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

[Phil O]

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.

Check out the link that I posted in my first reply.

Phil

[FMiguelez]

.

Duh! I can't believe I missed that

DP seems to score fairly high in that chart compared to the other apps, does it not?

I'll do tests using DP's automatic SRC at the highest possible quality, and hear what happens. I'll do careful a/b comparisons between the original and the SRCed.

[Phil O]

If you'll be using DP7.1, notice that it no longer has quality settings. It's always set at the highest quality. Please let us know what you come up with.

Phil

[newrigel]

.

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?

It all depends on the SRC algorithms... some are better written than others and cost more etc. because of the time and efforts involved.
The SRC will do the filtering and the truncation math for you to the specified target bit depth and sample rate. For the $ RX stomps ass! I can sample the noise before the audio on a file and train it and then batch process my files and it's incredibly artifact free... they compare it to NR costing $50K!!! The SRC is stellar too! I still for the life of me wonder why people sample so high? If your going to go HD I'd go with 88.2, or 176.4 etc. because the truncation is just a simple math function to the SF... but I tried the other day to hear an 18 khz tone and couldn't... most can't! I think what matters most is getting the dynamic range (through proper gain staging) to the converters (and the actual conversion chips and electronics themselves=$$$) and not slamming them to death.

[HCMarkus]

I think the word "trucation" is typically used when converting word-length, say from 24 bit to 16 bit. This is when dithering comes into play. 24 bit audio is preferred for tracking and mixing due to the available headroom and resultant reduced need to record at or very close to full-code.

When converting high sample rates to lower rates, the "extra" samples are esentially discarded, leaving the DA's filters to reconstruct an accurate analog output from sparser data. In digital audio's early days, these filters were a source of brittleness and harshness, hence the desire for higher sample rates which allowed a gentler filter to be used when reconstructing the analog output. With today's oversampling converters and very accurate brickwall filters most folks, if not all, would be hard-pressed to identify any sonic difference between rates of 192k and 44.1k.

If you convert each track independantly, I believe you'd be able to use almost any SRC without trouble, as errors introduced will be masked by other tracks. By using the recommended SRC's prior to mixing, including the Apple or current MOTU SRC, I believe you are assured a transparent and completely satisfactory result. In all honesty, you could probably do the conversion post-mixing (SRC the mix track) and have happy camper customers, but then you'd have to deal with mixing at 192!

Knock 'em dead with your mixes FM!

For additional reading, here's excerpted discussion from Wiki:

Example
CDs are sampled at 44.1 kHz, but a Digital Audio Tape, or DAT is usually sampled at 48 kHz. How can material be converted from one sample rate to the other?

First, note that 44.1 and 48 are in the ratio 147/160. If the original audio signal had been recorded at 7.056 MHz sampling rate, the process would be simple. Since 7.056 MHz is 160 x 44.1 kHz, and also 147 x 48 kHz, all we would need to do is take every 160th sample to get a 44.1 kHz sampling rate, and every 147th sample to get a 48 kHz sampling rate. Taking every Nth sample like this preserves the content provided the information (the audio signal) does not have any content above half the lowest sampling rate used (22.05 kHz) in this case.

So now the problem is how to generate the 7.056 MHz sampled signal, given that the original has only 1/160 of the samples needed. The somewhat surprising answer is to replace the missing samples with zeros. So if the original audio samples were ..,a,b,c,.., then the 7.056 MHz sequence is ..,a,0,0,0,...0,0,b,0,0...0,0,c,.., with 159 zeros between each original sample. This too will create extra high frequency content (in fact it is worse in this respect than linear interpolation) but at least the frequency response is flat. Then the digital filter removes the unwanted high frequency content. The work of this digital filter is also much easier if zeros are inserted, since the filter is basically an average and almost all of the samples are known to be zero.

So inserting the zeros, then running the digital filter (almost always an FIR filter since these can be designed to have no phase distortion) gives the needed signal - sampled at 7.056 MHz, but with no content above 24 kHz. Then just taking every 147th sample gives the desired output. Which sample to start with does not matter - any set will work as long as they are 147 samples apart.

[FMiguelez]

.

Hey, guys.
Thank you for the responses and info.

I tried doing the SRC in the iMac in the studio. I don't remember what DP version it has... I think it's DP7.01, and the conversions sounded just fine.
No artifacts I could hear. I closed my eyes A/B comparing, and frankly, I couldn't tell which was which at normal listening levels.

I tend to be obsessive, and before I start loosing sleep over this MAKING and forcing myself to hear a difference, I'll just trust what I casually heard and my first impression: a great DP SRC for individual tracks.

So for my purpose, DP7 did a great job.
There could be better 3rd party tools out there for this, but how much better will they make it sound? Is it really worth it? For this particular situation, I don't think it's worth it, and I want to stay away from diminishing returns.
Maybe if this were a huge project where I would get paid lots and lots and lots of money...
Oh, shut up, Fernando. Don't let that little nagging voice convince you you SHOULD have heard a big difference. Shut up!

I'll tell you what:
If she likes my trial mix and hires me, it will be because I did a great mixing job... but if she doesn't like it, then I will blame DP's SRC

Just kidding!

[newrigel]

DP will do just fine! I have to use RX to clean the room out of my vox tracks... and it works great for getting rid of that ambient "air" that's in there.