[Rob Tweed]

A couple of recent upgrades, one low cost, the other somewhat
expensive has got me thinking I've nearly reached audio nirvana.
Thought I'd share it:

Swapped out an Emu 404 USB for the Beresford TC-7520 DAC. An amazing
little box for a ridiculously small amount of money. I'll continue
using the Emu for recording. The Emu is a great DAC but to my ears
the Beresford is a definite improvement.

I use this to connect my Mac/iTunes/Apple Lossless seto-up to a
Behringer A500 power amp (again ridiculously cheap but a great amp).

As per recommendations here, where I've spent the big money is on new
speakers, and I splashed out on a pair of PMC TB2i's. As soon as I
heard them in the store I knew I had to have them, even though they
were considerably more expensive than I was planning on spending.
They just blew everything else I heard away - simply no contest. I
was shocked at the B&W floor-standers I heard: flabby overpowering
bass - what has become of B&W these days? The PMCs are as clean and
uncoloured a sound as I've ever heard, but bass aplenty too despite
their relatively small size.

The combination: as near faultless as I've every heard. A truly great
sound. I think I'll be happy with this set up for quite some time!

BTW no expensive cables or bi-wiring used :-)

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Sonnova]

On Fri, 7 Aug 2009 14:40:25 -0700, Rob Tweed wrote
(in article ):

A couple of recent upgrades, one low cost, the other somewhat
expensive has got me thinking I've nearly reached audio nirvana.
Thought I'd share it:

Swapped out an Emu 404 USB for the Beresford TC-7520 DAC. An amazing
little box for a ridiculously small amount of money. I'll continue
using the Emu for recording. The Emu is a great DAC but to my ears
the Beresford is a definite improvement.

I have one of those, only mine is called a "TEC" . Same model number and the
units are identical. I bought mine on E-bay for $90. It's pretty good but it
has a very pedestrian output stage. It uses an old 70's technology dual
op-amp (I forget the number) and that's its limiting factor. Otherwise, it
has a Crystal CS8414 input receiver and a Burr-Brown PCM 1716 24/96
converter. I bought it because my much more expensive (and admittedly better
sounding) Sonic Frontiers DAC 2.6/D2D upconverter lacks a headphone amp/jack.
I recently upgraded the DAC 2.6 with addition of a pair of the new National
Semiconductor LME49710NA output amps (in place of the original Burr-Brown
OPA134A (All the devices in the Sonic Frontiers converters are socketed
DIPS). Something I cannot do with the TEC because the dual op-amp is a tiny
surface mount part soldered directly to the circuit board.

I use the TEC it to listen to headphones late at night.

I use this to connect my Mac/iTunes/Apple Lossless seto-up to a
Behringer A500 power amp (again ridiculously cheap but a great amp).

I also have one of those! It IS a great sounding amp, as good as some I've
heard that cost TEN TO TWENTY TIMES it's $200 street price from places like
Zounds of Swee****er.

As per recommendations here, where I've spent the big money is on new
speakers, and I splashed out on a pair of PMC TB2i's. As soon as I
heard them in the store I knew I had to have them, even though they
were considerably more expensive than I was planning on spending.
They just blew everything else I heard away - simply no contest. I
was shocked at the B&W floor-standers I heard: flabby overpowering
bass - what has become of B&W these days? The PMCs are as clean and
uncoloured a sound as I've ever heard, but bass aplenty too despite
their relatively small size.

There we diverge. I haven't heard the PMCs. I use Martin-Logan Vantages
instead

The combination: as near faultless as I've every heard. A truly great
sound. I think I'll be happy with this set up for quite some time!

I'm sure you will. Any preamp?


BTW no expensive cables or bi-wiring used :-)

Good for you Bob!

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Rob Tweed]

On 8 Aug 2009 01:57:18 GMT, Sonnova
wrote:


I have one of those, only mine is called a "TEC" . Same model number and the
units are identical. I bought mine on E-bay for $90. It's pretty good but it
has a very pedestrian output stage. It uses an old 70's technology dual
op-amp (I forget the number) and that's its limiting factor. Otherwise, it
has a Crystal CS8414 input receiver and a Burr-Brown PCM 1716 24/96
converter. I bought it because my much more expensive (and admittedly better
sounding) Sonic Frontiers DAC 2.6/D2D upconverter lacks a headphone amp/jack.
I recently upgraded the DAC 2.6 with addition of a pair of the new National
Semiconductor LME49710NA output amps (in place of the original Burr-Brown
OPA134A (All the devices in the Sonic Frontiers converters are socketed
DIPS). Something I cannot do with the TEC because the dual op-amp is a tiny
surface mount part soldered directly to the circuit board.

From what I've read, the Beresford is a modded variant of the one you
have. See thread by cjcoss at
http://www.avforums.com/forums/hi-fi...-7510-a-3.html

You'll see quite a few suggested further modifications along the lines
you suggest at Stanley Beresford's site:
http://www.homehifi.co.uk/main/main.html and you'll also see his new
"Caiman" model which came out just days after I bought the 7250. It
would be interesting to compare it with the 7250.

There we diverge. I haven't heard the PMCs. I use Martin-Logan Vantages
instead

Nice! But somewhat beyond my budget!

Any preamp?

No. The Beresford has a variable output so I have it directly
connected to the Behringer amp.

Connections to the DAC a

- Mac Mini via USB
- Humax Freeview PVR via optical SPDIF
- old Toshiba DVD player via coaxial SPDIF for occasionally playing
CDs.

Here in the UK, BBC Radio 3 is transmitted on one of the digital
terrestrial TV channels (known here as Freeview) as a 192k MP2 signal
(see http://www.digitalradiotech.co.uk/di...o_samples.htm).
The Humax makes it very convenient to record programs such as the live
Proms concerts and the results are excellent.

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Sonnova]

On Sun, 9 Aug 2009 15:02:21 -0700, Rob Tweed wrote
(in article ):

On 8 Aug 2009 01:57:18 GMT, Sonnova
wrote:


I have one of those, only mine is called a "TEC" . Same model number and
the
units are identical. I bought mine on E-bay for $90. It's pretty good but
it
has a very pedestrian output stage. It uses an old 70's technology dual
op-amp (I forget the number) and that's its limiting factor. Otherwise, it
has a Crystal CS8414 input receiver and a Burr-Brown PCM 1716 24/96
converter. I bought it because my much more expensive (and admittedly
better
sounding) Sonic Frontiers DAC 2.6/D2D upconverter lacks a headphone
amp/jack.
I recently upgraded the DAC 2.6 with addition of a pair of the new National
Semiconductor LME49710NA output amps (in place of the original Burr-Brown
OPA134A (All the devices in the Sonic Frontiers converters are socketed
DIPS). Something I cannot do with the TEC because the dual op-amp is a tiny
surface mount part soldered directly to the circuit board.

From what I've read, the Beresford is a modded variant of the one you
have. See thread by cjcoss at

http://www.avforums.com/forums/hi-fi...-dac-dacmagic-
muc
h-better-than-beresford-tc-7510-a-3.html

You'll see quite a few suggested further modifications along the lines
you suggest at Stanley Beresford's site:
http://www.homehifi.co.uk/main/main.html and you'll also see his new
"Caiman" model which came out just days after I bought the 7250. It
would be interesting to compare it with the 7250.

Someone on the forum that you referenced said that the output stage on the
Beresford was different than the one in the TEC unit. I wonder what OP-amp
Bereseford uses? I'd change it in a minute to a LM4562 (the dual OP-amp
equivalent to the LME49710) if the output stage that's in there wasn't a tiny
surface mount part!

There we diverge. I haven't heard the PMCs. I use Martin-Logan Vantages
instead

Nice! But somewhat beyond my budget!

Any preamp?

No. The Beresford has a variable output so I have it directly
connected to the Behringer amp.

And the Behringer amp, of course, has it's own level controls on the front.
8^)

Connections to the DAC a

- Mac Mini via USB

Do you use this as an audio server? I have an Apple TV that I use in that
capacity. It's great!

- Humax Freeview PVR via optical SPDIF
- old Toshiba DVD player via coaxial SPDIF for occasionally playing
CDs.

Here in the UK, BBC Radio 3 is transmitted on one of the digital
terrestrial TV channels (known here as Freeview) as a 192k MP2 signal
(see http://www.digitalradiotech.co.uk/di...o_samples.htm).
The Humax makes it very convenient to record programs such as the live
Proms concerts and the results are excellent.

