• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. 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!

Review and Measurements and miniDSP 2x4 HD DSP and DAC

Another interesting things is that the miniDSP with Dirac engaged drops the volume by about 10db, they state that is is what dirac does to make sure the DSP does not clip or distort on any curve that might have frequencies being raised, but the measurements Amirm showed seem to indicate that with the volume lowered it has pretty low resolution.
 
Let's compare DAC linearity...

Curiosity:

How is the linearity measurement taken?

If it is a single frequency, and an amplitude sweep...

How is the linearity result compromised if there's an "idle tone" of a few dB at the selected test frequency?

Would you get two different linearity measurement results with different test frequencies?
 
How is the linearity measurement taken?

If it is a single frequency, and an amplitude sweep...

How is the linearity result compromised if there's an "idle tone" of a few dB at the selected test frequency?
A low frequency sine wave is used (200 to 400 Hz). This is fed to the DAC. The output is then bandpass filtered to the same frequency, excluding all the noise and distortion above and below the test frequency. This cleans up the response of the DAC and measurement ADC. Without it, you can't remotely go down to such levels.

Would you get two different linearity measurement results with different test frequencies?
I have tested both 200 and 400 Hz with the same outcome.
 
Thanks, I understood more of those measurements this time around! It may indeed be that it's the resampling which is the primary culprit here. No way to avoid that with toslink output, I guess?

Based on my highly attenuated tech knowledge in general and specifically about MiniDSP products -- Toslink (or indeed 75 ohm SPDIF or AES/EBU, in their devices that offer it) must go thru BB SRC4382 as input rcvr and, unfortunately, SRC to create 48K or 96K I2S (Sharc 48 or 96 based on "plug in" OS platform chosen to run on device) to "feed" to Sharc device...

-- this is why I was so anxious to hear just how super the USB decode quality would via the XMOS device straight-thru to Sharc platform....but it really isn't so super after all...insert 3 sad trombone notes here...

Corrections to the above welcomed!
 
Last edited:
I spent a lot of time thinking about buying miniDSP 2x4 or 2x4 HD, but after reading this review and I decided to buy Topping d10.
Thanks to Amirm.

ps. I expect SMSL AD18 review.
 
Last edited:
Hi there. Welcome aboard. My strategy is to get a great DAC and perform the DSP/Room EQ in software in a host PC. That way you don't have to optimize for two things at once (DSP and best in class DAC).
 
Hi there. Welcome aboard. My strategy is to get a great DAC and perform the DSP/Room EQ in software in a host PC. That way you don't have to optimize for two things at once (DSP and best in class DAC).

Fully agree with you!
Foobar2000 with MathAudio Room EQ was excellent
 
Hi there. Welcome aboard. My strategy is to get a great DAC and perform the DSP/Room EQ in software in a host PC. That way you don't have to optimize for two things at once (DSP and best in class DAC).

Same here, but the crossover functionality of the miniDSP can be attractive depending the type of loudspeakers in use.
 
I own a MiniDSP 2x8 that I use for an active crossover in a 2.2 system and can confirm the 10db drop in output but the rest of my system can compensate for it without introducing noise. I think others have found ways to keep the signal path purely digital from source to the final analog stage and bypassing internal DACs using i2s hacks but I haven't found it necessary as I can't discern any degradation and the sound (and measurements) is better than the passive XOs that were used prior. I find the MiniDSP 2x8 a pretty amazing piece of kit considering what you can do with it. Some people roll tubes and op amps, I like to roll speaker drivers and this tech makes it really easy to do that. Doesn't matter how expensive or unique the rest of your system is, the most important part are the speaker components as that's where the sound ultimately comes from....

I'm sure I'll try the i2S bit someday but at a certain point the audible gains are minuscule compared to the time and $$$ invested to make it maybe better.
 
Same here, but the crossover functionality of the miniDSP can be attractive depending the type of loudspeakers in use.
Agreed it is a good tool there. The solution with using a PC with multiple output channels is more challenging.
 
Amir, any chance you'd be willing to review and measure the original MiniDSP 2x4? I searched the site and didn't see it show up in case I missed it. I've got a pair of those in an active 3-way setup and I absolutely love it, but I'd be curious to see how it actually measures. I've been looking to get a new DAC for some upfront processing (streaming/USB stuff) but don't want to go too overboard seeing as how the MiniDSP converts everything back to digital and then analog again, doesn't make sense to get too fancy on the main DAC. I've ready through the whole data sheet on the ADAU1701 but there's little focus on the the ADC/DAC and more on the DSP side, maybe you've measured that IC in another unit? Thanks! And I love all the reviews, very informative stuff!
 
Welcome to the forum Aktiondan. I have not reviewed the original unit. And given the large backlog of purchased and loaned gear, I don't think it is in the cards for me to buy and review it.
 
miniDSP is a Hong Kong based company, and they don't do excellent job, so gotta avoid them at all cost, I know this from the start
 
miniDSP is a Hong Kong based company, and they don't do excellent job, so gotta avoid them at all cost, I know this from the start
That's an aweful harsh position. Quite a few of our members use miniDSP products and are quite pleased.
YMMV
 
Back
Top Bottom