Interesting topic: A/D converters and what makes them special?

  • Thread starter Thread starter chessrock
  • Start date Start date
C

chessrock

Banned
After doing a bit of research and reading some threads here, as well as at RAP and other boards, I've been noticing some factors regarding a/d converter quality that I find rather interesting. So I thought I might bring it up as a possible topic that some of you might have some comment to share . . .


As a lot of us know, conversion quality appears to be the hot area these days where we keep seeing all of these improvements. Or at least what we percieve to be improvements -- obviously sampling rates seem to be going up exponentially, and 24-bit should be almost universal before long. Convesion in a typical portable hard-disk recorder has gotten better, as have typical consumer sound cards. Interestingly enough, guys like Lynx have even managed to pack several professional-quality converters in to a PCI-based sound card . . . something that many people doubted was even possible, so it almost seems like they're defying physics. :D

Anyway, it seems like while everyone keeps trying to out-do the other guy in terms of sampling rates. And all of the focus seems to be on how the converter goes about converting sound in to these 1's and 0's and so on . . .

Meanwhile, it seems like the converters that really got it going on are the ones with the best analog sections, and the most stable word clocks. And this trend seems to be supported wherever it is that I might get in to this kind of discussion. It just seems like the converters that most people seem to like are the ones where those two issues have been addressed the most. And a lot of the designers and/or techie-types I talk to keep supporting that idea.

Anyway, for anyone out there who either knows anything about this subject or even cares . . . what do you think of all this? What's the biggest factor in the a/d conversion puzzle? Would you be happier with a 16, 18 or 20-bit, 44.1 converter that had a solid, audiophilish analog portion and a stable word clock? Or are the advances in the 1's and 0's, the sampling rate and higher bit-depths really making a major audible difference here, and worth pursuing further and paying more for?
 
chessrock said:

Anyway, for anyone out there who either knows anything about this subject or even cares . . . what do you think of all this? What's the biggest factor in the a/d conversion puzzle? Would you be happier with a 16, 18 or 20-bit, 44.1 converter that had a solid, audiophilish analog portion and a stable word clock? Or are the advances in the 1's and 0's, the sampling rate and higher bit-depths really making a major audible difference here, and worth pursuing further and paying more for?
Click to expand...

probably to take a lot of heat (or maybe not) I think it is a moot subject for 99.9% of the people here. I've commented about mic choices, preamps and the such. I think the differences you are talking about are even smaller. Sure... the notion (around here, that's for sure...and which I happen to believe also) is that a killer 20 bit will outperform a so so 24 bit all night.
However, there might be a couple, three people that can actually take advantage of that, that come around here, in my lame ass opinion.

And...I really am leary of peoples listening environments of whom claim to hear MAJOR differences. I am not going to dismiss the differences. I would only show my ignorance and say there are none. My point is the amount of difference.

And...how are you going to dispute the shitty little home reccor, who can't even play a barre chord and capture it with a s/n of more than 50 dB who says he HEARS a major difference and swears by 320kbps MP3 files still being crap and all the lot? He's right...but so am I right when I say "that mic sounds fuller"...only because someone stepped up to the mic instead of standing 3 feet away. Was it the mic?...was it the proximity effect? OR was it just standing closer to the mic?...or did the asswipe not notice the mic signal was now 3 dB hotter?... which one... ?? Oh...it was the mic... the V#6 mic is fuller sounding... Because I KNOW!!

Next question...you may say... Where's your friggin proof, Mixmkr?? My proof is the mp3 clinic and the variety of posts that are there. I get a chuckle at all the people that just say "I get great results with this and that....." (especially in the famed Mic forum)... go listen to their product.. case closed.

sorry...I probably got waaaay off topic... but what the hell, I knew you of all people would understand Chess... ..or maybe not...I guess we'll see ;)
 
You're really 100% right on this one, Mix.


I guess I'm kind of addressing this question more from a technical perspective to the guys who aren't necessarily hanging around the mp3 clinic. :D
 
admittedly there was a time in my life where I'd get wet spots talking about this stuff, and I certainly understand it about other people enjoying the subject too. Hell... recording equipment is A LOT of fun now that I can actually afford to buy it now! And it looks cool. Better than 15,000 dollar kitchen cabinets if you ask me.

