whats the less eye tiring font?
which for computer screen and which for printed media
Can you generalize more your question, it's too specific.
"whats the less eye tiring font?"
I'll assume this means to read.
"which for computer screen and which for printed media"
For screens users prefer to ¡see! either b & w, grey scale or ClearType but usually San Serif types. There is debate, indicating to me a wide spectrum of user preference (as well as a loud club of "invented HERE" pushers-in-action), over these three rendering techniques, and which is best for any given combination of user, screen, environment and reading preferences. When seeing is not enough and ¡reading! is desired, *things* change, but it's difficult to know exactly how since no one has done comparative studies of the same fonts and readers and content.
Print is easy, people like to read serif fonts. Adobe Garamond, Poynter Old Style or Century Schoolbook are three good examples of widely used text faces for print that do not tire users out as much as others.
What's the most nutritious food?
What's the most soothing music?
What's the most comfortable shoe?
Who's the sexiest actress (lab tests for that should be interesting).
Is there a pattern here?
Might it just be that there are no absolutes, and that people aren't all the same?
Results may vary.
The stuff people like is bad for them.
So, are the fonts we like bad for us?
What ARE the "side effects" of Comic Sans?
And more importantly:
Do fonts that read like butter cause hardening of the lenses?
To my knowledge, this kind of experiment has not been done yet.
Claude Hopkins was working "in the field" along those lines in the 'teens and 'twenties:
IIRC, he used split runs of direct mail promotions to refine which advertising techniques worked best.
The pieces were text heavy -- he was a copywriter. A great influence on another text-centric ad man, David Ogilvy.
It's been a while since I read the book, and I don't recall whether he tested different type treatments, or just different copy.
I do remember that Ogilvy used the split-run technique to test two identical versions of a charity ad, one with positive type and one with reverse, and the positive type raised way more money, not just a percentage, but something like ten times.
" ...two identical versions of a charity ad..."
"whats the less eye tiring font?"
I'm not sure how long winded your charities are up there, but I'd have trouble getting tired from just one.
Well, yeah, of course...
But what if the question is only, "which performance is louder?"... That's my point. In the same analogy, two orchestras with the same number of horns, violins and kettle drums have the potential to be equally loud, but one orchestra could be the grade school orchestra I was in once upon a time and the other could be the Toronto Symphony Orchestra. Obviously equal only if you are very selective and not at all rational in how you evaluate them. One could just forget about legibility for a moment and take some time to think about the "voice" you want to communicate with. This is a concept that is quite foreign among people I've met who test fonts for legibility.
Kevlar: "it is not advantageous to use fewer words." Ever? ;)
RenMan:"Has this developed into an abstract discussion or what?"
What, I think. ;)
"After scanning this thread, it looks like no one recommended a font!"
And it looks like, after your post, no one has yet. With the RFS we were given there is no answer.
All we can assume is that the script in this request is Latin.
"the following four are the ones we reach for most often"
the way people experience fatigue from reading is by not bothering to finish something... In a magazine, which as I recall is where Ogilvy placed those ads, there's always something else to move onto - Even if it's the next ad or piece of junk mail.
Nobody reads a half page and goes, "Man, I'm whacked!" They just turn the page, without even knowing why.
I think it would be next to impossible to pick one face as 'the least tiring'. But you could set some parameters that will tend to indicate if a font could be less tiring than some others. You could start with open shapes, large counters, Large(ish) x-height and tracking that isn't too tight and so on. In other words, almost any well designed typeface intended for imersive reading.
Nick, I was just involved in some 'legibility' testing for a signage typeface with a well know Canadian institute for the blind, here in Toronto... (round 2, if Joe is reading. I forced a second round of testing after pointing out some serious flaws in the first round). The trouble with this sort of testing so far as I have seen (with the notable exception of ClearviewHWY) is that nobody doing the testing knows anything about typography ("X-height is how big the letters are, right? - And kerning is how far apart the are, and serifs are..." "UH, yeahwhatever") and the discount the skill of the designer, and then make ridiculous comparisons of completely dissimilar typefaces the basis of their experiments.
>Do fonts that read like butter cause hardening of the lenses?
Whatever you do, when you turn 45 you get a card in the mail: "Congratulations, welcome to middle age! As your first gift, you are awarded presbyopia for both eyes, fee of charge. More wonderful surprises to come!"
Or maybe that's just my memory playing tricks on me :)
The ad was for Save The Children, and the black type "raised twice as much money", according to Ogilvy on Advertising. He wrote that book in 1983, retired in his chateau in France.
