Try out above. All platforms, including mobile; tablets are fine, smartphones too small, even if rotated.
Frozen versions: 2021-06 | 2022-03 | 2023-11 | 2024-04 | 2025-04 | 2025-06 | 2025-07 | 2025-08

FrACT is frequently updated. It could be a bug fix, more often a cool new feature or a new preset for some study. As programming is wont to be, problems, bugs, or undesirable side effects may sneak in. In spite of extensive testing (like using the testing suite above), problems may slip through. To protect you, before deploying a major update, I preserve the former version, thus "freezing" it and linking to it here. [Even if said link disappears, that version remains on the site.] So, if you routinely rely on FrACT, use the latest frozen version. It’s best installed as a Web App, so FrACT will still run even if this website is temporarily down.

"FrACT", a semi-automatic visual test battery available as a free app, employs psychometric methods combined with anti-aliasing and dithering to provide self-paced assessment of visual acuities (Bach 1996), contrast sensitivity (using optotypes or gratings) and Vernier acuity on standard equipment. FrACT complies with the international norm for acuity testing (EN ISO 8596) and is recognized as one of the "Recommended Methodolog(ies) for Assessment of Visual Acuity" by the Harmonization of Outcomes and Vision Endpoints in Vision Restoration Trials Taskforce. FrACT is used world-wide in vision labs, by optometrists, ophthalmologists, and in clinical trials (→Who used FrACT). It has been validated in independent laboratories (→Descriptions & Validations).

After rewriting the code to serve all platforms, I renamed FrACT to "FrACT10"; they are identical in basic test behaviour. FrACT10’s possibilities now surpass the "classic" FrACT; its code is →open source (→commit history). With FrACT10 moving forward online and as web app, "classic" FrACT will remain available here.

Key characteristics

• Test symbols / Optotypes: Sloan letters, Landolt rings, tumbling Es, symbols, gratings; Vernier patterns for hyperacuity. Gratings: sinusoidal or square-wave.
• Acuity / contrast thresholds: Assessed with a progression of test symbols, using a Bayesian approach and following a modified "Best PEST" strategy (see General Pertinent Sources for details).
• Results Display: As LogMAR, Snellen fraction, or decimal acuity; full details can be exported for further processing.
• Customization: Many parameters can be set, such as for testing low contrast acuity or colours, and results include not just acuity/contrast thresholds but also reaction time and more.
• Technicalities: Equipment resolution limits in space and luminance are mitigated by dithering and antialiasing.

Where and why to use?

• Participants’ screening in research settings: This is where I have personally used FrACT the most.
• Clinical Studies: Ideal for studies where visual acuity or contrast sensitivity is an outcome variable, aided by FrACT’s PRESETS for reproducibility, and Export options for reporting, e.g., Chang et al. 2022, Palanker et al. 2022, Nanoscope’s RESTORE Trial 2024, Mohanty et al 2025. It can even be embedded in larger systems and fully controlled via HTML messages.
• Acuity detail studies: Useful for research on specific acuity properties such as contrast, exposure time, reaction time, feedback, noise, and hyperacuity, as evidenced by numerous papers.
• Perfect for keeping your waiting patients entertained :) !

