Yet another A/B test of DACs

[raif71]

Hi all.
Look what just arrived last week. Following in the footsteps of @Pdxwayne , I will do an A/B comparison of two DACs ie the SMSL m400 (akm 4499) and Topping D10s (ess 9038). The amplifier that I will use is the SMSL sp400. Headphones will be Sennheiser hd600, hd800 and Hifiman Sundara.

The A/B switch has the following configuration:

The in/out jacks are 3.5mm female connection so, I’ll be using SE connections for the DACs and amp.

Basically I will route the gears as below:

I will control music files that are being played using my Hiby R5 Player to the D10s. I will then use the optical out of D10s to connect to the other DAC (smsl m400). Both of these DACs are then connected via their rca out to the 3.5mm input of the switch. Lastly the output of the switch will then connect to amp using also a 3.5mm to rca in of the amp.

I will attempt to do sighted tests first as I will need my children’s help to push the button between A or B for blind testing. Need to explain to them on how to go about doing this. This is just an explanation on what I’m going to do. Doing it and when is another matter lol. Feel free to comment. Thanks.

Edit: This is my 1st thread that I started after more than a year in ASR haha

 

[Doodski]

Great AB comparator you have there. Do you have a AC multimeter or a oscilloscope so that you can use it to match the output of each DAC so that one is not louder than the other? It will require adjusting the volume control for one DAC or even inserting a potentiometer inline after the DAC that has a higher output voltage. If your SMSL M400 has a volume control you should be able to trim the output voltage with that. I can't remember if it has volume control.

 

[raif71]

Great AB comparator you have there. Do you have a AC multimeter or a oscilloscope so that you can use it to match the output of each DAC so that one is not louder than the other? It will require adjusting the volume control for one DAC or even inserting a potentiometer inline after the DAC that has a higher output voltage. If your SMSL M400 has a volume control you should be able to trim the output voltage with that. I can't remember if it has volume control.

Yes, the smsl has a variable volume. It's at fixed volume now but will make it variable for the tests. No, I don't have multimeter/oscilloscope. Will have to play it by ears.

 

[preload]

I'd be curious if you can identify differences during your initial level-matched sighted comparison (I'm assuming if you can't hear differences at this point, it wouldn't make sense to proceed with a blinded comparison).

 

[preload]

Yes, the smsl has a variable volume. It's at fixed volume now but will make it variable for the tests. No, I don't have multimeter/oscilloscope. Will have to play it by ears.

Do you have a digital SPL meter? You can level match to 0.1dB that way. If you don't precisely level match, and you identify differences, nobody will accept your results (nor should you). Although if you match by ear and you don't identify differences, you can probably stop there.

 

[raif71]

I'd be curious if you can identify differences during your initial level-matched sighted comparison (I'm assuming if you can't hear differences at this point, it wouldn't make sense to proceed with a blinded comparison).

Yes, you are correct. Actually I'm expecting that I can listen differences between them sighted. I'm expecting I will be able to prefer one dac over the other. Ahemm

 

[Doodski]

Yes, the smsl has a variable volume. It's at fixed volume now but will make it variable for the tests. No, I don't have multimeter/oscilloscope. Will have to play it by ears.

Here's a nice sine wave generator. Use 1kHz, 3kHz or even 5 kHz to level match.

 

[raif71]

Do you have a digital SPL meter? You can level match to 0.1dB that way. If you don't precisely level match, and you identify differences, nobody will accept your results (nor should you). Although if you match by ear and you don't identify differences, you can probably stop there.

Don't have that either. I wouldn't bother it as of now... Perhaps along the road, I'll get a multimeter or spl meter... I suppose it's the journey that is what matters, right?

 

[Katji]

^^^Yes, it can serve as an example to other people, how to do it, how to start.
...So you could provide information about the comparator, like where to get it and how much. ...It needs to be inexpensive.

 

[pma]

Don't have that either. I wouldn't bother it as of now... Perhaps along the road, I'll get a multimeter or spl meter... I suppose it's the journey that is what matters, right?

Without exactly matched level from both DACs the test makes no sense, unfortunately. It would have the same validity as the sighted test. A multimeter would cost you $10.

 

[Thalis]

Don't have that either. I wouldn't bother it as of now... Perhaps along the road, I'll get a multimeter or spl meter... I suppose it's the journey that is what matters, right?

ya get a MM

 

[Katji]

How do we use a multimeter instead of an SPL meter?

 

[Jimbob54]

Yes, you are correct. Actually I'm expecting that I can listen differences between them sighted. I'm expecting I will be able to prefer one dac over the other. Ahemm

I've done similar and I couldn't tell even without accurate volume levelling. Try it with them both at max/fixed if they sound about right, get the kids to flick the switch. You might be in for a surprise. Might only be the flick that gives it away.

 

[Blumlein 88]

Please get a multi-meter. Otherwise you are just wasting time, twiddling thumbs, however you want to put it.

I might avoid the $10-$15 ones. But your local home improvement store or Walmart will have some plenty good enough for $30 and up.

Test with a 400hz tone or 1 khz tone and set so that both devices are outputting voltages within 1% of each other. A 3.5 mm extension cable would let you use the end to tap the output with the multi-meter leads.

 

[preload]

Please get a multi-meter. Otherwise you are just wasting time, twiddling thumbs, however you want to put it.

I might avoid the $10-$15 ones. But your local home improvement store or Walmart will have some plenty good enough for $30 and up.

Test with a 400hz tone or 1 khz tone and set so that both devices are outputting voltages within 1% of each other. A 3.5 mm extension cable would let you use the end to tap the output with the multi-meter leads.

And if you're not comfortable using a multimeter and taking a measurement via a tapped output (or you're too lazy = me), you can still achieve the necessary level matching accuracy via an inexpensive digital SPL meter.

 

[Blumlein 88]

And if you're not comfortable using a multimeter and taking a measurement via a tapped output (or you're too lazy = me), you can still achieve the necessary level matching accuracy via an inexpensive digital SPL meter.

A more inconvenient and easier to get wrong method vs a multimeter.

 

[preload]

A more inconvenient and easier to get wrong method vs a multimeter.

Speak for yourself. Anyone can turn on an SPL meter and place it inside their headphones.

OTOH, using a DMM isn't easy and straightforward for most people (i.e. which leads plug into which port, which measurement mode, which range to select, and nuances around proper contact with the spliced cable). How many newbies do you think are going to probe the spliced wire and get widely varying levels due to poor contact, instead of use alligator clips or jumpers? If this were an electrical engineering professionals forum, I would recommend the DMM or oscilloscope, obviously. But I recognize that not everyone here is an EE. In fact, someone 3 posts above just asked "How do we use a multimeter instead of an SPL meter?" That's not the say that the OP isn't an expert in circuits, but I'm going to venture a guess that most people who are facile with a multimeter already own one (or two or three).

 

[julian_hughes]

You can level match accurately enough by outputting into a recording application such as audacity & adjusting accordingly. I haven't done this for a long while but if I recall correctly you can *easily* match to 0.2 dB.

 

[preload]

You can level match accurately enough by outputting into a recording application such as audacity & adjusting accordingly. I haven't done this for a long while but if I recall correctly you can *easily* match to 0.2 dB.

OH, that's another good idea. So, you would just send the RCA out to an ADC? Or to the line-in on the PC sound card?

 

[raif71]

Thank you all for the DMM/SPL/Oscilloscope/recording suggestions to do level matching. Yes, I suppose this would not be a scientific endeavor if I at least cannot level match "precisely" between the 2 setups so I will spend some time to explore DMM or SPL and weigh the pro/cons of price and practicality of using either.