Most active commenters
  • (3)

←back to thread

631 points wojtczyk | 41 comments | | HN request time: 0.953s | source | bottom
1. dylan-m ◴[] No.41406819[source]
One of my favourite unreported MacOS issues comes from how, at some point, they changed the appearance of the window close button to be a particular shade of red with a tiny little X in the center. And if you happen to be using a particular kind of screen and possibly wearing glasses, that little X kind of wanders around in the button, appearing just slightly off center in a maddening way. Made only more maddening by the glasses component: https://www.robbert.org/2014/10/the-off-center-close-button/.

That post points out it’s probably just subpixel stuff causing the issue, but I think my thick, cheap glasses at the time were adding a layer of chromatic aberration to something that was already visually confusing.

I assume it’s kind of gone away at this point with all the high DPI screens these days. But I remember thinking at the time, if there was a public bug tracker, that issue would be a fun one.

replies(8): >>41406849 #>>41406994 #>>41407250 #>>41407351 #>>41408018 #>>41408369 #>>41408545 #>>41409242 #
2. trilbyglens ◴[] No.41406849[source]
This is not a software bug, but rather an optical phenomenon called "chromatic aberration". What's happening is that your glasses are bending light at different angles depending on the wavelength, to the red and blue and green are landing at slightly different places on your retina.

It's a hard problem to solve optically and requires specially shaped lens. It's a common issue in telescopes, with higher end expensive scopes having these specially shaped lenses to reduce this effect.

replies(4): >>41406888 #>>41406909 #>>41406914 #>>41406952 #
3. voctor ◴[] No.41406888[source]
From the post:

> In conclusion, the off-center “x” is real and probably an artifact of the display or how it is rendered. It is unlikely that it is the result of chromatic aberration.

replies(1): >>41406903 #
4. ◴[] No.41406903{3}[source]
5. meta-level ◴[] No.41406909[source]
from the comment:

> That post points out it’s probably just subpixel stuff causing the issue, but I think my thick, cheap glasses at the time were adding a layer of chromatic aberration to something that was already visually confusing.

6. dustincoates ◴[] No.41406914[source]
Is this why, when I'm reading text on a dark background, red will appear on a different plane than white? I was just wondering the reason last night.
replies(4): >>41407113 #>>41407296 #>>41407305 #>>41407321 #
7. almostnormal ◴[] No.41406952[source]
Chromatic aberration is mostly relevant further away from the center of vision. If there is an icon (or text) visually inspected carefully it is at the center where chromatic aberration matters least.

The icon is mis-aligned, or its the different color subpixels of the screen that are not produced at the same place. Tradidionally, red is to the left.

replies(1): >>41407178 #
8. mrob ◴[] No.41406994[source]
>More expensive lenses have a coating to compensate for this chromatic aberration.

You can't compensate for chromatic aberration with a coating. You need a compound lens made from multiple elements each with a different dispersion, e.g.:

https://en.wikipedia.org/wiki/Achromatic_lens

More expensive glasses lenses usually have worse chromatic aberration than cheap ones. The cheapest material for glasses lenses (PADC, often called by the brand name CR-39) has one of the best Abbe numbers (measure of dispersion).

https://en.wikipedia.org/wiki/CR-39

https://en.wikipedia.org/wiki/Abbe_number

replies(5): >>41407335 #>>41407534 #>>41407735 #>>41408026 #>>41409285 #
9. leereeves ◴[] No.41407113{3}[source]
A different plane?
replies(1): >>41407179 #
10. _ph_ ◴[] No.41407178{3}[source]
To be precise: chromatic abberation is lowest at the center of the lens. But with glasses we often don't look through the lens center even if we have something in the center of our vision.
11. amiga ◴[] No.41407179{4}[source]
The red text seems to be closer than the other text. As if it were floating above the other text.
replies(2): >>41407252 #>>41407503 #
12. jahnu ◴[] No.41407250[source]
Another irritating thing that is captured in that image is the single pixel gap between the top of the application window and the menu bar. If the desktop background is bright it is very distracting. Not a mistake like the off centre X but drives me mad, nevertheless.
replies(2): >>41407683 #>>41408208 #
13. ◴[] No.41407252{5}[source]
14. zimpenfish ◴[] No.41407296{3}[source]
Sounds a bit like https://en.wikipedia.org/wiki/Chromostereopsis

> Another interesting reversal effect was observed in 1928 by Verhoeff in which the red bars were perceived as farther away and the blue bars as protruding when the bars are paired on a white background instead of a black background.

15. zerocrates ◴[] No.41407305{3}[source]
I have pretty strong high-index lenses, and definitely can get a kind of 3D effect.

