Topping D90 Balanced USB DAC Review

[barrows]

@amirm , @WolfX-700 , can either of you test the D-90 with DSD 256/512 via the direct (no onboard oversampling) DSD mode? I am very interested in this DAC's performance with DSD input, and I would like to know the output level with direct DSD input.
Anyone here using the D-90 this way with HQPlayer or Roon oversampling, I would be interested in hearing your opinions on sound quality and output level. I am aware that the volume control will not work in this mode.
Thanks!

 

[Veri]

can either of you test the D-90 with DSD 256/512 via the direct (no onboard oversampling) DSD mode?

Does the D90 even have a direct DSD mode..? Pretty sure it is processed by the AK4499 in a "default" manner.

 

[barrows]

Contributors to this thread have suggested it does, a mode where the volume control does not work, i hope they will weigh in and confirm. the chip certainly allows for this, it is right in the data sheet, Direct DSD goes directly from the input to the converter stage, bypassing the DS modulator and DSP stages.
I did a quick search to try and find a D-90 manual but was not successful.

 

[tonapo]

@amirm , @WolfX-700 , can either of you test the D-90 with DSD 256/512 via the direct (no onboard oversampling) DSD mode? I am very interested in this DAC's performance with DSD input, and I would like to know the output level with direct DSD input.
Anyone here using the D-90 this way with HQPlayer or Roon oversampling, I would be interested in hearing your opinions on sound quality and output level. I am aware that the volume control will not work in this mode.
Thanks!

I use Roon, and HQ Player and I am currently upsampling to DSD 256. I use an Allo USBridge Streamer, which had some issues with popping, which seemed more prevalent with DSD 512. However I prefer the newish 'EC' modulator and there are no consumer CPU's powerful (so I understand) enough to run DSD 512 with it, but 256 sounds great. So, to confirm, I am running Poly-sinc-ext2, ASDM7EC and DSD 256. I am using the DAC in XLR and DAC only modes. I saw in another thread (audiophile style ??) that the DAC is in DSD mode automatically when configured this way.

 

[barrows]

Hi @rosscouk, Can you confirm that the volume control does not work on the D-90 when you play DSD? That would be a good indication that it is using direct mode. If the volume control works it is definitely not, as that would require the modulator to be the path.

I agree on HQP-4 EC. I listened here to it and loved the sound of ASDM7EC and Poly-sinc-ext2, but could only listen briefly with my machine (stutters). Part of my "list" is getting a good I9-990K machine built for HQP4.

 

[tonapo]

Hi @rosscouk, Can you confirm that the volume control does not work on the D-90 when you play DSD? That would be a good indication that it is using direct mode. If the volume control works it is definitely not, as that would require the modulator to be the path.

Hi @barrows , I had to dig out the Topping controller (I have just got hold of 2nd hand Logitech Harmony remote) to adjust the volume. When I try, up and down (multiple times), there is no effect to the playing track. So perhaps we can assume it is in direct mode??

I agree on HQP-4 EC. I listened here to it and loved the sound of ASDM7EC and Poly-sinc-ext2, but could only listen briefly with my machine (stutters). Part of my "list" is getting a good I9-990K machine built for HQP4.

I am using a i7 9700 (my desktop machine, Roon on a NUC) overclocked to 5ghz and it manages ASDME7 at about 40% load. I did have issues because I checked the 'multicore DSP' and I also then had stutters, but greyed out, it plays no problem.

 

[barrows]

Yeah, i would not say that is definitive proof of direct DSD, but it is a good indication, as why not allow volume control if you are not going direct? I guess I need to inquire direct with Topping if possible.
Yeah, is your machine I7-9700K? I heard that one works. I just figured if I am building new might as well go with I9-9900K to be certain all is good as the price is not much different there, and I would rather be safe. Are you running Roon on the same machine together with HQP (integrated)? I work for Sonore and we are semi partners with Small Green computer, so I would be using the proprietary linux based sonic orbiter OS, it has great integration with both HQP and ROON.