Shit...I'm about to blow some cash on power conditioners... how sexy is that?;)

(ask me about my banjo!!)
 
there ya go...post some proof that shows you how good 8 bit converters sound!!!:D
 
i agree that a superior analog section and stable word clock make waaaaaayyyyy more of a difference then sample rate or bit rate for that matter...i was floored how less crowded my mixes sounded when i stacked audio track upon audio track...24/48 on a lynxtwo smokes 24/96 on a delta any day of the week..u hear music on a delta...u feel music with a lynx...i just need some more pre amp flavors :mad:
 
Teacher said:
i agree that a superior analog section and stable word clock make waaaaaayyyyy more of a difference then sample rate or bit rate for that matter...i was floored how less crowded my mixes sounded when i stacked audio track upon audio track...24/48 on a lynxtwo smokes 24/96 on a delta any day of the week..u hear music on a delta...u feel music with a lynx...i just need some more pre amp flavors :mad:
Click to expand...

That's one of the best examples, actually, that I can think of . . . especially when you consider that the Lynx cards and the M-audio cards use the exact same AKM converters, if I'm not mistaken. Difference seems to be in the analog section and in the word clock stability.
 
chessrock said:
After doing a bit of research and reading some threads here, as well as at RAP and other boards, I've been noticing some factors regarding a/d converter quality that I find rather interesting. So I thought I might bring it up as a possible topic that some of you might have some comment to share . . .


As a lot of us know, conversion quality appears to be the hot area these days where we keep seeing all of these improvements. Or at least what we percieve to be improvements -- obviously sampling rates seem to be going up exponentially, and 24-bit should be almost universal before long. Convesion in a typical portable hard-disk recorder has gotten better, as have typical consumer sound cards. Interestingly enough, guys like Lynx have even managed to pack several professional-quality converters in to a PCI-based sound card . . . something that many people doubted was even possible, so it almost seems like they're defying physics. :D

Anyway, it seems like while everyone keeps trying to out-do the other guy in terms of sampling rates. And all of the focus seems to be on how the converter goes about converting sound in to these 1's and 0's and so on . . .

Meanwhile, it seems like the converters that really got it going on are the ones with the best analog sections, and the most stable word clocks. And this trend seems to be supported wherever it is that I might get in to this kind of discussion. It just seems like the converters that most people seem to like are the ones where those two issues have been addressed the most. And a lot of the designers and/or techie-types I talk to keep supporting that idea.

Anyway, for anyone out there who either knows anything about this subject or even cares . . . what do you think of all this? What's the biggest factor in the a/d conversion puzzle? Would you be happier with a 16, 18 or 20-bit, 44.1 converter that had a solid, audiophilish analog portion and a stable word clock? Or are the advances in the 1's and 0's, the sampling rate and higher bit-depths really making a major audible difference here, and worth pursuing further and paying more for?
Click to expand...

Well, 24 bit is inherintley better than 16 bit as there is more dynamic range available. 44.1Khz has a theoretical top end of 22.05Khz. 96Khz has a 44.1Khz theoretical top end. So, 24/96 probably will be the standard for a while. There really is no benefit to any higher sampling freq because the human hearing range is approx. 20-25Khz. Why bother with anything higher?
24 bits VS 16 bits.
If you record at 16/44.1 the quietest sine wave that you can, the A/D would put out 1 LSB of information. This, obviously would translate into a square wave! The worst possible distortion we want. 24 bits, same test, would look significantly more like a sine wave. The distortion "floor" would be waaaay down there.

Music and Computers.
Like computers, people like the newest and fastest. For many years, comp manufacturers sold "faster processor". The problem for years was that it was the BUS SPEED that really made a big difference in the overall speed of the comp. Now, the A/D, D/A manufacturers are betting that you will buy faster conversion speeds and MORE BITS! Looks good, but anyone "in-the-know" will tell you that it is all for naught. It is a sales gimmick. 24/96 is the top end for any real benefit to the recording world. CD and 44.1Kz is the REAL problem. You must dither the bits to make the conversion. Dithering is "adding noise" on purpose to avoid the low volume "square wave" distortion. This is why CDs are always compressed. To avoid all the distortion, the CD mastering labs make sure that the dynamic range is limited to the top of the spectrum. If the average volume of the material recorded to CDs is the same, the 44.1Kz/16bit format is OK. The next "big thing" is to go to 24/96 as the CD format. This would be like the computer manufacturers doubling and tripling the bus speeds. Where do you think the CPU speeds came from? All of a sudden the CPU speeds are triple what they were a year ago? They got smart and upped the bus speeds.
 
