Topping E30 DAC Review

[somebodyelse]

There are a number of designs around for devices that will allow for self-administered ABX testing. The switching and level matching are usually fairly similar for a given sort of device under test. The controllers have more variation, with software, discrete logic and mechanical switches having been used in different designs to produce randomness in X. See for example https://sound-au.com/abx-tester.htm or https://www.head-fi.org/threads/discrete-logic-abx-switch-box.625791/

The first problem is that they aren't readily available off the shelf. If they were available then they'd be both too expensive (you want me to spend how much to 'prove' I can hear this night-and-day difference???) and not expensive enough (clearly the device isn't transparent enough, which is why I can't tell the difference). Or we'd end up in arguments about whether you could hear a difference in the sound of the switching rather than in the device, as has been documented in at least one of the papers.

 

[samsa]

There are a number of designs around for devices that will allow for self-administered ABX testing. The switching and level matching are usually fairly similar for a given sort of device under test. The controllers have more variation, with software, discrete logic and mechanical switches having been used in different designs to produce randomness in X. See for example https://sound-au.com/abx-tester.htm or https://www.head-fi.org/threads/discrete-logic-abx-switch-box.625791/

The first problem is that they aren't readily available off the shelf. If they were available then they'd be both too expensive (you want me to spend how much to 'prove' I can hear this night-and-day difference???) and not expensive enough (clearly the device isn't transparent enough, which is why I can't tell the difference). Or we'd end up in arguments about whether you could hear a difference in the sound of the switching rather than in the device, as has been documented in at least one of the papers.

I looked at both of those.

The single blind part of the first one is OK, but there never was any serious issue about setting up the single blind experiment. The double-blind part is completely deterministic and hence could only be used once before you had to throw it away and have your buddy wire up a new one (and, even then, it's not even pseudo-random -- hence not capable of producing a scientifically-reliable result).

The second one was apparently never built; it's hard for me to tell, from the schematic, whether it would actually work as intended.
One of the commenters on the head-fi.org thread suggested using an Arduino, which would be the right approach, but no design was forthcoming.

The only working example I've been able to find is this one.

 

[yejun]

Or just use any preamp with multiple input and can be remote controlled by Logitech remote. Wrote a simple program do the input switch.

 

[samsa]

Or just use any preamp with multiple input and can be remote controlled by Logitech remote. Wrote a simple program do the input switch.

Personally, I don't see the issue with a single-blind ABX test, in this situation. Put the switcher in another room (with a dirt-cheap IR Remote extender) and there's no "Clever Hans" issue. It's only @BDWoody's insistence that only double-blind is valid that riles me. There's another thread where the (single-blind) loudspeaker testing work by Dr. Floyd Toole is being (justifiably!) extolled. I don't see anyone casting aspersions on his work because it's not double-blind.

 

[somebodyelse]

The single blind part of the first one is OK, but there never was any serious issue about setting up the single blind experiment. The double-blind part is completely deterministic and hence could only be used once before you had to throw it away and have your buddy wire up a new one (and, even then, it's not even pseudo-random -- hence not capable of producing a scientifically-reliable result).

From the Editor's Notes at the end:

Steven has done something I had thought would only really be possible using a microcontroller or a PC to achieve. The random switching is of such complexity that it would be virtually impossible for anyone to know and remember the combinations created by each switch. Especially if assembled according to the instructions - the switches are not only capable of giving an excellent randomisation of their own accord, but if they are wired in a random fashion as you build the unit, the likelihood of being able to remember the combinations is extremely low.

Granted it's not up to Enigma levels of switching complexity, but it's probably enough. I'd go down the arduino route, but some people are just allergic to computers.

That's missing the point though - even if we have an acceptable test it won't settle the argument, it'll just change the excuses.

 

[samsa]

That's missing the point though - even if we have an acceptable test it won't settle the argument, it'll just change the excuses.

That's where, philosophically, I disagree. I want to lower the barrier to applying scientific methods to address questions like this.

The biggest excuse is, "ABX testing is too hard." If you insist on ridiculously-difficult-to-implement conditions then, yeah, ABX testing is too hard. But, generally, to produce meaningful (if not necessarily airtight) results, it doesn't have to be so hard.

 

[Kane1972]

I would say all DAC's sound very close to one another and I think our sensory apparatus (eyes and ears etc) and brain are designed to "homogenize" things, to fill in gaps and keep things consistent etc. So, using two DAC's that are of similar design and specs etc and trying to determine sonic differences in double blind ABX tests, I think would be very hard to impossible if the switching is seamless and unknown to the listener.

I think the brain needs to know something has been switched for it to "reassess" things, to use it's full processing power and pay full attention to identify the changes. I think being able to switch between the two DAC's oneself, but not knowing which DAC is which is a better way. So you can listen to say a 2 bar loop and concentrate on one aspect at a time, like bass notes, then reverb tails etc etc. Make subjective notes about DAC A and DAC B.

Then change the music loop (or keep the same one) and get a friend to switch between A & B, telling you when they are switching, but your friend may or may not have switched the cables, so A COULD now be B. If you can then identify which is the original A and B with convincing accuracy, then I feel this means you can hear a difference.

Anyone else think this would be a valid test?

I have done similar tests online between file formats (MP3 vs WAV etc) where I was comparing X to A & B and I had to identify if A or B were the same as X.

 

[Kane1972]

Hopefully my E30 will be shipped today. Ordered it 12 days ago from HiFiStudio79, who had said they would have them in stock within 2 or 3 days, but obviously they didn't. However, the status says today "ready to ship", so fingers crossed.

