Industry standard dither?

[Unknown soldier]

Is the apogee UV22 considered the "best" dither/wordlength reduction? Almost all digital processors/software has dither these days. Just wondering what's considered the industry standard...

[malice]

I would say hear the contestants, and make your own opinion...

http://www.24-96.net/dither/

there is no standart so far, several company propose differents algorythms, they are different sounding ...

malice

[thethrillfactor]

Originally posted by Unknown soldier
Is the apogee UV22 considered the "best" dither/wordlength reduction? Almost all digital processors/software has dither these days. Just wondering what's considered the industry standard...

Probably the POW-R dithering scheme.

[e-cue]

POWr (noise shaping #3) and IDR are my most popular... Sometimes I'll try that crane song thingy out.

[doug_hti]

Originally posted by e-cue
POWr (noise shaping #3) and IDR are my most popular... Sometimes I'll try that crane song thingy out.

I had been using #3 most often, but recently I've found that type 1 is good for some stuff as well, depending on what song. The only way I really hear the change is in the vocal.

But POWr has got to be one of the best, over waves, apogee, etc.

[Renie]

Originally posted by thethrillfactor
Probably the POW-R dithering scheme.

Thrill,

Have you tried the Cranesong Analog dither from the CD?

[thethrillfactor]

Originally posted by Renie
Thrill,

Have you tried the Cranesong Analog dither from the CD?

Hi Renie,

I own the HEDD unit with the dither on it.

A year ago I tested every dither I could find.

The (2) that I liked the best were the POW-R dither (3) and the Spark XL dither. The Crane Song was pretty good.


I am testing out now the different CD burning/mastering programs on the MAC. So far Waveburner Pro is winning. I am going to try all of the PC programs this week. I've heard that the math is better on a PC and will see if this is true.

One thing that has become apparent with PT on the mac is that a recorded split file(instead of a bounce) master sounds way better when burned in Waveburner Pro(more depth and bottomn).

[NathanEldred]

POW-r is really great. It adds that certain something. Samplitude 7.0 makes it a dream to use too as it's built in as an automatic and changeable output function. I'm also usually defaulting to POW-r #3.

[Renie]

Originally posted by thethrillfactor
Hi Renie,

I own the HEDD unit with the dither on it.

A year ago I tested every dither I could find.

The (2) that I liked the best were the POW-R dither (3) and the Spark XL dither. The Crane Song was pretty good.


I am testing out now the different CD burning/mastering programs on the MAC. So far Waveburner Pro is winning. I am going to try all of the PC programs this week. I've heard that the math is better on a PC and will see if this is true.

One thing that has become apparent with PT on the mac is that a recorded split file(instead of a bounce) master sounds way better when burned in Waveburner Pro(more depth and bottomn).

Ah, I think you like Spark XL for SRC too.
I'll have to look into that and Waveburner Pro...
btw I've got the Massive Passive here, on loan.

[thethrillfactor]

Originally posted by Renie
Ah, I think you like Spark XL for SRC too.
I'll have to look into that and Waveburner Pro...
btw I've got the Massive Passive here, on loan.

Good luck with the MP.

The only thing i would say is try to approach it with out expectations.

A lot of people were dissapointed because they were expecting a Pultec and got something else.

It is a very versatile EQ.

If you need any pointers or ideas let me know.

[Bob Olhsson]

UV-22 is pretty long in the tooth at this point. I used to use it all the time and thought it sounded like tape. (To be fair, I haven't heard their latest version.)

There are really a couple schools of thought. People who like noise shaped dither pretty much prefer Pow-R. Others don't like any kind of noise shaping and prefer flat TPD which just about all gear offers as an option. I like WAVES IDR with the noise shaping off or Pow-R 3 depending on the material. I also try the Crane Song dither because sometimes it works absolute miracles.

[RKrizman]

Originally posted by Bob Olhsson
[ I also try the Crane Song dither because sometimes it works absolute miracles. [/B]

I've designed my own custom dither by taking some nature recordings I made at King's Canyon at sunrise (soft air sounds, streams, the occasional bird) and eq'ed it to resemble the curve of the Cranesong dither. It's absolutely beautiful, and the occasional "bird spike" adds a subtle bit of interest. I'd share it with you guys, but I'm not willing to give up the unfair advantage it affords me.