The classic terminal blue and green text colors on a black background is the situation where I first noticed it: moving my head makes them shift in different directions giving a parallax or depth effect.

16. krispyfi ◴[] No.41407321{3}[source]
https://en.m.wikipedia.org/wiki/Chromostereopsis
17. kortilla ◴[] No.41407335[source]
Can you link to something to buy?
replies(1): >>41407408 #
18. johnwalkr ◴[] No.41407351[source]
When I got an ultra wide monitor I also noticed this for the first time on the macOS circular red X button, when wearing glasses. I guess a wide monitor has you looking at things off-axis more. It is really remarkable how much you can make the X dance around the red circle by moving your head. There's something about the colours and simplicity of that particular icon that really cause the effect. The effect is almost non-existent if you're looking at a more complex image, so I guess it's also remarkable how much your brain compensates for chromatic aberration in most contexts.
19. mrob ◴[] No.41407408{3}[source]
I don't think anybody makes achromatic glasses lenses because they would be too thick and heavy.

AFAIK, every optician sells PADC (e.g. CR-39) lenses.

20. TeMPOraL ◴[] No.41407503{5}[source]
I get that a lot with default terminal colors - on black background, dark blue and dark red look shifted in opposite directions relative to baseline (white/light colors); when both colors are used in close proximity, it gives me a strong and quite distracting 3D effect.

I always thought this is specific to that color combination (red and blue on black) and LCDs, thus is perceivable by anyone, and could be used to create intentional 3D effects; I never considered glasses may be a factor too.

21. w4rh4wk5 ◴[] No.41407534[source]
I can confirm this. I had annoyingly bad chromatic aberration with my previous glasses. I specifically asked for CR-39 lenses for my next set of glasses and now it's barely noticeable at all.

I would recommend this to any programmer who uses high-contrast syntax highlighting. To me, it felt fatiguing every time I noticed differently colored words scrolling slight further than other words on a terminal screen on the same line.

replies(3): >>41407736 #>>41407764 #>>41409404 #
22. jeffhuys ◴[] No.41407683[source]
I knew from the very start of using macOS that it was designed around apps NOT being full-screen. Yeah, they are catering to the full-screen-apps people a little more nowadays, but embracing floating windows everywhere, and making good use of the distinction betweend CMD+Tab and CMD+~ makes it so much more powerful than just tabbing through full-screen apps, or three-finger-swiping. It also makes macOS way more beautiful to look at in my opinion.
replies(1): >>41408498 #
23. azlev ◴[] No.41407735[source]
TIL. Thank you.
24. agent86 ◴[] No.41407736{3}[source]
One thing to keep in mind is that CR-39 is not impact resistant. They will shatter and can do horrible things to your eyes when they do. Kids should always be put in impact resistant lenses.

If you’re a desk jockey, or impact resistance is not a concern, CR-39 will give the least aberration with the exception of crown glass.

The hidden hack here if you need/want impact resistance is to ask for Trivex lenses. Same impact resistance as polycarbonate but much better ABBE value. It’s often overlooked because it costs a little more than polycarbonate and most people don’t complain about the distortion.

Also, anecdotally, you get what you pay for with progressive lenses. I have a cheap lens in my sunglasses and a higher end lens in my daily drivers and I can easily tell the difference.

replies(1): >>41408408 #
25. jrockway ◴[] No.41407764{3}[source]
This really bothered me many years ago, and I tried CR-39 and even glass, just for fun. I was never that happy with the results. I could always distract myself with chromatic aberration, and I think I eventually decided not to care anymore.

But right now, I have high index lenses and am reading HN with Dark Reader, and even if I use the maximum strength of my glasses (progressive bifocals), I can't really see any chromatic aberration.

I'm not sure if I should be happy or worried.

26. trustno2 ◴[] No.41408018[source]
I mean horizontal and vertical centering is the hardest thing in computer science
27. germinalphrase ◴[] No.41408026[source]
I can’t speak to glasses, but limiting chromatic aberration in the binocular world does seem to involve coatings (at least as Swarovski, Leica, Zeiss present it).
replies(2): >>41408101 #>>41408106 #
28. 4gotunameagain ◴[] No.41408101{3}[source]
If you cut some wavelengths, you won't get their contribution to the distortions ;)
29. sitharus ◴[] No.41408106{3}[source]
You can’t eliminate chromatic aberration with coatings, it’s a physical property of how the lens interacts with light. The only way to fix it is to adjust your lens types or materials. Zeiss’ current marketing seems to agree https://www.zeiss.com/consumer-products/us/nature-observatio...

Coatings are still very useful to reduce other lens artefacts though.

