• 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!

Master Thread: Are measurements Everything or Nothing?

Hmmh yeah, maybe Topping D90SE measures nice here, but, it depends how you look at things.

I did run some measurements too.

ToppingD90SE_1k_44k1.png


OK, that looked decent, let's see how it looks like with -60 dBFS input:
ToppingD90SE_-60dB_44k1.png


OK, WTF is going on here? Let's try the same -60 dBFS signal with 352.8k rate:
ToppingD90SE_-60dB_352k8.png


OK, better, but something is still wrong. How about DSD256 input:
ToppingD90SE_-60dB_DSD256_ASDM5ECv2.png


Better still, but still some noise floor modulation issues.

If we then take a half price DAC with same test, same DAC chip:
SMSL_M500mkII-1k-352k8.png



SMSL_M500mkII-1k_-60dB-352k8.png


Now that's better!

And we can even go to about 250€ PCIe card based on AKM chip:
NUAudio_352k8_1k_-60dB.png


So maybe choosing a DAC based on the SINAD chart is not the best thing to do. I wasted 900€, and the D90SE stays in the box. I had totally opposite experience with the ADI-2 for example. Maybe more thorough testing by the manufacturer?
 
Last edited:
Hmmh yeah, maybe Topping D90SE measures nice here, but, it depends how you look at things.

I did run some measurements too.



OK, that looked decent, let's see how it looks like with -60 dBFS input:
View attachment 180266

OK, WTF is going on here?
That looks like a measurement glitch. And some of this is compounded by changing the x axis, not to mention looking at nanovolt stuff outside of the <20kHz use range and trying to tie that to anything audibly meaningful.
 
Why do people keep throwing out this strawman about buying based on SINAD? Just because the stuff is laid out on a chart according to their SINAD that doesn't mean we're buying them because we think something with -120db sinad sounds better (or even different) from something with -110db sinad.
 
Hmmh yeah, maybe Topping D90SE measures nice here, but, it depends how you look at things.

I did run some measurements too.

View attachment 180265

OK, that looked decent, let's see how it looks like with -60 dBFS input:
View attachment 180266

OK, WTF is going on here? Let's try the same -60 dBFS signal with 352.8k rate:
View attachment 180267

OK, better, but something is still wrong. How about DSD256 input:
View attachment 180268

Better still, but still some noise floor modulation issues.

If we then take a half price DAC with same test, same DAC chip:
View attachment 180273


View attachment 180269

Now that's better!

And we can even go to about 250€ PCIe card based on AKM chip:
View attachment 180272

So maybe choosing a DAC based on the SINAD chart is not the best thing to do. I wasted 900€, and the D90SE stays in the box. I had totally opposite experience with the ADI-2 for example. Maybe more thorough testing by the manufacturer?
Why are some of the graphs to 20 khz and some 80 khz. Was the ADC sample rate different for some of them?
 
Hmmh yeah, maybe Topping D90SE measures nice here, but, it depends how you look at things.

I did run some measurements too.

View attachment 180265

OK, that looked decent, let's see how it looks like with -60 dBFS input:
View attachment 180266

OK, WTF is going on here? Let's try the same -60 dBFS signal with 352.8k rate:
View attachment 180267

OK, better, but something is still wrong. How about DSD256 input:
View attachment 180268

Better still, but still some noise floor modulation issues.

If we then take a half price DAC with same test, same DAC chip:
View attachment 180273


View attachment 180269

Now that's better!

And we can even go to about 250€ PCIe card based on AKM chip:
View attachment 180272

So maybe choosing a DAC based on the SINAD chart is not the best thing to do. I wasted 900€, and the D90SE stays in the box. I had totally opposite experience with the ADI-2 for example. Maybe more thorough testing by the manufacturer?
Why the differences in frequency scale?
 
Why are some of the graphs to 20 khz and some 80 khz. Was the ADC sample rate different for some of them?

Scaled the graph to better see apart the suspect frequency peaks.

Another problem discovered with the same DAC.

705.6k source 1k tone looks fine:
ToppingD90SE_1k_705k6.png


Switching to 19+20 kHz IMD test tone however:
ToppingD90SE_IMD_705k6.png


Why did the noise floor jump up?

