Totaldac Reference D1

[cat6man]

The upgrade is internal and needs to be done by Totaldac . The USB power supply is managed differently it involves removing a few components' on the board and running a new lead .. Any more info best to contact Vincent . All current servers already have this done from what I understand .
Mine was close to a year old and I just missed this upgrade .

My totaldac is the twelve 4 box setup , when I tested the Rendu it did sound better than the Totaldac Server but once I had this upgrade done on the
Totaldac server the Rendu wasn't even close .

I hope this helps , any other questions about this upgrade contact Vincent

thanks rao, this is very informative.
so as a starting point, the microrendu outclassed last years totaldac server, which is pretty impressive. power upgrades to the server then reversed the comparison.
can you describe in what ways the sound quality of the upgraded totaldac jumped way past the microrendu?

it does seem clear now that power supply voltage, general cleanliness of the usb signal and careful attention to grounding are all critical here.
it has to help that vincent can put his server and dac together and jointly optimize these things as a combination.

how closely the microrendu can get with various levels of power supply sophistication is an open question (and floating the power supply may help as well).
the uptone audio super-cap supply, which swenson is doing for ifi and is specifically targeted for the microrendu, may be a particularly good match.........i guess we'll see in the fall when it becomes available.

if i can bug you for a few more details of your testing:
can you tell me how you had the microrendu configured and what you used to stream the music from? (e.g. NAS using LMS or mpd, JRiver on mac, hqplayer to NAA, i.e. where/how was it "rendered" and what file format did you stream, flac/wav?)
how did you connect the microrendu, via h/w adapter supplied or via an additional usb cable?



thanks!

 

[cat6man]

I currently have the Rendu in a headphone system (arrived this week), and last night its sound finally clicked to amazing. However, I'm looking to put together a system centered around a TotalDAC DAC in the near future, and lurk on here every now and then. I'm really interested in knowing whether the TotalDAC reclocker will best the Rendu, which you just stated it does. Could you expand on the tweaks you made to the reclocker, or are you sworn to secrecy? What sort of power were you feeding the Rendu - I use the Uptone Audio JS-2 linear power supply with the Rendu?

my totaldac d1-dual is in my headphone system as well (DAC output to pass pre-amp with balanced bypass output to BHSE/stax009.

previously i was streaming to a sonore sms-100 with battery power supply, then a split power LH usb cable and a separate battery on the power leg to the DAC. it was magical when streaming WAV files (big drop in SQ with flac).
then i started screwing around with some low cost usb conditioners (that will go unnamed) and moved some cables around...........and lost the magic...........took out the new usb conditioning stuff, but still no magic as i had also moved some cables and stuff around and didn't remember exactly how i had things configured before.....................ouch............i really believe that the grounding and cabling were important factors but i haven't had the time to figure out cause and effect there.

when i got the microrendu, i hooked it up with just the wall wart that powered the sonore battery supply and let it break in for a couple of days.............when i got around to listening to it, magic was back and i haven't even started to optimize things like power supply or streaming output mode yet.
I have a lot of optimization to go, both with power supplies and streaming output modes to the totaldac.
i'll update here when i have more concrete results.

 

[wstam]

The upgrade is internal and needs to be done by Totaldac . The USB power supply is managed differently it involves removing a few components' on the board and running a new lead .. Any more info best to contact Vincent . All current servers already have this done from what I understand .
Mine was close to a year old and I just missed this upgrade .

My totaldac is the twelve 4 box setup , when I tested the Rendu it did sound better than the Totaldac Server but once I had this upgrade done on the
Totaldac server the Rendu wasn't even close .

I hope this helps , any other questions about this upgrade contact Vincent

Thanks for the info. I will check with Vincent.

 

[paul79]

I also tried the MR on the Totaldac Twelve 4 box before the Server upgrade. My findings are exactly the same as rau's findings. After, the MR was not even close, before the upgrade, the MR was slightly better in all areas. This was with MPD, of which I find to sound better than even ROON for most endpoints. This upgrade to the D1 Server really fixes it.