Re: Re: Interesting topic: A/D converters and what makes them special?

acorec said:
... 44.1Khz has a theoretical top end of 22.05Khz. 96Khz has a 44.1Khz theoretical top end. So, 24/96 probably will be the standard for a while. There really is no benefit to any higher sampling freq because the human hearing range is approx. 20-25Khz. Why bother with anything higher?
Click to expand...

This discussion reminds me of an article that I read some time back. Rupert Neve speaks to his theories on this issue. Here's an excerpt:

There have also been studies of the possibility of sensing ultrasonic frequencies with our skin. Certainly, ultrasonics in high levels can cause dizziness and nausea, and thus there are workplace limits on how high ultrasonic sound levels can legally be. There has also been plenty of anecdotal evidence, and some tests that are only semi-scientific, but compelling nonetheless. Rupert Neve does a test where he changes sine waves to square waves with high fundamentals, and people can hear the difference when they should not theoretically be able to, as the only difference is in harmonics that are above the commonly accepted audible range. He also tells a story of Geoff Emmerick correctly pointing out a couple of improperly terminated channels just by listening to the console output when the differences were only a few db down at around 50 kHz. In both cases above, there may be other distortions at work that explain the differences heard, but it remains interesting nonetheless. It has also been pointed out that trumpet with a harmon mute has a harmonic near 50 kHz which is near the amplitude of the fundamental, thus the argument of the upper harmonics being so low as not to matter is not an entirely accurate statement. So it seems that there is sometimes significant energy above 20 KHz, and ultrasonics may in some way be perceptible to humans, or possibly have some affect on what’s in the audible band.

The full article is at http://www.promastering.com/pages/techtalk_mac/tt-3_mac.html
 
Re: Re: Re: Interesting topic: A/D converters and what makes them special?

Rev E said:
This discussion reminds me of an article that I read some time back. Rupert Neve speaks to his theories on this issue. Here's an excerpt:

There have also been studies of the possibility of sensing ultrasonic frequencies with our skin. Certainly, ultrasonics in high levels can cause dizziness and nausea, and thus there are workplace limits on how high ultrasonic sound levels can legally be. There has also been plenty of anecdotal evidence, and some tests that are only semi-scientific, but compelling nonetheless. Rupert Neve does a test where he changes sine waves to square waves with high fundamentals, and people can hear the difference when they should not theoretically be able to, as the only difference is in harmonics that are above the commonly accepted audible range. He also tells a story of Geoff Emmerick correctly pointing out a couple of improperly terminated channels just by listening to the console output when the differences were only a few db down at around 50 kHz. In both cases above, there may be other distortions at work that explain the differences heard, but it remains interesting nonetheless. It has also been pointed out that trumpet with a harmon mute has a harmonic near 50 kHz which is near the amplitude of the fundamental, thus the argument of the upper harmonics being so low as not to matter is not an entirely accurate statement. So it seems that there is sometimes significant energy above 20 KHz, and ultrasonics may in some way be perceptible to humans, or possibly have some affect on what’s in the audible band.

The full article is at http://www.promastering.com/pages/techtalk_mac/tt-3_mac.html
Click to expand...

Yes. This is true. 16/44.1 is not a great format and is the real problem with final music reproduction at this time. I wish that the CD standard would go 24/96 as this would solve the biggest problem we have in digital recording. Maybe back to vinyl?
 
What I've been hearing a lot of . . . and correct me if I've heard wrong . . . is that a lot of 24-bit converters aren't necessarily pumping out a full 24-bits worth of actual audio. So am I to assume that some of these bits are just "junk bits?" -- or as some people like to call them "marketing bits."


You also have to consider the whole "garbage in" rule, here, too, and it would stand to reason that the quality of the analog portion of the converter would play a huge role -- and it would seem like a pretty likely area for a manufacturer to cut corners.

As far as the whole ultrasonic thing goes . . . keep in mind that such a phenomenon would only affect less than 1% of the listeners out there. Maybe this type of thing has some merit in laboratories, but how many consumer speakers out there are capable of reproducing ultrasonics?
 
Re: Re: Re: Interesting topic: A/D converters and what makes them special?