"If you start your body copy with a drop initial, you increase readership by 13%"
"It is now known that widows (short lines) increase readership".
"This is 11-point, and about right"
-more gems from Ogilvy, the result of readership surveys. Note that the term emphasizes the reader (albeit as a marketing demographic), whereas our "readability" focuses on the text.
I've read the book several times, but not lately.
"But in this case it’s just a matter of using fewer words...when MS sets up tests, they use 100’s sometimes 1000’s of words. ClearView’s testing is better — as fewer words are used"
The Clearview testing used fewer words only because of the impracticality of creating dozens of signs. I prefer to counterbalance and use 20 or more words per condition (never thousands) to reduce word effects. The word table will always be recognized faster than the word yacht no matter what conditions I compare. Making similar word lists and showing different word lists to each reader reduces these kind of effects. Fewer words are not better.
"I was just involved in some ’legibility’ testing for a signage typeface with a well know Canadian institute for the blind, here in Toronto"
Could you tell us more about the mistakes they were making in round 1? There is an unfortunate lack of typographic knowledge among psychologists that needs to be corrected, but designing a good study can be difficult even with the help of typographic experts. I would be interested in hearing your thoughts about the best way to design this particular test.
"could you tell us more..."
in an firstname.lastname@example.org
"...the[y] discount the skill of the designer, and then make ridiculous comparisons of completely dissimilar typefaces the basis of their experiments."
Now now. The designer is unimportant to these readability or legibility issues. The issues are engineering problems, not design problems.
"...with the notable exception of ClearviewHWY..."
Though the face is a notable exception, the testing is hardly better than what anyone else with an agenda has done.
That's interesting. I've read the Ogilvy twice -- once in the 90s and once this century -- and I'm still not clear in my own mind what I think of his conclusions.
David wrote: "Make up your mind."
I'll try to be clearer. In the Clearview study they looked at 6 font conditions on 2 kinds of signage materials for a total of 12 conditions. They created a word list of 6 words and attempted to make the 6 words similar to each other. They created signs for each word they used for each condition - a total of 72 signs. Participants identified the same six words in each condition. I imagine that it's somewhat expensive to create each sign, and therefore impractical to create to use 20 words and 240 signs or 20 different counterbalanced words in each condition for 2880 signs. Since my testing is either of computer type or printed on regular paper, it is not much more expensive to use 20 or more counterbalanced words per condition.
The advantage of using more words per condition is that it's possible to be more confident in your result when you find that words from one condition consistently perform better than words from another condition no matter what the word used. When cost is not a concern there is an advantage for using more words; it is not advantageous to use fewer words.
Berlow is quite wrong about that, in theory and in practice, but since he’s been around forever the kids are likely to believe him. As are the managers, who cannot differentiate one font from another unless it’s a true apples/oranges comparison (Univers vs. Arial, no; Times [“New”] Roman vs. Comic Sans, yes). We’re trying to get the discussion beyond “Let’s compare our new legibility font and Comic Sans” to “Let’s compare our new legibility font against plausible alternatives.”
Well, dberlow, - hardly better is still better, isn't it? :)
now, I'm off to re-read Mr. Ogilvy.
Whoops - I forgot x in my equation. Here's the real graph:
Silly me. Without x, it doesn't even make sense!
Wow. Has this developed into an abstract discussion or what? After scanning this thread, it looks like no one recommended a font! (Sorry if I missed any.) The following (for print) are not my suggestions, but at least they name names.
To B&A's list I would add Stone's Cycles family; I'm using Cycles Eleven and am stunned by its legibility and readability. Also for print, Freight Text is a worthy contender.
From Before and After Magazine, “What’s the right typeface for text”, an excerpt from their book Before and After Graphics for Business.
While many typefaces meet the requirements of legibility, readability, and beauty, the following four are the ones we reach for most often:
•Adobe Caslon (most readable typeface);
•Adobe Garamond (easy to read and elegant, too);
•Stone Serif (boring to look at but buttery to read);
•Janson Text (middle ground between the earthy, workmanlike nature of Caslon and the high classiness of Garamond.)
In his own time, with his copy-heavy ads, Ogilvy was against the grain of the "word+image=big idea" school of advertising which won all the awards and made the history books. But he was very successful and not alone in his method.