Best of luck with your pale, 2-dimensional recordings.



-R

[RKrizman]

Originally posted by thethrillfactor

One thing that has become apparent with PT on the mac is that a recorded split file(instead of a bounce) master sounds way better when burned in Waveburner Pro(more depth and bottomn).

Way better? I'm willing to bet they are bit for bit identical. Does Waveburner somehow distinguish between them and screw one of them up? Could this have anything to do with the alleged one sample offset in your HEDD?

I'm in shock and awed.

-R

[thethrillfactor]

Originally posted by RKrizman
Way better? I'm willing to bet they are bit for bit identical. Does Waveburner somehow distinguish between them and screw one of them up? Could this have anything to do with the alleged one sample offset in your HEDD?

I'm in shock and awed.

-R

Hey Rick,

This was done with someone elses mixes(not through my poor HEDD unit!!!)

I tried it on a couple of different songs and to my ears(and everyone else's) the difference was there.

It was actually in the lows, the unbounced files had more depth in the bottomn, while the bounced files were clearer up top, but lacked the depth.

Now this was through the old mixplus system.

I will hit some of the PC programs toward the end of the week(starting with Wavelab Pro). I am curious to see if there is a difference in the math(I know Lynn is doing a CD that is similar) but I want to hear it in a way that I would normally do something.

[malice]

Originally posted by thethrillfactor
Hey Rick,

This was done with someone elses mixes(not through my poor HEDD unit!!!)

I tried it on a couple of different songs and to my ears(and everyone else's) the difference was there.

It was actually in the lows, the unbounced files had more depth in the bottomn, while the bounced files were clearer up top, but lacked the depth.

Now this was through the old mixplus system.

I will hit some of the PC programs toward the end of the week(starting with Wavelab Pro). I am curious to see if there is a difference in the math(I know Lynn is doing a CD that is similar) but I want to hear it in a way that I would normally do something.

I'm glad I'm not the only one to think there is such a problem with bounced files in PT ...

I experienced similar problems

malice

[RKrizman]

Originally posted by thethrillfactor
H

It was actually in the lows, the unbounced files had more depth in the bottomn, while the bounced files were clearer up top, but lacked the depth.

Now this was through the old mixplus system.

I will hit some of the PC programs toward the end of the week(starting with Wavelab Pro). I am curious to see if there is a difference in the math(I know Lynn is doing a CD that is similar) but I want to hear it in a way that I would normally do something.

I'm beta testing that CD right now and I can promise you it will provide for some interesting and, for once, informed discussion.

About this whole BTD question, I'm gonna go check it out right now. I'll be back shortly.

-R

[e-cue]

Originally posted by RKrizman
I've designed my own custom dither by taking some nature recordings I made at King's Canyon at sunrise (soft air sounds, streams, the occasional bird) and eq'ed it to resemble the curve of the Cranesong dither. It's absolutely beautiful, and the occasional "bird spike" adds a subtle bit of interest. I'd share it with you guys, but I'm not willing to give up the unfair advantage it affords me.

Best of luck with your pale, 2-dimensional recordings.



-R

I never thought a "Sounds of Nature" cd could come in so handy. jk. Interesting.

I have a complation of tape machine hiss with different tape stock, different heads, clean & dirty heads, the Crane song joint, various "air leek" sounds, etc... In all honesty, I mostly hear a difference during fades, and even then, I usually default to POWr #3 and only change it if I notice stairstepping during fades. Sometimes it 'fun' to try using the crane song during the verses, and tape hiss during the hooks and other different combos. Whether or not you hear a difference is usually hit or miss depending on the material.

[RKrizman]

Originally posted by RKrizman
I'm beta testing that CD right now and I can promise you it will provide for some interesting and, for once, informed discussion.

About this whole BTD question, I'm gonna go check it out right now. I'll be back shortly.

-R

Okay Thrill, I'm back. Don't think I'm dogging your heels on this, but I just don't like to hear this stuff floating around unsubstantiated. It's already hard enough to work in digital without worrying about weapons of mass destruction hiding behind every bush.

