• Welcome to ASR. 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!

Anyone Review Topping D90III Discrete / D90 III Sabre yet ?

Actually, I2S is WORSE, as it was never meant to be transmitted in a cable, ONLY in traces on the PCB.

I wonder whom invented that stupid thing (I2S on a cable).
Credible 'cable' implementations of I2S use appropriate buffers and line drivers...

It's also been in use for a while: Sonic Frontiers had a version of I2S on their transports and DACs many years ago.
 
I tried this new fw and they fixed the volume problems, but to me it still sounds dull and flat compared to 0.39 .... Maybe it's a placebo, but I like 0.39 better.
EDIT: I use the USB input and RCA output to the Burson CV2.
I can't edit the previous post.... I tried again with v0.51, and it must have been a placebo - it sounds fine now.
 
Hi, with the new firmware (0.51) the DAC doesn't remember the last input. Can anyone confirm this?
 
Last edited:
I just upgraded the D903D firmware to v0.51 and I can confirm that the output is significantly lower than v0.39. I run the DAC directly into my amps and use preamp mode to control volume. If the front panel readings are correct (and were correct with v0.39) then the volume on music I play often is easily 8db(ish) lower than it was previously. I never had the volume under -21db for the quieter recordings with v0.39. Those same recordings are now -13db for similar comfortable listening. Output is set to 5V. I can't comment about the sound since I have no way to really A/B the firmware versions.
 
Just gotten my Topping D90III Discrete, and to my misfortune I joined the group of folks that are facing clicking sounds using the DAC.

Clicking sound whenever the DAC isn't getting any audio signal and click whenever receiving signal, this means every 2 clicks every TV toggle sound.

Even during silence when the movie starts and the production house title animations are displayed, the silence between those segments will also cause clicks.

In between every change in music you get 2 clicks, I believe it is a faulty unit. My research across the forums shows a very clear divide of 2 groups of us, one enjoying the sound and the other complain about this problem.

Its really a pity because the soundstage, naturalness of sound, holographic are all real, but the click due to the voltage drops and peaks is driving me crazy.

It is confirmed a voltage issue as the click happens even at -99.0dB when the speaker is totally silent the click still happens whenever you have silence in tracks.

The crazier thing is that this is not my first time facing this problem, clicks are common when DACs change across formats/bitrates or when they are not "Always On" in the .xml configuration in the XMOS driver control panel settings.

This is extra saddening as I went into the script and made "Visible" the option to "Always On" which worked for my SMSL D6-S but nothing is working on the Topping D90III Discrete.

In the other DACs, they are just simply programmed to sleep during idle hence causing the click when the voltage drops, but for the Topping D90III Discrete, it is a worse problem than the above.

I tried RCA, XLR outputs. Bluetooth, Optical, USB inputs across 2 totally separate systems but the click still happens, this is 100% due to the DAC's voltage drops and peak that have gone faulty. And quite a high amount of failures at that according to the posts in the forums across the web.

Also tried high pass filters to no effects, again as it is due to a voltage issue when at -99.0dB hence this wouldn't fix it.

I've contacted a reseller and they told me that this is not a standard problem. Which further points me to the fact that this is a fault in the DAC.

The only other contrary post I read over here I think by Jeremy was that since this is a 1-bit DAC, the PCM has to be converted to DSD at every instance hence causing the DAC-Format-Switching that causes the click? (Though im not experienced enough to comprehend this).

I guess one of the solutions in a create a "constant sound" on your system when it is connected through USB so the DAC will not sleep. But I use it with its optical input setting from the WiiM so I cant generate a constant noise.

Also can i post a question to the folks enjoying this DAC, if you are actually experience the click, just that it is a full music listening set-up hence it is triggering lesser and you are fine with it? Or simply there isnt such thing as the clicking sound even happening in the first place.

Cant believe that I sank in 999usd to be working on all these debugging, Topping please step up.... the B200 is so awesome, but I guess a DAC has much more software, firmware controls that can go wrong.

It is the Lunar New Year now so I've got to wait till 4th February in order to receive contact from official topping channels.

Update #1: I bought a 2nd hand DDC just to test out all the other inputs/ouputs like COAXIAL, AES, I2S and all of it still persists with the problem. But a silver lining is that I've received information from Topping that they will be working on a firmware update after Lunar New Year.
I updated the firmware in mine and took care of 80% of this...I cant explain it but I put long hours of low to mid volume content through it (maybe +/-100hrs). It sounded fantastic at this point and I thought the limit had been nearly reached as Id been hearing gradual improvement in key areas...Yesterday before leaving for work, I turned my amps off and let aggressive dubstep pump through it at nearly full volume (8hrs). When I got home, every performance aspect increased three-fold or more as though a dam had broke. The discrete went from talented and articulate to absolutely putting on a show of a silky 3d tapestry of incredibly accurate and musical sound. Every sound was precisely neatly in its own lane as a singular happening. Everything was now spaciously arranged left/right high and low across the soundstage....Not a harsh or tense note to be found. Wide-ranged vocals now had way more dynamism and are now only a sliver away from inducing goosebumps, and even with higher-ranged vocals ('Ride On' Celtic Woman), not a single point of became shrill or unfocused and no longer sounded constrained and dry...Cymbals shimmer and splash in mid air with perfect attack/decay in a truly captivating way..The sound now incredibly fluid and extreeemely fast..I could go on with praise////////// After this change, all clicks and pops disappeared. I wonder if it was due to some sort of higher resistance in a circuit somewhere pushing back??
 
