Amplitude - A typeface designed for airplane cockpits

Alexander OrnAlexander Orn Posts: 15
edited March 11 in Type Design Critiques
Hi everybody, my name is Alex and I am doing my final design project at my university. 
I'm designing a typeface for airplanes, that is suppose to easier to read when the instruments are vibrating. 
I've done a lot of research on the subject and based pretty much all design-decitions in some kind of theory/study I've found. 

The biggest, and probably most noticeable thing is the way horisontal lines are almost completely gone in the second style. The reason for this is because diagonal lines are easier to perceive in an vibrating environment. 
I've tried to make every letter as unique and clear as possible, because I don't want the user to have a hard time recognising  or mixing up the letters up with each other.
I made it monospaced because a lot of the text in these instrument are set vertically, and there's a lot of acronyms/shortend words.

This is the first time I'm making a monospaced typeface, so any feedback about the proportions would be very appreciated!
If there's something else you notice that I could change I'd love to hear also.

Here's a link to the album with some images https://imgur.com/a/Ic9ZB

Comments

  • AbrahamLeeAbrahamLee Posts: 191
    What a great project! What do pilots think of it?
  • Craig EliasonCraig Eliason Posts: 736
    edited March 11
    /W/ is too dark. Maybe its center vertex doesn’t need to come all the way up. 
    Consider lowering the center vertex of /M/ a little, which might help disambiguate it from an /H/. 
    Curious, what do acronyms have to do with monospacing?
  • Thomas PhinneyThomas Phinney Posts: 1,096
    Interesting about the diagonal-lines thing. As aircraft vibration is often either vertical or horizontal, diagonal lines would be easier to read with either of those, and no more susceptible to oscillatory non-directional vibration than anything else.

    I will be very curious to hear whether the inherent reduced familiarity and hence reduced legibility of the diagonal lines outweighs any benefit from reduced vibration sensitivity? I suspect that across typical test conditions the answer is “yes” and hence a more “normal” font will be better.
  • Very interesting project... Have you thought of testing the font with a 'real' layout? I'd say that even a short animation could be important if you are able to emulate the vibrations and prove your point. 
  • Note that the name 'amplitude' is already taken: https://store.typenetwork.com/foundry/fontbureau/fonts/amplitude

    @Thomas Phinney Makes a good point. Jave you considered straight but diagonal lines instead of wavy lines? And could you share the research on which you based this decision? I'm interested.
  • What a great project! What do pilots think of it?
    Thank you! I have not yet showed it to any pilots, but I certainly will if i get the chance. Towards the end of the project I will test it as good as I can, and hopefully I'll get in contact with a pilot.

    /W/ is too dark. Maybe its center vertex doesn’t need to come all the way up. 
    Consider lowering the center vertex of /M/ a little, which might help disambiguate it from an /H/. 
    Curious, what do acronyms have to do with monospacing?
    Thank you! I will work on the blackness in the /W/, although I would like to keep the vertex at the top so that it doesn't look too much like the /M/, but I'll look into it!

    About the acronyms; I read that when having a font monospaced it puts the focus to the individual letters in a word instead of the "word image" produced by a proportional typeface. And since most of the words in the studied environment are shortened like this: Radar = RDR and Emergency = EMGY — I argued that focus on each individual letter would be more important than "word image" since these shortened word might not look like their original counterpart.

    Interesting about the diagonal-lines thing. As aircraft vibration is often either vertical or horizontal, diagonal lines would be easier to read with either of those, and no more susceptible to oscillatory non-directional vibration than anything else.

    I will be very curious to hear whether the inherent reduced familiarity and hence reduced legibility of the diagonal lines outweighs any benefit from reduced vibration sensitivity? I suspect that across typical test conditions the answer is “yes” and hence a more “normal” font will be better.
    Thank you for your input!
    According to the study that I based this whole diagonal lines idea on the most common type of vibration is vertical, but I get what you mean and to be honest I am thinking the same way as you about this. 
    If I were to do this as a project for a real client it would probably look more "normal", but as an experiment I enjoy the challenge of coming up with a design that fits within the bounds of my brief and having all the decisions I make based in the theory I've found.
    If it ends up non-functional, at least I've made a pretty funky looking typeface :)
    When I'm done with the project I could update you all on success or failure of my design!

    Very interesting project... Have you thought of testing the font with a 'real' layout? I'd say that even a short animation could be important if you are able to emulate the vibrations and prove your point. 
    Thank you! I've been tinkering a little in AfterEffects trying to come up with a test for the typeface. My idea of a test I could create with my resources is having some animated text vibrating on a screen, and having a person trying to read it before some time limit. Maybe having my typeface compared to some other in-use "aircraft typeface" in this manner could prove/rebut my point!


    As a sidenote I apologize for any bad grammar/terminology, I am very much a Swede.
  • Note that the name 'amplitude' is already taken: https://store.typenetwork.com/foundry/fontbureau/fonts/amplitude

    @Thomas Phinney Makes a good point. Jave you considered straight but diagonal lines instead of wavy lines? And could you share the research on which you based this decision? I'm interested.
    Oh, I'll find some other cool name for it!

    About the diagonal lines, I found it hard to make them straight without not having a shallow angle. I wanted the lines to be as close to 45deg as I could get, and when they were straight and ~45deg they ended up really short instead.

    The reference for the study I based the diagonal lines thin on is this:

    Readability of vertically vibrating aircraft displays. 

