Setting up a Blind Audio Test

BoredErica · May 25, 2014

Hi,

I want to see if I can pick out the differences in sound from my onboard MSI z87 g45 ALC1150 audio compared to my Objective amp/dac. There are a few difficulties however. First, the most convinient way to do this test is to plug the quarter inch to 3.5mm converter on my headphone connection and plug the 3.5mm plug into the front port of the computer. This of course has drawbacks but the biggest issue is that I probably forgot to wire up the front audio port when I first built the computer. Or something. So the back works well, but that puts the tester in an awkward place in my room, cramped. Although, out of visible sight and making it truly a blind test. Another issue is that the headphone jack itself is quarter plug, and the Objective has a quarter inch jack. But the onboard has 3.5mm. So to switch between onboard vs Objective the quarter inch to 3.5mm converter has to either come off or come on, making this very cumbersome.

So I wonder if there is an easier way to do this test? Maybe some sort of cable that runs into both Objective and Onboard, with the playback device/amp changed with a click of the mouse? That would make things much simpler and more bulletproof of a test.

I literally just plugged my HD800s into onboard and the volume is more than loud enough for Foobar playback. I'm hoping that I am able to do more in depth digging around.

Rorgash · May 25, 2014

also, make sure you use music that is actually well made and not shitty top 40 music that was put together without any effort so that it actually uses the highs and mids and lows and actually has sounds that takes advantage of the extra clarity

Stefky · May 25, 2014

Get a 3.5mm to quarter inch adapter lead. They aren't expensive.

The most important things to remember are:

Volume matching - the on board and ODAC need to have exactly the same perceptual volume for the test to work.

Eliminate the possibility of the single blind conditions interfering - your tester must flip a coin rather than choose which to put in. Regardless of whether the same is chosen or not, he still needs to do the same number of mouse clicks and headphone removals and replacements. You note which you prefer of each two samples on a piece of paper, the tester is silent.

Use a sample of music that you know extremely well, covers a broad range of the frequency response, and has well produced stereo effects (I.e. Different elements of the music spread wide across the stereo field).

One of the biggest audible and measurable differences in components is cross talk, which will affect sound stage and separation, difficulties worse components have with frequency response happen at the extremes. Besides distortion, you won't be able to tell any differences if you listen to a singer song writer with a soft Spanish guitar centred both centred in the stereo field. You need a sample that has multiple instruments placed around your head at different widths, uses the depths of sub bass as well as high frequency information.

I would recommend two tests, one with lossless or 320kbps cbr material one with a very lossy 128kbps mp3. It may be that the only difference is that the better is more resolving of artifacts and both are perfectly fine with good quality recordings.

Don't second guess and re-test. If there is no difference you can discern, it's a good thing. Your inboard audio is fine and you can make some cash back selling the O2.

creatip123 · May 25, 2014

I'm thinking an audio receiver with remote control would make AB tests much more simpler, and easier. All cables don't need to be plugged-unplugged-replugged, just fiddle with the remote, selecting which source it's passing to the output.

BoredErica · May 25, 2014

I'm thinking an audio receiver with remote control would make AB tests much more simpler, and easier. All cables don't need to be plugged-unplugged-replugged, just fiddle with the remote, selecting which source it's passing to the output.

How much would that cost?