Sounds like you have a nice system. Of course any system with the Behringer
A500 amp is going to sound mighty good because those amps are so excellent!

[Rob Tweed]

For anyone interested in more info on the Behringer A500, here's an
intesting review I found:

http://theaudiocritic.com/wordpress/?p=23

[ excessive quoting deleted -- dsr ]

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Rob Tweed]

On 10 Aug 2009 04:40:16 GMT, Sonnova
wrote:

Someone on the forum that you referenced said that the output stage on the
Beresford was different than the one in the TEC unit. I wonder what OP-amp
Bereseford uses? I'd change it in a minute to a LM4562 (the dual OP-amp
equivalent to the LME49710) if the output stage that's in there wasn't a tiny
surface mount part!

Found lots of modding info for the Beresford he

http://theartofsound.net/forum/showthread.php?t=2352

Looks like the 7520 (the DAC I have) is easier to upgrade than the
7510 model you have as the opamps are socketed in the 7520, so it
looks like it's a simple drop-in upgrade to the LM4562s. Quite a few
other opamps have been tried out and recommended in this thread too.

I've read that the stock opamps in the 7520 are 5532's

So it looks like I'll be ordering a pair of LM4562's and popping them
in to hear the difference.

Connections to the DAC a

- Mac Mini via USB

Do you use this as an audio server? I have an Apple TV that I use in that
capacity. It's great!

Yes. I use iTunes on the Mac. I have a 400Gb USB hard drive attached
to the Mac Mini and have all my CDs (and many LPs transferred to
digital too) on the drive. Although I started with straight WAV
ripping, I now use Apple Lossless. As the disc is now full, every now
and again I'll convert a few WAV tracks to lossless to gain some more
room. 10,000 tracks and counting now :-)


Sounds like you have a nice system. Of course any system with the Behringer
A500 amp is going to sound mighty good because those amps are so excellent!

Interesting video he

http://www.youtube.com/watch?v=RWj5CUl7rio

Looks like it's perhaps not such a good idea to use the Behringer's
gain controls!

Can't say I noticed any real audible difference to be honest - indeed
with zero input on the DAC and full gain on the Behringer, I get a
quiet hum through the speakers...a bit of an earthing issue perhaps?

I do have to say about the Behringer amp, I had to send two back
before I got one I was happy with. The first developed an audible
mechanical hum after a few days and the second developed a dangerously
crackly gain control after a few days. My conclusion was that the
price you pay for such a ludicrously low price was poor quality
control. However the retailer I used here in the UK (Digital Village)
was excellent and swapped the units without question - they even paid
for the couriers each time which impressed me greatly.

Anyway the A500 I have now has run faultlessly for a year and a half.

Yes, so all in all, a great sound now and I'd recommend this
combination of components to anyone interested in experiencing serious
high end sound on a relatively small budget (about 1600 UK pounds in
total for the DAC, power amp and speakers). I exclude the cost of the
Mac as I had it anyway and was surplus to work requirements.

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Codifus]

On Aug 7, 5:40*pm, Rob Tweed wrote:
A couple of recent upgrades, one low cost, the other somewhat
expensive has got me thinking I've nearly reached audio nirvana.
Thought I'd share it:

Swapped out an Emu 404 USB for the Beresford TC-7520 DAC. *An amazing
little box for a ridiculously small amount of money. *I'll continue
using the Emu for recording. *The Emu is a great DAC but to my ears
the Beresford is a definite improvement.

I use this to connect my Mac/iTunes/Apple Lossless seto-up to a
Behringer A500 power amp (again ridiculously cheap but a great amp).

As per recommendations here, where I've spent the big money is on new
speakers, and I splashed out on a pair of PMC TB2i's. *As soon as I
heard them in the store I knew I had to have them, even though they
were considerably more expensive than I was planning on spending.
They just blew everything else I heard away - simply no contest. *I
was shocked at the B&W floor-standers I heard: flabby overpowering
bass - what has become of B&W these days? *The PMCs are as clean and
uncoloured a sound as I've ever heard, but bass aplenty too despite
their relatively small size.

The combination: as near faultless as I've every heard. *A truly great
sound. *I think I'll be happy with this set up for quite some time!

BTW no expensive cables or bi-wiring used :-)

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site:http://www.mgateway.com

What a co-incidence. You were the one whose opinion made me get the
EMU-0404 and now I also have the TC-7520. I have even done the
LM4562NA upgrade with the MLC5/6 caps removal. The sonic improvements
with the LM4562 upgrade are perfect......almost!

For some reason, and I think if you'll read between the lines at the
AOS forums, the LM4562 with caps removed makes the bass a wee bit
lighter than normal. This is when using the line-out to a stereo, not
headphone use. From about 40 Hz and below there is definite
attenuation in my system. If not for that one minor issue my yearn to
upgrade would be satisfied.

Stan has mentioned yet another possible PSU upgrade for the 7520/
Caiman and hopefully that will bring the bass back. Interestingly,
people who have the Caiman with the LM4562 upgrade have observed that
the bass has returned, with heft.

Comparing the EMU-0404 to the 7520, what I noticed as the biggest
improvement was the detail/resolution. Listening to the 7520 I close
my eyes and not only hear the instruments but can practically point
them out individually in space. The EMU gave you everything, but it
was somewhat blended together. It all sounds very very good, but
resolution was a bit lacking. On another forum, someone described the
sound of delta-sigma DACs as substituting frequency resolution for
amplitude resolution, and that they always sound too polite. After
reading that observation I realized that he was in fact describing the
EMU-0404 to a tee.

Anyhow, I continue in search of the ideal opamp for my 7520.

CD

[Sonnova]

On Mon, 10 Aug 2009 06:01:35 -0700, Rob Tweed wrote
(in article ):

On 10 Aug 2009 04:40:16 GMT, Sonnova
wrote:

Someone on the forum that you referenced said that the output stage on the
Beresford was different than the one in the TEC unit. I wonder what OP-amp
Bereseford uses? I'd change it in a minute to a LM4562 (the dual OP-amp
equivalent to the LME49710) if the output stage that's in there wasn't a
tiny
surface mount part!

Found lots of modding info for the Beresford he

http://theartofsound.net/forum/showthread.php?t=2352

Looks like the 7520 (the DAC I have) is easier to upgrade than the
7510 model you have as the opamps are socketed in the 7520, so it
looks like it's a simple drop-in upgrade to the LM4562s. Quite a few
other opamps have been tried out and recommended in this thread too.

I've read that the stock opamps in the 7520 are 5532's

So it looks like I'll be ordering a pair of LM4562's and popping them
in to hear the difference.

The LME49710NAs certainly made an audible difference in my Sonic Frontiers
DAC 2.6!

Connections to the DAC a

- Mac Mini via USB

Do you use this as an audio server? I have an Apple TV that I use in that
capacity. It's great!

Yes. I use iTunes on the Mac. I have a 400Gb USB hard drive attached
to the Mac Mini and have all my CDs (and many LPs transferred to
digital too) on the drive. Although I started with straight WAV
ripping, I now use Apple Lossless. As the disc is now full, every now
and again I'll convert a few WAV tracks to lossless to gain some more
room. 10,000 tracks and counting now :-)

Wow, that's a lot! I have nowhere near that many tracks on my Apple
TV/iTunes.


Sounds like you have a nice system. Of course any system with the Behringer
A500 amp is going to sound mighty good because those amps are so excellent!

Interesting video he

http://www.youtube.com/watch?v=RWj5CUl7rio

Looks like it's perhaps not such a good idea to use the Behringer's
gain controls!

Can't say I noticed any real audible difference to be honest - indeed
with zero input on the DAC and full gain on the Behringer, I get a
quiet hum through the speakers...a bit of an earthing issue perhaps?

The guy in the video said that at it's worst, the distortion went up to about
0.4-0.5 %. I do mot consider this "unacceptable listening distortion" as it
seems as if the ear is not real sensitive to this kind of distortion which is
probably why neither of us have noticed it. There used to be a very expensive
French tube amplifier (I forget it's name) that TAS was nuts about. It
measured almost 2.0% THD and that didn't keep it from sounding excellent. I
heard one once, I never noticed any distortion!

I do have to say about the Behringer amp, I had to send two back
before I got one I was happy with. The first developed an audible
mechanical hum after a few days and the second developed a dangerously
crackly gain control after a few days. My conclusion was that the
price you pay for such a ludicrously low price was poor quality
control. However the retailer I used here in the UK (Digital Village)
was excellent and swapped the units without question - they even paid
for the couriers each time which impressed me greatly.

