Better Than Eye Chart?

Allan Hytowitz likes to send his Dyop measuring stuff every so often.

It’s over most people’s head (I say this because it’s obviously over my head, as most thing are). Brought it up a few times here and there and mostly it gets shrugs because people want simpler, not more complicated.

And they know how an eye chart works.

And Allan, much like Otis and the rest of that clan get on some old bearded dudes nerves a bit. Because they always seem to take this tone, of “let us tell you why your stuff works, because you obviously don’t know and we are super smart”.


Part of your success in reducing myopia is your elimination of the inherent overminus of the Snellen test.
for the explanation.

That’s an interesting link, btw. After my eye stop rolling from the stuff it came packaged with.

“Let me explain to you why your stuff works, though”.

:joy: Just kidding. He’s probably right. “The vibratory motion of the visual saccades?” You got me, Allan.

Check that stuff out, ya nerds.


I’ve tried it in the past, my biggest problem was that it’s hard to do it alone. And when I was able to make it work at last with a setup, then my AF kicked in so hard that I got total nonsensical results :smiley: with a good pc app with a good mobile app as a remote control tool it would be awesome. But with it’s current level user-friendliness I don’t expect that it will get widely used even by Endmyopians (of course I know their main goal are the optometrists).
But I definitely like the approach and I would love a vision assess tool which is more objective than Snellen or cm measurement. Because it has a “binary” aspect (you either see them rotate or not) it could be that. They should just somehow make it more easy to use.

That’s just a weird way to say that “part of your success is using normalized glasses” :smiley:


Hey, halmadavid, when you did that test, how did the results compare to your centimeter measurements or Snellen chart?

As I mentioned it was unusable, because AF / blur adaptation. I was able to discern the rotation even when measured -1.75 diopter without glasses (my full prescription was around -3.5 back then).

1 Like

Wow. I’ve now tried experimenting with it a little bit, but I’m unconvinced that I’m doing it right. Maybe I’ve missed it somewhere in their instructions, but it would seem to me that the level of brightness of your screen could be crucial to your results.

1 Like

There was some guy in the UK on optiboards who supposedly had a battery of tests in some kind of simulator that was much more like various real world vision problems, with various lighting levels and movement and atmospheric conditions. I’m not sure how we would use it though.

I would think that as people improve their vision as measured by Snellen that their vision as measured by other tests would also improve.

1 Like

More from Allan:

I also know that we are “on the same side” in wanting to help people improve the quality of their vision and overcome the damage from the “cult of Snellen.” While eye care professionals mean well, they are technicians trained to closely follow the mantra of 1862 technology rather than scientists who know that " The first rule of science is to challenge your own assumptions."

And, of course, the key to the functioning of Science is the concept of Pixology.

However, what is awesome is that I finally have INDEPENDENT validation that the Snellen test IS making people blinder.

Let me know if you are interested in playing with my “toy” and what I can do to be of assistance to you.

Your Fan Club might want to play with the Dyop test. I have a Consumer Version for an iPad, which uses an iPhone as the controller.

1 Like


Great… now I just have to buy an iPad :smiley:

I will play a bit around again with the PC version :slight_smile:

I’ve did a test now.
For dyop used my screen with max brightness, from 2m. I used the “precision” dyop menu.
For Snellen my usual setup, 3m and around 250 lux (measured with my own app) indoor artificial lighting.
Cm measured again on my screen (still max brightnes).

Measured with my differentials on (left: -1.5 / right: 1.75). Usually I measure without glasses, so this way my habits and to what I’m used to has less influence.


Dyop Snellen Cm
Right 20/62 20/100 60
Left 20/52 20/70 63.1
Both 20/50 20/50 63.6

Currently I’m using -3.00 / -3.25 glasses as normalized, but it’s pretty close to full correction at this point. So the converted cm numbers should be around good. “Converted Snellen” based on some randomly found table would be -2.75 / -3.00. And if I use the same rules to convert the dyop snellen equivalent then it would be somewhere around -2.50 / -2.75.
I don’t know how to assess this honestly :slight_smile: But given that there is definitely some blur challenge with my current correction, although in general case I would reduce at this point, I think the -2.75 / -3.00 correction (so what Snellen gave) would be a fine correction even as normalized. The dyop’s one feels a bit too low for me, although most likely still fine as Endmyopia normalized.

The Dyop experience was much better than I was remember to (although I had to a bit problem to get around the menu). I used my phone + (open-source free) to control the cursor on my PC. With the 2m distance and differentials on it was no problem to navigate on the dyop measuring screen. The large icons helped too, and you only need to press two buttons next to each other (increase / decrease size). So it was a fine experience now :slight_smile:
I’ve found it easy to discern when I have my results: there was a point where I simply didn’t see the ring rotating anymore, that’s when I’ve noted down the current numbers.

