D/A A/D Converter. Essential?

[toasteh]

How much does a good A/D D/A converter has to say for your music? I know most Audiointerfaces already have one, but how much better would it be if I bought a rack converter specifically for this task? Any opinions are greatly appreciated. Thanks!

[Ethan Winer]

Originally Posted by toasteh

how much better would it be if I bought a rack converter specifically for this task?

It would be 0.4 percent better.

Seriously, outboard A/D/A is probably the last thing you need to worry about, and it comes way after having good microphones and good monitoring in a good room.

--Ethan

[ColourSurround]

I'll second that. The top three things really are a great mic(s), a great monitoring system, and a solid pre-amp. The room is important too but definitely difficult to attain if you don't already have a solid place from the get go.

[Bierce85]

Originally Posted by Ethan Winer

It would be 0.4 percent better.

Seriously, outboard A/D/A is probably the last thing you need to worry about, and it comes way after having good microphones and good monitoring in a good room.

--Ethan

I would've agreed with that statement before I heard the shootout between the Digi002's converters and a Mytek stereo 192..

The difference was night and day, and quite simply, the 002 was f*cking up the sound. I don't know if the 002's are any worse than other cheapo interfaces like M-audio/E-Mu and the like, but it was a real eye opener as to what bad conversion does to a perfectly good signal passing through it.


That said, I think once you get to, say.. RME's level you're about 90% of the way to as far as good conversion can bring you and that last 10% is definitely massively outweighed by the other factors mentioned.

[Ethan Winer]

Originally Posted by Bierce85

I would've agreed with that statement before I heard the shootout between the Digi002's converters and a Mytek stereo 192.

I don't know what shootout you're referring to, but most of the ones I've seen are fatally flawed in one way or another. The most common mistake I see is recording two different performances in succession. The only way to truly compare converters is to split the preamp output from each microphone and one performance. Have any details on the shootout you heard?

Also, if the difference really was huge, is it possible one of the devices was defective?

--Ethan

[Bierce85]

Hey ethan,

Here's a link to the thread:

Mytek converter v.s. Digi 002 converter:A/B Test here!

The shootout is flawed in the way you mentioned and since it was only 2-3 tracks it could've definitely easily been split but the difference between the two converters is still very clear. IMO, for the price difference between high end and low end you shouldn't have to set up a perfect A/B test to be able to hear the increase in quality, and in this case, they certainly didn't. I think scientific A/B tests are more necessary when judging tiny differences, not big ones like the one demonstrated here..

When you listen to the digi002 sample, it doesn't sound "bad", but when you hear the mytek it immediately gains a huge amount of clarity and detail. The vocal has much more of an identity in the mytek file, and to me the thought of a converter taking away from what I put through it is very concerning. I suppose there's a chance the 002 was defective, but like I said it didn't sound "bad", just not "good". The Mytek sounded much better.

[Ethan Winer]

Originally Posted by Bierce85

The shootout is flawed in the way you mentioned ... you shouldn't have to set up a perfect A/B test to be able to hear the increase in quality

I fear you missed my point. This is exactly why a more scientific test is needed. No two performances are the same, and even if they were, having even an inch difference in microphone placement can make a huge change in the sound.

I have no direct experience with either of those converters, and I have no vested interest in either coming off better. However, I imagine the specs for both are very similar - flat to within 1/4 dB from 20 Hz to 20 KHz, very low distortion and ringing, and so forth. So if we perceive an obvious difference in sound, it is much more likely due to a performance or microphone spacing difference.

Also, the posted example files are MP3s. I have to think the degradation from MP3 encoding would mask any subtle differences between two competent converters. Which leaves frequency response differences due to different performances and phrasing and microphone placement etc.

--Ethan

[Bierce85]

Originally Posted by Ethan Winer

if we perceive an obvious difference in sound, it is much more likely due to a performance or microphone spacing difference.

--Ethan

What I heard in that shootout was a huge difference in the clarity and detail of the voice. Neither clip sounded further or closer to the mic to my ears which is likely what it would of sounded like if the singer had simply moved. I understand moving side to side and up and down can also effect the frequency response of a voice but it's not like low-right is gonna sound detailed and clear, and high-left is gonna sound like a foggy muck

As for the specs... sure, maybe they're close but specs cant really tell you anything about how detailed or "high quality" sounding something is. Only your ears can tell you that. And when you make the switch from cheap to high end gear, you shouldn't have to do any testing to hear the increase in quality, you should just make a few recordings and the difference should be apparent.

