Solid State Logic SSL 2 MKII vs. Topping Pro E2x2 OTG (Audio Interfaces) Comparison Review and Measurements

[jkim]

Excellent !

Nice to see systematic testing of those interfaces.

A few suggestions:

You may get more accurate results in mono.

For low level testing, a passive attenuator after the DAC is the most efficient.
My personal setup is to merge both of the DAC's balanced outputs with an Y cable (each output includes a resistor, so little risk to break it), followed by a Shure A15AS or Behringer Monitor1.

2 other interesting results are
- the sensitivity of the mic input for min and Max gain.
- Shorted input noise at Max gain
You need an XLR male plug, with pins 2 and 3 shorted for that measurement.
Combined with the sensitivity at max gain, you should get an approximation of EIN.
Even if you don't have an accurate True RMS Multimeter to evaluate the sensitivity accurately, you may, at least, compare the interfaces vs each other.

For DAC measurements, you may get best results by using an Y cable to send the same output to both ADC's inputs.
Then, use the ADCIso in Stereo and Cross correlation averaging on either REW or the very last Multitone (released this week).

Of course, all this works only in Mono.
And, to be frank, performance of both devices should not require that much, since they are more than 10dB worse than your ADC anyway.

And in case you want to test the headphones power, I can only recommend the excellent E1DA headphones load board.
Solder 2 XLR- terminated cables to it abd you'll have a perfectly capable load board for tests.

Thanks for this useful information! Appreciate it. When I get a chance I will have to ask some clarifying questions. Please stay tuned.

 

[jkim]

Thanks for great review! btw, would you test ADC on TRS, instead of XLR? Both of them use XLR as mic input, not as line input. It may not be critical, but I'd like to see results on TRS too!

It was a type. I would not have obtained those line-in results using the XLR mic input. Corrected in the post. Thanks for pointing that out!

 

[jkim]

Usually in this price range, the interfaces will use the same amplifying circuit as the mic input, but it will be padded down. The padding will increase the resistance. The increased resistance will raise the noise floor from thermal noise. Otherwise everything else will change very little.

PS- you appear to be posting for the first time. Welcome to ASR.

It was just a typo. Of course, I did not use the XLR mic-in for line-level signals. What you mean by "same amplifying circuit" does not include its mic preamp. Right? You mean op-amps as an input buffer, don't you? Or something else?

And thank you for your welcome note!

 

[jkim]

P.S. Can you check what DC bias voltage the ADC is getting on its inputs? This ESS chip is strange because it uses 1.9v bias. However even the datasheet seems confused because it shows V_ref 1/2 and V_ref Buf both of which are not needed. I see V-ref Buf on the PCB there which is part of the confused datasheet schematic. If this is wrong it could explain the issues at higher input level.

Interesting. Will check when I get a chance.

 

[Blumlein 88]

It was just a typo. Of course, I did not use the XLR mic-in for line-level signals. What you mean by "same amplifying circuit" does not include its mic preamp. Right? You mean op-amps as an input buffer, don't you? Or something else?

And thank you for your welcome note!

Yes it does include the mic preamp. At this price they usually use a mic preamp (often one from THAT corp). That then feeds the ADC. When you use the line in, it also goes thru that same mic preamp, but the signal is padded down. This gives the input impedance a larger number (higher input impedance less gain), and keeps line in signals from overloading the circuit of the preamp. As a result it travels thru additional resistors which have their own thermal noise which is higher from being of higher impedance.

You often see on recording forums people looking for interfaces that don't put line in signals thru the mic preamp believing doing so colors the sound vs not going thru it. They want the line in to bypass the mic preamp. With a few exceptions you normally will need to spend quite a bit more to get that in a device. Antelope has made some with their own mic circuits though not always. I think Apogee has, and maybe SPL uses a chip preamp, but you can also bypass it. It isn't common to bypass the mic pre when using line in.

Here is the company that makes the chip mic preamps often seen at the lower prices. Maybe look for one in the unit and see which they are using.