 

[Matias]

@amirm , @WolfX-700 , can either of you test the D-90 with DSD 256/512 via the direct (no onboard oversampling) DSD mode?

I suspect that both already returned the DACs to their owners a few days since the reviews, so they don't have them anymore to do more measurements.
I could be wrong though.

 

[barrows]

Well, I am very close to pulling the trigger on one anyway, just need to confirm thta it indeed operates in direct DSD mode. It is a very affordable way to audition the 4499 chip's ability with DSD input.

 

[tonapo]

I guess I need to inquire direct with Topping if possible.

Please do let us know if you do hear from Topping.

Are you running Roon on the same machine together with HQP (integrated)?

No, I have an i7 NUC which runs ROCK. HQ Player is on my office desktop. I have had both on the NUC running windows previously, but it was a pain to manage headless, so I went back to ROCK on the NUC and I prefer this way. However, if you have a machine to do both easily, then that is a more descrete and preferable option.

 

[barrows]

However, if you have a machine to do both easily, then that is a more descrete and preferable option.

Yeah, I would hope to run both a dedicated machine that does nothing else, including storage onboard, just to keep things simple. I use a Sonore Signature Rendu SEoptical as an Ethernet endpoint in my system (as NAA or RAAT). Yeah, I'll post whatever I find out. I need to learn if the the Topping USB interface is ok with native DSD/linux via the Signature, probably is being XMOS, but I need to be sure.

Edit: I was just looking at the AKM 4499 data sheet, note that DSD performance degrades at 512, but 256 achieves highest performance the chip is specified for, so I would recommend not using DSD 512 with this chip, just go to 256. This is also good news for use with HQPlayer and the EC modulators.

 

[tonapo]

Edit: I was just looking at the AKM 4499 data sheet, note that DSD performance degrades at 512, but 256 achieves highest performance the chip is specified for, so I would recommend not using DSD 512 with this chip, just go to 256. This is also good news for use with HQPlayer and the EC modulators.

Now that is very good to note! Thank you. Where abouts are you looking, as I wouldnt mind seeing?

 

[barrows]

Data sheet, just read it.

 

[JohnYang1997]

Subjective opinion. I found the D90 to be more dynamic in the low frequencies and better imaging, but the D70 has a bit more exciting midrange, touching vocals. I listened both DACs with AT-HA22TUBE and Sennheiser HD660S. For now I prefer D70, but I will still compare for a final decision.

By the way, my unit has 2x speed issue with 44.1kHz, it needs to be fixed. But I think this does not affect other modes.

It's probably the XO for the 44.1khz/88.2khz was misplaced with the one for 48khz/96khz.

 

[JohnYang1997]

Data sheet, just read it.

Are you talking about the output current? If so then it doesn't apply to the current D90 which uses low current mode for all formats.

 

[barrows]

Are you talking about the output current? If so then it doesn't apply to the current D90 which uses low current mode for all formats.

No, the distortion and noise specs for different rates of DSD. The performance degrades with DSD 512 vs. lower rates.

 

[JohnYang1997]

No, the distortion and noise specs for different rates of DSD. The performance degrades with DSD 512 vs. lower rates.

The lower specs are caused by lower output current. Lower output current = higher gain in the analogue output stage = more noise.

 

[777]

The lower specs are caused by lower output current. Lower output current = higher gain in the analogue output stage = more noise.

Lower current -----> higher resistor in I/V converter, not higher gain. Of course, bigger resistor value mean higher noise.

 

[JohnYang1997]

Lower current -----> higher resistor in I/V converter, not higher gain. Of course, bigger resistor value mean higher noise.

It's still gain. See it as inverting amp (voltage gain) or transimpedance amp (transimpedance gain). Also you don't necessarily increase the i/v resistor. LPF can have higher gain after i/v.

 

[777]

Not the gain generate the noise.