#21

Is it possible to cover LEDs with gels to improve color differentiation? I don’t understand how color blindness works well, so I apologize if this is an ignorant question. It seems unfortunate that so many instruments use red/green tri-color LEDs.

1 Like
#22

It could potentially work, but it could also look pretty janky. Definitely worth a try though.

#23

What I don’t get about the Octatrack is that the MKII has RGB LEDs - they light up all sorts of pretty colors on startup. They clearly have access to the whole rainbow, but we get: red, green, yellow.

1 Like
#24

Weird, I had no idea! The MK1 definitely only has RG, and they share the same firmware, so it might be a reverse-compatibility thing? You’d think there might be some hardware-enabled flag, though; maybe not.

#25

I wish I knew! Check out this still image from the startup sequence:

image
image781×429 65.4 KB

Blue, white, purple… then it sweeps through all the reds, yellows, and greens.

3 Likes
#26

Frustrating for us colorblind folk that module and synth manufacturers use rg leds, for sure. I made a post on reddit and got absolutely pooped on for daring to suggest people use blue leds :eyes:

Anyway, cool that the mk2 has rgb. Has anyone cracked the firmware?

#28

There’s probably 2 levels each of R, G, and RG. For almost all of the LEDs, in regular use, there’s no continuous “mixing” to create various orange/yellow-green hues like in e.g. the OG Launchpad: either R is off, G is off, or they’re both on at equal levels to make yellow.

The exception are the 5 LEDs that do seem to have a few different levels: the A/B/C/D input levels, and the CF Card Status indicator. There’s also a flashy startup sequence that has all kinds of continuous levels on the other LEDs, so they’re clearly capable of it.

#29

The answer on the unofficial forum is to wear those 3D paper glasses. Yeah. Sure. I hate blue leds as well to be honest and don’t have the understanding about the several states that the octatrack might display as this can only be deciphered intuitively by someone seeing these colours. Would be cool if they fix it on the mk2. It looks like I just sold my Octatrack mk1 and if I ever will give it a third try I would probably remember that I told elektron about the issue in 2014. :stuck_out_tongue_closed_eyes:

#30

Hey everybody! Since this is kind of derailing from the original topic, Are you all ok with me splitting this to a new topic?

1 Like
#31

Oh, no worries, I am done complaining, sorry :slight_smile:

#32

That really wasn’t intended to be me “scolding” anybody for taking the thread into a different direction, more me thinking: “hey this is a very interesting and important topic, it needs to be its own thread”.
This shouldn’t be a place to just complain about problematic designs, but rather one where we can share strategies to deal with those, discuss hardware hacking, or even just post ideas about how things could be better designed. In my experience lines can sometimes be a great place to collectively brainstorm solutions and visions.

But back to the topic. From my experience, red-green LEDs are so ubiquos because they are relatively easy to source and don’t need a dedicated driver circuit like RGB ones. They give you 3 colours: red, green and amber and with the proper circuit design you can also play with the intensity.

Needless to say, people with deuternapia or deuteranomaly, i.e. almost 10% of the white Western male population does have a hard time with these (I don’t have any data on how the percentages might be with other ethnical groups). The part in the DNA that is reponsible for this is bound to the Y chromosome, so it only affects the male population.

Red-blue LEDs do exist and afaik work more or less the same. People affected by some other forms of colourblindness do have issues with blue, but can see green pretty well, but these are less frequent. If one wants to be fair and inclusive to everybody any colour should probably be used considering that somebody might not see it that well.
RGB LEDs are great in that regards because in theory you can design the firmware to accomodate different colour schemes. On simpler designs I think the best option is always to have discreete LEDs.

With Emilie we did spend quite some time figuring out how to retro-fit Mutable Instruments modules with an alternate mode that would avoid the colour-related issues and newer modules avoid these completely, whenerver it’s possible. But it’s not always that easy, because of course there’s a lot of factors that come into play when designing something.

2 Likes
#33

This is basically the approach that Emilie took with Stages, described here. In my experience, the approach works ok. It’s definitely better than not having a color-blind mode at all, but I’m often having to cycle through all 3 options to verify if I’m interpreting the intensities correctly, and since the hardware has glyphs based on color, I often forget which intensity maps to which function

1 Like
#34

I’m not color-blind, but I’m not a fan of color-coding with LEDs, anyway, and I try to avoid tools that rely on it. These multicolor LEDs are very seductive to hardware designers because of their spatial and electronic characteristics, but color coding is encoding—an obstacle to communication—and therefore a poor information design choice, usually made for the sake of pushing miniaturization to the limit.

I still love blinkenlights for on/off, tempo, magnitude (even bicolor for positive/negative magnitude is fine, although R/G pairing is probably not), and position.

1 Like
#35

Didn’t sound like it to me, no worries. I also felt that the thread was getting off topic. I am happy to share more insight if it is helpful, but really the main point is: a colourblind person often is much more challenged to suggest a solution that works colourwise. Eg the rainbow colours in the 4ms Spectral Resonator that also actually was colourcoded info was too much for me but I get it that sometimes it is more economic or prettier than a textdriven display.
Also for the curious ones, being colourblind to me feels more like being colourdislexic, I see pretty colours but don’t pin me down if that dark blue is really dark blue or just a really dark lilac; same with brown and olive colours.
When I went to school in the 80s and 90s overhead projectors were still a thing, I did not understand the colourcoded way of writing down the binomial forumlars (dark green versus dark red… uff… black and blue, totally fine…) at all at first and it was when I first thought something is not quite right with me.
Haha…

2 Likes