 

[cat6man]

paul

so your test was with the MR was as the renderer in MPD mode.

i will be testing this week with MR as an NAA, using hqplayer with no signal processing.
some folks report that moving the renderer function off of the MR further improves the sound.........stay tuned.

 

[paul79]

Right, and I used a Paul Hynes Supply to feed the MR.

 

[flyer]

I also tried the MR on the Totaldac Twelve 4 box before the Server upgrade. My findings are exactly the same as rau's findings. After, the MR was not even close, before the upgrade, the MR was slightly better in all areas. This was with MPD, of which I find to sound better than even ROON for most endpoints. This upgrade to the D1 Server really fixes it.

i find the LMS option better sounding than the MPD. The latter sounding a bit too laid back, IMHO.

 

[cat6man]

well i got time to do my test of the uRendu in MPD and NAA mode.

I wanted to compared MPD with NAA mode since some have reported improvements by moving the renderer from the uRendu and having the uRendu be only the NAA (less processing clearly). i had no preconceived ideas on how this would turn out but spent a few days installing hqplayer desktop on my main linux machine and verifying i could get sound out. i uninstalled all modes except NAA and MPD from the uRendu. BubbleUPNP was used as the control point for MPD and hqplayer desktop for NAA mode. (note: this required me to run/limp between adjacent rooms to start the hqplayer which my recently repaired knee/meniscus did NOT appreciate..............the obsessive things we do for better sound). another variable was using the sonore battery supply instead of the sms-100's wall wart. lastly, i had audiobuddy AK over to give me a 2nd opinion. he and i have been in pretty good agreement on changes to sound and he has been very helpful as i optimized cables/speakers/DAC over the past few years.

The setup:
D1-Dual DAC to Pass Aleph pre-amp to Ayon monoblocks to Nola Metro-Grand speakers, cabling via Synergistic Research.
Music in WAV format on netgear NAS running minimserver, optical from netgear 48 port switch, converted back to cat8 cable to uRendu with adapter directly into USB input of DAC.

Music: Alison Krauss (Forget About It), Paul Simon (Diamonds on the Soles of her Shoes), Jennifer Warnes (Runaway Horse), Duke Ellington Orchestra (Concerto for Cootie, Great Paris concert), Keith Jarret Trio (Things Ain't What They Used to Be), Caroline No (Beach Boys, mono), Mahler's First (Bernstein), Ramirez (Missa Criola), What's New in Baltimore (Frank Zappa live)

first we compared the wallwart power to the sonore battery power. there may have been a small difference but it wasn't night and day so we left the battery in and went on to the MPD/NAA comparison.

this blew us away. the NAA was so much better than MPD mode, streaming from the same NAS that is was drop dead simple, game over. except for one song (i'll get back to that), it was no contest. the soundstage was huge, the bass was fuller, the detail was much more apparent though not in your face, but most important to me (and i find this almost always the key indicator) was the ability to hear deep into the decay of notes. on the jennifer warnes, the decay of each bass note went on forever and segued into the next lyric or bass note, each of her lyrics ended wth a slow decay of her voice/breath that was "in the room". the keith jarret had much more bass, exceptional drum impact and the sound of his piano was night and day with NAA having overtones and complexities to his playing that made us want to turn off MPD mode after 5-10 seconds. the soundstage was huge on missa criolla and the paul simon intro with ladysmith black mambazo's vocals.
the only song where it wasn't a slam dunk was the alison krauss track. NAA had more detail and separation, but it wasn't as blindingly better. AK felt he could like either.

note that I had hqplayer set to do nothing fancy, no upsampling, no dither, just render the native resolution audio so the uRendu NAA could pass data to the DAC. i have never tested any of those hqplayer options such as upsampling or DSD and may not (I don't think i want to go down a rabbit hole with that many variables).


p.s. by the way, everyone should try before buying the hqplayer (as Miska suggests) to see if this is your cup of tea or not. i am running the desktop linux version trial which plays only for 30 minutes, but that (and a few shut downs and startups) gave me enough to convince me there is something very special here.