Then dropped sampling rate to half:
ToppingD90SE_IMD_352k8.png


Competing DAC based on same DAC chip, at 705.6k IMD:
SMSL_M500mkII-IMD-705k6.png


Note, 1 kHz IMD tone is even below -140 dB here.
 
We've got quite a few of them about the place...(including the member you just 'uhm'd) there doesn't seem to be a 'secret recipe' aspect to it.

How well do you understand the measurements that are typically done?

Where do you get 7/8 from?
If you know for example that the return currents produced by the transducers can in some cases represent a problem, you will find a way to measure the effects of the problem. Ok. But you must know, otherwise you will not take that measure.
 
That looks like a measurement glitch.

Interesting that out of all the DACs I have, only this one has such glitch. Of course I've seen all kinds of other stuff though.
 
Could be verified by someone else measuring the D90SE.
That's the beauty of testing and standards and/or reporting what test conditions were.
 
But overall, if we compare chips with their modulator in play PCM inputs. AKM is less smooth and rounded off than TI chips, being more in the middle in this respect, Wolfson (now owned by Cirrus) is also similar. ESS has it's typical "on your face" flavor, with some hardness, more like Cirrus Logic. What I've heard, lot of people share similar experiences.

It doesn't show up in measurements under static signal conditions, so no point in looking for the reasons there.

It's been a long time since I've heard a TI based DAC. Who's using them these days? I may have an old ROTEL CD player that uses Burr Brown or Wolfson. I know Cambridge Audio was still using Wolfson last time I checked. Cirrus is based in Austin, I think - used to see their building every day.

When you say "less smooth and rounded off" what specific aspects or measurements do you mean? My experience with the TEAC DACs was that they had less 'crisp' treble than ESS and a generally more rolled-off high end response. I'm not saying that's a bad thing, but I - like you - do notice the difference. Of course, it's almost certainly NOT due to the DAC chip itself, but rather the surrounding circuitry and even partnering equipment.

I think it's pretty well established that ESS has that 'bump' and 'in your face' sound in many applications, so I'm in agreement with you there.
 
Looks like different FFT settings. Why would the -60dB signal have a more than 20 dB higher noise floor? Does not make sense.

For similar reasons noise floor seems to vary based on sampling rate in other measurements I've seen of the same DAC.

I suspect the reason could be different implementation on how the 4 DAC channels are ganged on the output. If there's no suitable mixing resistance between the channels, it can drive feedback loops of the separate I/V stages used on D90SE (but usually not used on other ESS implementations) nuts. Since output of four delta-sigma converters run from four different modulators is NOT the same signal at high frequencies (before analog reconstruction filter)! You need to remember that you are mixing four high frequency (tens of MHz) signals together.
 
We've got quite a few of them about the place...(including the member you just 'uhm'd) there doesn't seem to be a 'secret recipe' aspect to it.

How well do you understand the measurements that are typically done?

Where do you get 7/8 from?
Also, if your (partial) measurement tells you: "everything is ok", but your ear tells you: "everything is not ok here! Check the high frequencies better!", What do you do, try to control better or give the blame it on Madame Dopamina and Mr Placebo?
 
Interesting that out of all the DACs I have, only this one has such glitch. Of course I've seen all kinds of other stuff though.
Measurement far far more likely than DUT. Your as-yet-undisclosed methodology and setup appear to be a bit shaky.
 
Also, if your (partial) measurement tells you: "everything is ok", but your ear tells you: "everything is not ok here! Check the high frequencies better!", What do you do, try to control better or give the blame it on Madame Dopamina and Mr Placebo?
Try to control at all. Unless you're someone who doesn't trust his ears and needs to peek. If it holds up, it's worth chasing.

Narrator: it's unlikely to hold up if we're talking about electronics.
 
It's been a long time since I've heard a TI based DAC. Who's using them these days?

At least iFi and T+A for example. Also in TEAC UD-501 (PCM1795).

When you say "less smooth and rounded off" what specific aspects or measurements do you mean?

Not any measurement, listening. I could nicely measure (analyze) too, given access to the raw data going to conversion stage, from the DEM stage output. Some things are quite challenging to analyze from analog output, especially when it's a multi-element converter.
 
Measurement far far more likely than DUT. Your as-yet-undisclosed methodology and setup appear to be a bit shaky.

Ahh yeah, now I remember better why I don't bother posting here. Sorry, I go away.
 
Back
Top Bottom