Your standpoint seems to be that no one should bother getting Lavry or Prism converters because even if they can afford it they're throwing thousands of dollars out the window in exchange in a 0.4% increase in quality. I simply cannot believe that. Signals running through custom built, high end analog circuits are gonna sound better than the same signals run through the $8 (probably) analog stage of a piece of gear like the digi002. There are people who have devoted large portions of their lives to converter design and strive to make better products every day. I don't think they would bother if they were only improving audio quality by such a miniscule amount. That would pretty much render them scam artists and if they were scam artists, don't you think they'd find a better way of scamming people then selling boutique audio gear to a small group of engineers and studios?

[manthe]

I can also hear a pretty apparent difference that just does not seem like it can be attributable to variances in performance or mic placement.

I also believe that good converter (at least decent) converters are a very important part of the recording chain. Next to knowledge/placement/engineering/performance, then mics, preamps and decent instruments/amps/vocal chops, etc., the AD is is very important.

BUT, I do believe that at a certain level you are most definitely dealing with very, very diminished returns for the price. This is not a guess. This is based on hearing the results of audio recorded through expensive vs 'mid-range' converters (eg RME). The differences in actual audio quality were pretty tough to quantify. They are there and become a little more apparent after stacking...but just not many, many thousands of dollars worth of difference, IMO.

This does not mean that high-end converter developers are 'scamming' people in any way. It just means that the little bit of extra quality typically ends up costing up to 500% to 700% more! It is up to the individual if that price difference is really worth the gain.

The difference between an 002 or an Edirol 2-channel USB and an RME or higher-end Tascam (DM series, etc) is very noticeable and apparent. Even that price difference is HUGE, but it is still at a relatively manageable dollar figure.

So, even if the difference is 1000%, you're talking $150 vs $1500, and people can afford $1500. But, when you get to the Prism level, you're starting to get to dollar figures like $10,000 for 2 channels and THAT is insane, at least for me!

Originally Posted by Bierce85

What I heard in that shootout was a huge difference in the clarity and detail of the voice. Neither clip sounded further or closer to the mic to my ears which is likely what it would of sounded like if the singer had simply moved. I understand moving side to side and up and down can also effect the frequency response of a voice but it's not like low-right is gonna sound detailed and clear, and high-left is gonna sound like a foggy muck

As for the specs... sure, maybe they're close but specs cant really tell you anything about how detailed or "high quality" sounding something is. Only your ears can tell you that. And when you make the switch from cheap to high end gear, you shouldn't have to do any testing to hear the increase in quality, you should just make a few recordings and the difference should be apparent.

Your standpoint seems to be that no one should bother getting Lavry or Prism converters because even if they can afford it they're throwing thousands of dollars out the window in exchange in a 0.4% increase in quality. I simply cannot believe that. Signals running through custom built, high end analog circuits are gonna sound better than the same signals run through the $8 (probably) analog stage of a piece of gear like the digi002. There are people who have devoted large portions of their lives to converter design and strive to make better products every day. I don't think they would bother if they were only improving audio quality by such a miniscule amount. That would pretty much render them scam artists and if they were scam artists, don't you think they'd find a better way of scamming people then selling boutique audio gear to a small group of engineers and studios?

[lrieves]

Going back to the original question, if you don't have good mic's and solid pre's, you will only be hearing the flaws in your sub-par gear more accurately with a dedicated converter. Also, if you are only recording acoustic guitar and vocals, it won't make the biggest difference, but will become most apparent when tracking a lot of instruments -so that is where it can begin to be an issue.

[guard dawg]

how can we have good monitoring without a good DA converter it is like saying i have this super sound system that plays casette tapes,you must start directly at the source first.

[Texum]

My take on these kinds of questions, is to take into consideration that your recorded tracks will be affected by all links in the chain, for sure, from source to room, mic to pre and converter. It just doesn't make sense to ignore any one link.

Like Ethan Winer always says, too many ignore their room and monitoring.
But, like others have said, D/A matters for accurate reference (this was once brutally demonstrated to me when comparing the 002 converters with the Rosetta. And, without good mics, pres and A/D it becomes more challenging to get good sounds to "tape".

Certainly converters do affect sound quality. The jury here (and other places) seems to be out on exactly how important converters are, in relation to other links in the recording chain. But neither does it seem to exist any agreement on which link actually is the most important. Some will claim the microphone, some the pre, others think that gear is much less important, and focus on source, room or engineering technique. Possibly they are all correct.

Maybe the best is to always try to keep as balanced a picture as possible, meaning, among other things, that if you have for example only a 002 in an unreated room, it doesn't make much sense to buy a UA2192, if that's going to be your only upgrade. But if you would like to upgrade your entire recording chain, don't skip upgrading the converters.