I'll try to do a level matched comparison between it and another ifi DAC but I have no idea how I can do this in any elegant way. I'll have to keep selecting the DAC to output from (in VOX preferences). However, my headphone amp (Atom) does have two inputs but are they of equal quality? They will also be different cables too. I'll have to switch the inputs around for round two of the subjective listening.

 

[Deleted member 17960]

Despite the difference in SINAD, are the D10, D30, and E30 all audibly transparent? I read comments in the D10 suggesting so, but never heard it explicitly from Amir.

FWIW the Fiio D03K ($15) is not transparent and sounds veiled compared to the D10, in my volume-matched comparison. But this could be attributable to the tactile experience of using a knob with one and not the other.

 

[Deleted member 17960]

This is the best resource I have seen here to help answer that question: https://www.audiosciencereview.com/...-thresholds-of-amp-and-dac-measurements.5734/

Thank you for that link.

Strict thresholds guarantee that anything measuring better will be transparent.
Lenient thresholds are only lenient relatively to the strict thresholds. They are also "strict" from an audible point of view. Devices measuring worse than the lenient thresholds might still sound fully transparent to you or anyone else for that matter.

Should the lenient thresholds be considered enough for practical purposes? I'm a little confused on where to draw the line regarding buying audio products, given the wealth of information presented in that thread. I did another back and forth comparison between the Fiio and the Topping and had trouble distinguishing a difference this time.

This is becoming a rabbit hole... Ugh

 

[JohnYang1997]

Thank you for that link.

Should the lenient thresholds be considered enough for practical purposes? I'm a little confused on where to draw the line regarding buying audio products, given the wealth of information presented in that thread. I did another back and forth comparison between the Fiio and the Topping and had trouble distinguishing a difference this time.

This is becoming a rabbit hole... Ugh

Lenient is generally good enough. But you need to look at details.

What is important is, don't waste money.

 

[frogmeat69]

What is important is, don't waste money.

Now you tell me,

 

[voodooless]

Lenient is generally good enough. But you need to look at details.

Details indeed.. SINAD wise lenient would be about 2/3rds of the field, and it would start at the end of the "red" section. I'm pretty sure quite a few of them do not get a "recommended" due to bad measurements.

 

[zoozilla]

Got my E30 and am liking it, no problems so far. Was a little wary about getting another Topping product after my D10 crapped out, but crossing my fingers I have better luck this time.

One thing I do miss is having the sample rate show on the display at all times. Is there any way to change the display to that instead of the volume?

Oh, and it sounds fine.

 

[half_dog]

Got my E30 and am liking it, no problems so far. Was a little wary about getting another Topping product after my D10 crapped out, but crossing my fingers I have better luck this time.

One thing I do miss is having the sample rate show on the display at all times. Is there any way to change the display to that instead of the volume?

Oh, and it sounds fine.

I´m getting anxious waiting mine unit arrive =P
About the sample rate, you need setup your player (example Foobar2000) to play through ASIO or WASAPI, those APIs communicate with your DACS bypassing Windows API sending the signal with its original sample rate, but they will mute any other software which is not using the same API.

 

[zoozilla]

I´m getting anxious waiting mine unit arrive =P
About the sample rate, you need setup your player (example Foobar2000) to play through ASIO or WASAPI, those APIs communicate with your DACS bypassing Windows API sending the signal with its original sample rate, but they will mute any other software which is not using the same API.

Yeah, I have WASAPI set up and everything.

It's just that the D10 would show the sample rate whenever audio was playing, which I really liked. The E30's display will show the sample rate for a second if you switch between tracks with different sample rates but will then go back to the default volume offset display. I was wondering if there was a way to have it show the sample rate the entire time music is playing like with the D10 but I don't think it's possible.

 

[yejun]

Yeah, I have WASAPI set up and everything.

It's just that the D10 would show the sample rate whenever audio was playing, which I really liked. The E30's display will show the sample rate for a second if you switch between tracks with different sample rates but will then go back to the default volume offset display. I was wondering if there was a way to have it show the sample rate the entire time music is playing like with the D10 but I don't think it's possible.

It displays sample rate in DAC mode, volume in PRE mode. Holding power button for 3 second to switch between them when it's powered off.

 

[dpippel]

Yeah, I have WASAPI set up and everything.

It's just that the D10 would show the sample rate whenever audio was playing, which I really liked. The E30's display will show the sample rate for a second if you switch between tracks with different sample rates but will then go back to the default volume offset display. I was wondering if there was a way to have it show the sample rate the entire time music is playing like with the D10 but I don't think it's possible.

Yep, as mentioned above, if you want to see the sample rate displayed all the time you need to put the E30 in DAC mode. This will also fix the output volume, so you'll have to control it with your downstream device instead of the E30.

 

[BDWoody]

Thank you for that link.

Should the lenient thresholds be considered enough for practical purposes? I'm a little confused on where to draw the line regarding buying audio products, given the wealth of information presented in that thread. I did another back and forth comparison between the Fiio and the Topping and had trouble distinguishing a difference this time.

This is becoming a rabbit hole... Ugh

Here's another great post and thread to help alleviate fears of what you may be missing out on in those ever smaller noise and distortion numbers:

Credit to @RayDunzl who came up with the Shout-o-meter chart:

https://audiosciencereview.com/forum/index.php?threads/the-shoutometer.2555/

How good is your hearing?

 

[Aldoszx]

My E30 just came yesterday. It took only 20 days to arrive, from China I suppose.
Now, we are getting closer