Have a few questions regarding DAC's - Page 5 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #121 of 140 Old 04-17-2012, 07:31 AM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 6,014
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 268
Quote:


Really?

Cites?

Brian C.J. Moore, Introduction to the Psychology of Hearing. Final chapter, or appendix, depending on the edition.

If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
Sponsored Links
Advertisement
 
post #122 of 140 Old 04-17-2012, 07:36 AM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 6,014
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 268
From Moore, op. cit., 4th edition:

Quote:


CD and DAT players generally have a specification which is far better than that of other components in a hi-fi system, especially cassette decks and loudspeakers. Essentially, the output signal which they provide is indistinguishable from that which would be obtained from the master tape produced by the recording studio (studio recordings are now usually digital recordings). Thus, provided a CD or DAT player is working according to specification, it will produce no noticeable degradation in sound quality. It follows from this that most CD players and DAT players sound the same.

Quote:


The basic performance of even a moderately priced hi-fi amplifier is likely to be so good that improvements in technical specification would make little audible difference. For example, a moderately good amplifier will have a frequency response from 20 to 20,000 Hz +/– 1 dB, distortion less than 1% and a signal-to-noise ratio greater than 90 dB (for input signals with a reasonably high level, such as from a CD player). These values are better than the limits required by the ear.


If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
post #123 of 140 Old 04-17-2012, 07:52 AM
SDL
Senior Member
 
SDL's Avatar
 
Join Date: Oct 2001
Posts: 254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by mcnarus View Post

Human beings are not calibrated test instruments, not even close. They shouldn't pretend to be.

Now, obviously there are some things that cannot be left entirely to objective measurementsspeaker preferences, room set-up. There you have to trust your ears. But you should also be honest about their limitations.

I agree completely! Human beings are not calibrated test instruments and clearly have personal preferences about music and how they listen to it.

SDL
SDL is offline  
post #124 of 140 Old 04-17-2012, 08:02 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by mcnarus View Post

Brian C.J. Moore, Introduction to the Psychology of Hearing. Final chapter, or appendix, depending on the edition.

Looks like quite a book, if even a bit misnamed due to all of the information about physiology.

It appears that they devote a whole chapter to audio, both the highs and the lows (metaphorically speaking).
arnyk is online now  
post #125 of 140 Old 04-17-2012, 08:04 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by SDL View Post

I agree completely! Human beings are not calibrated test instruments and clearly have personal preferences about music and how they listen to it.

However, there is a hard and fast rule - to have a sonic preference, there first has to be reliable sonic detection.

When you're talking speakers and rooms, reliable detection is generally easy to establish. Indeed the one big problem is that audible differences abound to the point of being distractions.

When you are talking amps and DACs, not so much. Usually, not at all.
arnyk is online now  
post #126 of 140 Old 04-17-2012, 08:07 AM
SDL
Senior Member
 
SDL's Avatar
 
Join Date: Oct 2001
Posts: 254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I apologize for taking the DAC discussion off into the blind testing debate. It wasn't my original intent, but that's where a lot of the posters on this thread seemed to want to go.

In regard to that issue, what does everyone think about the comments on this link from a supporter of blind testing who sees both its strengths and limitations? http://trustmeimascientist.com/2012/...stening-blind/

SDL
SDL is offline  
post #127 of 140 Old 04-17-2012, 08:20 AM
 
diomania's Avatar
 
Join Date: Dec 2008
Posts: 1,389
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 49
Quote:
Originally Posted by SDL View Post

In regard to that issue, what does everyone think about the comments on this link from a supporter of blind testing who sees both its strengths and limitations? http://trustmeimascientist.com/2012/...stening-blind/

SDL

What do you think about this book by Robert Harley? Check out the reviews. http://www.amazon.com/Complete-Guide...owViewpoints=1
diomania is offline  
post #128 of 140 Old 04-17-2012, 08:37 AM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 6,014
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 268
Quote:


In regard to that issue, what does everyone think about the comments on this link from a supporter of blind testing who sees both its strengths and limitations? http://trustmeimascientist.com/2012/...stening-blind/

Thanks for the link. It should be noted that the limitations he discusses are not so much limitations of blind tests themselves as legitimate cautions about interpreting the results. (And I suspect he's misinterpreting the results of that hardware vs. software test he cites, though I can't tell for sure because his description of it is too sketchy.)

If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
post #129 of 140 Old 04-17-2012, 08:38 AM
AVS Special Member
 
walbert's Avatar
 
Join Date: Jun 2010
Posts: 1,811
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by arnyk View Post

However, there is a hard and fast rule - to have a sonic preference, there first has to be reliable sonic detection.

Exactly!

If you can't actually tell A and B apart, you can't like A better than B.
walbert is offline  
post #130 of 140 Old 04-17-2012, 08:47 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by SDL View Post

I apologize for taking the DAC discussion off into the blind testing debate. It wasn't my original intent, but that's where a lot of the posters on this thread seemed to want to go.

In regard to that issue, what does everyone think about the comments on this link from a supporter of blind testing who sees both its strengths and limitations? http://trustmeimascientist.com/2012/...stening-blind/

This article is very common sense, and thorough.

The rest of the site, especially some articles by other authors, not so much.
arnyk is online now  
post #131 of 140 Old 04-17-2012, 08:47 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by diomania View Post

What do you think about this book by Robert Harley? Check out the reviews. http://www.amazon.com/Complete-Guide...owViewpoints=1

People I know like to tear it apart.
arnyk is online now  
post #132 of 140 Old 04-17-2012, 08:52 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by walbert View Post

If you can't actually tell A and B apart, you can't like A better than B.

