Taking A Poll - 44.1kHz or 48kHz sessions?

[Splinter]

Not to get off topic here, but this thread does lend itself to the discussion of converters. Once again, I had a staunchy view against high end gear. I thought all this hi res big dollar converter biz was hocus pocus. I was happy with my 2408 (it was even a, wow, mkIII) at 44.1 and no one would convince me otherwise... till we happened to get a MOTU 192. We needed some more Analog IO and MOTU 24 IO was overkill. I hooked it up and set it as my default output. I was astonished. I had been listening to some music prior to setting things up. When I turned that same CD quality music back on, it was night and day. I couldn't believe it - and this was still at 16 bit. I started to AB with the 2408. When I listened through the 192, it was like someone had pulled a veil off the speakers. It was now deep, wide, open, and had much better bass response. The 2408 sounded like plastic in comparison - cheap, flat, mushy and dull. You can think something sounds fine till you get the opportunity to compare it with something better and then you get a serious wake up call. Then you factor in a high res sample rate and you start hear, well, nothing. Things just become amazingly transparent and I know converters get better than a MOTU 192. So, other than the fact that high res sample rates improve the sound, quality conversion makes a HUGE contribution at ANY sample rate.

My most recent purchase for my home workstation was an Apogee Duet. I'm now fully convinced that boutique converters make a huge difference. I was working on a piano project where we had recorded the same piano in the same room with the same mics a year ago through a Digi 002 rack and now wanted to add a few more musical segments to the project. But, this time I used my Duet. I my gosh. The piano was huge. Huge! The low end was gigantic. The high end was smooth without getting peaky. Tons of depth and detail. Compared to the 002, it sounded golden. The 002 sounded a bit like the 2408 (only worse) - plastic-like, cheap, flat, peaky, harsh... yuck. And both were at 96k. I'm not saying this to rag on the cheaper converters as much as to share my personal testimony of how important good conversion is regardless of sample rate.

[Armageddon]

Well, as far as I'm concerned, Apogee is about as good as it gets, and I was mightily tempted by the Duet when I bought my last interface. I wound up picking 192 kHz, more inputs and powered effects over quality, so I wound up with a TC Konnekt 24D, which isn't bad, but it ain't no Apogee. In the most ironic twist, it's not really feasible to record and operate at 192 kHz on my machine, so I record and mix at 96 kHz, and while the two powered effects seem okay, only one or the other is able to be used at 96 kHz and neither one of them can be used at sample rates beyond that, sooooooo ... I should have bought a damn Duet. Hurts to admit, but it's true.

[bootsncats]

I consider sample rate as potential. When mixing for film I do 96K because of the DTS encoding capabilities of surround sound these days. 44.1 for CD and 48 for anything else. Its all in the converters. I have heard a 48K awesome converter sound better than a 192K crap machine. I like apogee, in the high end. RME has some sweet stuff as well. I had to record on an 002 the other week... I'm still recovering.

[kassonica]

Yeech 02 sounds very BAD to my ears

I feel for ya

[Dave Connor]

MOTU switched to AKM converters in their higher end pieces several years ago ( don't know which ones exactly but the 192 for sure) but never in any version of the 2408. AKM are very good converters. I actually come out of my 308's SPDIF into an Apogee PSX100 SE for DA conversion. I also dither my bit rate from 24 to 16 bit with the Apogee which is perhaps it's most vaunted feature. I have three 2408's for ADAT inputs.

[Splinter]

Well, Apogee's are probably a 7 on a 10 point scale. If you want to look at 10s, try Prisms or Weiss. The next step down would probably be Lavrys or Myteks, but all of these are quite a bit better than Apogees. Nonetheless, I'm making great sounding music on the Duet and I'm quite happy.

Recording at 192kHz is complete overkill. It's like recording 96kHz of nothing... at least nothing that will every be reproduced in the audible or perceived range or by any set of speakers you or I own. Not to mention it require 4 times the amount of storage space and 4 time the amount of processing power. I've compared 192 to 96 and heard no perceivable difference at all. As I said before, the quality of the converter will make a far greater difference than the sample rate.

