Review and Measurements of Totaldac d1-six DAC

[Patatorz]

Mr Patatorz, would you have a personal fight with Mr Brient? Are you the one owning patatorz.com?

I am then very surprised to find on your website a lot of comparison by yourself of different « audiophile internet switches » with specific focus on « linear power supply » and test protocol taking into account the « running-in » time of these products, as well as « high-end ethernet cables ». These switches costing between USD 1000 to 2000, linear power supply an additional USD1000, and the internet cables USD199 for 1.2 meter. You share your experience on how well all that improves S/N. But no measurements shared in a domain where it’s a norm. Snake Oil?

(Moderator, please take this sentence to mean ' shoot in the foot' clearly a language barrier issue no one's advocating anyone killed themselves.)
Mr Brient killed himself well enough. Is it possible that with that background you do not contribute to the killing?

And ? No personal fight at all. I’m always learning and challenge myself on what I hear or not hear. This is also the reason why read this forum, i assume like you. My journey in audio is still long and still learning from everyone. The most important is remain open minded. So I’m not shooting in my foot, I have a blog and give my personal vision, a vision that can change all along a life. It seems that a lot of people in this forum experienced also the same journey.

Your post is out of the context and has nothing to do here. You are talking under emotion and all your assertions are wrong. Ask to yourself only one question : why totaldac put on the table the fact that I own a dCS network bridge and telling that the d1 server is better ? What is his interest ? Ask yourself about the marketing of TotalDac on HCFR and with big influencers in the audio world. Do you really think that the 2 or 3 guys making the thread alive are only customers

What I expect is a clear and decent discussion between totaldac and Amir to understand and learn from my side, enrich my knowledge and build my opinion. I think you are also looking for that no ?

We can continue this discussion in MP if you want but be sure that there is no personal fight and your post is « limite ». Very sorry if you and Andrex who liked your post misunderstood my expectations.

PS : what I wrote wrong in your quote : the server is the reclocker, the cubox and the remote ?

 

[daftcombo]

On the french forum, Vincent announced these specifications back in 2011 :

bandwidth: 0Hz-45kHz for a 96kHz signal
distortion: 0.002%
output level: 3.3Vrms
noise floor: -131dB
output impedance: 75ohms
conversion delay of S/PDIF to analogical of a 96kHz signal: 10ms
internal digital treatments: 69bits
conversion: R2R 24bits without oversampling

(my translation)

Source: https://www.homecinema-fr.com/forum...e-fidelite-et-dac/dac-totaldac-t29963819.html, page 1


He gives a few more details on how the measurements were done on this page:
https://www.homecinema-fr.com/forum...idelite-et-dac/dac-totaldac-t29963819-30.html

For instance, the 0.002% distortion was obtained "at low frequencies [up to 22kHz from what I understood from the rest] and at -1dB so as not to disturb the analyser by the NOS design".
"Over-sampling main goal is to achieve beautiful measurements on analysers designed for analogical sources."

Noise floor of -131dB was obtained with a "-80dBFS signal without ponderation".

We also have"channel separation: 90dB (at 1kH)".


Anyway, do those specifications match the actual measurements that will predict what one will hear?
If so, are they good enough for a 13K500$ DAC?

 

[Frank Dernie]

Well, having struggled through this thread it does seem to boil down to one of the two types of review thread.
1. works well, a few comments and people saying they will buy one or asking about reliability and company longevity.
2. Works poorly then endless critiscism of measuring technique and comments about the only way to evaluate is by ear, mainly from manufacturer and owners but also those of like mind.

 

[Azeia]

It’s not just this thread. The glee with which some people seem to sink the boot is just a bit rude.

Referring to it as "glee" is already a misrepresentation. It's more that people will attack companies with horrible business practices; there's plenty of other extremely expensive gear out there that doesn't get shit on, because it at least measures well, even we find some of it is vastly overpriced.

Going by your logic, any criticism or mockery of a product is never justifiable because it's just "glee"; you're criticizing behavior instead of examining the reasons why people are reacting this way. Your narrative paints a picture that people are always desperate to find something to hate on, which is complete nonsense. Here's a thought experiment, if the audiophile market was not full of snake oil, do you think anyone here would be shitting on products? of course not, there would be no bad products to "gleefully" attack in that alternate universe.

As for this thread, humor is the best medicine, so people are making fun of something they justifiably think is completely ludicrous; a $14,000 product that performs worse than ass (and not the good kind of ass, mind you).

(Can’t believe an Aussie has to pull people up for being impolite. What is the world coming to )

To make it even more ironic, I'm Canadian. =)

 

[Deleted member 2348]

2. Works poorly then endless critiscism of measuring technique and comments about the only way to evaluate is by ear, mainly from manufacturer and owners but also those of like mind.

IMHO that's not a fair description of some of the criticisms, in particular post #536 by @KSTR.

A NOS DAC without reconstruction filter is broken by design and its errors can be shown easily. These errors show up in @amirm's measurement, but IMHO they are just not described 100% correctly. The d1-six DAC produces a (unacceptably) huge amount of out-of-band noise (as is to be expected for a DAC without reconstruction filter), but this is not equal to non-linearity which causes harmonic distortion or intermodulation distortion.

Just to make this clear: I would never buy or use a DAC without reconstruction filter; not even for $100.