So, getting back to the earlier discussion of the totalDAC server. I have tremendous respect for Vincent and his updated server may in fact be better than the uRendu. I have not had the pleasure of hearing it myself but I am a believer in Vincent's work. However, if you have compared the uRendu in MPD mode, my experience suggests you have not heard the uRendu at anywhere near its capabilities. I have no basis for an opinion on which is superior but you really should try NAA mode before concluding that the uRendu is outclassed (since NAA easily outclasses MPD mode).

Of course, remember that synergy is god, every system is different and your mileage may vary.

 

[paul79]

Your description of the improvements NAA brought to the MR are very similar to my thoughts on the Totaldac Server upgrade, so this is interesting.

 

[VPN]

Hello,

There is now a new Totaldac Twelve SE, which I understand offers two options of output to the preamp, directly from the resistors as the standard Twelve, and through a class A output stage, so you can select the output that sounds best in your system. From the website:


"The d1-twelve-SE

The d1-twelve-SE is a special version which can be used as a d1-twelve or can be used with following characteristics:
-class A discrete transistor output stage
-3.1Vrms on the RCA output
-6.2Vrms on the balanced XLR output
-high output current capability
-even in this buffered mode all 12 ladders of resistors are used (for the RCA output and the XLR output), each DAC monobloc being roughly a d1-six DAC

So with the d1-twelve-SE you can listen to the direct un-buffered output or you can listen to the strong balanced or unbalanced outputs and choose the best matching in your system."

Has anyone heard this new version and compared to the standard Totaldac Twelve? How did it sound? Did you prefer with or without the output stage? Which preamp did you use?

Thanks,

VPN

 

[paul79]

I think the difference will be system dependent. The SE version using the buffered output will drive more amplifier choices directly better than the Twelve if you don't want a preamp. Should also provide a more energetic and lively performance vs the more intimate sound of the Twelve. Good thing is that you have both to use!!

Or maybe you could drive a long cable to a subwoofer system using the buffered output.

Just more freedom to fit more systems.

 

[CKKeung]

Hello,
There is now a new Totaldac Twelve SE, which I understand offers two options of output to the preamp, directly from the resistors as the standard Twelve, and through a class A output stage, so you can select the output that sounds best in your system. From the website:
"The d1-twelve-SE
The d1-twelve-SE is a special version which can be used as a d1-twelve or can be used with following characteristics:
-class A discrete transistor output stage
-3.1Vrms on the RCA output
-6.2Vrms on the balanced XLR output
-high output current capability
-even in this buffered mode all 12 ladders of resistors are used (for the RCA output and the XLR output), each DAC monobloc being roughly a d1-six DAC

So with the d1-twelve-SE you can listen to the direct un-buffered output or you can listen to the strong balanced or unbalanced outputs and choose the best matching in your system."

Has anyone heard this new version and compared to the standard Totaldac Twelve? How did it sound? Did you prefer with or without the output stage? Which preamp did you use?
Thanks,
VPN

Hi VPN,
I did compare both the normal mode and the output stage mode in a Twelve SE.
The magic/realness of Twelve is diminished in the output stage mode.
The absence of output stage is one of the most important advantages of Twelve over most other dac.
Only the MSB Select II and Trinity DAC have such a design.
Unless your preamp must need a high output from the Twelve, otherwise the normal Twelve is the preferred option.

Cheers,
CK

 

[VPN]

Hello CK,

Thank you. Your feedback is quite interesting.
Cheers,

VPN

 

[cat6man]

would i be correct that the d1-dual has an 'old version' output stage, the d1-six has a 'new version class A' output stage, the d1-twelse has no analog output gain stage and the d1-twelve SE has the option of no analog output gain stage or the 'new version class A' output stage?

if that is correct, would a d1-six without an analog gain output stage be an interesting sweet spot?

background: i have a d1-dual and am thinking about a d1-six upgrade.............i'm wondering now if the d1-twelve reports of sonic magic are more due to the twelve or the elimination of the output gain stage

inquiring minds want to know.

thanks for the feedback CK.

