kevinwal

fret_man
Responses 8 and 15 basically contradict each other. I wonder which is correct?

How so? I don't see that they are contradictory in any way.

I would agree regarding the recommended usage for pow-r 1 and pow-r 2, but with pow-r 3, you couldn't get a bigger discrepancy regarding intended use:
 
Pragi says - Pow 3 for metall, EDM,Hard Techno  aso
mikedocy says -  pow-r 3 is made for wide dynamic range classical music
 
Like I said, I use pow-r3 regardless of source material as this produces the most transparent result.
 
I once did a test of all the different algorithms using the longest reverb patch I could find, recorded a very loud snare crack and exported the results of each.
 
Loading them up one at a time and listening as loud as possible makes for some interesting listening, especially the tail/decay
 
Try it - should only take 20 minutes of anyone's time, then you can choose which type to use based on your own findings.
 
 

kevinwal

fret_man
Responses 8 and 15 basically contradict each other. I wonder which is correct?

How so? I don't see that they are contradictory in any way.

+ 1

Bristol_Jonesey

kevinwal

fret_man
Responses 8 and 15 basically contradict each other. I wonder which is correct?

How so? I don't see that they are contradictory in any way.

I would agree regarding the recommended usage for pow-r 1 and pow-r 2, but with pow-r 3, you couldn't get a bigger discrepancy regarding intended use:
 
Pragi says - Pow 3 for metall, EDM,Hard Techno  aso
mikedocy says -  pow-r 3 is made for wide dynamic range classical music
 
Like I said, I use pow-r3 regardless of source material as this produces the most transparent result.
 
I once did a test of all the different algorithms using the longest reverb patch I could find, recorded a very loud snare crack and exported the results of each.
 
Loading them up one at a time and listening as loud as possible makes for some interesting listening, especially the tail/decay
 
Try it - should only take 20 minutes of anyone's time, then you can choose which type to use based on your own findings.
 
 

Hello Jonesey,
if in your eyes (better ears,hehe ) the pow 3 algorythm sounds  best on all
source material -there is no matter why you should not use this algo.
For me it works best  the way I wrote.
I tested all the different algorithms and found out,
that the different pow algorithms are  basic principles which
give simply more effective selection and possibilities, but again, for me.
 
 
 
 
 
 
 

Bristol_Jonesey
I once did a test of all the different algorithms using the longest reverb patch I could find, recorded a very loud snare crack and exported the results of each.
 
Loading them up one at a time and listening as loud as possible makes for some interesting listening, especially the tail/decay
 
Try it - should only take 20 minutes of anyone's time, then you can choose which type to use based on your own findings.
 
 

Something to keep in mind here:
 
Advanced noise shaping algorithms are typically based in large part on equal loudness contours in the range of ATH (absolute threshold of hearing) to 15 phon, because this is the expected level of the noise at typical playback levels.
 
If one dramatically cranks up the volume of a low level signal to make things audible, then one is actually listening at a much higher phon level and the ear's sensitivity is much different, making any conclusions invalid. If someone wanted to create dither + noise shaping that would work better at a 40 phon listening level, then they would base their noise shaping curve on 40 phon and not ATH or 15 phon and it might impress people who audition at artificially high levels, but it would be inferior under real world conditions.
 
IOW, it doesn't make any sense to audition dither + noise shaping at dramatically louder listening levels because the noise shaping curves are deliberately calibrated to work best at real-world listening levels and the whole point of using equal loudness contours to create the noise shaping curve is that the ear hears differently at different listening levels.

My point was that post #8 claimed pwr-1 and pwr-2 were good for classical while post #15 suggested pwr-3 was good for classical. I'm not claiming right/wrong here, just trying to get some clarification and learn something. For my use it probably won't make much difference.

fret_man
My point was that post #8 claimed pwr-1 and pwr-2 were good for classical while post #15 suggested pwr-3 was good for classical. I'm not claiming right/wrong here, just trying to get some clarification and learn something. For my use it probably won't make much difference.

Oops, I misread my post as post #15, it's actually #16 so I will officially withdraw the inquiry once I find the appropriate forms. Damn paperwork.

fret_man
My point was that post #8 claimed pwr-1 and pwr-2 were good for classical while post #15 suggested pwr-3 was good for classical. I'm not claiming right/wrong here, just trying to get some clarification and learn something. For my use it probably won't make much difference.

To me... the best bet if it REALLY matters (like a commercial release or some such) is to let a proper mastering engineer make that decision.
 
By that I mean work in your usual working bit depth/samplerate (eg 24bit/48khz or 24bit/96khz or whatever) then send your export... without dithering/conversion to the mastering engineer. They'll know what to do (and if they don't they aren't qualified for the job).
 
For those of us just farting away at home making demos or sharing with our buds... might as well just use the Pwr Dither settings unless you hear a noticeable difference/problem that bugs you.
 
Actually since all the online music players and computer players (like WMP) can all accept/playback 24bit files anyway it doesn't even matter. It seems (AFAICT) that it only needs to really happen if you are burning a CD to be played on an actual physical CD player... and who uses those things anymore.
 
I haven't owned or even touched one for YEARS!
 
lol

drewfx1

Bristol_Jonesey
I once did a test of all the different algorithms using the longest reverb patch I could find, recorded a very loud snare crack and exported the results of each.
 
Loading them up one at a time and listening as loud as possible makes for some interesting listening, especially the tail/decay
 
Try it - should only take 20 minutes of anyone's time, then you can choose which type to use based on your own findings.
 
 

Something to keep in mind here:
 
Advanced noise shaping algorithms are typically based in large part on equal loudness contours in the range of ATH (absolute threshold of hearing) to 15 phon, because this is the expected level of the noise at typical playback levels.
 
If one dramatically cranks up the volume of a low level signal to make things audible, then one is actually listening at a much higher phon level and the ear's sensitivity is much different, making any conclusions invalid. If someone wanted to create dither + noise shaping that would work better at a 40 phon listening level, then they would base their noise shaping curve on 40 phon and not ATH or 15 phon and it might impress people who audition at artificially high levels, but it would be inferior under real world conditions.
 
IOW, it doesn't make any sense to audition dither + noise shaping at dramatically louder listening levels because the noise shaping curves are deliberately calibrated to work best at real-world listening levels and the whole point of using equal loudness contours to create the noise shaping curve is that the ear hears differently at different listening levels.

Yes, point taken Drew. One mitigating factor is something I didn't mention in my first post and that was, I set it up so I was only listening to the reverb tail itself and cranked the volume by using an identical envelope on all the test tracks I order to really listen to what was going on.
The volume in the room while performing these tests never exceeded 78dB and even with the envelopes, always faded into silence.

sorry wrong forum