Review and Measurements of Benchmark DAC3

[amirm]

Now how audible is a dip of a decibel or two below -90 dbFS?

Well, in the context of Benchmark as a company, they would be out of business if the answer to that is: it doesn't matter.

A DAC can be built for $10. As price goes up, there is expectation of higher and higher precision. These tests aim to show if that precision is achieved or not. If it is not, then it is an important datapoint in purchase decision.

I hope in time to create some sort of evaluation of sonic effects of these differences. For now, my objective is to provide measurable data that can differentiate products.

 

[gvl]

Good point. There is actually a "settling algorithm" behind the scenes in Audio Precision software. Indeed I have had to optimize this to get stable values at very low measured levels. It has choice of averaging or attempting to arrive at minimal error using multiple measurement point. There is also settling time to make sure everything becomes stable before being measured after a parameter change.

Thanks, interesting. Btw, is the noise you consider in your linearity experiments band - limited? I would think the ultrasonics can be safely ignored as they will be filtered by speakers and human ear and won't be audible.

 

[amirm]

Thanks, interesting. Btw, is the noise you consider in your linearity experiments band - limited? I would think the ultrasonics can be safely ignored as they will be filtered by speakers and human ear and won't be audible.

There is a 50 dB attenuation on each side of the 200 hz tone. Since noise is broadband in nature, it is hugely reduced in amount using this filter. This is why the output looks like a clean sine wave.

 

[Thomas savage]

(Snip)

And yes, I am not being contentious with Amir or anyone here. Hope it doesn't come off sounding that way. It is a matter of discussion about how things are done. I learn a lot that way (especially when I put up something wrong and it is pointed out). BTW, glad to see an explanation of how AP does linearity using 200 hz tones and 2 db steps.

This is what these threads are really for imo, that and educating folks so I thank you Dennis for you time , line of questioning and resulting challenge.

I would like to highlight this exchange as a ‘ perfect ‘ example of how we want to go about things, it’s what I want to see and it’s vital in terms of what we are trying to do here at ASR. .., peer review.

 

[gvl]

There is a 50 dB attenuation on each side of the 200 hz tone. Since noise is broadband in nature, it is hugely reduced in amount using this filter. This is why the output looks like a clean sine wave.

You said it before, I should have been paying more attention.

 

[rebbiputzmaker]

A DAC can be built for $10. As price goes up, there is expectation of higher and higher precision. These tests aim to show if that precision is achieved or not. If it is not, then it is an important datapoint in purchase decision.

I hope in time to create some sort of evaluation of sonic effects of these differences. For now, my objective is to provide measurable data that can differentiate products.

This precision may or may not actually be that important of a point to one's purchase decision.

It all depends, as you state the eventual need to evaluate the sonic effects of these differences. This also brings into play the other DACs that might have been tossed aside prematurely without this eventual sonic evaluation.

 

[Jimster480]

Wow very unexpected results, I didn't expect the ADI pro and the Topping D50 to measure up to the UDAC3...
I didn't read every page here, but did amir ever test the headphone output?

 

[Soniclife]

I didn't read every page here, but did amir ever test the headphone output?

He did, most interesting it was too

 

[bennetng]

There is a 50 dB attenuation on each side of the 200 hz tone. Since noise is broadband in nature, it is hugely reduced in amount using this filter. This is why the output looks like a clean sine wave.

Just a visualization of a 24-bit recorded 200Hz fade in tone before and after a narrowband filtering. The non-centered waveform became centered and the envelope became very clean.

Don't know if the AP is doing similar things or not but that's what I can do with Adobe Audition.

 

[Jimster480]

And for one more sanity check, here is the same FFT this time augmented with Topping D50 being tested at 203 Hz as to shift its response to the right so that we can see both:

View attachment 13409

Notice how the Topping peaks land on the correct values and the two channels are identical in value down to -120 dB line.

Crazy how well the Topping stuff is measuring!

 

[Jimster480]

He did, most interesting it was too

Do you know what page its on?

 

[jaynyc]

Crazy how well the Topping stuff is measuring!

perhaps it might also be interesting to see if the Topping stuff exhibits unit-to-unit performance variance. Sometimes lower price point products are vulnerable etc

 

[Jimster480]

perhaps it might also be interesting to see if the Topping stuff exhibits unit-to-unit performance variance. Sometimes lower price point products are vulnerable etc

Well I could surely send in my D30 to test and he has one too so he can see...

 

[amirm]

Do you know what page its on?

The earlier tests where to see if THD changed less on benchmark vs RME on real headphone loads (it didn't).

But I had not posted the standard dummy load measurements. So here they are, first the DAC3 measurements alone:

We see the typical load dependent graphs with clipping happening earlier and earlier.

Now let me overlay the output of RME ADI-2 Pro over the same chart:

The RME has higher noise level as evidenced by the line being above DAC3 but as levels are increased, except for 33 ohm where it all of a sudden shuts down, it has no clipping! As such, it beats the DAC3 above 1 volt output or so.

Mind you, the DAC3's worst case distortion is 0.004% so not an audible concern. Just that objectively after reading all the white papers on Benchmark site on their headphone performance, it is fascinating to see RME doing a better job overall. Maybe this is the reason they are adopting the THX technology for their next generation headphone amps.

 

[amirm]

And oh, here is the output impedance:

 

[Jimster480]

And oh, here is the output impedance:

View attachment 13437

Very good performance overall, but as you said it is surprising to see RME do a better job vs Benchmark.
I know you have a new analyzer now, but I think it would be cool to see some Graphs with DX7/DX7S/ADI DAC/UDAC3 on the same graph since they are all kinda the same combo type units.

 

[amirm]

Very good performance overall, but as you said it is surprising to see RME do a better job vs Benchmark.
I know you have a new analyzer now, but I think it would be cool to see some Graphs with DX7/DX7S/ADI DAC/UDAC3 on the same graph since they are all kinda the same combo type units.

I will be testing them again as I compare them to other products in the future. For now, I have too much fresh stuff to review.

 

[Jimster480]

I will be testing them again as I compare them to other products in the future. For now, I have too much fresh stuff to review.

Fair enough! Can't wait

 

[ManglerFi]

@Jimster480, maybe I've missed something but why do you keep calling the Benchmark DAC3 "UDAC3"?

 

[SIY]

You held the key to the issue Thomas and you just didn't know it!!!

...Thanks for staying on me folks to get this sorted out.

I'm not sure that it's sorted. Let's say one channel is off by a constant (say) 3uV. So 2.0 v in one channel corresponds to 1.999997 in the other channel. At that 2V, the difference is 1.3 exp -5 dB, which just won't be indicated on that graph. At 1mV, the "off" channel will be 0.997 mV, which is 0.02dB, which will show up. At 10uV vs 7 uv, it's 3.1 dB... you get the point. Both channels could be tracking perfectly, but the log nature of the plotting will make it look as if the offset is a nonlinearity.

Can you eliminate that possibility (or punch a hole in my logic)?