Review and Measurements of Totaldac d1-six DAC

[gvl]

The sampling rate is 44.1 kHz and the digital generator is programmed to go 22 kHz. I could go to 22.05 kHz if you want to be pedantic about it. Is this what you are asking?

There should be ~3db droop at 20kHz due to sinc envelope, you came up short. Dunno why. IMD products from images adding to the fundamental?

 

[typericey]

Okay, let's pretend for a moment that the TotalDAC measures well. Let's say as good as the Topping D50s...

...still not worth $13k.

 

[diegooo1972]

It seems to me that Totaldac It's more a matter of personal taste and hearing plus engineering skills with less appeal due to personalized measurements.
How can you compare measurements if everyone do something different ?
Measurements comparison became useless.
I'm not saying that people shouldn't like Totaldac and their sound signature of course but I think, in that case,
that selling personalized measurements is not really appropriate.
They are useless for comparison but people eat them as they are standardized. Obviously.
Not fair.
But we are there again.
Other human beeing can't simply understand but every time I can't find comparable scientific methods.
I'm not very happy to eat that again and again from high end audio world. And I'm not alone.
And if someone wonder why people here take measurements so seriously is because if someone sell a 10k$ plus equipments, the world expect state of the art in every single aspect from it.
Also scientific perfectionism.
Otherwise posting measurements is contradictory.

 

[amirm]

Don't mix up the DAC under test sampling frequency and the analyser sampling frequency. An image frequency around 25KHz is taken by the analyser if it is set internally to 96KHz or 192KHz.

You can't set the sample rate of the analyzer too low or its filter response will be combined with the response of the device under test. That is why I use 192 kHz sampling normally.

I can however limit the bandwidth manually to 20 kHz. Once done, this is the results:

So the overshoot is not there anymore with the correction. However, one does not get a flat response either way despite what your website says.

Please confirm the above measurement as valid and I will update the review with it.

Note that frequency response issues was the least of the worries with your DAC.

 

[amirm]

How can you compare measurements if everyone do something different ?

Everyone doesn't do them different. Besides, there is almost no measurements presented from TotalDAC to be in dispute with anything.

 

[amirm]

It seems to me that Totaldac It's more a matter of personal taste and hearing plus engineering skills with less appeal due to personalized measurements.

We don't live in the real of "seems to me." We live in the world of what you can demonstrate as fact. I am happy to accept blessings of this DAC if in a controlled listening test it shows any value whatsoever.

As it is, it produces stair-step response which any high-end audiophile would run away from, if they knew what it did.

 

[diegooo1972]

I have read that -120dbfs was used instead of 0. Wasn't that a different measument ?

 

[amirm]

I have read that -120dbfs was used instead of 0. Wasn't that a different measument ?

It was. One was made up, the other was industry standard. You have industry standard measurements from me across countless products here. And they match those produced by well-engineered audio products. The fact that a company creates a random test doesn't give you the ticket to say there are no standards.

Regardless, it was easy enough to replicate his measurements so what is your point again?

 

[gvl]

You can't set the sample rate of the analyzer too low or its filter response will be combined with the response of the device under test. That is why I use 192 kHz sampling normally.

I can however limit the bandwidth manually to 20 kHz. Once done, this is the results:
View attachment 30234

So the overshoot is not there anymore with the correction. However, one does not get a flat response either way despite what your website says.

Please confirm the above measurement as valid and I will update the review with it.

Note that frequency response issues was the least of the worries with your DAC.

Here you go, a very nice -3dB at 20kHz. Was it ultrasonic garbage that was confusing the analyzer? Isn't it supposed to use a steep filter for each measurement point during FR test?

 

[diegooo1972]

MMM i think you misunderstood. I was talking about bad position in measurements from Totaldac.
Not yours. I completely agree with you.

 

[amirm]

MMM i think you misunderstood. I was talking about bad position in measurements from Totaldac.
Not yours. I completely agree with you.

Ah, I see it now. Sorry about that.

 

[BDWoody]

I don't care your measurements, I just don't want to let some of you say that it is a poor DAC design and with fake graphs on Totaldac web site.

Just misleading and convoluted. Telling Amir he doesn't know how to take measurements isn't going to improve a DAC that likely wouldn't outperform and almost certainly wouldn't be distinguishable in a real listening test from the analog outs on one of my $15 Chromecast Audios.

Eschew Obfuscation. Publish stats to accepted standards, or leave all your numbers out. That would give more room for testimonials on soundstage and all anyway...

 

[Dialectic]

A remarkable failure of company crisis management on the part of TotalDac.

 

[FredYves]

Measurements capture everything you could tell in controlled, blind tests. They cannot by definition predict subjective listening tests conducted outside of the realm of audio science. You wouldn't ask an atheist to swear on a bible that God doesn't exist! Let's not mix religions here.

What we question is taking audiophile myths as gospel, and running after checklist after checklist to satisfy them, high-fidelity experience be damned.

We exist to bring sanity to high-fidelity audio.

I am not religious here, I like to listen to music. I am not considering that a set of curves will determine all what I experience in listening to music. I like very much your strong approach based on measures but pretend to be the one bringing « sanity » means other are insane. It feels a bit too much to me - a bit religious?

 

[PaulD]

No, Amirm is being completely objective, not religious at all. He stated very clearly "Measurements capture everything you could tell in controlled, blind tests." This is objectively true and tested. He also said, "They cannot by definition predict subjective listening tests conducted outside of the realm of audio science." This is also true. This is the Audio SCIENCE Review forum, not the Audio Emotion Review forum...

Your emotional response to music (or sound) is your business, it is a personal emotional response. Mine (everyone's?) will almost certainly be different.

Properly measured performance IS the only measure that counts and translates objectively. Someone may prefer the sound of some particular unit, but that does not make it "better", all they can say is that they prefer it, despite it's (possibly) poorer objective performance. If it is not a properly conducted blind (ABX) listening test, then perhaps they like the fancy case, or the the lights or the fact that it costs more than their car, or that some "reviewer" liked it... It is an emotional and therefore biased response - nothing wrong with that, but acknowledge it for what it is (I call it Audio Fashion)... That is also much closer to religion than what Amirim wrote - he was objective and factual.

 

[THW]

Did I read this correctly?

A manufacturer doubling down on not taking measurements in line with industry standards and we are to take him seriously?

 

[stringerbell]

You can't set the sample rate of the analyzer too low or its filter response will be combined with the response of the device under test. That is why I use 192 kHz sampling normally.

I can however limit the bandwidth manually to 20 kHz. Once done, this is the results:

what happens to the other graphs, if you also limit the bandwith to 20Khz for them ?

 

[amirm]

what happens to the other graphs, if you also limit the bandwith to 20Khz for them ?

Many of them are bandlimited. For example, the key data in dashboard is limited to 22.4 kHz:

For some of the others, it doesn't make sense to limit them as they are trying to show ultrasonic content and such.

Linearity is already filtered. SNR is also 20 kHz.

When I get a chance I will check anyway....

 

[Patatorz]

Not sure to understand the behavior of TotalDac. Seems quite agressive and turning around the questions without trying to give clear answer. Only pushing back on competencies of Amir and never providing guidance of information to better understand how to measure a NOS dac.
This is not a good attitude to me in front of people who try to understand.

 

[BillG]

Not sure to understand the behavior of TotalDac.

I suspect it's due to a rather healthy profit margin he's trying to protect...