I Can Hear The Difference Between Some DACs

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I'm leaving this discussion after this post because it's really a moot point for me. I'm sticking with playing all my music from hard drives through the Oppo 205 DAC. It's just too convenient and enjoyable vs. shunting discs. The Surround Master is coming:SG which will push the Marantz deck out of the system anyway.

The purpose of my participation in this discussion was mainly to determine if I would hear an improvement with the $4000 exaSound DAC vs. the Oppo 205 DAC...And if that improvement would be worth $4000 to me especially since I just got the Oppo 205 this past September. Some say yes. Sully seems to say that "it depends" and I'd have to do precise level matching and double blind A/B tests to make a judgement.
Too bad you are leaving the discussion. Sorry I wasn't able to respond to some of your posts in time. Take my opinion for what it's worth ( 0.02 cents ) . I have no idea how the exasound measures and whether they are evening aiming for a flat frequency response, but if they are aiming for a neutral, flat frequency response and measure as such, I have a hard time seeing how they would sound any different. All other variables have a far greater impact.

Personally I would love for some engineer to explain to me what electrical properties make a component have these audiophile attributes. Even more I would to learn about how they tested each piece of equipment before shipping. Then we can all stop arguing.

If you have a pile of money burning a hole in your pocket and enough time to twiddle with it go ahead and try it, at least it looks nice. I'd have to agree with ssully though. If you really want to be sure, you would have to at least level match.
 
I come to this thread whenever I need a pick-me-up :LOL:
Heh.
I missed whatever happened. But I know what that looks like well enough.

Just to play devil's advocate...

For genuine enthusiasts, it's probably not so much: "I disagree and want to tell you you're wrong!"
Probably more along the lines of: "Wait what? I thought I researched this! What did I miss? Did I buy the wrong thing and get screwed? God damnit..."
 
It's a balancing act...although science is definitive...it's not all encompassing...because each of us is a variable...we are blessed or cursed with different hearing...different listening environments...different equipment...there can't be duplicates among us...there will always be differences...and in that vein....science can't cover everything....
I completely agree with your statement "each of us is a variable" and the scientific approach takes that into account. It is precisely why the concept of DBT etc. exists. Different hearing based on the whether, time of day, our moods, what experiences we had before playing said piece of equipment and what not else around us. The rest of it though is pretty well understood in the scientific realm though and is very quantifiable and measurable though.
 
Too bad you are leaving the discussion. Sorry I wasn't able to respond to some of your posts in time. Take my opinion for what it's worth ( 0.02 cents ) . I have no idea how the exasound measures and whether they are evening aiming for a flat frequency response, but if they are aiming for a neutral, flat frequency response and measure as such, I have a hard time seeing how they would sound any different. All other variables have a far greater impact.

Personally I would love for some engineer to explain to me what electrical properties make a component have these audiophile attributes. Even more I would to learn about how they tested each piece of equipment before shipping. Then we can all stop arguing.


Not an engineer (though neither are most 'mastering engineers') but offhand, system properties would be: frequency response, noise level, distortion. And the magnitude of difference matters, since our hearing for any of these thing is not infinitely sensitive
 
Personally I would love for some engineer to explain to me what electrical properties make a component have these audiophile attributes. Even more I would to learn about how they tested each piece of equipment before shipping. Then we can all stop arguing.

Everything is manufactured to a tolerance. Including electronic components.
Audio still has analog signal paths. (If this isn't intuitively obvious... Digital audio is encoded. Digital audio would sound like the modem startup telltale from dialup days if sent directly to a speaker. We still have the analog paths on the recording end and then the playback end.)
The analog signal paths are as critical as they've ever been and subject to degradation as the components in the circuits drift from nominal tolerance.
There are compromises in design and manufacturing with lower budget products.

If you have a pile of money burning a hole in your pocket and enough time to twiddle with it go ahead and try it, at least it looks nice. I'd have to agree with ssully though. If you really want to be sure, you would have to at least level match.
There's literally no choice! If you want to compare two audio streams that are supposed to be identical with your ears to try to determine if one is compromised vs. the other or not, you must level match them to within 0.1db or less. Otherwise you WILL pick the louder one as better sounding. Even if it turns out it has genuinely had some damage done to it. You'll miss that and pick the louder one unless the damage is really obvious.