@jakey: Don’t they want to give some free licenc to your Fan Club? :smiley: Even some dumbed down version, I obviously don’t care about nothing just the dyop part (but I think they don’t have such kind of licencing).


@jakey: Don’t they want to give some free licenc to your Fan Club? :smiley: Even some dumbed down version, I obviously don’t care about nothing just the dyop part (but I think they don’t have such kind of licencing).

I’ll ask. I imagine he’ll be happy to share.

IMHO Allan is quite close to this thing and maybe doesn’t fully appreciate the competing strenghts of simple, known, ‘good enough’, vs. something maybe better but also way more complicated. At least the way it’s presented now.

This more than anything needs a short animation and some super super easy to use demo. I can just imagine the opticians being sent these things as they are today, and how quickly it all gets thrown out.

“Things to learn, process to change, more things to pay for? What’s my ROI on all this? Oh yea, more accurate measurements? Mmmmkay, think we’re all fine here.”

Further Allan thoughts:

Your feedback and insights are invaluable, and totally correct.

I have been “in sales” for 48 years, which includes the “four blind years” when I was physiologically impaired by my glasses with zero income (for FOUR YEARS) due to that disability. However, I have been working on my “research project” for the past 14 years and as the research time demands increased, it made it very difficult for the sales function as a recruiter, both mentally and logistically as to time.

The mental difficulty is that sales and research require an entirely different mindset. With sales you have a defined product, defined market, defined benchmark as to results, and a relative time frame to execute it so that the entire process can be optimized.

With research you have a concept but no defined product, do not know how or why it works which is why you are experimenting, and have no idea how long the process will take. The similarity of science and stupidity is that both entail you to do something over and over expecting a different result, except that in science you are taking notes. (I also discovered that the USPTO will issue you a Patent even if you don’t know how or why your “new art” works.

Any suggestions as to how to speed up or simplify the measurement of visual processes will be greatly appreciated.

A year ago there were comments wishing that the Dyop test was even more efficient. The previous Clinical Sequential Algorithm had sequential whole number diameters which required about 15 steps to go from 6/30 to 6/600 or from 6/30 to 6/2. It took about 5 months to create a Bracketing Algorithm which allows the diameter range to go from 6/120 to 6/600 or from 6/120 t0 6/5.5 in seven to eight steps and measure acuity in 20 to 30 seconds per eye versus 60 seconds with the Sequential Algorithm.


I am sure that despite the awesome (and overwhelming) functionality of the current Dyop test, I am sure that I will find additional ways to enhance it.

Much like the development of the eye and brain in discerning which inputs are NOT valuable but rather are distracting, and learning to ignore them, I am now trying to regain my mental focus and emphasize productivity rather than discovery.

But I even own THAT URL.

Induced Dyslexia

Except that I keep coming up with new ideas such as incorporating gaze detection utilizing a simple web camera rather than the EXPENSIVE Tobii Eye Tracker, and being able to determine visual cylinder on a two dimensional surface without the need of cylinder lense (or the assistance of an Optometrist).

(We already have a prototype which works. We just need to refine it and have additional validation.)

So PLEASE continue your criticisms. They are extremely helpful to me as I try to regain my sanity.


This is super interesting! And could be really useful for Endmyopians. While the cm measurement is a pretty good tool to assess visual acuity, so far we don’t have any easy to use measurement for cyl. Checked the paper and it seems to be really simple and straightforward.

I will play around the tool for some time, I really like that they are open for feedback and try to come up with some ideas :slight_smile:


I would suggest to Allan that he should upload a short video demonstration of his dyop test to youtube. I do buy into his basic premise that technology has improved a lot since Snellen’s time and it’s now possible to make eye-tests that work better (or at least test different things) than a simple black-and-white image of text. If I had an ipad I would try to install it and play with it. I looked for a moment on youtube and didn’t find any demos of his Dyop test.

You can download and try the PC version for 7 days.

1 Like

Here is a short demo of visual acuity testing.


@shaerah thank you for posting this!

He asked me if I could make one for him. Definitely not my area. He’s going to have to find a proper marketing minion.

I have a license. Can PM it to you.


I’m trying to figure out how to measure astigmatism with it. At a first glance I don’t understand it. Have to look more in depth.

As I understand it cannot do that yet, they are just trying to make it happen.

1 Like

So I did another round of testing. I’m starting to like this tool better and better.

