I don't hear it!!!! you?

[JulianBrightnes]

Hey slutz,

Just noticed this dither-question thread the other day: "which dithering was the best?"
To be quite honest: I don't hear any friggin' difference between POWR, IDR or none at all!
Here's what I did: there's a music clip I bounced down from 24bit/48K (sorry, I'm in broadcast mainly) to 16 bit/48K, one time using POWR, once using IDR (both type 1 noise shaping) and once using nothing at all.
I don't hear any difference, they even cancel out with one another. Can you guys tell which one is which? I know the theory behind it... but hear it ??

Cheers, Julian

[PoorGlory]

I haven't listened to that particular test (at work right now) but I have played around with different dithers, different dither types and noise shaping types.. They all sound a little different, enough to make you want to choose one over the other after A/B'ing them. However, I guarantee you no one ever listened to an album and went "Gee, I wish they used IDR Type 1 dither with the Ultra noise shaping". Its extremely subtle, and most people won't give a shit.

[Ethan Winer]

Julian,

> I don't hear any friggin' difference <

Neither do I. I'm convinced it's a case of the Emperor's new clothes. If you think about it, all the action happens more than 80 dB below the signal level. How could something that soft possibly make a difference with typical pop music pounding away 80 to 90 dB louder?

I'm sure people do hear a difference, but I suspect the difference they hear is due to something else. For example, if you move your head even an inch or two, the frequency response changes due to the different arrival times from each speaker. You can prove this easily by measuring the frequency response at the listening position, then moving the measuring mike an inch or two and measuring again.

Reflections off nearby surfaces also skew the response slightly over very small distances. I'm pretty sure that is the main reason people report hearing a change, regardless of what they thought they were comparing.

--Ethan

[mogWai]

Originally Posted by Ethan Winer

J
I'm sure people do hear a difference, but I suspect the difference they hear is due to something else. For example, if you move your head even an inch or two, the frequency response changes due to the different arrival times from each speaker.

Great. Every single blind test rendered obsolete.

[severe]

Yea...my iPod's not picking up any of this "friggin' difference".

[Sizzleboy]

Originally Posted by JulianBrightnes

Hey slutz,

Just noticed this dither-question thread the other day: "which dithering was the best?"
To be quite honest: I don't hear any friggin' difference between POWR, IDR or none at all!
Here's what I did: there's a music clip I bounced down from 24bit/48K (sorry, I'm in broadcast mainly) to 16 bit/48K, one time using POWR, once using IDR (both type 1 noise shaping) and once using nothing at all.
I don't hear any difference, they even cancel out with one another. Can you guys tell which one is which? I know the theory behind it... but hear it ??

Cheers, Julian

The only time you will hear the difference is at extremely low levels. These levels are usually only reached during fade outs of audio and reverb. 99% of the time, the dithering noise is well below the audio signal level. Technically, the best dithering noise would be that which is the quietest and most unobtrusive, yet masks the stepping of volume levels that occours at very low levels.

If you really want to here what a particular shape of dithering noise sounds like, check them out at 8 bits.

You could also reduce a file's signal to about -85dB, apply the dithering, then normalise the file back to 0dB.

[Ethan Winer]

> Great. Every single blind test rendered obsolete. <

ROF,L. That was the farthest thing from my mind when I wrote that. Indeed, this is exactly why proper blind testing is so useful. A valid test will switch back and forth repeatedly, and attempt to find a trend across multiple tests. That way, the small changes I mentioned, mostly due to comb filtering, will be averaged out.

--Ethan

[RainbowStorm]

I have tested different POW-r, Waves L2 and other dithering algorithms and I had a hard time hearing any difference at first, especially when applying different settings on the same dithering algorithm and comparing different POW-r versions. I don't think I would be able to notice any difference between these, at least not with my Mackies... I noticed a difference when comparing on heavy digital mixes, especially between the POW-r and the Waves L2/L3 on Type 1 Ultra. I used Mackie HR824 monitors. However, I have to add that according to a medical test of my hearing my doctor told me my hearing is unrarely good (which I'm very thankful for), so that might play a critical role. I have had sound over sensitivity but it's getting better. I developed over sensitivity during a period of half a year with sound quality testing going to the extreme. These tests included dithering. When I heard the Waves L2 Type 1 Ultra I was blown away. The sound went slightly more detailed and full, like it was getting heavier in weight without getting muddy and thick. It felt less jittery. I would say that the overall dynamic range of the mix increased with something like 0,2 - 0,4 dB, at least that is how I best describe the difference, like when comparing two different cables of different quality and length.

Even if you can't hear any difference I strongly recommend that you go with the Waves L2/Waves L3 set to Type 1 Ultra. I asked Bob Katz what he thinks about the quality of the this dithering effect and he also thinks it's good. I guess if you ask other professionals as well you'll find some credible comments about the high quality of the Waves L2/L3 dithering algorithm. Also check out the blind testing of different dithering algorithms at http://www.24-96.net/dither/results.htm . I have not heard the MegaBitMax Ultra, so I cannot comment on it.

[JulianBrightnes]

Thanks for the replies, everyone. I'll just keep on dithereing as I always did anyway (better safe than sorry)... I got the hardware L2 set top IDR type one now!

Cheers, Julian

[Jules]

Sometimes I just toggle through the available different flavors of dither.. and convince myself that one perhaps, just perhaps sounds better suited than the other.. Can you do more than that?

If you were dealing with classical music and long piano decay notes, then I would fully belive folks involved in that field do hear a difference and have a right to be fussy about it.. But as for us pop n rock n rap rock n rollers?... Naaaaaaahhhhh. not really... Mastering engineers probably have a keen ear for this stuff..