Samplitude: Best sounding Daw?

[hibidy]

Other than one person, who is disagreeing?

Now, where the tricky part starts is the couple of times I've heard people saying there is a slight difference. THEN we have a problem. Some say "well the problem is so small nobody can tell".........but then we have "that" problem.

Here's a fun fact: So what? I've been doing this a long time, I seen more arguments on more forums about this than many can possibly imagine. Far be it from me to claim I have "the answer" but it does beg the question eventually......so what?

Incarnate can jump all over me if he wants, it's the kvr way, but there is NO PROOF in the long long time I've been around that there is any evidence that one host sounds better than the next.

Of course, then again, there is a percent of percent that believes that NOTHING sounds any different than the next (cables anyone????) and they have the power to ban/lock/etc. So who am I to say?

I'll bow out now, should have never got sucked into the black hole in the first place. These threads NEVER EVER EVER lead to anything good.....they just meander endless and cause heartbreak.

[Burillo]

these random illusions are in effect random. simply because every one of us is different and different DAWs appeal to us visually, and we will do these tests in different circumstances (e.g. someone is more comfortable in the lab, someone would be more comfortable at home instead etc.). there are too many factors to take into account - time of day, hours of sleep, whatever. hence, it *is* random. not *completely* random but *mostly* random, with a few corellations that would be difficult to trace unless you have a sample size of tens of thousands of people.

hibidy, beyond a certain point (physical defects etc.) cables do really sound the same it's just like with audio - you can clearly hear a difference between 22 and 44 kHz, yet no one can hear the differences between sample rates above that point unless something bad happens (e.g. your DAC has jitter problems that manifest themselves on very high sampling frequencies and thus degrade the sound).

[como baila]

Burillo, it's a shame to confound 'the null hypothesis' with nulling audio. The first is scientific research paradigm generally buttressed by statistical analysis and the second is a simple fact, i.e., it nulls or it doesn't. Perhaps it's a language thing?

I now find myself quarreling with everyone ... with the possible exception of Hibiddy.

Nulling audio establishes that two (or more) sources of audio are identical, when one is phase reversed and they are then played back together, they cancel each other out and produce silence ... allowing for system noise independent of the audio sources.

The audio recorded could be sublime or horrid. But the sound as we experience it is the end result of a sensory-perceptual process, not the independent physical properties of the sound source.

What I am having a hard time articulating to IncarnateX is that we could remove the confounding of cross-modality perceptual factors and design an experiment in which only sound preference is evaluated. In theory we could establish whether one thing sounded better than another to some majority of people by some type of group experimental design. This, however, could not answer the question, since it would be a matter of preference, no different than polling people whether vanilla or chocolate ice cream tasted better.

Better would be to take individuals who purport that one thing sounds (to them) better than another in an identifiable manner. We could them let them listen to samples of the same file played on different DAWs and see if they could identify above chance the DAW they claim 'sounds better.'

A double blind test minimizes bias. We are now taking people who claim to like Haagen Daas vanilla ice cream better than Ben and Jerrys or some other brands. If they can repeatably and reliable identify the best sound as coming from their chosen DAW, we can now say their claimed preference is based upon their actual experience of the sound.

If they cannot, then we reasonably 'fail to reject the null hypothesis' and state that whatever other factors, cross-modal or otherwise, that shape their preference, 'sound' is certainly not one of them.

Philosophers of science since the early Greeks have produced weighty tomes on the translation of sensation to perception without final agreement. The problem is unlikely to be solved on this audio forum. But the recognition that sensation must be translated somehow to perception is unavoidable.

Which is why even horrid sounds can sound good.

Como

[Burillo]

it's certainly a language thing, my understanding of the "null hypothesis" was apparently different than what was yours i apologize for misunderstandings.

[IncarnateX]

these random illusions are in effect random).

