Best way to down-sample individual tracks in DP...

[FMiguelez]

.

Hi, friends.
Someone wants me to do some mixes. Her recording engineer captured everything at 24/192... even the maracas...
There is no way in hell my current computer can handle that....
I will be doing this mixes in DP 4.61, with my current system as is.

What is the best way to make the individual tracks be something more decent, such as 24/44.1?
If I remember correctly, if I setup the DP session at 24/44.1, it will automatically downsample the tracks when I import them in the session, correct?

Will the quality of the tracks suffer? I'm particularly concerned about CUMULATIVE artifacts...

Will DP do a decent job here?
Should I rather use a 3rd party app for this purpose?

Any other things I should mind about this?

Thanks!

[Phil O]

If I remember correctly, if I setup the DP session at 24/44.1, it will automatically downsample the tracks when I import them in the session, correct?

Yes, if you set it up to do so under preferences.

Will the quality of the tracks suffer? I'm particularly concerned about CUMULATIVE artifacts...

I don't know. This from what's new in 7.1:

"IMPROVED SAMPLE RATE CONVERSION
Digital Performer now uses a new sample rate conversion
algorithm which provides increased precision, efficiency,
and transparency, with a 138 dB signal-to-noise ratio.
The Quality options found in the Audio File Conversion
dialog (Figure 40) in previous versions of Digital Performer
are no longer needed — the same efficient, high quality
conversion is always used."

Also if you look here:
http://src.infinitewave.ca/

Apple Core Audio (Leopard) looks pretty damned good, but I don't know what apps use it.

Phil

[FMiguelez]

.

Hmmmm...
Interesting what you cited about DP7.
Actually, I could use DP7 for the down-sampling operation only. In my MacBook Pro.
Perhaps I could do it there and see what happens.

I have a very old version of Peak, but I think the algorithm is too old. Perhaps if I update it...

Where is this? I don't think I've ever seen it:

Apple Core Audio (Leopard) looks pretty damned good, but I don't know what apps use it.

[newrigel]

Just import and move on... 192 is a sales pitch. Anyone recording @ that rate is fooling themselves because I'll post up an 18k tone and see if anyone can hear it! Were not dogs and if 16/44.1 is where you end up what's the difference... especially today with the MP3 as a delivery medium.
24/48k is plenty.

[FMiguelez]

Just import and move on... 192 is a sales pitch. Anyone recording @ that rate is fooling themselves because I'll post up an 18k tone and see if anyone can hear it! Were not dogs and if 16/44.1 is where you end up what's the difference... especially today with the MP3 as a delivery medium.
24/48k is plenty.

Totally agreed.
You should listen to how good the maracas sound at 192KHz, though (just kidding).

I just want to make sure I create the least possible amount of artifacts when I down-sample these tracks.

I can't believe people buy into this marketing 192 hype...

Imagine... if I do a good job and they like my mixes, it will be too hard for me to resist not telling them I down-sampled all their 192KHz tracks to 44.1. And that they didn't catch it... and worse, that they actually liked it

[Splinter]

FMiguelez:

If you have Apple's Quicktime or Compressor, use that to downsample the files. It's very fast and probably the purest, distortion-free SRC algorithm. This may be what DP is using now as it is a part of CoreAudio.

By the way, how many tracks are you mixing? I've had big 40 track 24/96 projects on my Dual 1.8 G5 with DP4.61 without problems. It did get a little sluggish toward the end of mixes, but it did the job valiantly!

[Phil O]

If you have Apple's Quicktime or Compressor, use that to downsample the files. It's very fast and probably the purest, distortion-free SRC algorithm. This may be what DP is using now as it is a part of CoreAudio.

I was wondering about that. I can't imagine Quicktime would not use the Core Audio algorithm, but does anybody know for sure what DP is using?

Phil

[FMiguelez]

FMiguelez:

If you have Apple's Quicktime or Compressor, use that to downsample the files. It's very fast and probably the purest, distortion-free SRC algorithm. This may be what DP is using now as it is a part of CoreAudio.

Hi, Splinter.
Do you mean Quicktime Pro?
If so, yes, I have it. So how would I go about this? Opening in QT and then exporting the 192 files to 44.1?

Just like that?
I've seen those options like a million times, but never thought I could get great results with QT!

By the way, how many tracks are you mixing? I've had big 40 track 24/96 projects on my Dual 1.8 G5 with DP4.61 without problems. It did get a little sluggish toward the end of mixes, but it did the job valiantly!

There are a lot of songs to be mixed. Some will be 4-5 tracks only, but some others will be around 30-40 tracks.
To be completely honest, I don't know if what bothers me is my machine having trouble handling the tracks at such rates, or simply the CONCEPT of the 192...

