KVR Audio

Mutant · Post by **Mutant** » Sat Aug 22, 2015 10:55 pm

aciddose wrote:Define "noise".

Diva A_B Test.wav between 23 and 27.500 seconds

A good sample for some noise removal plugin like REAPER's ReaFIR in substract mode for example.

IncarnateX · Post by **IncarnateX** » Sat Aug 22, 2015 10:55 pm

aciddose wrote:I'm perfectly calm and your continued accusations that my responses to your posts are emotional is also appearing as an attempt to undermine my arguments by attacking me personally.

Know how feel aciddose! When I argue with people they keep referring to paranoia. I have no clue who that is but he is probably one of those stalking me.

Teksonik · Post by **Teksonik** » Sat Aug 22, 2015 10:56 pm

aciddose wrote:Is it possible we can stop with this?

Hey you started it with other member....if you can't take it don't dish it out.

aciddose wrote:This forum can be so much more than this. This forum is the sum of its members. We can make it more.

Yes it can be so much more than constantly arguing over and over and over about hardware vs software. Day after day month after month year after year.

All this pseudo scientific nonsense over a couple of audio clips.

Really ?

Teksonik · Post by **Teksonik** » Sat Aug 22, 2015 10:58 pm

trimph1 wrote:I'm tempted to bring ampsim tests into this mess....

Or "Does it sound like a Virus" or the golden oldie "What sounds like a 303 ?" Or the ever popular and equally unending "PC vs Mac".............

IncarnateX · Post by **IncarnateX** » Sat Aug 22, 2015 11:01 pm

You seem manic Teksonic. What are you on and where can I get some?

SWAN808 · Post by **SWAN808** » Sat Aug 22, 2015 11:02 pm

IncarnateX wrote:It is a pity you cannot see the point here for I would say you could actually make a more bullet proof scientific design by taking my points seriously - if that is what you want.

I said that is not what I wanted. At the start of the test I offered people the best way to test themselves should they want to - ABX comparison with the wavs to a high level of confidence. This as Ive said - is a discussion piece. Frankly its mental that you are now bringing up publication.

Teksonik · Post by **Teksonik** » Sat Aug 22, 2015 11:04 pm

IncarnateX wrote:You seem manic Teksonic. What are you on and where can I get some?

It's called "Reality"....you should try it sometime. It will blow your mind...........

IncarnateX · Post by **IncarnateX** » Sat Aug 22, 2015 11:05 pm

analoguesamples909 wrote:
IncarnateX wrote:It is a pity you cannot see the point here for I would say you could actually make a more bullet proof scientific design by taking my points seriously - if that is what you want.
I said that is not what I wanted. At the start of the test I offered people the best way to test themselves should they want to - ABX comparison with the wavs to a high level of confidence. This as Ive said - is a discussion piece. Frankly its mental that you are now bringing up publication.

Too late. He is long gone. You will have to deal with me now...and it won't be in scientific terms....dickhead!

Teksonik · Post by **Teksonik** » Sat Aug 22, 2015 11:07 pm

Ok I'm out.....but let me say in parting.....DIVA is a wildly popular synth. The Obie is as well but there aren't too many of them left and those that are still functioning tend to be priced out of all but the collectors market.

So if you've got an Obie use it.......if not use DIVA and be well happy.........

And for God's sake people.......lighten up.......

aciddose · Post by **aciddose** » Sat Aug 22, 2015 11:07 pm

Mutant wrote:
aciddose wrote:Define "noise".
Diva A_B Test.wav between 23 and 27.500 seconds

A good sample for some noise removal plugin like REAPER's ReaFIR in substract mode for example.

Well, I wasn't asking for a concrete definition

Yes this is perfectly valid, but my point was that some things that might be considered noise such as high-frequency leakage (15khz buzzing while cutoff = minimal) or crossover distortion (sort of like an analog bit-crusher that only applies to the 1st bit) are entirely desirable. Especially if one is attempting to reproduce the response of particular circuit as closely as possible in software.

I'm not just grabbing at straws here, these are two reasons I find my analog synthesizers extremely preferable vs. software, due to the fact these properties are very expensive to reproduce in software and that no software I've ever seen actually does reproduce them.

So this is based upon subjectivity, but the argument does not rely upon it.

1. It is possible to have a preference for properties that lead to the inclusion of "noise" in a synthesizer's output.
2. (more just a note, not premise) It is expensive or impractical for software to trivially reproduce these properties and effects.
3. <insert plugin here> doesn't reproduce these effects and is differentiated by the presence or lack of these effects from other synthesizers.
1. Therefore, it is possible to have a preference for a synthesizer based upon these facts.
2. Therefore, it is incorrect to state that one or the other is more desirable in all cases.

edits: the list syntax isn't exactly a piece of cake for me. I hope it helps people understand my argument. It would help if phpBB didn't remove/add new line endings.

SWAN808 · Post by **SWAN808** » Sat Aug 22, 2015 11:08 pm

KVR...what a funny place...

trimph1 · Post by **trimph1** » Sat Aug 22, 2015 11:11 pm

analoguesamples909 wrote:KVR...what a funny place...

Reveal the test results!!

aciddose · Post by **aciddose** » Sat Aug 22, 2015 11:16 pm

The test is easy to predict. It will approximate a normal distribution without bias in the data. If the data is biased the distribution will shift due to those biases.

So this test assuming it was set up correctly (no biases), assuming it was designed to demonstrate no predictable difference between the two will provide a result equivalent to a coin toss.

If bias were introduced, the result will show that bias.

This is ground-level stuff and the explanation that it's intended as an advertisement for a bank is the only way I can understand the desire to include such sensationalism in the title or posts.

To claim a result demonstrates anything other than the expected result of a coin toss is fraud, intentionally misleading. If people wanted to identify sources of bias that is something else, but in that case the thread title should be changed!

aciddose · Post by **aciddose** » Sat Aug 22, 2015 11:28 pm

Also although I'm saying "if bias exists it will show up in the result" that doesn't mean it will be significant. People might be surprised at this fact, but when it comes to identifying differences using subjective means we humans are absolutely horrible at this.

In many cases you can intentionally introduce a bias that you consider objectively, relatively (using concrete measurements) "significant" and find that the result of the experiment yields no significance.

So I just want to note that I'm not claiming there is any certainty that a concrete bias in the data will translate to a bias in the subjective interpretation of that data.

trimph1 · Post by **trimph1** » Sat Aug 22, 2015 11:31 pm

Teksonik wrote:
trimph1 wrote:I'm tempted to bring ampsim tests into this mess....
Or "Does it sound like a Virus" or the golden oldie "What sounds like a 303 ?" Or the ever popular and equally unending "PC vs Mac".............

That'll work as well.

Diva vs Analogue - a real world test