Topping D10s USB DAC and Bridge Review

[Rasto78]

I'm still wondering about DSD audio (I only have one song in DSD). Is there a difference in playing DSD or PCM ? I have not been able to switch this converter to DSD mode. I use AIMP player (I also tried Foobar2000 and followed the instructions in the driver manual, I could not switch to DSD).

 

[staticV3]

I'm still wondering about DSD audio (I only have one song in DSD). Is there a difference in playing DSD or PCM ?

Not in the format itself, but DSD releases tend to be mastered differently than their PCM counterparts and that can definitely be audible.

I have not been able to switch this converter to DSD mode. I use AIMP player (I also tried Foobar2000 and followed the instructions in the driver manual, I could not switch to DSD).

Here's how to play DSD natively using foobar: https://www.audiosciencereview.com/...sl-actually-support-dsd512.42194/post-1491753

 

[Jimster480]

There's no deeper meaning behind these values, they're effectively random and you can change them as you see fit.


No. Anything beyond 48kHz serves zero purpose for music playback.


Rasto can see all sample rates with both DACs, it's just that they're set to those values by default.

Ideally you want to use like 44.1/24bit so that audio quality isn't limited and it also isn't resampled. Since the way the audio stack works is that if it is 16bit it just adds 0's which doesn't do anything to the quality of the music. If it is on 48k then it has to resample all music to a different sample rate.

 

[Foxxy]

Just a question: how susceptible is a USB powered DAC to the USB power quality. I know it has voltage regulators and such but how much can they do? Asking because I am hanging on a powered USB hub with a questionable power brick.

 

[gvl]

Just a question: how susceptible is a USB powered DAC to the USB power quality.

Depends on a DAC, some are better than others. The question you should be asking is if it’s audible or not in your case, and there’s only one way to find out which is to try it. Audible USB power issues usually present themselves as hum or buzz, if there’s none it’s all good.

 

[Foxxy]

Depends on a DAC, some are better than others. The question you should be asking is if it’s audible or not in your case, and there’s only one way to find out which is to try it. Audible USB power issues usually present themselves as hum or buzz, if there’s none it’s all good.

Thanks, silent as a rock

 

[waspa]

Hi.I broke my LME4970NA and have now OPA 2604.Can replace without any solution?

 

[Descartes]

Just bought one looking forward to using it for my iPad

 

[artismo]

Hi.I broke my LME4970NA and have now OPA 2604.Can replace without any solution?

John Yang, the lead designer, specifically suggested the OPA1612 or OPA1656 for a reason. While the OPA2604 is a classic, it is worth keeping a few technical realities in mind before you commit to that swap. I run the OPA1656 in mine for uniformity electrically and sonically. Maybe I'll get bored and do some experiments with my 1612 in there.

The OPA2604 is objectively noisier than the modern alternatives, and it has significantly less gain-bandwidth product, meaning it has less "room" for error correction in complex feedback loops. While the OPA2604’s slew rate is technically sufficient for audio, as you typically only need about 2V/us to hit 20kHz without distortion, that isn't the whole story.

You have to consider the weakest link rule. Since your IV stage is likely already optimized for a high-performance chip - the OPA1656, you are ultimately at the mercy of the upstream circuitry. Replacing a transparent, high-precision part with a 30-year-old FET design will likely make that stage the bottleneck of your entire signal path- if you're soldering. But since you're swapping the socketed op-amp, there's more below.

Then there is the headroom trap. You need to look at your specific supply rails. If your internal rails are tight, like 5V, that OPA2604's output swing limitations will severely restrict your dynamic range. It is "doable" with a massive asterisk—you might get sound, but you aren't getting the design's intended performance. If the goal is to get the device back to its design spec, stick to what the designer recommended. If you are just experimenting for a different "flavor," just be aware that you are trading measured performance for a different aesthetic.

Swapping that socket, the LPF/buffer side with the OPA2604*:
Speed mismatch - You’re following a high-speed, wide-bandwidth I/V stage OPA1656 with a significantly slower, vintage-spec buffer OPA2604. This can create phase shifts and distortion artifacts that the original design was meant to avoid.
Loading effects - The OPA2604 has different input capacitance and bias requirements compared to the OPA1656. Depending on the feedback network the designer used in that LPF stage, the OPA2604 might not have the loop gain necessary to maintain stability or low distortion at higher frequencies.

Which is why I just listen to the designer, and we are lucky enough to have him in the forums. So don't take just my word for it. I stuck with the OPA1656/OPA1612 - he knows what works and makes electrical sense.

[edited for clarity and punctuation]

Parameter	OPA2604	LME49720	OPA1656	OPA1612
Input Topology	JFET	Bipolar	SoundPlus FET	Bipolar
THD+N (1 kHz)	0.0003%	0.00003%	0.000032%	0.000015%
Input Voltage Noise	10 nV/rtHz	2.7 nV/rtHz	2.9 nV/rtHz	1.1 nV/rtHz
Gain-Bandwidth Product	20 MHz	55 MHz	53 MHz	80 MHz
Slew Rate	25 V/us	20 V/us	24 V/us	27 V/us
Input Bias Current	100 pA	10 nA	10 pA	60 nA
Input Offset Voltage	1 mV	0.1 mV	0.05 mV	0.1 mV
Supply Voltage Range	+/- 4.5V to +/- 24V	+/- 2.5V to +/- 17V	+/- 2.25V to +/- 18V	+/- 2.25V to +/- 18V
Output Current	+/- 35 mA	+/- 26 mA	+/- 100 mA	+/- 60 mA