That doesn't seem to stop a lot of people.

The history of DBTs is that we invented ABX thinking that it would be a good qualification test for some other tests we wanted to do for preference.

Some of us, but not all of us were blown away when all the stuff we wanted to rate by preference turned out to be indistinguishable.

Truth is, there is some stuff that sounds different, but the differences are so small that it seems pointless to even try to form a preference.

Of course you read all of these audiophile sites where everybody seems to think that everything sounds different. As the old saying goes: "Ignorance is bliss", except of course for many of us it isn't.
arnyk is online now  
post #133 of 140 Old 04-17-2012, 08:53 AM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 6,014
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 268
Quote:


If you can't actually tell A and B apart, you can't like A better than B.

But you can think you do!

If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
post #134 of 140 Old 04-17-2012, 01:44 PM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
Hi Arny,
Quote:
Originally Posted by arnyk View Post

I should invoke TOS 8 based on listening tests using a general run of commercial recordings. ;-)

Sorry, I don't know what TOS 8 is. That's why I'm here - In many respects, I'm still an audio newbee. Can you point me?

Quote:


Here's the reason why - clipping or at least significantly higher distortion just below FS is not that uncommon. However, its relevancy to listening enjoyment is most people's primary concern. . .
. . . But its audible consequences for clipping in the last dB before FS? Not so much!

Ok, I can accept that.

However, in the case I cited, the clipping was serious - far from FS. I wouldn't have tested otherwise, but as soon as I plugged it in, I could hear something was wrong. The audio was seriously distorted. I would have just trashed it, but I bought it to test something else, and had it apart anyway.

Long ago, I generated a bunch of full-scale sine waves in FLAC that I use for testing the hardware that I design. On the scope, the 1kHz wave was clearly flat-topped both top and bottom. Reducing the gain of the output buffer was all it needed.

So maybe the hardware I've been building is not correct? I always leave headroom such that a full-scale signal will never clip, even with worst-case tolerances on the components. Am I being too conservative?
MarkHotchkiss is offline  
post #135 of 140 Old 04-18-2012, 12:05 AM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,174
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 44
Quote:
Originally Posted by MarkHotchkiss View Post

So maybe the hardware I've been building is not correct? I always leave headroom such that a full-scale signal will never clip, even with worst-case tolerances on the components. Am I being too conservative?

Most DACs have a DC offset at ~2.5V so the signal swings 0-5V effectively. If Vcc is 5V which it is in USB, and the buffer has any gain then the signal swing sould easily exceed the voltage rails. Even if it didn't, there are fewer opamps that will actually swing signal rail to rail, typically CMOS units. Most will need a volt or so clearance between signal and rail to not clip.
A9X-308 is offline  
post #136 of 140 Old 04-18-2012, 04:42 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by MarkHotchkiss View Post

However, in the case I cited, the clipping was serious - far from FS.

Audio is a lot about quantification - how far below FS was it clipping?

0.1 dB?

1.0 dB?

3 dB?

The differences above can be determined from a sine wave in a glance.

Quote:


I wouldn't have tested otherwise, but as soon as I plugged it in, I could hear something was wrong. The audio was seriously distorted. I would have just trashed it, but I bought it to test something else, and had it apart anyway.

If you would, please PM me the make and model. I might buy one just because its been so long since I saw anything that seriously broken!

Quote:


Long ago, I generated a bunch of full-scale sine waves in FLAC that I use for testing the hardware that I design. On the scope, the 1kHz wave was clearly flat-topped both top and bottom. Reducing the gain of the output buffer was all it needed.

Sounds like someone was not on their toes. An overconfident designer? Tolerance build up? This enquiring mind wants to know! ;-)

Quote:


So maybe the hardware I've been building is not correct? I always leave headroom such that a full-scale signal will never clip, even with worst-case tolerances on the components. Am I being too conservative?

Not at all. It's just that I've been testing too much good hardware lately - everything seems to be working well.
arnyk is online now  
post #137 of 140 Old 04-18-2012, 10:50 PM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
Quote:
Originally Posted by A9X-308 View Post

Most DACs have a DC offset at ~2.5V so the signal swings 0-5V effectively. If Vcc is 5V which it is in USB, and the buffer has any gain then the signal swing sould easily exceed the voltage rails. Even if it didn't, there are fewer opamps that will actually swing signal rail to rail, typically CMOS units. Most will need a volt or so clearance between signal and rail to not clip.

Hi A9X-308,

Exactly right. This was a S/PDIF to analog dongle. The supply was 5 volt, down from a 9 volt AC wall-wart transformer through a bridge and a 7805. Besides the power supply, the only circuitry was an SoC and the output-buffer opamps. I suspect that the opamps were not rail-to-rail.
MarkHotchkiss is offline  
post #138 of 140 Old 04-18-2012, 11:12 PM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
Hi Arny,

PM sent.
MarkHotchkiss is offline  
post #139 of 140 Old 04-19-2012, 06:13 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by MarkHotchkiss View Post

Hi Arny,

PM sent.

Received - awaiting part 2! ;-)
arnyk is online now  
post #140 of 140 Old 04-19-2012, 06:14 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 437 Post(s)
Liked: 1059
Quote:
Originally Posted by MarkHotchkiss View Post

Hi A9X-308,

Exactly right. This was a S/PDIF to analog dongle. The supply was 5 volt, down from a 9 volt AC wall-wart transformer through a bridge and a 7805. Besides the power supply, the only circuitry was an SoC and the output-buffer opamps. I suspect that the opamps were not rail-to-rail.

Someone went cheap on the op amp! :-(
arnyk is online now  
Reply 2 Channel Audio

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off