That is the thesis to be tested with my paradigm. If there is a correlation between the subjects visual and auditory ratings even though there are no difference in auditory output, they are apparently not random, but systematic. Mind you that this only tells us that the visual stimulation influence their ratings but not whether this influence are due to the fact that they really rate the visual preferences and not the auditory or, as my hypothisis goes, that the visual prefences actually affects their attention levels. But this would be as close you can get to testing my hypothesis with those means. It would have to be further consolidated by involving for instance different kinds of measuring of brain activity, that is whether the ratings only are correlated with activity in active visual ares or also correlates with activity in auditory areas to dive further into the two explanations given in the above.

[IncarnateX]

What I am having a hard time articulating to IncarnateX is that we could remove the confounding of cross-modality perceptual factors and design an experiment in which only sound preference is evaluated. In theory we could establish whether one thing sounded better than another to some majority of people by some type of group experimental design. This, however, could not answer the question, since it would be a matter of preference, no different than polling people whether vanilla or chocolate ice cream tasted better.

I don' t know if I enter a paradox now and start siding with Burillo, but I do not see any obvious reason to blind test the predictions of the DAWs null test in respect the auditory modality only. We already know from the DAW null test (not to be confused with the null hypothesis) that there ARE not any differences in the produced output, so if the ratings are not random, we would have no alternative hypothesis other than speculation because we haven't taken other factors into the experimental design (e.g. the visual stimulation) that could account for this result. It would get us nowhere in respect to what is already given by the DAW null test. The prediction of the DAW null test would be that the ratings of auditory material in a double blind test are random, if they are not, any further analysis is prevented because lack of measured factors.

[IncarnateX]

Incarnate can jump all over me if he wants, it's the kvr way, but there is NO PROOF in the long long time I've been around that there is any evidence that one host sounds better than the next. .

I would not jumb over you in any contexts, Hibidy. However you also misunderstand my point. The aim is not to show that DAWs produce differences in output. The DAW null test is enough to prove that they don't, just like Burillo states. The question is whether there are others factors of perception, which can account for the fact that people actually say they hear differences other than claiming they are lying to themselves or hallucinating.

Another point: In science "no proof" does not mean that we shouldn't keep investigating a matter. If this were the rule we would soon be out of business, know what I mean?

And come here and get a hug mate, I am not as bad as I eh...sound on paper

[rifftrax]

...is that we could remove the confounding of cross-modality perceptual factors...

LOL. Seriously wtf.

We talking about DAWs sounding different, not "how many crazy scholar/scrabble/thesis-paper dictionary words we could combine in a loose-fitting uber phrase".

"Today class, we examine the effect of how some people think 1 + 1 = 2 and how others believe that 1 + 1 = 0. Does anyone have any pertinent pontifications on the matter?"

[BertKoor]

Hallucination indeed. Anyone ever found themselves in the following scenario? I know I have, about once a year at least.

Tweaking a (hardware) EQ or compressor to make it sound just a bit better. Wow, that's a subtle knob, doesn't do much... Oh wait, that's the sweet spot! Far better than it ever was.

Hold on, there's still that dip/peak/thing I supposedly took care of 10 minutes ago. [look better] Ah, that explains it! It was on bypass all the time

[LawrenceF]

These "DAW sounds better" debates are pointless without very strictly controlled double blind testing, as someone here stated. The reasons are not only pretty clear, but scientifically valid and already proven and tested... that our senses all work in concert and and our brains take shortcuts for efficiency.

There have been a multitude of experiments that demonstrate beyond any doubt how easily human senses are influenced by each other. None of them (not any single one of them) work in isolation and all of them can be (and regularly are) influenced by any of the others.

Play the exact same file from the exact same place the exact same way and ask people to pick a difference and many will. That's mostly where the psychology of it all comes into play, with expectations. The other stuff isn't psychology at all, it's biology. Stimuli comes into the brain from the various senses and the brain builds the most efficient and useful result. It's an inherent human survival mechanism that's not necessarily accurate, it sums 1+1 to make the most likely "2", for speed and efficiency... when the real result sometimes actually isn't 2.

It's kind of the only reason why magicians exist, they take direct advantage of that because it's quite well known.