The first one that Behringer sent me was likewise defective and hummed, but
the second one was perfect.

[quoted text deleted -- deb]

[Rob Tweed]

On 10 Aug 2009 13:01:35 GMT, Rob Tweed wrote:

Yes, so all in all, a great sound now and I'd recommend this
combination of components to anyone interested in experiencing serious
high end sound on a relatively small budget (about 1600 UK pounds in
total for the DAC, power amp and speakers). I exclude the cost of the
Mac as I had it anyway and was surplus to work requirements.

Actually I got that wrong - should be 1400 UK pounds. OK throw in the
cost of some decent speaker stands (I picked up some Atacama ones
second hand a few years ago on EBay - filled the tubes with sand to
deaden them)....and cables and call it 1500 :-)

Regarding the Behringer amp, further research makes me wonder whether
the EP2500 isn't a better and more reliable (and apparently a very
highly regarded) option
(http://forums.audioholics.com/forums...d.php?t=35767). Only
potential problem is an apparently noisy fan, though mods are
possible, eg see
http://forums.audioholics.com/forums...ad.php?t=45137

The EP2500 is about twice the price of the A500: about 280 UK pounds
here - still a ludicrously low price for what appears to be quite a
monster.



---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Sonnova]

On Mon, 10 Aug 2009 15:57:54 -0700, codifus wrote
(in article ):

On Aug 7, 5:40*pm, Rob Tweed wrote:
A couple of recent upgrades, one low cost, the other somewhat
expensive has got me thinking I've nearly reached audio nirvana.
Thought I'd share it:

Swapped out an Emu 404 USB for the Beresford TC-7520 DAC. *An amazing
little box for a ridiculously small amount of money. *I'll continue
using the Emu for recording. *The Emu is a great DAC but to my ears
the Beresford is a definite improvement.

I use this to connect my Mac/iTunes/Apple Lossless seto-up to a
Behringer A500 power amp (again ridiculously cheap but a great amp).

As per recommendations here, where I've spent the big money is on new
speakers, and I splashed out on a pair of PMC TB2i's. *As soon as I
heard them in the store I knew I had to have them, even though they
were considerably more expensive than I was planning on spending.
They just blew everything else I heard away - simply no contest. *I
was shocked at the B&W floor-standers I heard: flabby overpowering
bass - what has become of B&W these days? *The PMCs are as clean and
uncoloured a sound as I've ever heard, but bass aplenty too despite
their relatively small size.

The combination: as near faultless as I've every heard. *A truly great
sound. *I think I'll be happy with this set up for quite some time!

BTW no expensive cables or bi-wiring used :-)

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site:http://www.mgateway.com

What a co-incidence. You were the one whose opinion made me get the
EMU-0404 and now I also have the TC-7520. I have even done the
LM4562NA upgrade with the MLC5/6 caps removal. The sonic improvements
with the LM4562 upgrade are perfect......almost!

For some reason, and I think if you'll read between the lines at the
AOS forums, the LM4562 with caps removed makes the bass a wee bit
lighter than normal. This is when using the line-out to a stereo, not
headphone use. From about 40 Hz and below there is definite
attenuation in my system. If not for that one minor issue my yearn to
upgrade would be satisfied.

Stan has mentioned yet another possible PSU upgrade for the 7520/
Caiman and hopefully that will bring the bass back. Interestingly,
people who have the Caiman with the LM4562 upgrade have observed that
the bass has returned, with heft.

Comparing the EMU-0404 to the 7520, what I noticed as the biggest
improvement was the detail/resolution. Listening to the 7520 I close
my eyes and not only hear the instruments but can practically point
them out individually in space. The EMU gave you everything, but it
was somewhat blended together. It all sounds very very good, but
resolution was a bit lacking. On another forum, someone described the
sound of delta-sigma DACs as substituting frequency resolution for
amplitude resolution, and that they always sound too polite. After
reading that observation I realized that he was in fact describing the
EMU-0404 to a tee.

Anyhow, I continue in search of the ideal opamp for my 7520.

When you installed the LM4562s did you notice any improvement in the noise
floor? When I replaced the Burr-Brown OP-amps in my DAC 2.6, i certainly
notice that the music appears from a much "blacker" background than before. I
wonder if you noticed a similar phenomenon?

CD

[Codifus]

On Aug 10, 11:08*pm, Sonnova wrote:
Anyhow, I continue in search of the ideal opamp for my 7520.

When you installed the LM4562s did you notice any improvement in the noise
floor? When I replaced the Burr-Brown OP-amps in my DAC 2.6, i certainly
notice that the music appears from a much "blacker" background than before. I
wonder if you noticed a similar phenomenon?



CD

That was the other thing I noticed. Definitely a blacker background on
the 7520 when compared to the 0404. This improved noise floor was
apparent even before I started upgrading. The LM4562NA upgrade
improved the musicality of the 7520.

CD

[Rob Tweed]

On 10 Aug 2009 22:57:54 GMT, codifus wrote:

Anyhow, I continue in search of the ideal opamp for my 7520.

I think it's great that Stan has seen fit to socket the opamps in the
7520 to allow this kind of experimentation (a bit like the old
turntable tweaking of old, only now in the digital domain!):

Interesting observation and comparison with Benchmark here
(http://community.whathifi.com/forums...howThread.aspx) :


"You have to wonder why, if this op-amp swap is such a revelation,
[Beresford] doesn't use them in the first place. Don't you?"

to which a response was....:

"Yes I think the same ought to be asked of Benchmark too.

They did not start using the LM4562 until the £1300 DAC1 Pre. (The
£900 Benchmark DAC1 USB had the same standard NE5532 opamps that the
Beresford gets shipped with.)

The difference is that Beresford made it convenient and easy to
upgrade. "



---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Sonnova]

On Tue, 11 Aug 2009 06:02:38 -0700, codifus wrote
(in article ):

On Aug 10, 11:08*pm, Sonnova wrote:
Anyhow, I continue in search of the ideal opamp for my 7520.

When you installed the LM4562s did you notice any improvement in the noise
floor? When I replaced the Burr-Brown OP-amps in my DAC 2.6, i certainly
notice that the music appears from a much "blacker" background than before.
I
wonder if you noticed a similar phenomenon?



CD

That was the other thing I noticed. Definitely a blacker background on
the 7520 when compared to the 0404. This improved noise floor was
apparent even before I started upgrading. The LM4562NA upgrade
improved the musicality of the 7520.

CD

Yes, I noticed that as well.

[Sonnova]

On Tue, 11 Aug 2009 06:15:58 -0700, Rob Tweed wrote
(in article ):

On 10 Aug 2009 22:57:54 GMT, codifus wrote:

Anyhow, I continue in search of the ideal opamp for my 7520.

I think it's great that Stan has seen fit to socket the opamps in the
7520 to allow this kind of experimentation (a bit like the old
turntable tweaking of old, only now in the digital domain!):

Interesting observation and comparison with Benchmark here
(http://community.whathifi.com/forums...howThread.aspx) :


"You have to wonder why, if this op-amp swap is such a revelation,
[Beresford] doesn't use them in the first place. Don't you?"

to which a response was....:

"Yes I think the same ought to be asked of Benchmark too.

They did not start using the LM4562 until the £1300 DAC1 Pre. (The
£900 Benchmark DAC1 USB had the same standard NE5532 opamps that the
Beresford gets shipped with.)

The difference is that Beresford made it convenient and easy to
upgrade. "

[quoted sig deleted -- deb]

Yes, the LM4562/LME49710 is a significant improvement over earlier designs in
such areas as noise (only 2.5 nanovolts/root Hz typical) and symmetrical slew
(meaning that the slew rate is identical for both the positive and the
negative-going halves of the waveform) and extremely low total distortion. In
most op-amp designs, the slew is asymmetrical because the path through the
op-amp itself, for the negative going half of the waveform, is usually longer
than the path for the positive-going half. This is probably not important for
other than audio applications, but it is quite important for audio for the
slew to be symmetrical for bot halves of the waveform. IOW, no audio designer
worth his salt would ever design a discrete-component amp with the
characteristic of an asymmetrical slew rate. The actual slew rate of the
LM4562/LME49710 isn't all that out of the ordinary at 20 volts/microsecond
(the Burr-Brown OPA134As that I replaced with the LME49710s were also 20
volts/microsecond slew and 8 nanovolts/root Herz of noise typically).