Microphone Preamps – THAT Corporation

thatcorp.com

 

[Rja4000]

They want the line in to bypass the mic preamp.

As far as I understand the block diagram, the RME 802fs-UFX II-UFX III use such an architecture.

Line inputs are inputs 1-8, in top left corner.
Mic inputs are inputs 9-12, just below.
(Extracted from the UFX II user manual)

 

[Bacchusoo7]

Forgive me if I missed it, but what was the status of the "4k" buttons during these tests? The SSL has an "analogue colour enhancement" mode which, as far as I'm aware, means adds harmonic distortion.

 

[Rja4000]

At this price they usually use a mic preamp (often one from THAT corp). That then feeds the ADC. When you use the line in, it also goes thru that same mic preamp, but the signal is padded down. This gives the input impedance a larger number (higher input impedance less gain), and keeps line in signals from overloading the circuit of the preamp. As a result it travels thru additional resistors which have their own thermal noise which is higher from being of higher impedance.

Which seems confirmed by the Specs for the Topping E2x2 OTG :
Gain range is the same, with increased max level and impedance.

1. Microphone Inputs:
Maximum Input Level | 8.6dBu
Input Impedance | 1.5k Ohms
Available Gain | 58dB + 20dB (20dB digital gain)

2. Line Inputs:
Maximum Input Level | 23.9dBu
Input Impedance | 6k Ohms
Available Gain | 58B + 20dB (20dB digital gain)


For SSL, it's less clear, since gain range is different. So it probably doesn't use the preamp for line Inputs.
Which seems confirmed in the user manual:

The LINE input bypasses the pre-amp section, making it ideal to connect the output of an external preamp to if you so wish.

Microphone Inputs
Maximum Input Level: +9.7 dBu
Gain Range: 64 dB
Input Impedance: 1.2 kΩ

Line Inputs
Maximum Input Level: +24 dBu
Gain Range: 27dB
Input Impedance: 14 kΩ

 

[Blumlein 88]

Which seems confirmed by the Specs for the Topping E2x2 OTG :
Gain range is the same, with increased max level and impedance.

1. Microphone Inputs:
Maximum Input Level | 8.6dBu
Input Impedance | 1.5k Ohms
Available Gain | 58dB + 20dB (20dB digital gain)

2. Line Inputs:
Maximum Input Level | 23.9dBu
Input Impedance | 6k Ohms
Available Gain | 58B + 20dB (20dB digital gain)


For SSL, it's less clear, since gain range is different. So it probably doesn't use the preamp for line Inputs.
Which seems confirmed in the user manual:


Microphone Inputs
Maximum Input Level: +9.7 dBu
Gain Range: 64 dB
Input Impedance: 1.2 kΩ

Line Inputs
Maximum Input Level: +24 dBu
Gain Range: 27dB
Input Impedance: 14 kΩ

Could be the SSL uses less of the gain at line level as the higher impedance would result in more noise and at higher levels than they deem necessary.

 

[jkim]

Which seems confirmed by the Specs for the Topping E2x2 OTG :
Gain range is the same, with increased max level and impedance.

1. Microphone Inputs:
Maximum Input Level | 8.6dBu
Input Impedance | 1.5k Ohms
Available Gain | 58dB + 20dB (20dB digital gain)

2. Line Inputs:
Maximum Input Level | 23.9dBu
Input Impedance | 6k Ohms
Available Gain | 58B + 20dB (20dB digital gain)


For SSL, it's less clear, since gain range is different. So it probably doesn't use the preamp for line Inputs.
Which seems confirmed in the user manual:


Microphone Inputs
Maximum Input Level: +9.7 dBu
Gain Range: 64 dB
Input Impedance: 1.2 kΩ

Line Inputs
Maximum Input Level: +24 dBu
Gain Range: 27dB
Input Impedance: 14 kΩ

Could be the SSL uses less of the gain at line level as the higher impedance would result in more noise and at higher levels than they deem necessary.