Nowadays, would anyone dream of putting so much text in an ad? Very rarely, but that's not to say that text-centric marketing has vanished, as there is a whole genre of advertorial newsletters in print and online, and often the editorial in periodicals is thinly disguised advertising. Just look at the massive auto and real estate sections in daily newspapers, hardly objective journalism.
Plausible? Well, that's the "agenda" of which David speaks.
Different stakeholders have different notions of what is plausible, different agendas (quite often, hyping their new "readability" design).
Me, I can't take the science seriously unless the starting point is testing the types that have become the norm -- and yes, that includes Adobe Garamond -- not because of prejudices that engineers, artists, designers, neurologists, typographers, capitalists, whoever, bring to the table about what is "plausible", but because those are the fonts that are used most.
1. Test the most used fonts, both sans and serif
2. Test them in different settings: vary line length, leading, size
3. Test them against different demographic groups
That should keep someone busy for a while.
"- hardly better is still better, isn’t it? :)"
I s'pose. But in this case it's just a matter of using fewer words...when MS sets up tests, they use 100's sometimes 1000's of words. ClearView's testing is better — as fewer words are used, less needs to be made up as a result.;) On the other hand, there is nothing wrong with the fonts in the Clearview case, just the spin is a little wobbly.
Highway Gothic was a quirky icon of the "American Road", but it had a unit system for an interface and billboard proportions for its letter drawing. Clearview, might be miles of bad kerning. Hap'nin aleady, (in the hands of expert composers?) places like Hellertown and Bethlehem. Even though the old sign's been sprayed with acid for some number of years, the spacing is better for the task(s).
In short, one must be very careful in accepting agate proportions in type without giving such type proper composition, something that's impossible to spread out over a million uses without a unit system and firm guidance, isn't it?
Kev: "...Clearview testing used fewer words only because of the impracticality of creating dozens of signs."
They had better refine the definition of practical then, hu?
"Making similar word lists and showing different word lists to each reader reduces these kind of effects. Fewer words are not better." Make up your mind.
I've made fonts for Yachts and Tables this month, so I'm aware of the need for design solutions to these problems of readability. It hardly matters. e.g. what one does or tests in this area if the problem is defined by the client as strictly an "engineering problem". Fortunately, I don't let clients stick to that broken philosophy for long, or they don't stay clients for long ;)
Rus: " Ultimately, the differences between two or more well designed typefaces used appropriately will be vanishingly small, and probably not worth the effort of testing."
I think what you mean to say is: the differences between two typefaces well designed for the same purpose and both used appropriately will be small, and not worth the effort of testing. That is not what we are looking at in Clearview though, where "used appropriately" is a huge issue, not so in the case of MS's tests which are laughably short of fonts "well designed for the same purpose..."
I confess, I've been kind of shocked at the shoddiness of some 'science' I've seen in this area, so Id' have doubts in the results no matter how long the folk in lab coats were busy.
I think that what should be possible is the definition of some basic parameters of what will work well under certain circumstances and what won't. Ultimately, the differences between two or more well designed typefaces used appropriately will be vanishingly small, and probably not worth the effort of testing.
Steve... this happens all the time on Typophile.
the differences between two typefaces well designed for the same purpose and both used appropriately will be small, and not worth the effort of testing.
The fact that testing will not pick up the difference doesn't say much about that difference.
Two performances of the same piece of music with all the notes played correctly, at the same tempo, yet one moves and the other bores. Isn't type like that? Isn't that why we agonize over different versions of Garamond, idolize Helvetica and scorn Arial?
Why do we bother if it amounts to nothing? Is out sophistication so extremely redundant?
You have to decide what quality of difference you're trying to discern before you can test it.
And what if that concept only exists in the language of type so far, a pure visuality, not in words or numbers?
Here we have it: David Berlow advertising Adobe Garamond as some kind of appropriate font.
> a wide spectrum of user preference
More like a bipolar disorder. Mac-heads used to think b&w bitmaps were better than moderate anti-aliasing, now they think FullFuzz™ is best. Windozers mostly just take whatever you throw at them with a "Huh?" Actually the latter is a bit less bad, because you don't want religion interfering in the optimization of reading.
Me, I invent what I push, not push what I invent:http://www.themicrofoundry.com/manademo/
Colore, I have suggested before in this forum a measure for reading fatigue. My thought is that it is not so much the eyes as the brain that is tired when type and typography is less than optimal. My suggestion was to measure how soon reading comprehension declines over time--from fatigue--under varying conditions.
Bunches of things haven't yet been done, and
won't be, as long as the bouma model is denied.