[Codifus]

I've tried the LME49720s, the dual opamp versions, that is equivalent
to the LM4562NAs. From my basic understanding, the LME49720s are
basically a metal can version of the LM4562NAs. Sonically, though,
they are very different.

Keep in mind that my observations are with my TC-7520 with the MLC5/6
caps removed, as shown in the 2nd diagram from mod page here;

http://www.beresford.me/others/7520mods.html


For me, the LM4562NAs really really shine. They are almost perfect,
but their bass is just a bit light.

The LME49720s, which are basically a metal can version of the
LM4562NAs, are quite a different story. They have deep bass, but they
don't have the musicality of the LM4562NAs. Not even close. It's funny
how the plastic "inferior" quality chip out shines the "better" metal
can version. I think it all comes down to the synergy of all the
components. The TC-7520 may be better suited to plastic opamps.

I plan to try the Burr brown OPA2134s next. I hope they take well to
the MLC5/6 removal. Putting those caps back is not really an option
for me.


CD

[Sonnova]

On Tue, 11 Aug 2009 20:10:08 -0700, codifus wrote
(in article ):

I've tried the LME49720s, the dual opamp versions, that is equivalent
to the LM4562NAs. From my basic understanding, the LME49720s are
basically a metal can version of the LM4562NAs. Sonically, though,
they are very different.

Keep in mind that my observations are with my TC-7520 with the MLC5/6
caps removed, as shown in the 2nd diagram from mod page here;

http://www.beresford.me/others/7520mods.html


For me, the LM4562NAs really really shine. They are almost perfect,
but their bass is just a bit light.

The LME49720s, which are basically a metal can version of the
LM4562NAs, are quite a different story. They have deep bass, but they
don't have the musicality of the LM4562NAs. Not even close. It's funny
how the plastic "inferior" quality chip out shines the "better" metal
can version. I think it all comes down to the synergy of all the
components. The TC-7520 may be better suited to plastic opamps.

I plan to try the Burr brown OPA2134s next. I hope they take well to
the MLC5/6 removal. Putting those caps back is not really an option
for me.

HMM. The OPA2134 is essentially the same as the OPA134 in specs:
20V/microsecond slew rate, 8 MHz gain bandwidth, 8 nanovolts/root Hertz
noise. These are old op-amps (10 years old) and have much more noise (8
nanovolts/root Hertz vs 2.5) than do the LME49710 family. I also believe that
the Burr-Brown op-amps do not have symmetrical slew (I deduced this from the
large signal step response oscilloscope photo in the data sheet. The falling
edge of the square wave exhibits an offset of almost 4 microseconds at the
'0' crossing point. An offset that is not there on the rising edge. This is
indicative that the falling edge (negative going) portion of the signal takes
a longer path through the IC than does the positive going rising edge) and
the Burr-Brown OPAs have orders of magnitude more THD than do the newer
National chips. I replaced OPA134s in my DAC with the Natty chips and noticed
a big improvement. It costs little to experiment this way, but in my humble
opinion, replacing members of the LME49710 family with members of the OPA134
family is taking a big step backwards. Your milage may vary 8^)

[Codifus]

On Aug 12, 8:58*am, Sonnova wrote:
On Tue, 11 Aug 2009 20:10:08 -0700, codifus wrote
(in article ):



I've tried the LME49720s, the dual opamp versions, that is equivalent
to the LM4562NAs. From my basic understanding, the LME49720s are
basically a metal can version of the LM4562NAs. Sonically, though,
they are very different.

Keep in mind that my observations are with my TC-7520 with the MLC5/6
caps removed, as shown in the 2nd diagram from mod page here;

http://www.beresford.me/others/7520mods.html

For me, the LM4562NAs really really shine. They are almost perfect,
but their bass is just a bit light.

The LME49720s, which are basically a metal can version of the
LM4562NAs, are quite a different story. They have deep bass, but they
don't have the musicality of the LM4562NAs. Not even close. It's funny
how the plastic "inferior" quality chip out shines the "better" metal
can version. I think it all comes down to the synergy of all the
components. The TC-7520 may be better suited to plastic opamps.

I plan to try the Burr brown OPA2134s next. I hope they take well to
the MLC5/6 removal. Putting those caps back is not really an option
for me.

HMM. The OPA2134 is essentially the same as the OPA134 in specs:
20V/microsecond slew rate, 8 MHz gain bandwidth, 8 nanovolts/root Hertz
noise. These are old op-amps (10 years old) and have much more noise (8
nanovolts/root Hertz vs 2.5) than do the LME49710 family. I also believe that
the Burr-Brown op-amps do not have symmetrical slew (I deduced this from the
large signal step response oscilloscope photo in the data sheet. The falling
edge of the square wave exhibits an offset of almost 4 microseconds at the
'0' crossing point. An offset that is not there on the rising edge. This is
indicative that the falling edge (negative going) portion of the signal takes
a longer path through the IC than does the positive going rising edge) and
the Burr-Brown OPAs have orders of magnitude more THD than do the newer
National chips. I replaced OPA134s in my DAC with the Natty chips and noticed
a big improvement. It costs little to experiment this way, but in my humble
opinion, replacing members of the LME49710 family with members of the OPA134
family is taking a big step backwards. Your milage may vary 8^)

On a technical front, perhaps it may be a step backwards to try the
OPA2134, but I'm listening to the music, not looking at waveforms on
an oscilloscope

Can you recommend other Opamps I could try? I've picked up from AOS
that I should avoid high slew rate opamps like the THS4032. They don't
seem to fit well on my 7520 with the MLC5/6 caps removed. Despite the
very positive reviews of the 4032, I have been avoiding it because it
is specified that it is not ideally suited for audio only
applications, something about lacking unity gain.

Also, Beresford.com, the american distributor of Beresford DACs, is
working on a discrete opamp upgrade for the 7520. I can not wait to
try it. It's going to be a bit pricey (approx nearly half the cost of
the actual unit), but honestly, I'm not totally enjoying the 7520 at
the moment. The LM4562NA upgrade has a sweet, sweet, top end, from
about 40 Hz on up it sounds amazing. Anything below 40 Hz leaves me
wanting, though. Just not enough heft that I'm used to. The TC-7520
with the original opamps (NE5532s) has plenty of heft, but the
midrange was a bit lifeless.

I have even started using the EQ in iTunes just to compensate for the
LM4562NAs, and I hate adding to the signal path

Technically, discrete opamps also spec out to be inferior to the
modern integrated opamps like the LM4562NA, but many audio enthusiasts
have tried them, like the Burson series, and haven't looked back.

In fact, have a look at this article, you'll find it very interesting;

http://bursonaudio.com/burson_opamp_pc_tuner_review.htm

CD

[Sonnova]

On Wed, 12 Aug 2009 07:59:06 -0700, codifus wrote
(in article ):

On Aug 12, 8:58*am, Sonnova wrote:
On Tue, 11 Aug 2009 20:10:08 -0700, codifus wrote
(in article ):



I've tried the LME49720s, the dual opamp versions, that is equivalent
to the LM4562NAs. From my basic understanding, the LME49720s are
basically a metal can version of the LM4562NAs. Sonically, though,
they are very different.

Keep in mind that my observations are with my TC-7520 with the MLC5/6
caps removed, as shown in the 2nd diagram from mod page here;

http://www.beresford.me/others/7520mods.html

For me, the LM4562NAs really really shine. They are almost perfect,
but their bass is just a bit light.

The LME49720s, which are basically a metal can version of the
LM4562NAs, are quite a different story. They have deep bass, but they
don't have the musicality of the LM4562NAs. Not even close. It's funny
how the plastic "inferior" quality chip out shines the "better" metal
can version. I think it all comes down to the synergy of all the
components. The TC-7520 may be better suited to plastic opamps.

I plan to try the Burr brown OPA2134s next. I hope they take well to
the MLC5/6 removal. Putting those caps back is not really an option
for me.

