• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Phase shift and echo trails length illusion

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,226
Likes
17,805
Location
Netherlands
Curious, have you ever found brand new DAC sounded a bit off, but improved over time?

Never.. but then again, I’m not buying DAC’s every month ;) Once was a very early ESS adopter. Back then there were not commercial products yet (at least not for mere mortals), had/have miniDSP and now the E30. None ever improved over time. Nobody ever came up with a hypothesis on what physical principles underlay this supposed burn in.. Other than temperature warmup of components.. something that happens in a few minutes.
 
Last edited:
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
Never.. but then again, I’m not buying DAC’s every month ;) Once was a very early ESS adopter. Back then there were not commercial products yet (at least not for mere mortals), had/have miniDSP and now the E30. None ever improved over time. Nobody ever came up with a hypothesis what physical principles underlay this supposed burn in.. Other than temperature warmup of components.. something that happens on a few minutes.
Actually, maybe I can be the first? in ASR who could check if a brand new DAC like X16 needs a break in process.

I first powered it on long-term on the morning of January 30th.

I noticed I like E30‘s highs better with my stereo setup and decided to take voltage measurements the same day.

My measurements can be seen here:
https://www.audiosciencereview.com/...x16-balanced-mqa-dac-review.17419/post-663459

My measurements saw odd dip in 2Kh and lower voltages at 5+kHz as compared with E30, when voltage matched at 1Khz.

Now that I have used X16 for 3+ weeks, it is time to measure again.

Edit: you can disregard my comment about bass in that post. I have found out the issue is Tidal app on my laptop strangely turned down e30's volume by ~12db when using forced volume mode and playing back CD quality music.
 
Last edited:

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,226
Likes
17,805
Location
Netherlands
I’m sorry to say so, but those measurements just don’t have enough rigor to count as real proof. There are just to many factor where this could have been botched, specially looking at your inexperience. No offense here though! You’ve got to start somewhere. I think it’s great to see you put an effort in it an learn from it. So don’t let yourself be discouraged in any way.
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
I’m sorry to say so, but those measurements just don’t have enough rigor to count as real proof. There are just to many factor where this could have been botched, specially looking at your inexperience. No offense here though! You’ve got to start somewhere. I think it’s great to see you put an effort in it an learn from it. So don’t let yourself be discouraged in any way.
Yeah, my multimeter not accurate, but can show a trend and pretty much consistent to ~0.001V (measure same freq for same DAC multiple times and on different days).

See https://www.audiosciencereview.com/...topping-e30-listening-tests.17988/post-665605

Regarding experience using multi meter, I had plenty of experience during my E30 vs KTB listening test period last year.

I measured x16 2Khz multiple times as it was not as expected as compared to KTB and E30. I then measured multiple times again with E30 just to make sure the multimeter was consistent.

Heh, I may still get same voltage numbers for x16 like before, don't need to put me down that early yet.
: P
 
Last edited:
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
Wow.....Indeed burn in happens for X16.....

Again, my multi-meter is not accurate, but it is consistent within 0.001V, and can show trend. See my KTB vs E30 thread posts, like https://www.audiosciencereview.com/...topping-e30-listening-tests.17988/post-665605 for more information.

Again, I measured multiple times when X16 was basically brand new because I could not believe the 2Khz difference and +5KHz difference as compared to E30. 100% confidence with my previous measurements.

This is comparison between KTB (measured at about 2 months old), E30 (measured at about 2 weeks old) and X16 (measured about 10 hours old):
KTB_E30_X16_voltages_X16_10_hours_old.PNG


This is comparison captured today. E30 (measured at about 2 months old) and X16 (measured at about 3 weeks old):
E30_X16_voltages_X16_3_weeks_old.PNG
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
Did my brain got burn in
Yes.

If 30 years of playing around with audio have taught me anything, it's that psychological effects are extremely strong must be taken as the default explanation for any audio phenomenon that hasn't been strictly controlled. Some days my system sounds amazing and I can't believe it's possible to get any better no matter how much money you throw around. On other days the exact same system is good, but has niggling issues that make me wonder if I need to adjust the EQ or tweak it in some other way. It's all in my mind and just a function of my mood, so much so that I can put on my headphones and use them to diagnose whether I'm feeling slightly depressed.
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
Wow.....Indeed burn in happens for X16.....

Again, my multi-meter is not accurate, but it is consistent within 0.001V, and can show trend. See my KTB vs E30 thread posts, like https://www.audiosciencereview.com/...topping-e30-listening-tests.17988/post-665605 for more information.