[sergioelectro]

Originally Posted by manthe

I can also hear a pretty apparent difference that just does not seem like it can be attributable to variances in performance or mic placement.

I also believe that good converter (at least decent) converters are a very important part of the recording chain. Next to knowledge/placement/engineering/performance, then mics, preamps and decent instruments/amps/vocal chops, etc., the AD is is very important.

BUT, I do believe that at a certain level you are most definitely dealing with very, very diminished returns for the price. This is not a guess. This is based on hearing the results of audio recorded through expensive vs 'mid-range' converters (eg RME). The differences in actual audio quality were pretty tough to quantify. They are there and become a little more apparent after stacking...but just not many, many thousands of dollars worth of difference, IMO.

This does not mean that high-end converter developers are 'scamming' people in any way. It just means that the little bit of extra quality typically ends up costing up to 500% to 700% more! It is up to the individual if that price difference is really worth the gain.

The difference between an 002 or an Edirol 2-channel USB and an RME or higher-end Tascam (DM series, etc) is very noticeable and apparent. Even that price difference is HUGE, but it is still at a relatively manageable dollar figure.

So, even if the difference is 1000%, you're talking $150 vs $1500, and people can afford $1500. But, when you get to the Prism level, you're starting to get to dollar figures like $10,000 for 2 channels and THAT is insane, at least for me!

Hello,

That is a nice opinion. I am going to read your post again, because I was planning about buying a new converter like a HEDD or a 2192 to upgrade from my RME Fireface400.
Some people here said that upgrade is going to be night and day. I make electronic music and after reading your post I am not so sure about that much difference.
Thanks again

Sergio

[rainy-taxi]

@Ethan:

Grab a M-audio FW1814 somewhere, grab a Presonus Firestudio 26 somewhere. Connect a pair of monitors to it and listen to a cd. Then change the connection to the other card and compare (or use a monitor controller between both). The difference in sound is night and day. And we are still talking in "Lo-fi" gear here.

Only thing used by playing a cd is the DA convertor. Those alone can be very different sounding on devices. I doub't this is a 0.4% difference.

Of course, knowing what you are doing, placing mics well, having good instruments etc. All those things help making things sound better. But if we look at for example having a drumloop from a sampler and record it with 2 different convertors, you have 2 different results.

[Unclenny]

I put RME ADI-2 in front of my MBox 1 before my last project. Didn't change anything else....decent mic's/pre's.....passable room and monitoring.

The difference was more than just noticeable to me.

[fooman]

I didn't read the 20 other replies... but I'd have to say that it does matter. Obviously like anything else you buy, there IS a point you can get to where it simply doesn't matter much to upgrade. It's the point of diminishing returns or whatever, where another $1000 will get you barely anything in return.

So anywho, I went from two Delta 1010's to an Aurora 16. Good lord there was a major step up right there. The digital sock that was over my audio was lifted. It was seriously a really good feeling to know I didn't waste my cash because so many people said "AD/DA isn't a big deal".

Oh, and the 3124+ i upgraded to for the rock music I do didn't hurt ;)

[narcoman]

Originally Posted by ColourSurround

I'll second that. The top three things really are a great mic(s), a great monitoring system, and a solid pre-amp. The room is important too but definitely difficult to attain if you don't already have a solid place from the get go.

Just have to re-order that a little...


the single most important yet overlooked feature of a recording or/and environment is the room in which you are monitoring. Without that, your monitors, mics, amps and even recording area are a waste of time.

You CAN make good records with poor gear in a great room, at least your decisions are based on a good point of reference - but it is MUCH harder to do the converse.

[Ethan Winer]

Originally Posted by Bierce85

it's not like low-right is gonna sound detailed and clear, and high-left is gonna sound like a foggy muck

Originally Posted by manthe

I can also hear a pretty apparent difference that just does not seem like it can be attributable to variances in performance or mic placement.

Nobody can sing or play a guitar exactly the same way twice, and changes in frequency response can also be huge with tiny positional changes. Look at the first graph below to see hard proof of what I'm talking about. This is the frequency response in a well-treated room with the microphone moved in one-inch increments. Further, the response is averaged in third octaves which minimizes the displayed differences. The lower graph below shows the true (not averaged) response for two locations four inches apart (in a different room).

It may be difficult to accept that the lower graph shows the same microphone and room only four inches away, but it's true! I am convinced this is the real reason people report large changes in devices that should sound more or less the same. If you get Mix Magazine, in the current (April) issue columnist Paul Lehrman mentions my research these graphs are from. Full article HERE.