HMM. The OPA2134 is essentially the same as the OPA134 in specs:
20V/microsecond slew rate, 8 MHz gain bandwidth, 8 nanovolts/root Hertz
noise. These are old op-amps (10 years old) and have much more noise (8
nanovolts/root Hertz vs 2.5) than do the LME49710 family. I also believe
that
the Burr-Brown op-amps do not have symmetrical slew (I deduced this from the
large signal step response oscilloscope photo in the data sheet. The falling
edge of the square wave exhibits an offset of almost 4 microseconds at the
'0' crossing point. An offset that is not there on the rising edge. This is
indicative that the falling edge (negative going) portion of the signal
takes
a longer path through the IC than does the positive going rising edge) and
the Burr-Brown OPAs have orders of magnitude more THD than do the newer
National chips. I replaced OPA134s in my DAC with the Natty chips and
noticed
a big improvement. It costs little to experiment this way, but in my humble
opinion, replacing members of the LME49710 family with members of the OPA134
family is taking a big step backwards. Your milage may vary 8^)

On a technical front, perhaps it may be a step backwards to try the
OPA2134, but I'm listening to the music, not looking at waveforms on
an oscilloscope

Ah, but while good on paper doesn't necessarily translate to good on music,
bad on paper often does. At any rate my experience is that the National chips
sound better than the older Burr-Brown OPA134 family.

Can you recommend other Opamps I could try? I've picked up from AOS
that I should avoid high slew rate opamps like the THS4032. They don't
seem to fit well on my 7520 with the MLC5/6 caps removed. Despite the
very positive reviews of the 4032, I have been avoiding it because it
is specified that it is not ideally suited for audio only
applications, something about lacking unity gain.

The problem with many op-amps is that they need external compensation to be
stable under audio conditions and therefore are not pin-for-pin replacements
for the GP op amp families that we are talking about. Be careful. But the
THS4032 looks pretty good. It is symmetrical in slew, very fast, with
extremely low noise. It also COULD be unstable in audio applications due to
its wide bandwidth unless externally compensated (the data sheet shows how).
But I don't know for sure that it would be necessary. OTOH, I don't see how
you can use it as it' seems to only be available in surface mount packages.

Also, Beresford.com, the american distributor of Beresford DACs, is
working on a discrete opamp upgrade for the 7520. I can not wait to
try it. It's going to be a bit pricey (approx nearly half the cost of
the actual unit), but honestly, I'm not totally enjoying the 7520 at
the moment. The LM4562NA upgrade has a sweet, sweet, top end, from
about 40 Hz on up it sounds amazing. Anything below 40 Hz leaves me
wanting, though. Just not enough heft that I'm used to. The TC-7520
with the original opamps (NE5532s) has plenty of heft, but the
midrange was a bit lifeless.

I have even started using the EQ in iTunes just to compensate for the
LM4562NAs, and I hate adding to the signal path

Technically, discrete opamps also spec out to be inferior to the
modern integrated opamps like the LM4562NA, but many audio enthusiasts
have tried them, like the Burson series, and haven't looked back.

In fact, have a look at this article, you'll find it very interesting;

http://bursonaudio.com/burson_opamp_pc_tuner_review.htm

Thanks, I will.

[John Stone]

On 8/12/09 7:58 AM, in article , "Sonnova"
wrote:

HMM. The OPA2134 is essentially the same as the OPA134 in specs:
20V/microsecond slew rate, 8 MHz gain bandwidth, 8 nanovolts/root Hertz
noise. These are old op-amps (10 years old) and have much more noise (8
nanovolts/root Hertz vs 2.5) than do the LME49710 family. I also believe that
the Burr-Brown op-amps do not have symmetrical slew (I deduced this from the
large signal step response oscilloscope photo in the data sheet. The falling
edge of the square wave exhibits an offset of almost 4 microseconds at the
'0' crossing point. An offset that is not there on the rising edge. This is
indicative that the falling edge (negative going) portion of the signal takes
a longer path through the IC than does the positive going rising edge) and
the Burr-Brown OPAs have orders of magnitude more THD than do the newer
National chips. I replaced OPA134s in my DAC with the Natty chips and noticed
a big improvement. It costs little to experiment this way, but in my humble
opinion, replacing members of the LME49710 family with members of the OPA134
family is taking a big step backwards. Your milage may vary 8^)

Discussing the theoretical performance of one op-amp vs another is all well
and good, but this is all based on spec sheet evaluation, not real life
application. What is most relevant is the performance of the op-amp in the
actual circuit application, which itself is determined heavily by the
components and circuitry used around the chip. And while you make a case
that the distortion spec in the National chip is "orders of magnitude lower"
than that of the Burr Brown,(.00003% vs.00008%) the fact is that in either
case we are talking about distortion levels that are way below any possible
level of audibility.

If I look at technical reviews of highly respected DAC's such as the
Benchmark DAC1 or Logitech Transporter, what I see is SOTA performance in
virtually every significant performance aspect being measured. Ruler flat
response, near perfect linearity, miniscule levels of THD and IM even into
low impedance loads, and noise levels well below the noise floor of the
signal itself. These DACs also get high marks from subjective reviewers. Yet
they accomplish that performance using the lowly, ancient NE5532 in the
analog stage.

So my question is, what accounts for the substantial improvement you claim
to hear by substituting excellent performing op amps with really excellent
performing op amps? You mention symmetrical slewing, but realistically, if
this was an actual problem, it would have to show up somewhere else in the
specs. I doubt any of these chips come anywhere close to being pushed into
slew induced distortion.

[Sonnova]

On Wed, 12 Aug 2009 16:22:12 -0700, John Stone wrote
(in article ):

On 8/12/09 7:58 AM, in article , "Sonnova"
wrote:

HMM. The OPA2134 is essentially the same as the OPA134 in specs:
20V/microsecond slew rate, 8 MHz gain bandwidth, 8 nanovolts/root Hertz
noise. These are old op-amps (10 years old) and have much more noise (8
nanovolts/root Hertz vs 2.5) than do the LME49710 family. I also believe
that
the Burr-Brown op-amps do not have symmetrical slew (I deduced this from the
large signal step response oscilloscope photo in the data sheet. The falling
edge of the square wave exhibits an offset of almost 4 microseconds at the
'0' crossing point. An offset that is not there on the rising edge. This is
indicative that the falling edge (negative going) portion of the signal
takes
a longer path through the IC than does the positive going rising edge) and
the Burr-Brown OPAs have orders of magnitude more THD than do the newer
National chips. I replaced OPA134s in my DAC with the Natty chips and
noticed
a big improvement. It costs little to experiment this way, but in my humble
opinion, replacing members of the LME49710 family with members of the OPA134
family is taking a big step backwards. Your milage may vary 8^)

Discussing the theoretical performance of one op-amp vs another is all well
and good, but this is all based on spec sheet evaluation, not real life
application. What is most relevant is the performance of the op-amp in the
actual circuit application, which itself is determined heavily by the
components and circuitry used around the chip. And while you make a case
that the distortion spec in the National chip is "orders of magnitude lower"
than that of the Burr Brown,(.00003% vs.00008%) the fact is that in either
case we are talking about distortion levels that are way below any possible
level of audibility.

You put too many zeros in the 0.08 number.

If I look at technical reviews of highly respected DAC's such as the
Benchmark DAC1 or Logitech Transporter, what I see is SOTA performance in
virtually every significant performance aspect being measured. Ruler flat
response, near perfect linearity, miniscule levels of THD and IM even into
low impedance loads, and noise levels well below the noise floor of the
signal itself. These DACs also get high marks from subjective reviewers. Yet
they accomplish that performance using the lowly, ancient NE5532 in the
analog stage.

So my question is, what accounts for the substantial improvement you claim
to hear by substituting excellent performing op amps with really excellent
performing op amps?

Darker silence, and cleaner, more silken highs. Others have noted this as
well. I wished I had another DAC 2.6 so that I could do a real DB test on
them. The best I could do is take turns with a friend where one of us would
swap the op-amps (or not) while the other was out of room. Upon returning
neither of us had any trouble discerning which op-amp set we were listening
to, or whether the op-amps were swapped at all. I know that this kind of test
is not very accurate because ONE of us knows which is which and there is
always the possibility of body language or other cues giving away what was
done. But if so, I'm not aware of it. Still, a DBT is preferable.

You mention symmetrical slewing, but realistically, if
this was an actual problem, it would have to show up somewhere else in the
specs. I doubt any of these chips come anywhere close to being pushed into
slew induced distortion.