 

[CKKeung]

would i be correct that the d1-dual has an 'old version' output stage, the d1-six has a 'new version class A' output stage, the d1-twelse has no analog output gain stage and the d1-twelve SE has the option of no analog output gain stage or the 'new version class A' output stage?

if that is correct, would a d1-six without an analog gain output stage be an interesting sweet spot?

background: i have a d1-dual and am thinking about a d1-six upgrade.............i'm wondering now if the d1-twelve reports of sonic magic are more due to the twelve or the elimination of the output gain stage

inquiring minds want to know.

thanks for the feedback CK.

Hi cat6man,

Yes, the classA output stage inside the Six is newer and better than that in the old Dual.

The Twelve SE has an selector switch. The output stage can be activated or inactivated.

I think the output of a Six without output stage may be too low. The Twelve has 6 resistor arrays working together per channel so that the output stage can be omitted.
Please email Vincent for confirmation.

Cheers
CK

 

[Gladiator]

If anyone is looking to upgrade to the d1-twelve's and looking to sell a d1-six let me know. I am in the market for a used totaldac.

 

[arnies]

Hi VPN,
I did compare both the normal mode and the output stage mode in a Twelve SE.
The magic/realness of Twelve is diminished in the output stage mode.
The absence of output stage is one of the most important advantages of Twelve over most other dac.
Only the MSB Select II and Trinity DAC have such a design.
Unless your preamp must need a high output from the Twelve, otherwise the normal Twelve is the preferred option.

Cheers,
CK

I own the TotalDAC D1-Twelve and purchased it prior to the availability of the Twelve SE. After reading about the Twelve SE, my thoughts about what it most likely sounds like with the active output stage engaged mirror what CK actually heard. I agree that if your preamp has enough gain then the standard version of the d1-twelve is the way to go. Also, I agree with CK, in that the Total D1-six cannot generate enough voltage output from its reduced number of resistors to be very usable without its active output stage.

 

[cat6man]

I own the TotalDAC D1-Twelve and purchased it prior to the availability of the Twelve SE. After reading about the Twelve SE, my thoughts about what it most likely sounds like with the active output stage engaged mirror what CK actually heard. I agree that if your preamp has enough gain then the standard version of the d1-twelve is the way to go. Also, I agree with CK, in that the Total D1-six cannot generate enough voltage output from its reduced number of resistors to be very usable without its active output stage.

hi

could you elaborate on the last comment. unless i misunderstand, the twelve has 2x the resistor/ladders as the six, which is 2x the voltage or 6 db power. as you saying a 6db difference in output could make the direct output unusable? or that the 6dB drop would mitigate the gain in noise floor obtained by doubling the number of resistor ladders?

 

[CKKeung]

I listened to the Totaldac Six dac paired with the SGMS2015 server at Volent HK, the official SG dealer in Hong Kong last week.
All-rounded and very musical/analogue-like.
An excellent matching!
I prefer this combination than SGMS's usual partner of T+A DAC8 DSD dac.

 

[EuroDriver]

After 5 days working with Ben at Volent's Hong Kong office office, workshop and demo room, we have been able to deliver from the Sound Galleries music Server, a sound that is simply stunning from the D1-six and D1-twelve with upsampled bits delivered from Roon > HQ Player > D1

Not many audiophiles in Hong Kong were taking TIDAL lossless streaming from the Internet very seriously due to the SQ results they were getting. When we played TIDAL streaming to a D-12 owner who bought the first SGM, his jaw dropped and said " this is a game changer"

This happy first purchaser is inviting his friends over for a listen to the SGM playing local music files, and music from Tidal

This trip was my first sonic encounter with the D1-six and twelve, and I have to say to Vincent, I kiss you on both cheeks