Doing a shootout between converters is not trivial to setup! You kind of need to setup dueling systems and level match them in the analog domain. Otherwise you're always introducing other variables.

But then what does "sound better" mean?
If you literally have the master on a reference system, you can listen to a copy and compare it. Pretty straightforward there.
But what if you have two copies? Which one sounds more like the master? You don't know because you don't have that master! Maybe one of them sounds "better" to you, but which one sounds most like the actual master?

See, it's not even so much that there's literal disagreement. It's more that you really have to remind yourself that some of the variables are big enough that you can't attribute "better" in some scenarios to the variable you thought you were looking at.
 
Last edited:
Everything is manufactured to a tolerance. Including electronic components.
Audio still has analog signal paths. (If this isn't intuitively obvious... Digital audio is encoded. Digital audio would sound like the modem startup telltale from dialup days if sent directly to a speaker. We still have the analog paths on the recording end and then the playback end.)
The analog signal paths are as critical as they've ever been and subject to degradation as the components in the circuits drift from nominal tolerance.
There are compromises in design and manufacturing with lower budget products.
I am not sure if you are trying to provide an explanation to audiophile properties, but I can say it is not the explanation I was looking for. Sure, I agree every component electrical or otherwise is manufactured to within certain tolerances. I am probably preaching to the choir here, but most equipment ( even relatively inexpensive, though not all, there is some stuff that is just cheap shit to be sure ) is designed to work around those tolerances within its specifications. As an example, if an amp is designed to produce a flat frequency response from 10- 20KHz, it will provide that within the range of the manufacting tolerances of the components chosen over its expected lifespan. If one of the pieces doesn't meet this requirement for whatever reason (component tolerance/failure etc.), it fails spec, that's it. Good forbid somebody is actually listening to every receiver/amplifier that comes off the manufacturing line for QC testing. That is what engineering is all about after all.
 
Not an engineer (though neither are most 'mastering engineers') but offhand, system properties would be: frequency response, noise level, distortion. And the magnitude of difference matters, since our hearing for any of these thing is not infinitely sensitive
Now, you are not talking like an "audiophile" engineer, are you ?
:p
 
Repeatedly stating you're not bringing something up again is kind of bringing it up again. If I'm going to be accused of slander I think it's reasonable to respond.
My only point in the post was to let @bluelightning know what happened to that thread. He asked a direct question and I gave a direct answer. Past that you'll have to take your beef to the moderators.
 
"Audiophile" to me means the reproduction system can produce 1:1 what the engineer heard on their system.
Maybe a little more to it...
This also suggests the recording was somewhat sonically ambitious to begin with. (Reproducing some distorted drek 1:1 isn't that meaningful.)

But an "audiophile" playback system is one that can reproduce 1:1 what was heard in the studio (for good or bad).
That's my interpretation. Anything beyond that is too subjective from an engineering perspective and into creative territory IMHO.

There's not much mystery in some of the cheap gear. The designs are limited in scope right on paper on day one. They still deliver a certain level of operation and a bunch of people agreed they still had a relevant product even though it was theoretically compromised. Diminishing returns arguments and all that.
 
Repeatedly stating you're not bringing something up again is kind of bringing it up again. If I'm going to be accused of slander I think it's reasonable to respond.


My only point in the post was to let @bluelightning know what happened to that thread. He asked a direct question and I gave a direct answer. Past that you'll have to take your beef to the moderators.

Did you consider a DM? Because there's quite a lot of drawing attention now.
 
There's literally no choice! If you want to compare two audio streams that are supposed to be identical with your ears to try to determine if one is compromised vs. the other or not, you must level match them to within 0.1db or less. Otherwise you WILL pick the louder one as better sounding. Even if it turns out it has genuinely had some damage done to it. You'll miss that and pick the louder one unless the damage is really obvious.
I cant even do that on the equipment I own. Volume steps are in 0.5dB increments.

