Home Theater Forum and Systems banner
1 - 11 of 11 Posts

·
Registered
Joined
·
191 Posts
Discussion Starter · #1 ·
Once again it's in the thread's title!

I mix in analog, and I like to record my mixes at higher sampling frequencies (usually 96kHz), as they are usually mastered in the analog domain.

Yes, I know the best would probably be to mixdown to an analog deck, but of course it's another real mess...:innocent:

How do you like to do?
 

·
Registered
Joined
·
103 Posts
If I had the opportunity for super analoge mixing, I would like to record the stereo bus output at 192KHz/32bits :D
There's nothing to be gained from 192:
http://www.lavryengineering.com/documents/Sampling_Theory.pdf

At 88.2/96 you are quite high enough to deal with any aliasing problems created by plugins which don't oversample.

I tend to prepare my mixes for either 44.1/32 (for CD audio) or 48/32 (for video). I leave the dither to the mastering engineer. Since I record at 44.1 (or 48 for video projects) there are never any SRCs in my process. I'm not sure how crucial that is anymore, probably a bias I still hold over from when SRCs were kind of iffy. I'm told most SRC processes nowadays are not really too audible. My workflow works well enough that I don't really feel the need to reinvent that wheel.
 

·
Registered
Joined
·
79 Posts
There's nothing to be gained from 192:
http://www.lavryengineering.com/documents/Sampling_Theory.pdf

At 88.2/96 you are quite high enough to deal with any aliasing problems created by plugins which don't oversample.

I tend to prepare my mixes for either 44.1/32 (for CD audio) or 48/32 (for video). I leave the dither to the mastering engineer. Since I record at 44.1 (or 48 for video projects) there are never any SRCs in my process. I'm not sure how crucial that is anymore, probably a bias I still hold over from when SRCs were kind of iffy. I'm told most SRC processes nowadays are not really too audible. My workflow works well enough that I don't really feel the need to reinvent that wheel.
Nice. Never quite thought of doing that way. But the mastering house is probably going to do a better job of converting/dithering than the studio.
 

·
Registered
Joined
·
193 Posts
I don't mix (export) to higher rates, unless I have a project that was originally tracked at 44.1 K and I later find I need a video mix. Aside from matching a neccessary format in another domain, it seems pointless to me. If I record at a particular rate and bit depth, I keep it there. With today's oversampling DAWs, the processing is quite robust at 24 / 32 (floating).... there are people who claim they can hear the difference. I'm not them so I can't argue what somebody else hears. But I'm not really ashamed to say, I'm not one of them.

Though I've struggled to hear the difference of higher sampling rates, I must admit I cannot. but I can, and HAVE heard the difference between 16 bits and 24 bits. (24 bit fixed point, that is, which is practically the same as 32 bit floating point.) I have not had the opportunity to listen to a 32 bit fixed point (48 bit floating point) system, but I'm willing to bet it would not be noticeable, either, which is likely why we don't see such converters on the market.

Those who claim they hear the difference in the 192KHz, say they only hear it in the realm of "directionality" and "openness". From a logical standpoint, this would make sense, because this is, in fact how the music would sound in the improvment category IF we could ALL hear it.

Perhaps if I created an album for my cat, he would appreciate my willingness to kick up the high end fidelity a little. But I don't think he'd be willing to drop me a tip for the extra headroom involved....

That's actually the other reason I don't use higher bandwidth... I back up every project to DVD disks in the form of raw data (so I can keep my hard drives relatively clean). All such data takes up space. Some of my projects require 5 DVDs of raw data. More data requires more time for backup, more hard drive wear, costs me channel processing bandwidth, and ultimately, more money. Because time is money, when analyzing all the extra expense over the long haul, .... well, it just doesn't add up if I can't hear it. ... So headroom bulk is just that. Extra, unneeded weight (in my PERSONAL opinion).
 

·
Registered
Joined
·
291 Posts
I really can't fathom why anyone would MIX DOWN to a higher sampling rate. You start a project and stop it with the same sample/bit depth settings unless as ej pointed out you find out it's going on a DVD or something as well, even then you can still keep it at 44.1/16k.

