you take A LOT of sine waves.
No idea what you're trying to say, using a few more words and less assumptions might help a discussion.
No, that is not a square wave. add in some real world parasitics and physics into it and there is no perfectly on, or perfectly off, nor a perfect transition between the 2.
a computer can be told what voltage level (with some margin) to call on, or off, 1, or 0, but that does not make it so.
Last edited:
I am lost, we are talking about what artifacts sampling can give?
A con of less sensitivity is that you need more power.
Since that goes with a 10LOG, you need A LOT more power to get to the same SPL.
(twice the power for only 3dB extra
)Which means everything gets more warm, which means more things like power compression etc.
Although for home use that is a little far fetched.