its not fair to use RMAA to say the frequency response is flat when soundcards can inflate the harmonics.
a simple sine wave is totally different than anything with an echo of some sort.
if you have the knowledge and power, try those frequency sweeps with any reverb effect like 'cave' or 'auditorium' and watch those frequency responses alter.
also, despite the voltage being the same level throughout the frequency sweep.. that doesnt mean the timing is the same.
anybody with basic electric circuit building know-how can build an amplifier to clamp tight for a frequency response sweep.
it's all about how you excite the amplifier that can change the results.
different release timings for different frequency ranges can cause the sound to be warm or cold.
the voltage can stay the same and be registered as flat, but that doesnt mean things like attack or sustain are being recorded to showoff the quality of the soundcard.
technology is getting upgraded often.. and these things wont necessarily change.
the experts dropped out of the game too soon, and the masters of today will do the right thing to keep people constantly wanting to upgrade.
their wanting to upgrade will translate to other people wanting to buy their first piece of hardware.
the better it looks, the more likely someone is going to buy it even if they dont use it much.
i cant argue much about software touching the audio and adding noise.. because that may be true down to 0.00000005 %
the shape is what is important, if you raise the amplitude.. they can take a picture of what the piece of audio looks like before it was boosted.. then make the piece of audio look exactly like it did before it was touched.
just think about it.. have you seen the digital art created on computers from scratch?
the pictures (and video) can totally fool the person to have absolutely no idea if it was taken with a camera or made on the computer.
construction quickly brings upon re-construction.
if you can see it before you alter it, you can shape and mold it back to what it was.
have a look on youtube and watch artists sketch a 3d model of a persons face.
you can watch it come together from a round blob, up until they are tapping paint onto the cheeks to make it look perfect.
the better we can see the audio signal in its digital form.. the better it can be boosted or cut without any damage.
yes, some of the pieces might be fake when you are done.. but when it plays across the digital to analog convertor.. that chip isnt going to know any different.
math is a gift from nature, and it stems into the high degrees of complexity that can make a grown make collapse without warning.
that means a $10,000 microphone isnt sensitive enough to capture the difference a software equalizer makes.
as i said, there are different qualities.
have you ever watched a cartoon where a farmer feeds a pig?
the audio industry is feeding us the same scrap with these audio plugins.
(empty bones and apple cores)
only the most expensive stuff has meat on the bone and some actual apple left to eat.
but those pieces of software cost thousands of dollars.
we are still at the very lowest point of digital audio.
they continue to lie to us about sample rates.
it is impossible to record 20khz when the sample rate is only 44.1khz
you have TWO chances per second to capture the 20khz signal.
each of those chances only last 1.360544218 milliseconds
go here and see how fast you can click on the mouse button when the screen changes:
http/www.humanbenchmark.com/tests/reactiontime/index.php
most people score 215 milliseconds.
i just scored a 189ms
to try and capture some treble, you only get 1.3ms
that is super fast !!
ONE chirp of 20khz lasts 3ms
two chances @ 1.3ms each = 2.6 ms NOT 3ms
that is what they mean by Nyquist–Shannon sampling theorem.
2.6ms does not fit exactly (or with extra) when compared to 3ms
i've learned...
each sample has a shape like this ^v
and it actually looks something like this .......... '
that shape can have 65,536 different variations.
now, this might not be absolutely perfect.. but it is easier than thinking each shape has it's own address in a compiler somewhere.
some soundwaves have an echo.. and that means the shape is more complex than a solid '
it looks smeared as if you scribble with a pencil and wipe your thumb across it.
that is how you realize the number 65,536 can potentially get used up really quick.
because you have to respect the dots per inch when you are looking at that shape.
you can have a large grid or a grid that has absolutely no gaps (ray tracing)
the smaller the grid.. the more those numbers get used up faster.
same thing with dots per inch with an optical or laser mouse.
now, when you are looking at the audio with that kind of magnifying scope.. you can then use math to boost the amplitude and keep its shape by going in there and manually re-shaping it.. just like a 3d modeler shapes a persons head or body.
it is bloody perfect.. and electronic components that go onto a circuit board are also right behind in perfection.
computers run on ultra low sonic when they want to build one like that.
the smaller the voltage changes, the smaller the entire voltage requirement.
you might see a computer running exactly like your computer (or some brand new super fast gaming computer) running on only .0002 millivolts
static is a severe concern and you might have to wear an anti-static body suit to even get close to it.
software plugins are not the enemy.. the cheap effort put into creating the software plugin is why they dont work very good.
i mean, when a plugin costs more than your soundcard.. some time and effort went into creating it.
when audio software costs more than the entire computer.. some serious time and effort went into it.
i've seen computer software that costs more than my car..!!!
there are probably people programming computer software that costs more than the price of a house.
if you give up re-creating the sound sample.. then yes, it would then have some noise in it.