And of course it's the brain that gets tired and not not the eyes.
Otherwise we couldn't keep looking at stuff all day without having to close them.
However, measuring reading comprehension is unnecessarily complex and brittle.
Comprehension is self-regulating - all you really need to measure is speed.
>all you really need to measure is speed.
That is already being done, as I understand from Kevin Larson's posts here.
I think measuring decline of comprehension with time might better discriminate between better and worse type and typography.
I don't know whether it will turn out to be a more sensitive measure, but it seems to me worth a try.
One might also measure decline of speed with time, which so far as I know has also not been tried.
I would suggest trying both to see whether there is a difference. The key thing in either case is to give time for fatigue to set in. It may be that having demanding reading material will also help get more sharp discrimination to show up.
> That is already being done
Non-immersively. No good.
> decline of speed with time
No, just speed to complete a long reading task.
> having demanding reading material will also
> help get more sharp discrimination to show up.
No, that moves the measurement from perception to comprehension,
with the latter being outside the scope of typography and type.
The content has to be "normal".
>No, just speed to complete a long reading task.
Why not measure the decline with time? The profile may tell us something.
The main thing is to push the length of time to where fatigue sets in, and there is a decline in comprehension or speed or both to varying degrees.
>No, that moves the measurement from perception to comprehension,
with the former being outside the scope of typography and type.
Did you mean instead 'the latter', that comprehension is outside the scope of typography? I don't agree, but surely perception is within the scope of typography.
> Why not measure the decline with time?
You could. Like you say, you could measure many things,
and each will give some bit of insight. But as usual it's a
matter of priorities.
> The main thing is to push the length of time
Yes, but equally important is to foster deep immersion, via non-invasive
testing and... good typography! We need to measure the difference between
good and exceptional, not good and horrible, because in the latter boumas
never have a chance anyway.
> Did you mean instead ‘the latter’
Yes, sorry - I edited it, but apparently not fast enough.
I think there’s a strong need to understand eye fatigue. When someone breaks a finger there is an understanding that the bone in the finger fractured and that it can be repaired by stabilizing it and giving it stress-free time to repair. With eye fatigue we don’t seem to agree even if the fatigue is in the eye or brain.
I’ve been partnering with optometrist James Sheedy’s visual ergonomics lab to learn more about eye fatigue. I’m putting together a talk for ATypI to talk about some of the findings from his lab.
Certainly saccading is fatiguing - and really that's the
whole point. I guess that doesn't count as brain fatigue,
but neither does it count as retina fatigue... However I
do see that what I wrote above is misleading; possibly I
was too eager to dampen William's contribution - sorry.
Anyway to me it's moot what part is getting tired first.
Instinctively I simply work from the assumption that
fixation duration should decrease and saccade distance
should increase, and the type can help, by making it
easier to map the shapes to words.
>I’ve been partnering with optometrist James Sheedy’s visual ergonomics lab to learn more about eye fatigue.
That's great. But if there is a real difference between legibility [ability to distinguish letters quickly and accurately when flashed at you] and readability [ease of absorbing meaning] then increased brain fatigue as opposed to eye fatigue must be involved in equally legible but less readable material.
Looking only at eye strain would seem to prejudge the issue that there is no valid distinction. Why not look actively for this, and actually test it rather than prejudging it?
After all, typographers for over fifty years have thought the distinction is valid. They may be wrong, but such a prevalent view is at least worth testing, it seems to me.
Also it seems to me that the mental fatigue factor should be pretty readily evident, as you can always lengthen reading time to where everybody gets fatigued and performance--either speed or comprehension or both--go down. Then you can see whether there is a difference between, eg, the same font with different leading and measure, or different fonts that test as equally legible.
> increased brain fatigue as opposed to eye fatigue must be
> involved in equally legible but less readable material.
Sure, but how is that related to typography?
> such a prevalent view is at least worth testing
Here I totally agree.
And the same thing applies to the serif, concerning which I've tried to
convince Kevin not to simply ignore the mountain of anecdotal evidence.
> After all, typographers for over fifty years have thought the distinction is valid. They may be wrong, but such a prevalent view is at least worth testing, it seems to me.
I think mental fatigue is also an interesting area of investigation. For me it made sense to start with eye fatigue as people complain about eye fatigue from computer usage, and I’d like to do something to reduce that problem, and Sheedy is a top optometrist interested in solving user problems related to computer vision syndrome.