Slew induced distortion and asymmetrical slew are not the same thing. Nor am
I intimating that slew induced distortion is at work in any of these designs.
Asymmetrical slew indicates that the path a negative-slewing signal takes
through the op-amp is longer than the path the positive-slewing signal takes.
This makes the fall time of a square wave lopsided. I.E. the rising edge is
more vertical than the falling edge. In the case of the Burr-Brown OPA
family, the picture shown of a oscilloscope photo of a large swing square
wave centered about "0" volts shows a sharp, non-linear break in the falling
edge of the square wave as the waveform crosses the zero-volt reference line.
Not only is the rate of slew slower at that point, but it is non-linear as
well; starting out much slower than on the positive side and then getting
more vertical as the negative swing increases. This shows a characteristic
non-linearity in the long-tailed pair current source for the negative half of
the op-amp. While probably not significant audibly, it does nonetheless
illustrate that the OPA134 family is an older design than the National family
and COULD account for the cleaner highs heard with the newer op amps

[Arny Krueger]

"John Stone" wrote in message

On 8/12/09 7:58 AM, in article
, "Sonnova"
wrote:

HMM. The OPA2134 is essentially the same as the OPA134
in specs: 20V/microsecond slew rate, 8 MHz gain
bandwidth, 8 nanovolts/root Hertz noise. These are old
op-amps (10 years old) and have much more noise (8
nanovolts/root Hertz vs 2.5) than do the LME49710
family. I also believe that the Burr-Brown op-amps do
not have symmetrical slew (I deduced this from the large
signal step response oscilloscope photo in the data
sheet. The falling edge of the square wave exhibits an
offset of almost 4 microseconds at the '0' crossing
point. An offset that is not there on the rising edge.
This is indicative that the falling edge (negative
going) portion of the signal takes a longer path through
the IC than does the positive going rising edge) and the
Burr-Brown OPAs have orders of magnitude more THD than
do the newer National chips. I replaced OPA134s in my
DAC with the Natty chips and noticed a big improvement.
It costs little to experiment this way, but in my humble
opinion, replacing members of the LME49710 family with
members of the OPA134 family is taking a big step
backwards. Your milage may vary 8^)

Discussing the theoretical performance of one op-amp vs
another is all well and good, but this is all based on
spec sheet evaluation, not real life application. What is
most relevant is the performance of the op-amp in the
actual circuit application, which itself is determined
heavily by the components and circuitry used around the
chip. And while you make a case that the distortion spec
in the National chip is "orders of magnitude lower" than
that of the Burr Brown,(.00003% vs.00008%) the fact is
that in either case we are talking about distortion
levels that are way below any possible level of
audibility.

I suspect that we are actually seeing the results of a war between Op Amp
and DAC chip designers, to see who is the weakest link. At this point it is
very much a war of numbers, because even cheap DACs and op amps are
overkill, if actual sound quality while listening to music is the criteria.

If I look at technical reviews of highly respected DAC's
such as the Benchmark DAC1 or Logitech Transporter, what
I see is SOTA performance in virtually every significant
performance aspect being measured. Ruler flat response,
near perfect linearity, miniscule levels of THD and IM
even into low impedance loads, and noise levels well
below the noise floor of the signal itself. These DACs
also get high marks from subjective reviewers. Yet they
accomplish that performance using the lowly, ancient
NE5532 in the analog stage.

The NE5532 has been around for what must be 30 years now. It shows up all
over the place. While it only costs about $0.50 in small quantities that any
basement manufacturer can benefit from, its still overkill for most
applications where it is used.

So my question is, what accounts for the substantial
improvement you claim to hear by substituting excellent
performing op amps with really excellent performing op
amps?

Expectation bias, pure and simple.

You mention symmetrical slewing, but realistically,
if this was an actual problem, it would have to show up
somewhere else in the specs. I doubt any of these chips
come anywhere close to being pushed into slew induced
distortion.

That's easy to figure - just calculate the peak-peak voltage at 2.8 times
RMS, and multiply by 6.28 times the highest frequency of interest. 2 volts
per microsecond is overkill for any consumer audio application, and not by a
little.

[Rob Tweed]

A somewhat cheeky observation:

- we seem to be saying that minute differences in the distortion and
noise characteristics of opamps make clearly audible differences to
the sound of the DAC, and yet quite big measurable distortion effects
in the Behringer A500 aren't going to have any audible consequences?

Anyway, be that as it may....I modded my Beresford DAC with the
LM4562NAs. I *believe* it changed the sound - not such splashy treble
and the sound isn't quite so "in your face" is how I'd describe it
now. Bass seems fine IMO. In fact I'm not sure I didn't prefer the
sound with the original opamps! :-)

However I'd say any audible difference is pretty subtle, and nowhere
near as marked as the improvements I've achieved with getting the PMC
speakers.

Whatever the differences related to the opamps, real or imagined, I'm
still VERY happy with the setup I now have.


---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Codifus]

On Aug 13, 3:42*pm, Rob Tweed wrote:
A somewhat cheeky observation:

- we seem to be saying that minute differences in the distortion and
noise characteristics of opamps make clearly audible differences to
the sound of the DAC, and yet quite big measurable distortion effects
in the Behringer A500 aren't going to have any audible consequences?

Keep in mind that when the signal is travelling thru the opamp, it is
very very small. That electrical audio signal, having been slightly
bumped up by the opamp, is then massively amplified by your Behringer
to drive the speakers. Consequences on the audio are bound to vary.


Anyway, be that as it may....I modded my Beresford DAC with the
LM4562NAs. *I *believe* it changed the sound - not such splashy treble
and the sound isn't quite so "in your face" is how I'd describe it
now. *Bass seems fine IMO. *In fact I'm not sure I didn't prefer the
sound with the original opamps! :-) *

Did you remove the MLC5/6 caps or leave them intact? It has been said
that leaving them intact gives a somewhat boomy, one-note bass with
the LM4562NA upgrade.


However I'd say any audible difference is pretty subtle, and nowhere
near as marked as the improvements I've achieved with getting the PMC
speakers.

Whatever the differences related to the opamps, real or imagined, I'm
still VERY happy with the setup I now have.


Cool, and like you titled this thread, I am close but not quite to
audio nirvana with my 7520

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site:http://www.mgateway.com

CD

[John Stone]

On 8/12/09 11:15 PM, in article , "Sonnova"
wrote:

On Wed, 12 Aug 2009 16:22:12 -0700, John Stone wrote
(in article ):

On 8/12/09 7:58 AM, in article , "Sonnova"
wrote:


And while you make a case
that the distortion spec in the National chip is "orders of magnitude lower"
than that of the Burr Brown,(.00003% vs.00008%) the fact is that in either
case we are talking about distortion levels that are way below any possible
level of audibility.

You put too many zeros in the 0.08 number.

Hmm. I copied and pasted the numbers right out of the datasheet.

http://focus.ti.com/lit/ds/symlink/opa2134.pdf


So my question is, what accounts for the substantial improvement you claim
to hear by substituting excellent performing op amps with really excellent
performing op amps?

Darker silence, and cleaner, more silken highs. Others have noted this as
well. I wished I had another DAC 2.6 so that I could do a real DB test on
them. The best I could do is take turns with a friend where one of us would
swap the op-amps (or not) while the other was out of room. Upon returning
neither of us had any trouble discerning which op-amp set we were listening
to, or whether the op-amps were swapped at all. I know that this kind of test
is not very accurate because ONE of us knows which is which and there is
always the possibility of body language or other cues giving away what was
done. But if so, I'm not aware of it. Still, a DBT is preferable.

What I was asking for was not what you say you hear, but to what measurable
technical parameters you attribute this audible difference. Or is it your
position that easily audible differences aren't easily measured? In previous
threads you seem to have been an avid believer that basically all power amps
are the same once certain minimal conditions are met. But when it comes to
small signal amplifiers you seem to take an opposite stance. I'm just
curious as to why your opinions are split in this way.
FWIW, some years back I was part of an ABX test comparing a first generation
op-amp (ua741) with a straight wire bypass. Picking one from the other was
extremely difficult.

[Sonnova]

On Thu, 13 Aug 2009 12:42:19 -0700, Rob Tweed wrote
(in article ):

A somewhat cheeky observation:

- we seem to be saying that minute differences in the distortion and
noise characteristics of opamps make clearly audible differences to
the sound of the DAC, and yet quite big measurable distortion effects
in the Behringer A500 aren't going to have any audible consequences?

