FAQ

Congratulations! Don’t hesitate to alert me: bach@uni-freiburg.de or open an issue on GitHub. Will fix ASAP.

Congratulations! Contact me: bach@uni-freiburg.de and we’ll see. Has frequently worked ;).

Short version: Nothing, just implementation details.

Long version: Development of FrACT goes back to 1984; the dynamics of computer technology and specifically of web browsers enabled it to go more and more online, culminating in FrACT₁₀ around 2020. Now FrACT₁₀ has every feature of the “classic” FrACT, and more. The basic algorithm (Best PEST) uses practically the same code base as three decades ago. Since FrACT₁₀ as a progressive web app can now be installed for offline use, there is no reason to use “classic” FrACT.
Any new development (e.g. plastic card calibration, noise embedding, Presets, etc.) is only available in FrACT₁₀.

First, regarding difficulty: While FrACT homes in on the threshold, and intentionally occasionally goes beyond it, it is normal and necessary that the trials become challenging. The “homing in” aspect can be appreciated when using the All trials Plot button after completing a run. It is beneficial to reassure participants that it is normal to guess and that “errors are removed by the computer”.

Second, regarding “easy trials”: On purpose, on trial #12, 18, 24, …, an “easy” optotype (large or high contrast) appears. These trials are designed to be reinforcing and reassuring for the testees; especially if the number of trials is set such that the last trial is an easy one.

If this feature is not desired, there are pertinent checkboxes “Occasional easy trials” found in Settings▸Acuity… and Settings▸Contrast… to switch off the easy trials.

Assuming both are done “right”, there is no difference. Try yourself: e.g., use plastic card first, then measure the blue ruler. Use whatever you prefer.

The best possible acuity is based on “smallest stroke = smallest Landolt gap” being 1 pixel. However, antialiasing allows to present fractional pixels, and the smallest possible element size can be set at: Settings▸Acuity▸min. stroke [px]. The default value there is 0.5 px. Since max possible decimal acuity is based on 1 pixel (thus demanding good test properties, safe side), the result can be a better value. The note on display resolution… in the manual provides more detail.

Yes, but you need a visual display unit (VDU) with really tiny pixels (4K is probably fine). In the Manual, there is a note on display resolution… on this, with examples of suitable VDUs. The accessible acuity range is given on the Settings▸General screen at the bottom right.

As an aside: “Reading acuity” and “near acuity” are rather different things; they “only” have in common a low viewing distance of typically 30–50 cm.

The technical limitations are screen size and distance. Going close and using a large screen can bring you into the ultra-low vision range. Then the problem arises whether the observer has sufficient vision. Here is a discussion on this. Also, the accessible acuity range is given on the Settings screen at the bottom right. When the top two are green, you’re good for normal acuity.

Very demanding, an expert field by itself; so just some thoughts:

• Will need a very large screen
• Should run in ︎Fullscreen mode to not waste screen estate
• If the low vision condition is congenital, Sloan letters will not work, so consider Landolt rings (possibly with only 4 alternatives, even if that needs more trials).
• Set Settings▸Acuity▸Starting LogMAR to a higher value than the default 1.0, e.g. 2.5, to avoid initial frustration.
• Settings▸Acuity▸Margin for biggest optotype: Is ½ stroke width enough for you?
• When ceiling effects occur, namely the largest optotype is not seen, the result will indicate this (details here in the manual)
• Run an acuity test run yourself wearing strong blur filters (e.g., foils that simulate cataracts). Really! All personnel, including principal investigators, should do this once. This experience will change your outlook.

Well, I admit it’s a bit hidden; there are some idiosyncrasies in FrACT₁₀, suggestions welcome…
In  Settings▸Gratings▸What to sweep  select  Spatial frequency, "Acuity".

Absolutely, you can do that. Use Settings▸Display Transformation to compensate for the image mirroring/inversion. [BTW: surface mirrors are needed to avoid double images.]

Enter a negative value for the Weber contrast in Settings▸Acuity>Optotype contrast, e.g. with the max– button.

A good question… and the answer is complex. First: it depends on the size (Diameter), default 60’. Second: an optotype (be it letter, Landolt ring, tumbling E…) does not have a spatial frequency, rather it is a mixture (I’ve calculated the spectrum but never published it).

I had hoped to have a setting for this. However, after extensive experiments with the Multi-Screen Window Placement API, I conclude that it is not possible to programmatically move to 2nd screen without user interaction. So use these simple solutions:

• If you’re using FrACT₁₀ in a browser, move the browser window to the 2nd screen.
• If you’re using the FrACT₁₀ progressive web app, move the window to the 2nd screen (and, depending on browser, it will even remember that after restart).
• Using the older stand-alone app? Update to the progressive web app :).

When the window is on the second screen, you can use the ︎Fullscreen button to go, you know, fullscreen. Also note: Need to repeat size calibration (“plastic card / blue ruler”) on the 2nd screen – its pixel size probably is different.

So the plastic card fits vertically but not horizontally (or vice-versa)?

In that case the pixels on your VDU are not square. While modern VDUs all have square pixels, it can happen that the screen resolution is set to a value not compatible with square pixels. We want to avoid that anyway, because it will lead to blur. Look for “Display resolution” (or a similar term) on your computer and experiment with different combinations.

I see two possible scenarios where this could occur:

• The settings for screen calibration (Length of blue ruler) and Observer distance combine to a very large optotype, and you see only its empty center.
• The Optotype eccentricity setting is too large, shifting the optotype off-screen.

If nothing rings a bell, use Settings▸PRESET▸Standard Defaults: Now you should see something, and you can proceed to apply your preferences in Settings.

The version number will only increment when I make major changes or add a feature. In contrast, the release date is updated whenever anything changes. Find them both at the main screen, top right; or use the About button – there not only FrACT₁₀’s version date is given but also the versions of embedded frameworks. The export format has its own version scheme, the exported data set starts with the current version number; this allows your analysis software to more easily deal with changes here.

Please include the version when citing FrACT. Thank you!

The most likely cause: you did not install it from a “stand-alone” page, like from here.

1. Take a look at the Manual 😇
2. If that does not answer your question, don’t hesitate to email me: bach@uni-freiburg.de.