What I wrote in the last paragraph in my post above would give me ENORMOUS pleasure
Maybe a little bit irrational on my part, but I can't help it. I mean, this singer ain't in her 20's at all, you know? An the recording engineer, while very respected, I'd bet almost anything he records like that due to marketing hype and NOT because it makes a difference... I don't know why I feel like this, but can't help it...

Thanks for the recommendations, Splinter!

[Phil O]

I think this is an important question to get answered, FM. It's not very often that I need SRC, but when I do I want to know it's the best available to me. I guess if QuickTime and DP both use Core Audio, then the results would be the same, but if they don't, the questions becomes which one is better? It would be nice to have Magic Dave chime in on this. How do we summon him? We need a Magic Dave light like the one used in Batman. Yeah, that's the ticket.

BTW, I hope your project goes well. I've had both good and bad experiences mixing other people's stuff. I guess when I start a project, I'm thinking about the mix as I track. How I record something is as important to the mix as how I mix it. When another engineer just hangs some mics and records with the "We'll fix it in the mix" mentality, things don't always go well. On the other hand, there's this one studio (I won't name them) that does a great job tracking, but the engineer there can't mix to save his life. Several people have brought me his stuff to see if I can "fix" it. Boy, I wish I had his studio. The sound there is awesome. Too bad he doesn't know what to do with it! But I digress. Good luck with this and if you get any more info on SRC, please report back.

Phil

[FMiguelez]

.

Hey, Phil:

This would be my first "real" mixing-for-a third-party gig. I say real because I've done it before, but for friends who need "rescuing"... for people who spent 3 bucks on recording, got terribly sounding tracks, and wonder why their music doesn't sound like a Hit



But this particular recording engineer is very well respected, and I'm sure he did a great job recording.
To tell you the truth, it is I who is feeling a bit insecure about this gig... I am a composer, and not a mix engineer. I've learnt a lot, and I think my mixes keep sounding better every year, but I would never call myself a mix engineer... But she likes my stuff, and she wants to give me a shot with this gig.

I proposed to her I can do one trial mix for her, and if she likes it then I will mix the whole project. This would empower me to set my terms, and most importantly, she would get MY sound out of me. She would need to trust my taste, and give me an "open card" to my interpretation of things (based on hers, of course).

I would like to have a definitive answer too about what is the best tool out there to do sample rate conversions. The one that introduces the LEAST amount possible of digital artifacts.

So, I will rephrase my question:
Those of you who mess with sample rate conversions, what is the BEST app out there for such purposes? The most transparent one?

Also, does DP use the CoreAudio algorithm, as suggested up-thread?

Also, I suppose there is a difference between downsampling individual tracks, and doing so for a full mix track.

I heard Peak Pro 6 had one of the best algorithms on the market, but I found people either, LOVE Peak or HATE Peak.

Please give us your recommendations!

Thanks!

[magicd]

The new SRC in DP7 is a proprietary MOTU process. From the MOTU web site:

"A new ultra-fast, CPU-efficient sample rate conversion algorithm delivers increased precision, efficiency, and transparency, with an astonishing 138 dB signal-to-noise ratio."

Dave

[Phil O]

The new SRC in DP7 is a proprietary MOTU process. From the MOTU web site:

"A new ultra-fast, CPU-efficient sample rate conversion algorithm delivers increased precision, efficiency, and transparency, with an astonishing 138 dB signal-to-noise ratio."

Dave

Woah! The Batlight worked! Thanks for that response Dave. But with all due respect, I have to ask this question: But is it better than Core Audio? (If I'm putting you on the spot, you can just like wink or somethin')

Phil

[FMiguelez]

.

There you have it, ladies and gentlemen.
No more speculation regarding DP's SRC.

Thanks for chiming in, Dave

I'm curious about what Phil just asked too.

Wait! Now that I remember, didn't Shooshie post a website link where they compare different DAW's SRC algorithms?
If memory serves me well, DP scored fairly high in there...

I'll see if I can dig it up and find it. The thing is that it wouldn't include DP7, but only DP5 or DP6.

[newrigel]

If you have Apple's Quicktime or Compressor, use that to downsample the files. It's very fast and probably the purest, distortion-free SRC algorithm. This may be what DP is using now as it is a part of CoreAudio.

I was wondering about that. I can't imagine Quicktime would not use the Core Audio algorithm, but does anybody know for sure what DP is using?

Phil

Every application goes through CA to talk to the world... what the app uses is a different story. Izotope has good SRC so I do my batch converts with RX because the NR is awesome and works fantastic! I get rid of that "air" broadband distortion with it and it works fantastic for that. I don't have that luxury as some to record in an anechoic chamber so it works great for getting rid of the colorization to that "dead quiet" state.

[FMiguelez]

.

Hey, Newrigel.

I'd like to understand this a bit better...
If I do a SRC from 192 to 44.1, what happens to all that "extra" information of the original file? Does it get truncated?

Would it be a good thing to put a LPF around 22 KHz just before I do the SRC? Or that is exactly what SRC does anyway?