I sure didn't say that. I did say that I've never noticed it. The amp sounds
fine at any volume setting on the front-mounted level controls. Perhaps its
more a question of what KIND of THD (I.E. what's the cause of it) rather than
the amount that's audible in one instance and not in another. Usually noise
and harmonic distortion are lumped together in one percentage reading and not
separated into noise, even order harmonic distortion, odd order harmonic
distortion, or some kind of non-correlated distortion. Some will obviously
sound worse than others.

Anyway, be that as it may....I modded my Beresford DAC with the
LM4562NAs. I *believe* it changed the sound - not such splashy treble
and the sound isn't quite so "in your face" is how I'd describe it
now. Bass seems fine IMO. In fact I'm not sure I didn't prefer the
sound with the original opamps! :-)

The highs sounded better with the National op-amps when I changed them too,
so it isn't just you that noticed the difference. But only a true DBT can say
for sure. Unfortunately, that's not really practical in this case (hell, I've
never even SEEN another DAC 2.6, much less had access to one for a DBT!).

However I'd say any audible difference is pretty subtle, and nowhere
near as marked as the improvements I've achieved with getting the PMC
speakers.
\
No, certainly not. Transducers are the most different of all audio
components. amps, preamps, DACs, and CD players are today, much more alike
than different and the differences, if discernible at all, ARE very subtle.

Whatever the differences related to the opamps, real or imagined, I'm
still VERY happy with the setup I now have.

As am I. And really, that's the important thing. I feel that I have taken an
8-year old, high-quality, up-sampled 24/96 DAC and brought it up to date with
better op amps. I hear a difference and still heard it in a single blind
test. That suits me for now.

[Sonnova]

On Thu, 13 Aug 2009 15:34:11 -0700, codifus wrote
(in article ):

On Aug 13, 3:42*pm, Rob Tweed wrote:
A somewhat cheeky observation:

- we seem to be saying that minute differences in the distortion and
noise characteristics of opamps make clearly audible differences to
the sound of the DAC, and yet quite big measurable distortion effects
in the Behringer A500 aren't going to have any audible consequences?

Keep in mind that when the signal is travelling thru the opamp, it is
very very small. That electrical audio signal, having been slightly
bumped up by the opamp, is then massively amplified by your Behringer
to drive the speakers. Consequences on the audio are bound to vary.

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know that
the ear exhibits different sensitivities to different kinds of distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more sensitive
to noise modulation than to either even or odd order distortion. It is also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it is of
distortion at higher frequencies. So the perception of distortion depends on
what KINDS of distortion as well as HOW MUCH distortion is present in any
component.

[Codifus]

On Aug 13, 3:42*pm, Rob Tweed wrote:
........Whatever the differences related to the opamps, real or
imagined, I'm
still VERY happy with the setup I now have.

---

Rob Tweed

I think I see why you may not have noticed a difference with the
LM4562NA upgrade. I looked up the specifications on your speakers and
they respond down to 40 Hz. They use a ported 7 inch woofer to deliver
the bass. Their setup makes me think of the legendary Acoustic
Research AR 18, a 2-way bookshelf design which were a lot of fun to
listen to and sounded a whole lot bigger than they were. The AR 18s
weren't ported but perhaps their single 8 inch sealed woofer produced
about the same bass as a ported 7 inch woofer? Anyhow, those AR18s are
still coveted today, and they are over 20 years old.

My speakers now, the Cambridge Soundworks Tower IIs, use two 8 inch
woofers with a bass port in the back. They produce strong bass down to
30 Hz. This is getting into the bass "you-can-feel" territory. The
spiked feet to firmly place them on the carpet are absolutely
essential. That bass region is pretty much where I notice the somewhat
light behavior of the LM4562s.

I just want to add that I'm realizing more and more that I owe my
development of the appreciation of high fidelity audio to the late
Henry Kloss, a man who strived to bring high end audio to the regular
guy, like me

CD

[bob]

On Aug 13, 7:44*pm, Sonnova wrote:

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know that
the ear exhibits different sensitivities to different kinds of distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more sensitive
to noise modulation than to either even or odd order distortion. It is also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it is of
distortion at higher frequencies. So the perception of distortion depends on
what KINDS of distortion as well as HOW MUCH distortion is present in any
component.

But you can still measure it, which means you can still correlate
measurement with perception. But when you do, guess what you find?

bob

[Codifus]

On Aug 13, 10:10*pm, bob wrote:
On Aug 13, 7:44*pm, Sonnova wrote:

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know that
the ear exhibits different sensitivities to different kinds of distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more sensitive
to noise modulation than to either even or odd order distortion. It is also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it is of
distortion at higher frequencies. So the perception of distortion depends on
what KINDS of distortion as well as HOW MUCH distortion is present in any
component.

But you can still measure it, which means you can still correlate
measurement with perception. But when you do, guess what you find?

bob

Do you concede that there are devices that measure extremely well but
sound only so-so? And that there are also devices that measure poorly
but sound great? In my experience, technical measurements have only
given you an idea of how good a device can sound, and still to this
point, never the whole story. 20 years ago digital audio was
introduced as "perfect sound forever," and yet here we are today.

CD

[Rob Tweed]

Anyway, be that as it may....I modded my Beresford DAC with the
LM4562NAs. *I *believe* it changed the sound - not such splashy treble
and the sound isn't quite so "in your face" is how I'd describe it
now. *Bass seems fine IMO. *In fact I'm not sure I didn't prefer the
sound with the original opamps! :-) *

Did you remove the MLC5/6 caps or leave them intact? It has been said
that leaving them intact gives a somewhat boomy, one-note bass with
the LM4562NA upgrade.

I left them intact. As I understand it they're required if you use
the variable output on the Beresford, which is what I do: the DAC is
connected directly to the Behringer power amp, effectively acting as a
"digital pre-amp"

I think I'll get familiar with the modded sound for a few weeks then
try a switch back to the original opamps and see what I think. I'll
report back in due course.

I was considering getting a Caiman as well as the TC-7250 that I
already have, which would allow for quicker and easier comparisons,
but I decided that, relatively cheap though these DACs are, I couldn't
really justify it right now :-)

---

Rob Tweed
Company: M/Gateway Developments Ltd
Registered in England: No 3220901
Registered Office: 58 Francis Road,Ashford, Kent TN23 7UR

Web-site: http://www.mgateway.com

[Sonnova]

On Thu, 13 Aug 2009 19:10:26 -0700, bob wrote
(in article ):

On Aug 13, 7:44*pm, Sonnova wrote:

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know that
the ear exhibits different sensitivities to different kinds of distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more sensitive
to noise modulation than to either even or odd order distortion. It is also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it is of
distortion at higher frequencies. So the perception of distortion depends on
what KINDS of distortion as well as HOW MUCH distortion is present in any
component.

But you can still measure it, which means you can still correlate
measurement with perception. But when you do, guess what you find?

bob


You can measure it, but what are you measuring? Most THD measurements lump
all the different types of distortion and noise together.

[Arny Krueger]

"John Stone" wrote in message


FWIW, some years back I was part of an ABX test comparing
a first generation op-amp (ua741) with a straight wire
bypass. Picking one from the other was extremely
difficult.

To say the least. However, you had to hold your mouth right with 741s to get
that sort of performance. Set the gain to high, the signal too low or too
high, and their poor bandwidth, and noise performance would come and bite
you, even audibly.

Move forward to the now nearly 30-year-old 5532, and you've got your work
cut out for you!

[Arny Krueger]

"codifus" wrote in message


Do you concede that there are devices that measure
extremely well but sound only so-so?

Not if measured completely and thoroughly.

And that there are
also devices that measure poorly but sound great?

That can easily happen because of the rediculously high standards some
people have for measured performance.

In my
experience, technical measurements have only given you an
idea of how good a device can sound, and still to this
point, never the whole story. 20 years ago digital audio
was introduced as "perfect sound forever," and yet here
we are today.

There is no evidence that the CD format is inherently audibly flawed. The
first generation CD players came very close to exploiting the potential of
the format, and were monumental advancements over LPs, and the second
generation CD players basically provided all the audible performance that
will ever be needed.

The worst things that has happend to music on CDs happened during
production, not optical encoding or playback on good players.

[bob]

On Aug 13, 11:15*pm, codifus wrote:

Do you concede that there are devices that measure extremely well but
sound only so-so?

Not if you're doing the right measurements.

And that there are also devices that measure poorly
but sound great?

Sure. Even mediocre stuff can sound just as good as SOTA.