[carrythebanner]

I was about to type up a response when I remembered a thread from yester-year which said everything I wanted to:
http://www.motunation.com/forum/viewtop ... 054#p57054

[Armageddon]

Well, Apogee's are probably a 7 on a 10 point scale. If you want to look at 10s, try Prisms or Weiss. The next step down would probably be Lavrys or Myteks, but all of these are quite a bit better than Apogees. Nonetheless, I'm making great sounding music on the Duet and I'm quite happy.

Recording at 192kHz is complete overkill. It's like recording 96kHz of nothing... at least nothing that will every be reproduced in the audible or perceived range or by any set of speakers you or I own. Not to mention it require 4 times the amount of storage space and 4 time the amount of processing power. I've compared 192 to 96 and heard no perceivable difference at all. As I said before, the quality of the converter will make a far greater difference than the sample rate.

I was actually comparing interfaces in the same price range, and for what I paid for the Konnekt 24D with all the bells and whistles, I could have bought a Duet ... not quite a top-end converter in either instance, but on a sliding scale, a Duet at 96 kHz beats out a Konnekt 24D at 192 kHz any day. Neither are going to stack up to a top-end converter.

I'm not quite sure 192 kHz is useless -- there's a camp that would argue 96 kHz is useless and that you're not gaining anything by recording over 48 kHz sample rates -- but the quality of the converter is a major factor, regardless of what you're recording at.

[Splinter]

I'm not quite sure 192 kHz is useless -- there's a camp that would argue 96 kHz is useless and that you're not gaining anything by recording over 48 kHz sample rates -- but the quality of the converter is a major factor, regardless of what you're recording at.

Well, I use to think that about 96K, too, but have heard the improvement. Going above that I hear no improvement at all. Besides, the whole antialiasing filter issue is legit at 44.1 and 48k, but when you double those frequencies it no longer is a problem in marring the high end of a 20Hz to 20kHz range.

[HLStudios]

48k here!!

[Armageddon]

Well, I use to think that about 96K, too, but have heard the improvement. Going above that I hear no improvement at all. Besides, the whole antialiasing filter issue is legit at 44.1 and 48k, but when you double those frequencies it no longer is a problem in marring the high end of a 20Hz to 20kHz range.

It does raise an interesting question: who is actually recording and mixing at sample rates above 96 kHz right now?

On a sliding scale, 1-bit DSD comes closest to the fidelity of analog tape (and I feel pretty comfortable in assuming that no one is recording and mixing in this format, though some may be mastering in this format), then, a fair way down is 192 kHz and so on. I remember having 96 kHz capability was a big deal only a few years ago (of course, I also remember being excited when 24-bit capability was introduced to most DAWS towards the end of the Nineties), so as computers get faster and hard drives get bigger, will we start poking around at 192 kHz? Are people with ProTools HD setups doing it already?

[bayswater]

On a sliding scale, 1-bit DSD comes closest to the fidelity of analog tape

What measures of fidelity would that entail? Why would you use analog tape as the benchmark? Did I make a huge mistake when I stopped using tape nearly 20 years ago?

[Mr_Clifford]

On a sliding scale, 1-bit DSD comes closest to the fidelity of analog tape

What measures of fidelity would that entail? Why would you use analog tape as the benchmark? Did I make a huge mistake when I stopped using tape nearly 20 years ago?

Exactly. Let's not forget why they spent so much time and money developing digital in the first place.

Noise floor, wow, flutter, generational degredation, bias, tape stretch, bleed, print-through, tapehead wear, physical track limits (and we think Pro Tools LE is bad!)........

[bayswater]

On a sliding scale, 1-bit DSD comes closest to the fidelity of analog tape

What measures of fidelity would that entail? Why would you use analog tape as the benchmark? Did I make a huge mistake when I stopped using tape nearly 20 years ago?

Exactly. Let's not forget why they spent so much time and money developing digital in the first place.

Noise floor, wow, flutter, generational degredation, bias, tape stretch, bleed, print-through, tapehead wear, physical track limits (and we think Pro Tools LE is bad!)........

... and then we spend a whole lot of money on plugs and boxes to simulate all that non linearity.

[bongo_x]

On a sliding scale, 1-bit DSD comes closest to the fidelity of analog tape

What measures of fidelity would that entail? Why would you use analog tape as the benchmark? Did I make a huge mistake when I stopped using tape nearly 20 years ago?

Yes.