Can you spot the difference in this chromatic aberration test?

Hey everyone! Can I ask for your help with a little test?

I’m trying to design a measurement target for focus reach (aka centimeters to blur) that uses edges of specific colors, but I’m not sure how easy or hard this is to see for most people. If you use Differentials / computer glasses, or have a natural blur horizon around screen distance, could you please have a look at this image:

focusreachtest-2019-06-24-smaller

If you try to find your edge of blur with this, in theory, you should be able to see that it’s closer for cyan (FOCUS) than magenta (REACH), as in, magenta blurs earlier when moving back. But I’ve run into people who say they can’t see this. Which obviously is bad for trying to use it as a standard. So I’m wondering if it’s just the odd person without proper computer glasses, or if this is hard to spot for some reason.

How do the two lines blur for you as you check your edge of blur with computer glasses? Please try both eyes individually if it’s inconclusive.

  • The magenta clearly blurs earlier.
  • It’s hard to notice, but I think magenta does blurs earlier.
  • I can’t tell whether the cyan or magenta line gets blurry first.
  • I think cyan blurs earlier.
  • I don’t have the undercorrection for a screen blur horizon.
  • None of the above

0 voters

2 Likes

i can’t tell.
as i blink the blur shifts from magenta to cyan and vice-versa.

1 Like

That is awesome!!! Varakari —> Endmyopian with skills!

I must have been reading at edge of blur when I first saw the focus/reach, because I literally thought magenta was printed blurry in absolute. Blew me away when I inched closer and saw it wasn’t.

2 Likes

@Ponder Nice! That’s what’s supposed to happen! :nerd_face:

@miocardio, this sounds like you still aren’t challenging the focus hard enough. Could you try again, going up with the distance or down with the correction until it’s hard to hold focus? There should be a point where it’s much harder for an eye to switch focus to the magenta line.

Thank you to everyone who participated so far!

But why in the world is it failing on a considerable amount of people? We have almost a quarter answering “I think cyan blurs earlier” – which really shouldn’t be happening, not even for colorblind people.

:thinking::thinking::thinking: What’s going on here?

@Varakari
I happen to have high astigmatism (2.5 and 3)
In addition to -4.5 spherical.

It is hard to determine which blurs earlier with single eye, because when magenta blurs, cyan has a lot of double vision.

Using both eyes it is more noticable that magenta blurs earlier but it is very subtle.

Why not use red and violet? They are the edges of the visual spectrum so should show the most difference in blur horizon if i understand correctly.

1 Like

For me it seems that they are both beginning to blur at roughly the same time. However the magenta has higher contrast on the white background, so I’m noticing it first. As soon as magenta begins to blur, if I look at cyan closely i will see the same double-vision blur, but it is so faint that i hardly notice it.

1 Like

i just put my glasses further from my eyes and i’ve notice cyan is more sharper than magenta. Magenta gets blur before cyan. Cyan gets clear before magenta. But it’s still hard to spot the difference.

1 Like

Random data point : I put my phone on inverse colors (white text on black) and the FOCUS is now red and REACH is now green, but REACH is blurrier than FOCUS, still.

1 Like

Don’t your computer glasses correct for most of the astigmatism?

The cyan one is actually an inverse red edge. I don’t want to involve blue edges, because they are way off and using them as reference could lead to using too much lens power.

But you can look at this:

focustest-chroma-visionlogtool

It should show the effect more strongly. I’m now wondering if I should use the dark background after all; it greatly increases contrast.

You figured out why cyan and magenta are used! I wanted a white background for the surrounding brightness, so I used the negative of red and green on black.

But now I wonder if we should just draw a black circle in the center and make the target in red and green.

To anyone looking at my post above this one, is the chromatic effect between red and green (ignore the blue stuff below) easier to spot than the cyan magenta one this thread started with?

1 Like

It’s equally easy for me. Not apples to apples comparison though, since the original cyan/magenta appears much smaller font in this thread than the red/green on black.

1 Like

I did not use my computer glasses when testing, i tested with no correction

I happened to be really pushing the distance with my differentials when I first looked at the targets and the difference was ridiculously obvious. Then I read your actual post, and trying the moving closer/farther away to find the actual edge, it was trickier to tell the difference between the colours.

1 Like

Cyan & Magenta is only slightly easier for me to see the chromatic effect than Red & Green.

The effect is trippy!

1 Like

I’m assuming that cases where cyan&magenta seem easier than red&green on black, like for @deadpan, are my mistake, because I didn’t adjust the font size for the older image. The large font can make it harder to detect blur.

So far, my takeaways from this test are (I also aked some people directly, friends and Reddit etc):

  • The chromatic effect is visible and should be useful for the majority; more than half see it immediately.
  • About a quarter see something else and may measure not useful results because of this. The reason is unclear, but mistaking low contrast for blur is the most likely culprit.
  • The remaining cases struggle to see the effect. In most cases when I try to make the instructions more explicit and ask for some retries, they see the effect eventually.

@itamar, if you try with your computer glasses, does the effect become easier to see? If you have the time, please also look at the image with text on black that I posted in my previous answer to you.

These results aren’t terrible, but having around 30% problem cases needs to be addressed. I think for the test release, I’ll switch to a red/yellow/green on black test target that uses the smaller font size of the cyan/magenta test, and write instructions for people who can’t see the effect (yet) to use yellow on black for measurement for the time being.

Thanks again for all your input! Let’s see if the next version can cover more cases well.

@Varakari
With the Differentials it is much clearer
magenta clearly blurs earlier
green blurs earlier than red

1 Like

It worked for me immediately with my differentials but for curiosity I tried it with my normalised and the magenta seemed easier to see. I’m guessing it’s the contrast against the white background that gives the impression.
I really like this!

1 Like

@itamar Awesome! Lesson learned, for the logging software instructions, I’ll make sure to communicate clearly that you need to use Differentials / computer-adjusted glasses. Glad to hear that all cases work as intended. Thank you very much for sticking with it and troubleshooting the cause! It’s often hard to guess what may have gone wrong on the other side of a vote/post.

@Mag: I’m thinking the same thing. This test is supposed to validate that you are at the edge of blur. If you put on glasses for distance in front of the computer, it should not work.

This is the point I want to get at in the end. The test is intended to first help to validate that you’re at the edge of blur, and then you go a bit closer, to the green edge of focus, for a precisely defined measurement that’s used in the log. The code and form for logging is already done; just the test target is missing. Right now, I’m pondering this one:

focusreach-v3-2019-06-26

As I’m on a different desk – I’m in Frankfurt right now – I notice that the effect is much harder to spot when there’s high background brightness.

I think this thread has turned out super useful. We already found a number of things that could have gone wrong with this.

Your question reminds me of the eyeQue Personal Vision Tracker. They have red and green figures on black to see.

1 Like

With computer glasses on, the magenta and the red clearly blur faster than the cyan and the green.

Just pointing out that saying “clearly blur” is funny. :slight_smile:

1 Like