In my experience, technical measurements have only
given you an idea of how good a device can sound,

True of transducers, not true of DACs.

and still to this
point, never the whole story. 20 years ago digital audio was
introduced as "perfect sound forever," and yet here we are today.

Check your math.

bob

[Arny Krueger]

"Sonnova" wrote in message
...
On Thu, 13 Aug 2009 19:10:26 -0700, bob wrote
(in article ):

On Aug 13, 7:44 pm, Sonnova wrote:

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know
that
the ear exhibits different sensitivities to different kinds of
distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more
sensitive
to noise modulation than to either even or odd order distortion. It is
also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it
is of
distortion at higher frequencies. So the perception of distortion
depends on
what KINDS of distortion as well as HOW MUCH distortion is present in
any
component.

But you can still measure it, which means you can still correlate
measurement with perception. But when you do, guess what you find?

You can measure it, but what are you measuring? Most THD measurements lump
all the different types of distortion and noise together.

Arguing that THD is potentially a highly flawed means for characterizing the
sound of something is old news.

If all THD is very, very low then the very, very low measurement makes a
significant point - the device has no characteristic sound due to *any* kind
of HD. That starts happening someplace around 0.1 % THD, which is well
north (like 3 orders of magnitude or more) for the IC chips being discussed
here.

The fact of the matter is that as a rule, when someone plugs a modern
audio-grade IC into a socket in a DAC (usually just a low gain, nominal
impedance buffer) and starts talking about some characteristic sound,
they're talking about hyperbole, rumor, and/or their personal biases.

[Scott[_6_]]

On Aug 14, 1:00*pm, ScottW2 wrote:
On Aug 14, 6:00*am, "Arny Krueger" wrote:

"codifus" wrote in message

And that there are
also devices that measure poorly but sound great?

That can easily happen because of the rediculously high standards some
people have for measured performance.

*Or the ridiculously low standard for "sounds great". *Reference SET
lovers.


Have you never heard a SET based system sound great?


[ I cautioned ScottW2, and I'm cautioning everyone else:
please keep this discussion civil. -- dsr ]

[Sonnova]

On Fri, 14 Aug 2009 16:34:25 -0700, Arny Krueger wrote
(in article ):

"Sonnova" wrote in message
...
On Thu, 13 Aug 2009 19:10:26 -0700, bob wrote
(in article ):

On Aug 13, 7:44 pm, Sonnova wrote:

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know
that
the ear exhibits different sensitivities to different kinds of
distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more
sensitive
to noise modulation than to either even or odd order distortion. It is
also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it
is of
distortion at higher frequencies. So the perception of distortion
depends on
what KINDS of distortion as well as HOW MUCH distortion is present in
any
component.

But you can still measure it, which means you can still correlate
measurement with perception. But when you do, guess what you find?

You can measure it, but what are you measuring? Most THD measurements lump
all the different types of distortion and noise together.

Arguing that THD is potentially a highly flawed means for characterizing the
sound of something is old news.

I'm not really arguing anything. I'm merely pointing out that there are
different kinds of noise and distortion components inherent in a general THD
spec or measurement and that some may be more discernable, and therefore more
annoying to the human ear than are others. This would explain why many people
find that in some components, rather large amounts of THD are not noticeable
at all while in other circuits or components, relatively low amounts are
easily heard.

If all THD is very, very low then the very, very low measurement makes a
significant point - the device has no characteristic sound due to *any* kind
of HD. That starts happening someplace around 0.1 % THD, which is well
north (like 3 orders of magnitude or more) for the IC chips being discussed
here.

Again, the human ear is more sensitive to some types of distortion than it is
to others, but what you say is basically true. Any differences heard between
devices are likely to be extremely subtle (if there at all), and in the
overall scheme of things, not terribly important after one has been listening
for more than a few seconds,

The fact of the matter is that as a rule, when someone plugs a modern
audio-grade IC into a socket in a DAC (usually just a low gain, nominal
impedance buffer) and starts talking about some characteristic sound,
they're talking about hyperbole, rumor, and/or their personal biases.

You are certainly entitled to that view.

[Fred.]

On Aug 14, 7:34*pm, "Arny Krueger" wrote:
"Sonnova" wrote in message

...





On Thu, 13 Aug 2009 19:10:26 -0700, bob wrote
(in article ):

On Aug 13, 7:44 pm, Sonnova wrote:

Very true, plus the fact that there are a number of different kinds of
distortion and noise components that are all lumped together in a THD
measurement. Perhaps one device has more of one than another and I know
that
the ear exhibits different sensitivities to different kinds of
distortion. I
know, for instance that the ear is less sensitive to even-order harmonic
distortion than it is to odd order harmonic distortion and is more
sensitive
to noise modulation than to either even or odd order distortion. It is
also
said that how one perceives distortion also depends upon frequency. For
instance, the ear is more tolerant of low-frequency distortion than it
is of
distortion at higher frequencies. So the perception of distortion
depends on
what KINDS of distortion as well as HOW MUCH distortion is present in
any
component.

But you can still measure it, which means you can still correlate
measurement with perception. But when you do, guess what you find?
You can measure it, but what are you measuring? Most THD measurements lump
all the different types of distortion and noise together.

Arguing that THD is potentially a highly flawed means for characterizing the
sound of something is old news.

If all THD is very, very low then the very, very low measurement makes a
significant point - the device has no characteristic sound due to *any* kind
of *HD. *That starts happening someplace around 0.1 % THD, which is well
north (like 3 orders of magnitude or more) for *the IC chips being discussed
here.

The fact of the matter is that as a rule, when someone plugs a modern
audio-grade IC into a socket in a DAC (usually just a low gain, nominal
impedance buffer) and starts talking about some characteristic sound,
they're talking about hyperbole, rumor, and/or their personal biases.- Hide quoted text -

- Show quoted text -

I'm sure that a large number of differences in sound people claim they
can hear are based on ego and/or bias as you suggest. But, I'm
eaually sure the human ear is remarkable in its capabilites, and in
the differences in capabilities between people. We have only the
roughest idea what the typical human ear is sensitive to, let alone
the range of the atypical.

While tend to reserve judgement, I'm truly reluctant to reject any one
claim on nothing more than what I personally hear, or on an electrical
measurement, which itself is based on certain expectations concerning
*normal* human hearing.

Fred.

[bob]

On Aug 15, 1:21*pm, "Fred." wrote:

I'm sure that a large number of differences in sound people claim they
can hear are based on ego and/or bias as you suggest. *But, I'm
eaually sure the human ear is remarkable in its capabilites, and in
the differences in capabilities between people. *We have only the
roughest idea what the typical human ear is sensitive to, let alone
the range of the atypical.

Human hearing has been a subject of scientific study for a good
century and a half. We have quite a good idea of what the human ear is
sensitive to. Some audiophiles may not have a good idea of what the
human ear is sensitive to, but that's a reflection on them, not on the
state of scientific knowledge.

While tend to reserve judgement, I'm truly reluctant to reject any one
claim on nothing more than what I personally hear, or on an electrical
measurement, which itself is based on certain expectations concerning
*normal* human hearing.

Claims are easy to reject when they fall well out of the range of even
"exceptional" hearing. There's a gray area, of course, where only
careful listening tests can tell you whether something is or is not
audible to a particular person. None of the claims in this thread come
anywhere near that gray area.

bob

[Scott[_6_]]

On Aug 15, 9:05*am, ScottW2 wrote:
On Aug 14, 6:16*pm, Scott wrote:





On Aug 14, 1:00*pm, ScottW2 wrote:

On Aug 14, 6:00*am, "Arny Krueger" wrote:

"codifus" wrote in message

And that there are
also devices that measure poorly but sound great?

That can easily happen because of the rediculously high standards some
people have for measured performance.

*Or the ridiculously low standard for "sounds great". *Reference SET
lovers.

Have you never heard a SET based system sound great?

*I have with significant constraints on it's use, the major one
being SPL. *Increase the volume only slightly to well below
anything approaching live levels and
audible increases in distortion resulted in completely
different sound results. * IMO, such a system has a very low
standard for "sounds great".

I can see how that can be an issue with most speakers. I have heard a
few SET based systems that used very efficient speakers that sounded
pretty darned good. even great in some cases. SPLs were not an issue
in those systems. I fail to see the "lower standard." If something
sounds great it sounds great. No? What is the 'lower standard?" Is it
just your particular experiences and the issue of SPLs?

Note to the moderator. I assure you I will remain civil with Scott W.
I am confindent he will do the same with me.