The honest truth is that nobody has the same ears, everyone hears subjectively and it frankly doesn't matter if you mix down to 44.1 or 192, the average music consumer isn't going to notice a difference.

The only reason anyone needs to be recording to or upsampling to 96/192 is if there's seriously delicate, dynamic music being performed, whether it be a whole symphony or a small quartet of strings. IMO 88.2 is great for everything else.

On the flipside, I'm the opposite of ej when it comes to drive space and whatnot. hard drives are so inexpensive nowadays there's no reason ANYONE shouldn't have less than 500gbs. I mean, when you can get 1TB hard drives for 99 dollars now, what's the point in going less? a few TB drives in RAID 0 or 0+1 and you're good for years. Even just one TB will last you years unless you've got stuff going on every single day of the year.

I'm all about quality over loudness (as I'm hoping all of us here are :p) higher sampling rates capture MORE action, which means you get the most out of the dynamics handed to you which is highly important. The downside? Sometimes it's just a wee bit too sensitive and captures certain sounds you may not want, extra clicks from musicians hitting things and whatnot.

Getting back to the original question, no way. It makes no sense. At all.

I work with what I'm given when all is said and done though. Everything comes down to a preference and if I had a building and a board
I would probably be recording everything at 88.2 as opposed to 44.1.
 

·
Registered
Joined
·
103 Posts
I really can't fathom why anyone would MIX DOWN to a higher sampling rate. You start a project and stop it with the same sample/bit depth settings unless as ej pointed out you find out it's going on a DVD or something as well, even then you can still keep it at 44.1/16k.

The honest truth is that nobody has the same ears, everyone hears subjectively and it frankly doesn't matter if you mix down to 44.1 or 192, the average music consumer isn't going to notice a difference.

The only reason anyone needs to be recording to or upsampling to 96/192 is if there's seriously delicate, dynamic music being performed, whether it be a whole symphony or a small quartet of strings. IMO 88.2 is great for everything else.

On the flipside, I'm the opposite of ej when it comes to drive space and whatnot. hard drives are so inexpensive nowadays there's no reason ANYONE shouldn't have less than 500gbs. I mean, when you can get 1TB hard drives for 99 dollars now, what's the point in going less? a few TB drives in RAID 0 or 0+1 and you're good for years. Even just one TB will last you years unless you've got stuff going on every single day of the year.

I'm all about quality over loudness (as I'm hoping all of us here are :p) higher sampling rates capture MORE action, which means you get the most out of the dynamics handed to you which is highly important. The downside? Sometimes it's just a wee bit too sensitive and captures certain sounds you may not want, extra clicks from musicians hitting things and whatnot.

Getting back to the original question, no way. It makes no sense. At all.

I work with what I'm given when all is said and done though. Everything comes down to a preference and if I had a building and a board
I would probably be recording everything at 88.2 as opposed to 44.1.
One reason you would mix at a higher sampling rate than you recorded is to control aliasing from digital plugins. It's not crazy at all. Whether or not those aliasing errors are audible enough to be 'necessary' is certainly debatable but the idea that higher sampling rates mitigate aliasing errors is a simple fact.
 

·
Registered
Joined
·
191 Posts
Discussion Starter · #10 ·
I really can't fathom why anyone would MIX DOWN to a higher sampling rate. You start a project and stop it with the same sample/bit depth settings unless as ej pointed out you find out it's going on a DVD or something as well, even then you can still keep it at 44.1/16k.
Of course my question was linked to the fact that I do mix in the analog domain, so I can chose whatever SR I want for the mixdown A/D conversion.
 

·
Registered
Joined
·
291 Posts
Of course my question was linked to the fact that I do mix in the analog domain, so I can chose whatever SR I want for the mixdown A/D conversion.
Gotcha. I think I went on a little rant though lol, even though I meant everything I said. I don't concern myself with aliasing or anything because I believe that pretty much all of nobody except extreme audiophiles with magical ears and 80k sound systems would be the only ones to really care/whine about anything like that.
 
1 - 11 of 11 Posts
Top