I took Lynn's excellent tracks he provided for the AWSUM-DAWSUM project, which consists of 22 stereo tracks of a gospel choir and band, with horns, recorded 24/48. I did a BTD, also at 24/48, stereo interleaved, as well as an internal mix to 2 mono tracks, panned hard. I pulled the BTD back into Protools and compared. They nulled out completely. I then printed the null signal to it's own stereo tracks and upped the gain by 76 db, just to see what was lurking in there. Nothing. Absolutely nothing audible, with my LSR 28's cranked all the way. In my book that means they are the same.

There are all kinds of ways you can get different results. Of course, your time-based effects will be different with each pass. Also, it's possible to mistakenly feed the BTD with a different source that what you are monitoring the session with. Or maybe you're comparing a 16 bit BTD to a 24 bit session. Also, many times I have been thoughtless enough to drag a mix back into a session and ultimately discover that it sounds different because it's going through the compressor I have on the master fader for the second time. Doh! In short, it's a slippery slope, even for skilled practitioners.

The good news is that it only took me 5 minutes to check it out for myself in a very controlled fashion. If you still think there's a difference I implore you to demonstrate it with 2 short files. I'm totally willing to be proven wrong, but I have yet to hear any shred of evidence to suggest I am.

-R

[thethrillfactor]

Originally posted by RKrizman
Okay Thrill, I'm back. Don't think I'm dogging your heels on this, but I just don't like to hear this stuff floating around unsubstantiated. It's already hard enough to work in digital without worrying about weapons of mass destruction hiding behind every bush.

I took Lynn's excellent tracks he provided for the AWSUM-DAWSUM project, which consists of 22 stereo tracks of a gospel choir and band, with horns, recorded 24/48. I did a BTD, also at 24/48, stereo interleaved, as well as an internal mix to 2 mono tracks, panned hard. I pulled the BTD back into Protools and compared. They nulled out completely. I then printed the null signal to it's own stereo tracks and upped the gain by 76 db, just to see what was lurking in there. Nothing. Absolutely nothing audible, with my LSR 28's cranked all the way. In my book that means they are the same.

There are all kinds of ways you can get different results. Of course, your time-based effects will be different with each pass. Also, it's possible to mistakenly feed the BTD with a different source that what you are monitoring the session with. Or maybe you're comparing a 16 bit BTD to a 24 bit session. Also, many times I have been thoughtless enough to drag a mix back into a session and ultimately discover that it sounds different because it's going through the compressor I have on the master fader for the second time. Doh! In short, it's a slippery slope, even for skilled practitioners.

The good news is that it only took me 5 minutes to check it out for myself in a very controlled fashion. If you still think there's a difference I implore you to demonstrate it with 2 short files. I'm totally willing to be proven wrong, but I have yet to hear any shred of evidence to suggest I am.

-R

Hey Rick,

First of all when I read your post it made me laugh a little.

Yes i do check for those things(if not I probably won't get hired anymore).

Did you burn your examples to a CD and listened?

That was my point, after its bounced down, how does it sound compared to the two tracks burned as a split file.

Also Lynn's test doesn't take into account the math involved when doing fades or panning.

That's why i said I wanted to hear a "real world"test for myself in a way where I would probably use it.

[Ted Nightshade]

The dither on the HEDD 192 is so pleasing to me, everything is mixed down (from analog) straight to 16 bit. 24 bit does not sound nearly as nice to me (at least on the acoustic /percussion, etc. stuff we're up to).

My old time engineer says it sounds like germanium to him, and listening to his old 1/4 full track 1958 Sony reel to reel with germanium transistors, I will say the same. Sweet.

I wonder are we totally out to lunch? Dave Hill says in the very brief manual that a lot of prized old equipment was listened to...

At this point, I can't really imagine mixing to 24 bit and dithering later, it's so satisfying to mix to 16:44.1 and be looking at exactly what you'll end up with.

[malice]

Also Lynn's test doesn't take into account the math involved when doing fades or panning.

And how the host program takes care of Pluggins latency for instance ...

Is it perfectly accurate ?

For experiencing problems with that, I can imagine big phase issue that could act like a filter on some freqs, I don't know ...

All I know is there are some differences.

I guess I will too make a test involving some pluggins pan, eqs.

Some of them exists in every platform, would be interesting to try that ...


malice

[RKrizman]

Originally posted by thethrillfactor

Did you burn your examples to a CD and listened?

