DAC - Going Stupid Cheap

[Thomas savage]

Atomic clocks and GPS just introduce more digititis. Future upgrades will allow linkages to analog clocks;

This needs blu tack to eliminate pre ringing..

 

[watchnerd]

This needs blu tack to eliminate pre ringing..

Pre-chiming in this case.

 

[amirm]

Have not opened the boxes yet but the DAC and S/PDIF adapters have arrived!

 

[watchnerd]

Have not opened the boxes yet but the DAC and S/PDIF adapters have arrived!

 

[watchnerd]

Have not opened the boxes yet but the DAC and S/PDIF adapters have arrived!

ETA for first tests?

 

[amirm]

ETA for first tests?

Not sure. Busy making Pizza tonight.

Only have tomorrow before going to CES. Maybe I get some data then.

 

[amirm]

OK, just wasted more than an hour debugging why Windows would no longer recognize my Audio Precision analyzer Have to pack for trip tomorrow morning so here is quick data for you all to chew on .

Test file is J-test run at 48 Khz sample rate. Comparison is between iFi iDAC 2 ($300) DAC against the $20 Fiio Taishan. Interface is USB on the computer to iFi DAC driving its internal DAC, or using its digital output to drive the FiiO externally. In other words the digital interface is the same for both.

As you see performance is fair bit worse on FiiO. Its noise floor is some 25 dB worse (notice that FiiO output is slightly lower than iFi). And it has two jitter sidebands that don't exist in iFi DAC (although the iFI has low frequency random noise around its main 12 Khz tone).

On the positive side it is not doing anything horribly wrong. At less than one tenth the price and less than a quarter in size and weight, it is doing pretty well.

USB power by the way is from an adapter that came with my USB microscope (i.e. some random Chinese design). iFi is self powered through my laptop.

This is preliminary/hurried measurement . So don't run with it too much until I get back from CES....

 

[watchnerd]

OK, just wasted more than an hour debugging why Windows would no longer recognize my Audio Precision analyzer Have to pack for trip tomorrow morning so here is quick data for you all to chew on .

Test file is J-test run at 48 Khz sample rate. Comparison is between iFi iDAC 2 ($300) DAC against the $20 Fiio Taishan. Interface is USB on the computer to iFi DAC driving its internal DAC, or using its digital output to drive the FiiO externally. In other words the digital interface is the same for both.

View attachment 4229

As you see performance is fair bit worse on FiiO. Its noise floor is some 25 dB worse (notice that FiiO output is slightly lower than iFi). And it has two jitter sidebands that don't exist in iFi DAC (although the iFI has low frequency random noise around its main 12 Khz tone).

On the positive side it is not doing anything horribly wrong. At less than one tenth the price and less than a quarter in size and weight, it is doing pretty well.

USB power by the way is from an adapter that came with my USB microscope (i.e. some random Chinese design). iFi is self powered through my laptop.

This is preliminary/hurried measurement . So don't run with it too much until I get back from CES....

Thank you so much for doing this!

But I think perhaps something is wrong with the screenshot as I only see the iFi data?

EDIT: LOL...NVM..it's because it's nighttime and I'm running Flux, which filters out blue light, and of course the line is super blue...

 

[amirm]

Yes, sorry. I usually pick brighter colors but I was in a hurry to get this posted.

 

[watchnerd]

While definitely inferior, it looks like the jitter of the cheap-as-hell Taishan is still below the audible threshold for jitter.

 

[watchnerd]

OK, I bought that one too. Will be interesting to use this for USB bridge to the other DAC.

Just checking, did we measure @RayDunzl 's cheapie DAC, too, yet or did it just burst into flames / disintegrate / die of shame?

 

[RayDunzl]

Just checking, did we measure @RayDunzl 's cheapie DAC, too

I don't think I have a measurement noise floor low enough - PC on-board Sound ADC - to do anything low level.

 

[amirm]

Here you go: http://www.audiosciencereview.com/f...view-and-measurements-of-budget-20-dacs.1316/

 

[watchnerd]

Here you go: http://www.audiosciencereview.com/f...view-and-measurements-of-budget-20-dacs.1316/

Wow, thanks so much!

 

[watchnerd]

Okay, now that I've read the test:

Does anyone know what sample rate is usually used by the optical out of a 4k TV?

 

[amirm]

99% of TV and film programming use 48 Khz sampling rate. So not good news on that front.

 

[watchnerd]

99% of TV and film programming use 48 Khz sampling rate. So not good news on that front.

So given FiiO seems to be targeting the Taishan at TV scenarios, they seem to have goofed.

 

[amirm]

So given FiiO seems to be targeting the Taishan at TV scenarios, they seem to have goofed.

I think for general TV listening it is probably fine. But yes, as a general rule if I were to build something for TV use, I would optimize 48 Khz. Likely that is not general knowledge for people not in the video world.

 

[Blumlein 88]

I think for general TV listening it is probably fine. But yes, as a general rule if I were to build something for TV use, I would optimize 48 Khz. Likely that is not general knowledge for people not in the video world.

I don't think it is a big secret that the video world is 48 khz. If you market something to the TV market it should have been based upon 48 khz if you are only doing one of them correctly.

 

[Thomas savage]

I think for general TV listening it is probably fine. But yes, as a general rule if I were to build something for TV use, I would optimize 48 Khz. Likely that is not general knowledge for people not in the video world.

I knew that, I'm a pretty low base mark for knowladge so...