30. sph ◴[] No.41408208[source]
That gap provides contrast and separation between two similarly-coloured-but-not-quite grey objects. It would look worse without it, though I agree it is silly.

This is the same reason why window gaps are so popular in all tiling window managers. It just looks better.

31. amelius ◴[] No.41408369[source]
Honestly what annoys me most is that it looks like a sideways traffic light.
32. xattt ◴[] No.41408408{4}[source]
> It’s often overlooked because it costs a little more than polycarbonate and most people don’t complain about the distortion.

I just started wearing glasses. I asked about fringing and they had brushed my concerns off as me being new to glasses.

replies(1): >>41410150 #
33. skydhash ◴[] No.41408498{3}[source]
It’s quite maddening if you use spaces (as workspaces) and same windows across them. An option like CMD-Tabbing being restricted to only applications with windows in the current space will go a long way.

Power users is an unknown concept at Apple.

replies(1): >>41408927 #
34. ◴[] No.41408545[source]
35. jwells89 ◴[] No.41408927{4}[source]
I’m not sure that they’re an unknown concept, because macOS has lots of little things tucked away all over the place for power users. It’s one of the things I miss most when using other desktops, particularly those that go maybe a little too far on the minimalism thing (like GNOME).

It’s just that they expect these users to have fairly specific usage patterns and design around those. The further one’s personal patterns deviate from that expectation, the higher the level of friction encountered.

replies(1): >>41409069 #
36. skydhash ◴[] No.41409069{5}[source]
> It’s just that they expect these users to have fairly specific usage patterns and design around those.

That’s how you design generic appliances, not professional tools. While macOS is great for the users it caters to (that only use a handful of apps), it’s not for people that use their computers as computers (making it do pretty much everything).

replies(1): >>41409399 #
37. hedora ◴[] No.41409242[source]
My glasses cause a bit of chromatic aberration, but not enough that I'd expect to see this sort of effect except at the edges of their field of view.

Now that you point it out, the X is way off center on my up-to-date M2, so I took a screenshot with default display settings and zoomed in to look at the pixel work.

The X is rendered asymmetrically. It appears to be about 0.1 pixels too far to the left and down, since the antialiasing has shaded pixels "outside the X" but only on those sides. The antialiased render of the red circle is symmetric. This matches what I see without zooming in and rules out my glasses.

I wonder if someone fixed the bug for low-dpi displays where subpixel rendering mattered a lot, but did so in a way that hard-coded whatever Apple shipped 10 years ago. Maintaining tall piles of hacks is hard.

Alternatively, maybe their font renderer is getting wobbly in its old age. The window manager is my #1 complaint about this laptop, but crappy font rendering vs. well-configured Linux is also on my list.

38. cubefox ◴[] No.41409285[source]
I thought I got used to the color fringes in my glasses, but the real problem is that they actually reduce image clarity away from the center of the lenses. If you look e.g. at white text on a dark background from an angle, the chromatic aberration blurs (the color components of) the letters together. You can't really see clearly by moving the eyes to the edge of the FoV of your glasses; you have to turn your head instead.

This is directly contradicting the main purpose of glasses: to see clearly. So it's actually somewhat less safe to e.g. drive with glasses that have major chromatic aberration. No idea why optometrists brush it off as a minor glitch.

39. jwells89 ◴[] No.41409399{6}[source]
It’s kinda tough, because a clean-cut, coherent vision of how the OS is intended to be used is necessary to build a great experience. The more you try to accommodate ways of usage beyond that, the more the vision falls apart and you end up with checkbox waterfalls and branching tunnels of config dialogs added in the pursuit of making everybody happy.

So realistically, judiciousness is required to keep it all glued together, and some usage patterns just won’t be accommodated.

For example, Apple doesn’t seem to be bending over backwards to make former Windows users happy, because the way that desktop works is just too different from what they’ve envisioned and what their long time users are used to. If they add a series of toggles to support Windows usage patterns, that’s a sudden 2x multiplier on the behaviors and UI that needs to be tested.

That said, I don’t necessarily agree with all of Apple’s decisions (I’ve never liked the linear representation of virtual desktops that in place since 10.7 Lion that well and preferred 10.6’s Snow Leopard’s 2D grid, for example), but the lines have to be drawn somewhere.

40. kps ◴[] No.41409404{3}[source]
For those with stronger prescriptions who want higher index lenses to reduce thickness (and weight), look at http://opticampus.opti.vision/tools/materials.php and/or talk to your optician about available materials. (Personally, I've settled on MR-8 for my last couple pairs of computer glasses.)
41. w4rh4wk5 ◴[] No.41410150{5}[source]
Deal with it for a few weeks, your brain will likely adapt.