 

[Sir Sanders Zingmore]

Going by your logic, any criticism or mockery of a product is never justifiable because it's just "glee"

That’s not my logic at all.
Criticise away, mock too. Just be a bit less crass about it. A bit of decorum please gentlemen

 

[THW]

On the french forum, Vincent announced these specifications back in 2011 :

bandwidth: 0Hz-45kHz for a 96kHz signal

this just looks like a really strange test condition to me, do people really send 96 kHz signals down their systems?

unless I’m reading this wrongly that is.

 

[graz_lag]

That’s not my logic at all.
Criticise away, mock too. Just be a bit less crass about it. A bit of decorum please gentlemen

Within a scenario where a $13,000 DAC compares to a $9 Apple dongle there is a VAST prairie for the horses to gallop ...

 

[Andrex]

this just looks like a really strange test condition to me, do people really send 96 kHz signals down their systems?

unless I’m reading this wrongly that is.

Some details on the measures and conditions.

Bandwidht ; 0Hz-45 kHz for a 96kHz signal. Done with a correction of 0db at 0hKz, and -1db at 45kHz
Distorsion: 0.002%, 1 kHz at 0 dBFS and 0 and _"dbfs from 20 to 20000 Hz. With a specific note "I don't have a device that correctly suppresses energy above 22KHz, so measurements over the entire spectrum are not possible".
Channel separation : 90dB at 1KHz

Some people are more qualified than I am to assess the relevance of these data.
I hope that this will help to move the debate forward in a calm manner.

 

[Snafu]

this just looks like a really strange test condition to me, do people really send 96 kHz signals down their systems?

unless I’m reading this wrongly that is.

For example Tidal uses 96kHz with their "master" audio:
https://support.tidal.com/hc/en-us/articles/115000397069-TIDAL-MASTERS

 

[THW]

For example Tidal uses 96kHz with their "master" audio:
https://support.tidal.com/hc/en-us/articles/115000397069-TIDAL-MASTERS

ok, but human hearing is limited to 20 - 22 kHz, i’m not entirely sure how a 96 kHz test signal is in any way relevant to our listening experience.

there also isn’t much content up at the higher frequencies in music IIRC anyway.

 

[Azeia]

That’s not my logic at all.
Criticise away, mock too. Just be a bit less crass about it. A bit of decorum please gentlemen

Can you cite an example of where people have gone "too far"?

Keep in mind that it doesn't help when the engineer behind the product comes on this thread and stirs up more shit by acting childish, defensive, and belligerent.

 

[guenthi_r]

It´s only a "workaround"
"Analog filtering at low sample rates (44100, 48000 Hz, as example) is one of problems of these R2R-DAC. To solving the issue, low sample rate is upsampled and digitally filtered before analog filtering"
Source

 

[Sir Sanders Zingmore]

Can you cite an example of where people have gone "too far"?

Keep in mind that it doesn't help when the engineer behind the product comes on this thread and stirs up more shit by acting childish, defensive, and belligerent.

“Too far” is your term, not mine (at least I don’t think I used it ).

I’m not talking about the engineer’s comments.

 

[KSTR]

As for published vs. measured specs:
- Distortion matches.
- Noise floor doesn't seem to match but noise floor specs *without* specifing bandwidth are useless. Quite probably it matches when the same bandwidth is used. Noise floor means no signal, if a signal is present (0dBFS, notably), this changes into a dynamic range measurement which can have higher noise floor from signal dependant modulation noise (aka noise floor modulation).

So in terms of these measurements which are considered some of the more important ones, it all looks OK.

 

[Snafu]

ok, but human hearing is limited to 20 - 22 kHz, i’m not entirely sure how a 96 kHz test signal is in any way relevant to our listening experience. there also isn’t much content up at the higher frequencies in music IIRC anyway.

Tidal has +16.300 "master/mqa" releases and for example my br-player (spdif out to dac) plays cd's 44 - 192kHz (if i want to).

Latest mqa file : http://www.meridianunplugged.com/downloads/MQA_List.csv
[/offtopic]

 

[Andrex]

As for published vs. measured specs:
- Distortion matches.
- Noise floor doesn't seem to match but noise floor specs *without* specifing bandwidth are useless. Quite probably it matches when the same bandwidth is used. Noise floor means no signal, if a signal is present (0dBFS, notably), this changes into a dynamic range measurement which can have higher noise floor from signal dependant modulation noise (aka noise floor modulation).

So in terms of these measurements which are considered some of the more important ones, it all looks OK.

Maybe it can help you :

Noise floor: -131dB, "it's just a measurement of the signal graph at -80dBFs, without any weighting. Done with a audio analyser R&S UPL.

Voilà.
Finally, we obtained the basic data.

 

[KSTR]

A NOS DAC without reconstruction filter is broken by design and its errors can be shown easily.

Technically speaking, yes, but perception-wise (the speakers and the ears being the reconstruction filters) the picture is not that clear.
Can be easily demonstrated with D/S-DACs / HP amps that have chips with NOS digital options like the RME-Adi2 (DAC & PRO) which has subsequent electronics free of distortion (no IMD problem). When the HF drop is corrected for to better than 0.1dB precision, absolutely no chance to hear a difference between the NOS setting and the standard Short Delay Sharp filter, yet the NOS mode measures as expected (and with much lower real harmonic distortion than the TotalDac, that is).

 

[KSTR]

Noise floor: -131dB, "it's just a measurement of the signal graph at -80dBFs, without any weighting. Done with a audio analyser R&S UPL.

Yes, I saw that, but bandwidth statement is missing.

 

[rmo]

Totaldac Disaster