I have had a standard Radio Shack SPL meter for many years, and have used it often to set levels. But it will only get you so close because of the bouncy reading. You would really need recorded sine waves to use it properly. Noise and music just have too much variation. In the end I would always set levels using the SPL meter, and then fine tune by ear, then re-check the meter. Ironically the auto level set routines in todays AV gear, even using those relatively cheap puck mikes, can do a very good job of it. REW can also provide some nice tight measurements for SPL.

A sales rep that I got to know from a local audio salon many years ago took the level match issue seriously. His method was to set levels as closely as possible when A/Bing components. If you think you hear a perceptible difference, focus in on the difference, turn down the level on the preferred component a tiny bit to assure it was slightly lower in level, then repeat the A/Bing, and see if you can still detect the difference.
 
Just a quick note about using sine waves for listening tests: Be careful. They tend to sound quieter than they actually are, and there is no more efficient way to roast a speaker's voice coil than with a continuous sine wave. (I've learned this the hard way.)
 
I cant even do that on the equipment I own. Volume steps are in 0.5dB increments.
In the analog or digital domain?

Hope this quick math is right...
I believe the smallest volume alteration possible with 24 bit audio is something like 0.000008db
16 bit would still give you something like .00146db

0.1 adjustments are no problem digitally
I believe the smallest fader moves in Reaper DAW are 0.06db (at least without looking up any 'micro adjust' modifiers I may have forgotten about).
If some software is restricting that further... it's all artificial and being done by whatever software we're talking about.

If we're talking about an analog system adjustment, volume trim in analog is as infinitely small as you move the knob... unless the knob is detented.

Anyway...
Setting up a genuine shootout between different DACs would require dialing in dueling systems and matching their analog outputs to within 0.1db.
This would be a big fiddly experiment to setup accurately. Your 0.5db volume knob restriction (wherever that's coming from) further illustrates the challenge.
 
"Audiophile" to me means the reproduction system can produce 1:1 what the engineer heard on their system.
Maybe a little more to it...
This also suggests the recording was somewhat sonically ambitious to begin with. (Reproducing some distorted drek 1:1 isn't that meaningful.)

But an "audiophile" playback system is one that can reproduce 1:1 what was heard in the studio (for good or bad).

1:1 is impossible, because what was heard in the studio is the result of the system, which includes the room. And that cannot be modeled 1:1 at home.

You can aim for your hardware (and software, and even rooms) to introduce as little unintended coloration as possible.

Beyond that it's all preference. For example, I like to introduce a *lot* of controlled coloration to two-channel sources, by upmixing via Dolby PLII.
 
Were the accusations made in a DM? I really think you should let it go now.

I think you misunderstood me. I meant had you considered explaining what happenned to the thread via DM so you could 1) describe more clearly what happened as you see it without repeating more info in this thread, and 2) not draw attention to it. I'm not feeling sore.
 
In the analog or digital domain?

Hope this quick math is right...
I believe the smallest volume alteration possible with 24 bit audio is something like 0.000008db
16 bit would still give you something like .00146db

0.1 adjustments are no problem digitally
I believe the smallest fader moves in Reaper DAW are 0.06db (at least without looking up any 'micro adjust' modifiers I may have forgotten about).
If some software is restricting that further... it's all artificial and being done by whatever software we're talking about.

If we're talking about an analog system adjustment, volume trim in analog is as infinitely small as you move the knob... unless the knob is detented.

Anyway...
Setting up a genuine shootout between different DACs would require dialing in dueling systems and matching their analog outputs to within 0.1db.
This would be a big fiddly experiment to setup accurately. Your 0.5db volume knob restriction (wherever that's coming from) further illustrates the challenge.

I didn't say it wasn't possible, but my home equipment just wont do it. I'm trying to think how long its been since I had an analog preamp with a continuous volume adjustment. Maybe a Marantz receiver way, way, way back in the day. Maybe even an Adcom unit after that (cant remember). But its been at least 30 years. Everything since has used a stepped ladder type system. Analog yes, but not a true rheostat. So volume is adjusted in discrete 0.5 dB steps.

Does any modern gear actually have a continuously variable volume pot any more?
 
Back
Top