> Certainly saccading is fatiguing
How do you know? What would you measure to detect fatigue? How to measure eye fatigue is one of our key questions.
I actually wouldn’t have guessed that saccading is fatiguing as we our eyes constantly saccade when open (and sometimes when closed too). Is breathing fatiguing?
>people complain about eye fatigue from computer usage
Absolutely this is an important area worthy of research, and am glad you are looking into--just as a suffering user!
I look forward to research on the other issue of brain fatigue as well, because I think it will capture a lot of the issues that traditionally print typographers have been concerned with. For example the sans vs serif in extended text--which I think is mixed up with the % black and spacing issues--will I think more likely yield to the extended test looking into the fatigue factor.
edit: Upon reflection, I think that the key thing in distinguishing legibility and readability is not whether the fatigue is due to eye fatigue or brain fatigue, but simply that fatigue is a variable somewhat independent from legibility.
I love reading Minion!!! Very easy on the eyes!
> people complain about eye fatigue from computer usage
But extrapolating that to print is dangerous.
> How do you know?
A key to good design.
But, of course, I do not Know. I do however Think.
> What would you measure to detect fatigue?
Speed. Just leave it all up to speed.
With the right testing parameters, it's all you need.
How do I "know"? Instinct.
> Is breathing fatiguing?
If you do it fast enough for a long enough time, of course. Just like there's strolling to admire the scenery versus running to catch the bus, there's reading a magazine in a waiting room versus reading a research paper for tomorrow's meeting.
And of course it’s the brain that gets tired and not not the eyes. Otherwise we couldn’t keep looking at stuff all day without having to close them.
I'm sorry, Hrant, I can't tell whether you are being facetious or actually mean this. Maybe my brain is tired :)
Of course, by this silly argument, we couldn't keep thinking stuff all day without having to shut down our brain from time to time. Hmm.
Eye strain is one of the easiest, least subjective aspects of reading to measure, because eye movement is controlled by muscles and we can compare how those muscles move when performing different tasks, including reading different typefaces. Obviously there is going to be greater variability in eye strain between larger factors, e.g. different sizes of type, quality of reproduction, etc.
Muscles get tired, so eye-tiredness is, if you'll excuse the phrase, a no-brainer. The brain doesn't have moving parts, so in what sense does it get tired? Obviously not in the same sense that eyes get tired. What happens is that we get tired -- get holistically tired, if you like --, and lack of rest starts to affect the brain's ability to function properly. Unlike muscle fatigue, which can be addressed by resting or refocusing the particular muscle groups -- e.g. closing your eyes, looking at more distant objects --, brain fatigue requires overall rest (during which the brain, ironically, may be quite active for long periods).
So long as I am well-rested, my brain doesn't get tired from reading, and if it does get tired -- i.e. my attention wanders, my content retention diminishes -- that is symptomatic of general exhaustion, not because I've overexhausted my synapses. My eyes always give out before my brain does.
Link to blurb on Kevin's ATypI talk - love the title...
Thanks, Si. Okay, so correct my previous statement that 'eye strain is one of the easiest, least subjective aspects of reading to measure'. It should be one of the easier aspects to measure, but apparently researchers have only just started trying.
>So long as I am well-rested, my brain doesn’t get tired from reading, and if it does get tired — i.e. my attention wanders, my content retention diminishes — that is symptomatic of general exhaustion, not because I’ve overexhausted my synapses.
In general, any repetitive task, mental or physical, I think tires one quickly. For example, if you do a series of sums non-stop I predict that in not such a long time you will start to make mistakes, and not because of general fatigue. --If you switch to push-ups, you won't do fewer than you would be able to at the beginning. I may be wrong, but that's the kind of thing I'm talking about.
If you don't experience this, you are blessed. I personally cannot keep at a single task non-stop without experiencing 'brain fry' at some point within an hour. So I stop, take a short break, read Typophile :), and then return to the work task.
According to the book Fatigue as a Window on the Brain "cognitive fatigue" is a reality, and I see from googling that there are scientific articles on cognitive fatigue in astronauts, soldiers, victims of multiple sclerosis, stroke and others. Some of this relates to the impact of physically induced fatigue on mental performance, but others seem to look at the impact of mental tasks.
Kevin Larson above seems to think this is a legitimate area of research also.
I agree with you that eye fatigue is certainly real--I've had headaches from wrong lens prescriptions, for example--but so is cognitive fatigue, both in my experience and it seems from the scientific literature.
edit: This article reports a test of mental fatigue due to visual attention tasks.