Rev E said:
This discussion reminds me of an article that I read some time back. Rupert Neve speaks to his theories on this issue. Here's an excerpt:

There have also been studies of the possibility of sensing ultrasonic frequencies with our skin. Certainly, ultrasonics in high levels can cause dizziness and nausea, and thus there are workplace limits on how high ultrasonic sound levels can legally be. There has also been plenty of anecdotal evidence, and some tests that are only semi-scientific, but compelling nonetheless. Rupert Neve does a test where he changes sine waves to square waves with high fundamentals, and people can hear the difference when they should not theoretically be able to, as the only difference is in harmonics that are above the commonly accepted audible range. He also tells a story of Geoff Emmerick correctly pointing out a couple of improperly terminated channels just by listening to the console output when the differences were only a few db down at around 50 kHz. In both cases above, there may be other distortions at work that explain the differences heard, but it remains interesting nonetheless. It has also been pointed out that trumpet with a harmon mute has a harmonic near 50 kHz which is near the amplitude of the fundamental, thus the argument of the upper harmonics being so low as not to matter is not an entirely accurate statement. So it seems that there is sometimes significant energy above 20 KHz, and ultrasonics may in some way be perceptible to humans, or possibly have some affect on what’s in the audible band.

The full article is at http://www.promastering.com/pages/techtalk_mac/tt-3_mac.html
Click to expand...


The skin sensing is interesting. When I used to play 5 nights a week during college, I would experience what I referred to as "sound hangovers". It would take me days to recover my energies after a week long gig. I even resorted to full stop earplugs and even though my ears felt better, I would still be dogged tired from the sound pressures which I attributed to nerve endings in my skin that would be overstimulated from the sound levels. Once I stopped playing fulltime I never experienced the fatigue again.

Sound is a very complicated beast.
 
What makes AD converters special is not the sample rate, not the bit rate, but the design of the filter.
 
chessrock said:
What I've been hearing a lot of . . . is that a lot of 24-bit converters aren't necessarily pumping out a full 24-bits worth of actual audio. So am I to assume that some of these bits are just "junk bits?" -- or as some people like to call them "marketing bits."
Click to expand...

That's all converters chess. At 24 bit the theoretical dynamic range is 24*6 = 144 dB. Even the better converters are doing in neighborhood of 118-120 dB. I'm not sure of the science so I'll defer to the more scientific among us. But even if soundcards could produce the theoretical dynamic range none of our music has that kind of range. Nor are there any monitors that could produce that kind of dynamic range. Plus we'd blow our ears out listening to that kind of dynamic range.


You also have to consider the whole "garbage in" rule, here, too, and it would stand to reason that the quality of the analog portion of the converter would play a huge role -- and it would seem like a pretty likely area for a manufacturer to cut corners.
Click to expand...

The folks at Lynx Audio have seemed to take this view about the audio section. I'm so sorry that I haven't gotten their card yet/sooner. I'm in the process of revamping my studio efforts and starting with 2 really great channels of mic, pre and converter. This approach and a lot of the stuff that you're saying is where I am in my studio efforts. I'd rather have 2 great channels than 8 passable and mediocre ones like I have now.
 
I've got an RME DSP Multiface here right now. It's fairly good. It's incredibly stable. I would recommend it for most any home recordists.

I think the conversion issue comes down to a couple of issues. First, I think Chessrock nails it with clock stability and low jitter. Second, I think it comes down to the quality of the D/A conversion. I think that for whatever reason it's a bigger trick to get it right going out as opposed to going in. I've had stuff recorded even on the lowest cost M-Audio card mix fairly well with improved D/A. Anyone else notice this D/A phenomenon???

Steve
www.mojopie.com
 
ozraves said:
I've got an RME DSP Multiface here right now. It's fairly good. It's incredibly stable. I would recommend it for most any home recordists.

I think the conversion issue comes down to a couple of issues. First, I think Chessrock nails it with clock stability and low jitter. Second, I think it comes down to the quality of the D/A conversion. I think that for whatever reason it's a bigger trick to get it right going out as opposed to going in. I've had stuff recorded even on the lowest cost M-Audio card mix fairly well with improved D/A. Anyone else notice this D/A phenomenon???

Steve
www.mojopie.com
Click to expand...

Bingo! D/A conversion and making it sound right is far more of an engineering challenge than A/D conversion. The secret is in the LP filter design. The better the design, the "warmer" and more "analog" the music will sound. The harshness that is common to cheap-o systems is because the LP filters are really cheaply designed.
 
You must log in or register to reply here.
Back
Top