First I started with normalized. I measured my eyes to left: 20/20.5, right: 20/24.5. I really like that you can do much more precise measurements than with Snellen. And the measurement is really straightforward and easy in the sense that there is no subjective decision making. Can I see rotating, or I don’t see any movement? If the latter, then use the switching buttons (switch the rotation and/or which one rotates). Usually it’s the same result, rarely a bit additional refinement with those.

What I also really love that you don’t need large distances to do this. I think it can be really useful for most Endmyopians, especially people not living in those countries where the standard size of flats starting at 90 m2 :slight_smile:
I could do the normalized testing with 2m away from the screen, which I think easily doable to almost everyone. And I would assume that the distance doesn’t really matter until the screen resolution is high enough to be able to show the small features of the circles. I think it would be worth to test this from mobile screen distance too, I can imagine it could work. I’ve did a quick test on my computer (with is a standard full HD screen with 96 ppi), and without glasses it works. With differentials I cannot go low enough, but mobiles has much higher ppi than a usual computer screen, so it can be ok. Especially if we don’t want to go below 20/20. Or even full hd laptop screens should do better (again, higher ppi).

Of course a bit drawback that to get 100% precise measurement you should have a calibrated screen (the app support screen calibration hardwares). But I think it’s still more precise than you usual printed Snellen or printed cm measurements, because at least your screen brightness is constant (I think generally the brightness should be maximised when doing such tests).

Also what is great that you have absolutely no chance to active AF in this test. There is no contrast to follow, nothing to AF on. You can just stare and either see the rotation or not.

What I don’t understand yet is that it gives overinflated numbers when I try to test without glasses. I can see the rotation with 20/102.2 left, 20/109.7 right without glasses. Which should be around -1.50? Which is waay too low. But maybe it is only for me because of this double visions issues? Because what I can see in this case is like 6-8 circles and only 1-2 of them is moving. Or to be more precise: I see two large overlapping grey blobs, and there is 1-2 moving part of the large grey blob :slight_smile: I may need to actually read the manual if it talks about this case :smiley: But given that it’s intended for optometrists use I can imagine they don’t even tried to deal with this case, because such high myopia the optometrists would anyway put some glasses on the subject.

So the huge advantages of this test:

  • No need to have large distances
  • No (or way less) chance to AF during measurement
  • No subjective decision making, it’s either rotates or not

The only drawbacks I see so far:

  • The PC program is a bit unstable and not too intuitive and absolutely bloated for our purpose (it’s intended for optometrist use). But this doesn’t influence how good the method works
  • You need a wireless mouse or keyboard or set-up that you can control your PC with your phone to adjust the settings. Neither is a really big problem, but you won’t buy a wireless input just for this, and the latter may be a bit too tech-savvy for many people. I mean it’s just “download this program, next-next-next install” on PC, and downloading an app for mobile (works on both android and ios), but yeah… Their iPhone - iPad pair programs solves this, but then you need both an iPhone and an iPad.
  • For some reason for me it doesn’t give proper results when I try to use without glasses. I suspect I am the exception. Or maybe it only problem for high(er) myopia and should be ok when my uncorrected vision is closer to -2.00? (so to where my differentials correct me) But maybe other people could test it :slight_smile: @NottNott have you gave a chance to this? :slight_smile: I know you measure regularly too.

The first two problem is “just” user interface problem an can be solved, has nothing to do with the actual method. The latter is a bit bigger problem, but honestly even if I can only test my eyes with glasses I don’t find that a really big problem, because for glassless testing cm measurement anyway more practical at my myopia.

So for what to use this?

  • I think a good practical usage is to assess objectively how good you can see with your normalized. Based on my testings you cannot deceive it by more light, or trying to look a bit or “yeah, maybe that’s not really blurry”. It either rotates or not. If I would have used it to assess my vision with normalized when I’ve reduced too fast, I would have definitely not reduced. @NottNott it can be useful for you too I think :slight_smile:
  • Obviously it’s good for regular distance vision testing. Seems to be pretty consistent for me so far. As mentioned I can only do the testing with some glasses which is a drawback, but anyway at my myopia it would be more precise to use differentials for measuring. So it can be used a “distance vision” testing pretty well, with similar precision than what we can achieve with cm measurements in “close-up”.
  • Theoretically it would be also a great tool to assess how good your differentials. But on my screen I cannot go small enough at my close-up distance to don’t see rotation. I have to go back to ~80-90 cm. But it’s an only 96 PPI screen, so… it can be also useful for this based on the setup.

Could we adapt it to the meow sure app, you think? Or otherwise remove the layers of complexity for end users?

I’m sure I could wrestle some sort of doable endmyopia license agreement out of him.