That was my point, after its bounced down, how does it sound compared to the two tracks burned as a split file.

Also Lynn's test doesn't take into account the math involved when doing fades or panning.

That's why i said I wanted to hear a "real world"test for myself in a way where I would probably use it.

I didn't burn samples to a CD because then I would be comparing 44/16 to 48/24, unless I misunderstand you. I gather you're talking about the difference between bouncing to disc versus mixing to two mono tracks inside the session itself. So I bounced to disc at 48/24 then dragged it back into the original session. Or are you comparing BTD as stereo-interleaved to BTD as split files?

Of course it's just as easy to do with real world files, with fades and pans.

I'd like to continue the discussion, but at this point I'm not sure what it is you are seeking to compare. What do you mean by "2 tracks burned as a split file"?

-R

[thethrillfactor]

Originally posted by RKrizman
I didn't burn samples to a CD because then I would be comparing 44/16 to 48/24, unless I misunderstand you. I gather you're talking about the difference between bouncing to disc versus mixing to two mono tracks inside the session itself. So I bounced to disc at 48/24 then dragged it back into the original session. Or are you comparing BTD as stereo-interleaved to BTD as split files?

Of course it's just as easy to do with real world files, with fades and pans.

I'd like to continue the discussion, but at this point I'm not sure what it is you are seeking to compare. What do you mean by "2 tracks burned as a split file"?

-R

My original post was about CD mastering/burning programs.

The idea was a CD burned from a bounced file from a PT session, compared to a CD burned from PT split files.

I was testing the different Mac CD programs to see which one changed the file the most(or the least). Technically they should all sound the same, but they don't.

I could hear though that a stereo interleaved bounce file, compared to a tracked split file came out different.

[RKrizman]

Originally posted by thethrillfactor
My original post was about CD mastering/burning programs.

The idea was a CD burned from a bounced file from a PT session, compared to a CD burned from PT split files.

I was testing the different Mac CD programs to see which one changed the file the most(or the least). Technically they should all sound the same, but they don't.

Sorry, I misunderstood. Did you have any problems with Jam? I've never noticed a problem, but next time I have a chance I'll check it out.

-R

[jho]

Originally posted by thethrillfactor
Hi Renie,

I own the HEDD unit with the dither on it.

A year ago I tested every dither I could find.

The (2) that I liked the best were the POW-R dither (3) and the Spark XL dither. The Crane Song was pretty good.


I am testing out now the different CD burning/mastering programs on the MAC. So far Waveburner Pro is winning. I am going to try all of the PC programs this week. I've heard that the math is better on a PC and will see if this is true.

One thing that has become apparent with PT on the mac is that a recorded split file(instead of a bounce) master sounds way better when burned in Waveburner Pro(more depth and bottomn).

Thrill,
A couple questions about above.
1. On the recorded split file, are you coming back into PT from the hedd and recording to L/R files in PT to bring into waveburner? As I have this setup I am curious. If it's not what you are doing then what is your preferred method these days?
2. Are you giving waveburner pro a 24 bit file and using it's dither?

Thanks

[Bob Olhsson]

Some years ago I noticed a difference between interleaved and non-interleaved files on CD-Rs made with MasterList CD. We ran tests comparing manufactured CDs and found that the CD-Rs from interleaved files were closer to the manufactured product.

[thethrillfactor]

Originally posted by jho
Thrill,
A couple questions about above.
1. On the recorded split file, are you coming back into PT from the hedd and recording to L/R files in PT to bring into waveburner? As I have this setup I am curious. If it's not what you are doing then what is your preferred method these days?
2. Are you giving waveburner pro a 24 bit file and using it's dither?

Thanks

Sometimes I do the first if the Masterlink is not around. I normally go through the HEDD when i am adding some kind of Comp/EQ to the 2 mix. If its an all digital affair(which is rare for me these days) than I won't even bother with the HEDD.

Yeah it is one of the prefered methods.

I actually prefer to go to the Masterlink so I can do the final 2 mix at a higher rate than my session(88.2 or 96K) especially when i am doing the analog stuff.


Waveburner Pro uses the POW-R dither. When burning CD's I go either #1 or #3.