Again, I measured multiple times when X16 was basically brand new because I could not believe the 2Khz difference and +5KHz difference as compared to E30. 100% confidence with my previous measurements.

This is comparison between KTB (measured at about 2 months old), E30 (measured at about 2 weeks old) and X16 (measured about 10 hours old):
View attachment 114479

This is comparison captured today. E30 (measured at about 2 months old) and X16 (measured at about 3 weeks old):
View attachment 114480
The difference between 0.377V and 0.381V is 0.09dB (20*log(V1/V2)). I think it's safe to say this can be disregarded as inaudible.
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
The difference between 0.377V and 0.381V is 0.09dB (20*log(V1/V2)). I think it's safe to say this can be disregarded as inaudible.
In my other thread about burn in, with solderdude suggestions, I have found out the change was more about x16 temperature change effect.

E30 very stable. No change measured from cold.

X16 went back to original numbers (like when measured new). 4 hours later, still original numbers.

My investigation will continue in https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-684261

You might as well skip everything.

But it is odd that everyone dismissed the usefulness of my meter right away. I keep telling them it can be a good story teller, but no one listened.

See https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-687522

And
https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-687717

What do you think?
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
In my other thread about burn in, with solderdude suggestions, I have found out the change was more about x16 temperature change effect.

E30 very stable. No change measured from cold.

X16 went back to original numbers (like when measured new). 4 hours later, still original numbers.

My investigation will continue in https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-684261

You might as well skip everything.

But it is odd that everyone dismissed the usefulness of my meter right away. I keep telling them it can be a good story teller, but no one listened.

See https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-687522

And
https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-687717

What do you think?
I'd have to say the first thing I'd consider is that we're looking at a change of 4mV here, which is pretty small. My first priority would be make sure I knew what the measurement error of my multimeter was. Since the E30 seems stable, repeat the same measurement on it 10 times in the morning, then 10 times at midday, when it's warmer. Check that you get the same reading when you move the probe cables around a bit, etc.

But let's say all those come up the same and 4mV is outside the measurement error of the device. The primary environmental effect that will change the way a circuit works is temperature, as when a wire heats up its resistance rises. Modern high-quality resistors are explicitly designed to minimise this effect, and a good circuit design will balance out thermal changes as well. But it's worth getting out a hair dryer and using it to heat up the X16 then see if this causes any change in the numbers you're getting (I remember amirm did this on an Yggdrasil he'd been loaned and the guy who'd lent it had a conniption, but you can't do any damage unless the unit is exceptionally fragile).

I think the most important thing, though, is to consider whether the numbers you're getting really correlate with a perceivable difference. Get a track and apply a 0.1dB low-shelf filter to it in Audacity or another audio editor. I'm willing to bet that you won't be able to tell the difference between that and the original file.
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
I'd have to say the first thing I'd consider is that we're looking at a change of 4mV here, which is pretty small. My first priority would be make sure I knew what the measurement error of my multimeter was. Since the E30 seems stable, repeat the same measurement on it 10 times in the morning, then 10 times at midday, when it's warmer. Check that you get the same reading when you move the probe cables around a bit, etc.

But let's say all those come up the same and 4mV is outside the measurement error of the device. The primary environmental effect that will change the way a circuit works is temperature, as when a wire heats up its resistance rises. Modern high-quality resistors are explicitly designed to minimise this effect, and a good circuit design will balance out thermal changes as well. But it's worth getting out a hair dryer and using it to heat up the X16 then see if this causes any change in the numbers you're getting (I remember amirm did this on an Yggdrasil he'd been loaned and the guy who'd lent it had a conniption, but you can't do any damage unless the unit is exceptionally fragile).

I think the most important thing, though, is to consider whether the numbers you're getting really correlate with a perceivable difference. Get a track and apply a 0.1dB low-shelf filter to it in Audacity or another audio editor. I'm willing to bet that you won't be able to tell the difference between that and the original file.

Thanks. I have measured E30 in January, February, and March. I have measured KTB in December, various time of day.

On both, I got mostly exactly same readings. Variant was 0.001.

So I have ruled out meter issue in normal home usage.

There are more updates about measurements in https://www.audiosciencereview.com/...my-gustard-x16-measurements.20679/post-688971

It would seem the difference between the filter of E30 and X16, plus the channel imbalance of x16 itself, could add up to 0.3db.

Now the issue is if I can sense that difference using test files. The funs never end, it would seem
 
Last edited:
Top Bottom