Hacker News Reader

Comment by bsaul | Hacker News Reader

I onced interviewed for a job at a software company developping virtual instruments plug-ins. At some point in the interview, the guy told me this great story :

"We made some test once, and we changed the skin of the software, to a new color. Every people said the software sounded better with that new skin. Yet we changed absolutely nothing except the color."

I'm still wondering if some graphical configuration ( such as a bright color) wouldn't stimulate the brain more, making it more receptive in general, and to sound in particular, letting people "hear" better.

View on HN

While I totally agree with the moral of this story, I'd like to point some reasons why the visuals of a software might interfere with audio:

1. just check this website: http://thume.ca/screentunes/ there is a large chance that your computer videocard and/or LCD will make a noise on that page

2. when the videocard makes such a noise, it can interfere with the PSU and thus affect the soundcard as well.

3. Thus, it's actually possible that changing a "high-frequency stripy" interface to a solid color will lower the noise-floor on the sound-card a bit.

I can confirm that. On most computers I could tell what is happening on screen just by listening. Scrolling, video playback, task switching all makes different noises. I have decent headphones and amplifier.

Any external DAC will eliminate this for you.

Even decent internal soundcard does.

Nope. I'm running into an external DAC and then a Rega integrated amp. Makes noise her as the bars get smaller. The noise is independent of amplifier volume, including with the amplifier volume all the way down, and even occurs with the amplier and DAC switched OFF. The noise is actually coming from the LCD.

This is a ground loop. Small part of the current returning from CPU to PSU goes through "common ground" connection to the DAC, then through the analog DAC-amp ground conductor, causing some voltage to develop across this connection, and then back to your computer through grounding cables in the walls.

The voltage across DAC-amp ground causes a DC offset seen by the amp (likely less than 1mV). The exact amount of voltage varies in time with CPU power consumption, producing AC that you hear.

You need to eliminate the ground connection between computer and DAC (toslink, USB optoisolator, etc.).

RE: #1, it definitely made noise on my LCD. It's an older CCFL type backlight, I don't know if that makes a difference but I do know that fluorescent lighting can emit similar sounds.

Can confirm this: Just by listening at the static through my head phones, I can tell what setting my laptop's display brightness is set at.

We find whenever we significantly change the look of our website or our emails, conversion goes up significantly for a couple of weeks and then most the time goes back to the normal levels.

I think a difference just stimulates the brain, along the same lines of why supermarkets infuriatingly keep moving the products around so you can't just auto-pilot your shop.

Yet another thing to consider when doing A/B testing. Looks like you'll need active controls as well as controls!

There's a reverb plugin I use that has 70s 80s and "now" modes, which drastically affect things like internal sampling rates (bandwidth) and noise. It also completely changes the color of the large GUI from deep red/brown to cool blue to bright white.

I find that I have to shut my eyes or look away from the screen when toggling the modes to avoid biasing my perception of how the audio is changing. It's particularly bad because the visual changes are designed to evoke the effect it has on the sound.

Sensory experiences are incredibly subjective and different modalities interrelate in all kinds of complex ways. It could be said that everybody is synaesthetic to some extent, some just more so than others.

In many contexts outside science, whether some sensory difference exists objectively or not may not be relevant at all as long as there is a perceivable difference in subjective experience.

Reply on news.ycombinator.com