    Patrik Andersson, Claesvon Hofsten. 1999. 
    Displays Volume 20, Issue 1, 25 February 1999, Pages 23-30



  • James MontalbanoJames Montalbano Posts: 853
    edited March 11
    If any of the digital Flight Deck developers show interest in your design, be warned that the liability issues regarding licensing your design will be place fully in your corner. That means taking out insurance to the tune of many millions of dollars and showing their company as the beneficiary in your CofO (Certificate of Insurance). You will also have to create a shell company to sign all the licensing documents to protect you as an individual from any liability claims. Welcome to the world of font licensing. Sleep well at night.
  • SiDanielsSiDaniels Posts: 265
    As "Amplitude" is taken how about "Kenneth" as in  ?
  • @Alexander Orn: What is the study you mention above (in your original post)?
  • I dig the Vibration style, but it's really impossible to judge if that makes any actual sense for the application. What studies did the diagonal logic originate from?

    If we forget about the reasoning behind it, it can be appreciated in its own right, which might lead to interesting ideas also.

    On the point of legibility though. For me the whole point of adding both i serifs and and l tail is to distinguish the two from each other and from the 1. You've negated that differentiation by adding both a i bottom serif, and an l top serif. Especially imagining some vertical vibration the three letters will be very hard to tell apart. The large x-height equally turns n and h pretty close to each other. All the bowled characters like p d b q could probably be more clear had they no spur, since the stem is so short the eye might sort the bowl to stem position either as belonging up or down based on some vibration or quick glancing up and down. Personally I find that the Vibration style also adds so much noise that the added benefit of distinction of individual letter might be lost to overall more strain on deciphering the words. The logic of making all horizontals zigzag also means: There is no more a visual group of "letters with horizontals".
  • Alexander OrnAlexander Orn Posts: 15
    edited March 13
    @Rob McKaughan  @Johannes Neumeier

    Readability of vertically vibrating aircraft displays. 

    Patrik Andersson, Claesvon Hofsten. 1999. 
    Displays Volume 20, Issue 1, 25 February 1999, Pages 23-30
  • I'm not sure replacing a horizontal stroke with a wavy horizontal stroke is an improvement in legibility. I suspect it would be more helpful to redesign the glyph architecture to avoid horizontals, e.g. with a /C-shaped /E.
  • @Alexander Orn: While I think this is very interesting project, the scientific basis is a bit thin. It would be very impressive if you could experimentally show that your typeface works well in airplane-like conditions. To do this, you could probably start with randomly vibrating some text on screen, in accordance with an airplane vibration pattern (so, mostly in vertical/horizontal direction?), and letting participants read short texts. If you could create such animation(s), I could help you with setting up the experiment.
  • @Johannes Neumeier Thank you for your input! I am totally going to go over the typeface now that you mention these problems!

    The problem with not having serifs on /i, /l and /1 is that they become so thin, because of the monospace. Without the serifs they leave a huge gap in the text, but maybe the problem of them being hard to tell apart is bigger than having gaps in the text.

    The second problem with /n, /h, /p, /b, /d and /q could maybe be fixed by increasing the length of the ascenders and descends and removing/changing the spurs from the bowled letters!


  • The problem with not having serifs on /i, /l and /1 is that they become so thin, because of the monospace. Without the serifs they leave a huge gap in the text, but maybe the problem of them being hard to tell apart is bigger than having gaps in the text.

    If you add a bottom left serif to the i it might help to better differentiate between i and l.

    André
  • @Jasper de Waard Thank you! I'm planning on doing such tests, and I have already started tinkering with some animations just like you described. 
    Your help would be very welcome and I will soon have some animations done.
    How do you imagine setting up this experiment?  
  • I am only familiar with Open Sesame, which is free open source software (Mac and Windows) for creating and running psychological experiments. Participants can see the videos and give a response, and the results are output to an excel file. You would just need a laptop to test on :) It's fairly straightforward to create such an experiment, you could give it a try yourself. The tutorials are of great help. If you have any technical or experimental design questions, feel free to send me a pm. I'm no expert, but I think this shouldn't be too hard.

    With regards to the experimental design, the options are endless. I'd say: display short clips of moving words (in both variants of your typeface and a more 'standard' typeface), and measure participants' reaction times on a task that requires identifying the word.

    On a different note: do you think pilots actually read the text on their displays? Wouldn't they know what each number/color represents from experience, without having to actually read the dashboard? Perhaps you should give special attention to designing effective numbers, but I don't know enough about airplane dashboards to be sure.
  • Be ware, though: designing, running, and analyzing experiments can take a lot of time
  • James MontalbanoJames Montalbano Posts: 853
    edited March 13
    >>Be ware, though: designing, running, and analyzing experiments can take a lot of time<<

    Yes indeed. 
    An aerospace  company has been testing and evaluating a suite of Terminal Design fonts for use in their flight deck for over a year. They are close to a decision but still are mulling over the final weights and widths they require. Since these implementations have a useful lifespan of decades, they certainly are within their rights to take their time.
  • However, testing the basic effect doesn't have to be complex.

  • An aerospace company has been testing and evaluating a suite of Terminal Design fonts for use in their flight deck for over a year.
    Which ones, James?
  • @Rafael Saraiva Sorry, can't say. NDAs.
  • Well, I hope it's not Badinage  :) 

    btw under nda i would not post what you posted but it is certainly another topic. Good luck with the project!
  • James MontalbanoJames Montalbano Posts: 853
    edited March 14
    What I posted has not violated any of the terms of the NDA I signed.
Sign In or Register to comment.