• 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!

Topping E50 Review (Balanced DAC)

a 1m or shorter regular RCA cable should work fine, I've had issues with longer ones.
If you have a video RCA cable (with a yellow connector typically), those are 75 ohm too and work just as well.


Yes, but it doesn't switch automatically to the live input. You'll have to switch to the input by pressing the multifunction touch button at the front, or left/right on the remote.
i reconnect back both the optical and coaxial inputs and do the manual switching, both are working now :=)
thanks guys :=)


another question, as Topping E50 only has one optical and one coax inputs, is it better for the cd-player to use the optical and the streamer on coax?
 
I have E50 but after some times it turns off . When I try to turn it on again it automatically turn it self off again in a second or so
 
I just bought the E50 and it's an improvement over the original Khadas ToneBoard. I didn't think I'd be able to hear the difference between the two?
 
I just bought the E50 and it's an improvement over the original Khadas ToneBoard. I didn't think I'd be able to hear the difference between the two?
Hearing a difference is easy unless you match the output volumes exactly using a multimeter.
 
Excuse me for the newbie question: can I use the coaxial plug of the E50 to plug the DAC to a set of studio monitors that have coaxial input? I would plug my tablet to the E50 via USB cable, and then the E50 to the studio monitors through coaxial.
 
No, the coaxial is an input only. What other input connectors are on your studio monitors?

If possible, you want the RCA or TRS from the E50 going straight into the monitors (active I assume). The signal going out of the E50 this way is exceptional quality.
I can get a set of Presonus e3.5 monitors that have a TRS balanced input, so I guess that would work then! :)
 
  • Like
Reactions: GPJ
Based on Amir's review, I purchased an E50 with an eye toward using it as the signal source with an E1DA Cosmos Scaler, ADCiso, and APU (in some cases). Unfortunately, I'm not getting the same results as Amir got. Primarily, the 3rd harmonic of the 1000 Hz tone is much higher than what he published. Mine is about 10 dB worse; around -123 dBc under the same conditions as Amir tested. The other harmonics measure very close to his results. Same for the noise level.

To verify the rest of the test system, I measured a D10s. My results were almost exactly what he got.

I also tried using the APU to notch the 1000 Hz fundamental. Same high 3rd harmonic.

Finally, I used a "Victor" test generator, which has outstanding harmonic performance. This really convinced me that the spectrum analysis part of the system is running as it should.

So, it's the E50.

My question is: Does anybody know how to tweak the distortion cancellation coefficients in the ESS ADC chip inside the E50 to minimize distortion? Has somebody figured out how to access those?

Thanks!
 
Based on Amir's review, I purchased an E50 with an eye toward using it as the signal source with an E1DA Cosmos Scaler, ADCiso, and APU (in some cases). Unfortunately, I'm not getting the same results as Amir got. Primarily, the 3rd harmonic of the 1000 Hz tone is much higher than what he published. Mine is about 10 dB worse; around -123 dBc under the same conditions as Amir tested. The other harmonics measure very close to his results. Same for the noise level.

To verify the rest of the test system, I measured a D10s. My results were almost exactly what he got.

I also tried using the APU to notch the 1000 Hz fundamental. Same high 3rd harmonic.

Finally, I used a "Victor" test generator, which has outstanding harmonic performance. This really convinced me that the spectrum analysis part of the system is running as it should.

So, it's the E50.

My question is: Does anybody know how to tweak the distortion cancellation coefficients in the ESS ADC chip inside the E50 to minimize distortion? Has somebody figured out how to access those?

Thanks!
What can cause this sample to sample variation?
10db worst is no joke when your purpose is exactly this performance and not audibility.
Maybe send it to Amir for verification?

(we already know that 1 sample is never enough but here the difference is huge,we're talking about 120db vs 110-112db SINAD perhaps? )
 
If you're interested, I could post my test results. That might be interesting, but won't fix anything.

Your suggestion about Amir is a good one, but I know that he's very busy and I wouldn't want to ask him to do something that isn't part of his mainstream activity - reviews. Besides, he may not know how to correct this anyway.

This unit probably meets the published specifications, so it wouldn't be covered under warranty. I'm just hoping to improve the performance for this application. I suspect it could be that the distortion correction is misaligned. With the right software to tweak the registers, I could fix that here.

If I can't I'll probably give the DAC away and stick with the D10s. It's got worse noise, but the distortion is better. No reason to keep a DAC for measurements when it doesn't do what you want. As you say, it's almost certainly audibly perfect.
 
Aww, what the heck...

This is a comparison I made this morning of the D10s and the E50. It's not a direct analog to Amir's measurements, because it's done at 384 KHz sampling rate and I did not optimize the levels for noise. It was intended to compare distortion levels.

Red is the E50 unbalanced output. Black is the D10s output. No notch of the fundamental. Same levels. Just a swap of the DACs.

Topping v Topping.png
 
Both look like they'd do the "job".
Well, maybe.

Imagine that you saw a review of a sports car from a very reputable tester. The car was great in every way. In addition, it was an absolute bargain. The tester found that the car could get to 100km/hr in 4.0 seconds, which isn't in the 3 second range of the really expensive guys, but for the price it's still magnificent.

So, you buy one with the intent of converting it to a race car in the appropriate category. Your first step is to test the car to get a baseline of where you're starting from.

You discover that no matter what you do, the fastest it will do 0-100km/hr is 4.8 seconds. This wouldn't matter for your ride to work, to the beach, or to the grocery store. But, it might not be so competitive as a race car.

Now, do you have a bad sample? Is something misaligned? Is it fixable? Was the sample provided to the tester a "ringer" and not representative of actual production units?

These are my questions.

I'm not suggesting that anything is underhanded or corrupt here. I just think that there may be something wrong with my unit. OK, maybe not wrong, but not optimized. Since both channels test pretty much the same, it doesn't seem likely to be a bad DAC chip or opamp. Besides, the noise performance is really great. Here, the 2nd and 3rd harmonics are worse than in the D10s sample I own. That doesn't seem right.

I know that ESS DAC chips use distortion compensation that can be tweaked and saved in the firmware associated with the chip. At least one manufacturer provides a means to do that at home. I'm hoping that Topping has something similar, or that somebody has figured out how to do it unofficially.
 
Back
Top Bottom