I have updated D90 Discrete III to the 0.51 release.
The output levels that were too low in 0.49 are corrected in the new firmware. 0.39 and 0.51 have the same output levels.
Also corrected, the output level issue when switching to and from the IIS input.

- Rich
 
Last edited:
I have updated D90 Discrete III to the 0.59 release.
The output levels that were too low in 0.49 are corrected in the new firmware. 0.39 and 0.51 have the same output levels.
Also corrected, the output level issue when switching to and from the IIS input.

- Rich
Thanks; I will definitely give this a try first chance I get. I am wondering if it might resolve faint ticking that occurs when music is not playing.
 
Thanks; I will definitely give this a try first chance I get. I am wondering if it might resolve faint ticking that occurs when music is not playing.
I have had cases of faint ticking on multiple components that seems to be caused by RF of some kind.
The Oppo Sonica DAC did this until I moved it away from other components.

- Rich
 
My D90 III Discrete are used with Raspberry Pi5 (I have also used Pi4s) as Roon endpoints.
Recently, I added Pi2AES hats to both of my D90s which allows simultaneous connection of USB, COAX, AES, and I2S inputs.
In this configuration, the ROON endpoint is directly converted to each type of input without any changes in the source.
Switching between inputs is pretty fast, allowing comparison. The sound quality of all inputs is different.

I find, from best to worse, I2S, COAX, AES, and then USB.
The internal processing of D90 each input is different, I2S seems to have the most direct path since additional processing is not available.

If you are into home experiments, I recommend giving a Pi2AES a try, certainly, if you are using Roon endpoints.

- Rich
 
Switching between inputs is pretty fast, allowing comparison. The sound quality of all inputs is different.

I find, from best to worse, I2S, COAX, AES, and then USB.
The internal processing of D90 each input is different, I2S seems to have the most direct path since additional processing is not available.
Is this for PCM playback? Native DSD512 in DAC mode should sound identical via I2S and USB. Or does it not?
 
Is this for PCM playback? Native DSD512 in DAC mode should sound identical via I2S and USB. Or does it not?
It does not sound the same in my system.

Roon->Raspberry Pi5 Pi2AES->D90III Discrete->Benchmark AHB2->Revel 126BE

-Rich
 
I updated the firmware in mine and took care of 80% of this...I cant explain it but I put long hours of low to mid volume content through it (maybe +/-100hrs). It sounded fantastic at this point and I thought the limit had been nearly reached as Id been hearing gradual improvement in key areas...Yesterday before leaving for work, I turned my amps off and let aggressive dubstep pump through it at nearly full volume (8hrs). When I got home, every performance aspect increased three-fold or more as though a dam had broke. The discrete went from talented and articulate to absolutely putting on a show of a silky 3d tapestry of incredibly accurate and musical sound. Every sound was precisely neatly in its own lane as a singular happening. Everything was now spaciously arranged left/right high and low across the soundstage....Not a harsh or tense note to be found. Wide-ranged vocals now had way more dynamism and are now only a sliver away from inducing goosebumps, and even with higher-ranged vocals ('Ride On' Celtic Woman), not a single point of became shrill or unfocused and no longer sounded constrained and dry...Cymbals shimmer and splash in mid air with perfect attack/decay in a truly captivating way..The sound now incredibly fluid and extreeemely fast..I could go on with praise////////// After this change, all clicks and pops disappeared. I wonder if it was due to some sort of higher resistance in a circuit somewhere pushing back??
Which version, 0.59? Or 0.51?
 
Sure thing bro
I do many level matched listening tests with fast switching. Though, blind testing is tough. Though I have two D90s, a web controlled Shelly switch controlling an ARX trigger controlled XLR switch.

At some point, I may write an blind testing app but I have no time for that right now.

However, bias is a funny thing. There is bias that leads to hear differences that do not exist. Others, refuse to hear differences that do exist.

It would be poor science IMO to consider that all environments have been tested at ASR.
I appreciate what has and what has not been tested.

- Rich
 
It would be surprising if level matched coax and AES always sounded the same. Most AES transmissions have pulse transformers on input and output.
The reviews test for jitter and coax matches USB. Optical tests a little worse usually. Few tests for AES. I2S even fewer tests.
There may be solid reasons the inputs do not sound the same. Jitter is only one of these.
 
Back
Top Bottom