So even if a performer could duplicate the same sound twice, which is impossible, moving even one inch changes the response a lot. I'm not talking about sounding off-mike, just a large change in the response at mid and high frequencies. This is infinitely more change than the difference between any two A/D/A converters at any price point. And this is exactly why all A/B comparisons of different performances are fatally flawed, no matter how compelling the audible differences may seem.

--Ethan

[theblue1]

Originally Posted by Bierce85

What I heard in that shootout was a huge difference in the clarity and detail of the voice. Neither clip sounded further or closer to the mic to my ears which is likely what it would of sounded like if the singer had simply moved. I understand moving side to side and up and down can also effect the frequency response of a voice but it's not like low-right is gonna sound detailed and clear, and high-left is gonna sound like a foggy muck

As for the specs... sure, maybe they're close but specs cant really tell you anything about how detailed or "high quality" sounding something is. Only your ears can tell you that. And when you make the switch from cheap to high end gear, you shouldn't have to do any testing to hear the increase in quality, you should just make a few recordings and the difference should be apparent.

Your standpoint seems to be that no one should bother getting Lavry or Prism converters because even if they can afford it they're throwing thousands of dollars out the window in exchange in a 0.4% increase in quality. I simply cannot believe that. Signals running through custom built, high end analog circuits are gonna sound better than the same signals run through the $8 (probably) analog stage of a piece of gear like the digi002. There are people who have devoted large portions of their lives to converter design and strive to make better products every day. I don't think they would bother if they were only improving audio quality by such a miniscule amount. That would pretty much render them scam artists and if they were scam artists, don't you think they'd find a better way of scamming people then selling boutique audio gear to a small group of engineers and studios?

I don't know HTF you get that out of what Ethan wrote.

I think you should read what he said and really think about it, if you think such a "shootout" can tell you anything significant at all.

[stagefright13]

Ethan is right about multiple takes they are all different. I didn't listen to any of the sample tho. Vocalists warm up as they go. Repeating the track makes it easier each time etc.

But if you have an 002 like I do even a modest converter upgrade makes quite a difference. If ya already have a good chain going in. I bucked the trend of upgrading converters for quite some time. But they did push it to the next level. Try to rent something first. I rented some Apogees for a few days and I was sold on a converter upgrade.

The D/A is worth it alone.

[rainy-taxi]

Originally Posted by theblue1

I don't know HTF you get that out of what Ethan wrote.

I think you should read what he said and really think about it, if you think such a "shootout" can tell you anything significant at all.

A shootout like that has arguable results where its the question if those are valid and useful at all to base your opinion on. However, there still is a lot of difference between convertors...

[Astrain]

I have both the digi002 and the mytek stereo 96 AD, and my daily experience with both is very similar to the one in the shootout.

Ive also tested the internal clock vs mytek clock by playing the same segment of a recording and you can clearly hear a diference, to put it this way with the mytek clock the recording sounds tight, I would say it sounds pro, with the digi clock it sounds not as defined more prosumer,

Even worst I ve been pre-mixing some work that Im recording with a mbox2 and Adam A7 and sub8 and it sucks , harsh is the best word to describe the DA in the mbox, of course I know mbox its aimed to the prosumer market, but my point is that converters do make a big diference, maayybee a little less important than room, mics, pres, but very important.

[Bierce85]

Originally Posted by theblue1

I don't know HTF you get that out of what Ethan wrote.

I think you should read what he said and really think about it, if you think such a "shootout" can tell you anything significant at all.

Ethan makes very good and valid points but I still think I'm hearing better conversion in the mytek.


Ethan: in your test, how far was the microphone initially from the source? How loud was the source? Wouldn't both these things affect the results? As I understand it, the more direct sound hitting the mic as opposed to reflected sound, the less room acoustics are going color the sound, and the quieter the sound, the less reflected sound there will be.

The singer sounded relatively quiet to me (he definitely was not belting it out), and he was probably only a few inches from the mic meaning the mic was getting mainly direct sound. This makes the test seem more repeatable to me.

Please correct me if I'm wrong.

[allencollins]

Originally Posted by Ethan Winer

It would be 0.4 percent better.

Seriously, outboard A/D/A is probably the last thing you need to worry about, and it comes way after having good microphones and good monitoring in a good room.

--Ethan

Conversion is as as important as all the other elements.
A/D D/A is an essential part of 'good monitoring in a good room'

What good is a 'good' room or good nearfield monitors if your D/A is not on the level? It's no good at all. It'