Now I understand it. Great information. Topping's implementation explains why its gain knobs are SO sensitive with line inputs, compared to SSL's. 60 dB gain is unnecessary for line level signals. But if line and mic ins had to share the same gain structure (i.e., same preamp circuit), they didn't have a choice. I believe many low-cost audio interfaces have the same issue.

 

[jkim]

Forgive me if I missed it, but what was the status of the "4k" buttons during these tests? The SSL has an "analogue colour enhancement" mode which, as far as I'm aware, means adds harmonic distortion.

I immediately disregarded the "4K" effect. It must be simply for fun experiments. It adds a huge amount of harmonics.

 

[Bacchusoo7]

I immediately disregarded the "4K" effect. It must be simply for fun experiments. It adds a huge amount of harmonics.

I figured as much, but with the results of the sine sweep, it made me wonder.

Thanks

 

[andivax]

Remember. It's not a distortion. It is the legendary SSL saturation! Just kidding.
It's really frustrating that cheap interfaces are so bad these days.
Interesting how stable the drivers and how handy is the mixer (routing, loopback).

 

[GWolfman]

Thanks! I’m just getting into interfaces so I’m happy to see these tests. Appreciated!

 

[Arthur Edwards]

Your SINAD test for the E2x2's ADC was 101dB, worst case, which does not appear to correlate with the Audio Science result of 109 dB.

You provided a voltage test level for the SINAD test, but it is unclear if this is for -0.5dBFS. ADCs should not be tested at 0dBFS. REW should indicate the ADC's digital output level.

This review was informative. Thanks for the work you put into this.

 

[jkim]

Your SINAD test for the E2x2's ADC was 101dB, worst case, which does not appear to correlate with the Audio Science result of 109 dB.

You provided a voltage test level for the SINAD test, but it is unclear if this is for -0.5dBFS. ADCs should not be tested at 0dBFS. REW should indicate the ADC's digital output level.

This review was informative. Thanks for the work you put into this.

Good observation. Multitone Loopback Analyzer does indicate ADC's digital output level. If you enlarge the chart, you will see "Lvl" at the bottom status bar. It was -0.9 dBFS for that test.

There must be two reasons why my SINAD measurement of E2x2's ADC is worse than Amir's. One is the quality of the test tone for which I used Topping D 50III's DAC output. It should add some harmonics compared to AP's test tone. Another reason---this is my guess---is that Amir must have used somewhat lower input gain on E2x2. As my test shows, there is some saturation close to ADC's full scale output. I wish Amir's AP result showed ADC's digital output level as well..

 

[Rja4000]

I wish Amir's AP result showed ADC's digital output level as well..

Well, it does show the digital input level and voltage :
4Vrms, -0.14/0-0.16dBFS for line input

Also, distortion is hardly increasing with level

To compare with Amir's results, you should switch to line input.

The distortion of the D50 III is vanishing low, well below -130dB, so that can't have much impact on what you see here, which is much higher. (Although the input impedance here is lower)

 

[Sokel]

Your SINAD test for the E2x2's ADC was 101dB, worst case, which does not appear to correlate with the Audio Science result of 109 dB.

You provided a voltage test level for the SINAD test, but it is unclear if this is for -0.5dBFS. ADCs should not be tested at 0dBFS. REW should indicate the ADC's digital output level.

This review was informative. Thanks for the work you put into this.

For the ADC DAC part Topping states an 100dB SINAD (A) weighted @ 24bit 96kHz,and the same for the aux and headphone output.

Does not state the frequency though,is that maybe 20Hz-20kHz?
If not the results are ok for ADC.

 

[Rja4000]

For the ADC part Topping states an 100dB SINAD (A)

This is the DAC (Output).
The ADC is at 110dB.

 

[Sokel]

This is the DAC (Output).
The ADC is at 110dB.

Oops!Coffee didn't work yet,you're right,I'll edit it.