[ThoughtExperiment]

wibble

[IncarnateX]

This is manily adressed to Burillo and Como:

Here are the predictions of our different stands on this matter according to the suggested design.

The independent variable is the visual preferences. Lets say we make people rate the daws on a scale from 1-6 for visual pleasentness.

The dependent variable is the auditory rating (1-6) of exactly the same audio material on the different daws.

There are three conditions:

1. The subjects rate the auditory material for pleasentness while looking at the DAWs.

2. The subjects rate the visual pleasentness of the daws while looking at them but not hearing them.


The last is the control condition.

3. We provide them with a blind test in which they rate the same audio material without seeing the DAWs.



Burillo's hypothesis: Any differences in auditory ratings are random illusions. Predictions:

1. The visual and auditory ratings are not positively correlated but random.
2. In the blind test the subjects are not able to pick out their favorite DAW; effects are random.



My hypothesis: Any differences in auditory ratings are not random illusions but systematically influenced by visual preferences.
Predictions:

1. The visual and auditory ratings are positively correlated and not random.
2. In the blind test the subjects are not able to pick out their favorite DAW; effects are random.

Note that we share the expected outcome of the control condition (3). In both hypotheses, it is taken for granted on basis on the DAW null-test that the DAWs produce no actual differences in stimulation and therefore we both expect that the subjects are not able to pick out their favorite DAW beyond chance on basis of the auditory material alone.

Now could someone please do this test in their sparetime and come back and solve the matter?


....and to avoid any bias: Please do not use KVR members, choose your subjects with care.

[ThoughtExperiment]

This is manily adressed to Burillo and Como:

Here are the predictions of our different stands on this matter according to the suggested design.

The independent variable is the visual preferences. Lets say we make people rate the daws on a scale from 1-6 for visual pleasentness.

The dependent variable is the auditory rating (1-6) of exactly the same audio material on the different daws.

There are three conditions:

1. The subjects rate the auditory material for pleasentness while looking at the DAWs.

2. The subjects rate the visual pleasentness of the daws while looking at them but not hearing them.


The last is the control condition.

3. We provide them with a blind test in which they rate the same audio material without seeing the DAWs.



Burillo's hypothesis: Any differences in auditory ratings are random illusions. Predictions:

1. The visual and auditory ratings are not positively correlated but random.
2. In the blind test the subjects are not able to pick out their favorite DAW; effects are random.



My hypothesis: Any differences in auditory ratings are not random illusions but systematically influenced by visual preferences.
Predictions:

1. The visual and auditory ratings are positively correlated and not random.
2. In the blind test the subjects are not able to pick out their favorite DAW; effects are random.

Note that we share the expected outcome of the control condition (3). In both hypotheses, it is taken for granted on basis on the DAW null-test that the DAWs produce no actual differences in stimulation and therefore we both expect that the subjects are not able to pick out their favorite DAW beyond chance on basis of the auditory material alone.

Now could someone please do this test in their sparetime and come back and solve the matter?


....and to avoid any bias: Please do not use KVR members, choose your subjects with care.

Or alternatively the three of you could get a room

[como baila]

Or alternatively the three of you could get a room

Who can afford that? The street works for me.

Como

[como baila]

...is that we could remove the confounding of cross-modality perceptual factors...

LOL. Seriously wtf.

We talking about DAWs sounding different, not "how many crazy scholar/scrabble/thesis-paper dictionary words we could combine in a loose-fitting uber phrase".

"Today class, we examine the effect of how some people think 1 + 1 = 2 and how others believe that 1 + 1 = 0. Does anyone have any pertinent pontifications on the matter?"

Look, I don't resent it that you're philosophically unsophisticated. Perhaps you can design high end audio circuits from the ground up? I know, I can't.

But if you get a chance, check out Aristotle's four types of causation. It will help you get a grip on how to think through problems like whether or why DAWs might sound different to different people.

Your sarcasm does reflect some promising command of language, if not of reasoning.

Como