• One particular advantage of FrACT: its ultra-wide range with continuous coverage
  • For (ultra) low vision: FrACT can reproducibly assess CF and even HM: Counting Fingers, CF≈1.9 LogMAR; Hand Movement, HM≈2.3 LogMAR. →Sources and more detail.
  • For normal vision: FrACT can assess acuities beyond 2.0 decimal = 40/20Snellen = –0.3 LogMAR.
    [Although 20/20 vision is commonly considered as "normal," numerous studies (e.g., Frisén & Frisén (1981), Elliot et al. 1995, Pointer 2006) suggest that the actual average acuity among the general population is nearly double that figure. Consider: If Snellen had adopted the population average as the standard for "normal" vision, approximately half of the population would have fallen below this threshold.]
  • This wide range needs appropriate visual display units, see "A note on display resolution..." in the Manual.
• Low contrast optotypes for Acuity: Settings▸Acuity▸Optotypes/background color>b/w
• Color for Acuity: Settings▸Acuity▸Optotypes/background color▸Color
• Mirror support: Settings▸General▸Display transformation
• Gratings: Can sweep (find threshold) for contrast or spatial frequency: Settings▸Gratings▸What to sweep:
• Eccentricity for Acuity and Contrast: Settings▸General▸Optotype eccentricity [°]
• Timing control: ISI, display time and response time: Settings▸General▸Timeouts
• Noise embedding for Acuity and Contrast: Settings▸Misc▸Noise embedding
• For studies: reproducible across time and locations through the Settings▸PRESETS mechanism or the All settings export/import, and multiple result export options
• Explore the Settings for more information. Something unclear? Simply hover your mouse over any questionable GUI element to reveal tooltips which tell you more.
• Full programmatic control of FrACT with HTML messages for embedded installations.

Cost, in one word: Nothing.

Yes, free als in beer. What’s more, my apps don’t contain any code to monetize their use (e.g. selling your location). However, should you desire to contribute: donations are gladly accepted via Buy Me a Coffee.

Please remember to cite it when publishing (with version number). Since this is free software, there’s no warranty for anything. Your feedback is warmly welcomed and has frequently improved FrACT.

So where can I get it?

FrACT10: Use online (e.g., as running embedded above or alone) or as Web App. Since the Web App can be installed on all platforms, the former downloads for MacOS & Windows are deprecated.

FrACT10 is a Web App (details: blog) that can be installed for offline use like so: First open it in its own window/tab (or open a "frozen" version) and look for your browser’s "3 dot" or sandwich menu:

• Other desktop browsers
  • Safari (MacOS ≥ Sonoma): File > Add to Dock ...
• In all cases the Web App will be saved here: «user»/Applications/...
• Firefox doesn’t (openly) support Web Apps
• Linux: depends on your specific flavor. I found various recipes, e.g. for "Web App Manager"
• iOS: Add to Home Screen in the Share menu and proceed as suggested
• Android: depends on the specific brand.

Over 1700 papers cite FrACT, see adjoining figure spanning nearly three decades.
Examples of FrACT in clinical trials: Zrenner et al. 2010, Stingl et al. 2013, Stingl et al. 2021, Chang et al. 2022, Palanker et al. 2022, Nanoscope’s RESTORE trial (NCT04945772, NCT04919473, Mohanty et al 2025).
Recommended in this review: Schmetterer et al. 2023.

FrACT used as outcome measure in these registered trials (compiled 2024):
NCT03326336, NCT03913143, NCT00515814, NCT01024803, NCT01497379, NCT01614158, NCT01673503, NCT01876147, NCT02391298, NCT02720640, NCT03392324, NCT04111068, NCT04184037, NCT04284293, NCT04417959, NCT04642495, NCT04762368, NCT05655689, NCT05921162, NCT06425666, NCT06704737.

• FrACT10 online: Any modern browser on Android, iOS, Linux, MacOS or Windows
• FrACT10 Web App: Chrome/Edge/... browsers on Windows, same + Safari on MacOS, also Android, iOS, Linux

Also nearly any modern visual display unit. For remote input see the response box below in Usage Details. Please observe the Checklist for reliable results.

Clinical trials demand greater rigor than what’s outlined in the Checklist. These trials will be professionally managed, and I’ve had the pleasure of collaborating with numerous exceptional clinical scientists, statisticians and other experts. Many of my suggestions will state the obvious and here appear in random order, they’re not intended to be exhaustive.

• Run every condition twice. That yields a quality marker (measure of reproducibility, e.g., the Bland-Altman limits of agreement), then average for any further processing.
• After you have established all desired settings (that differ from FrACT’s defaults), let’s make a PRESET for your study. This ensures reproducible settings across time & centers.
• Document the version of FrACT, don’t change it during the course of your study (see "frozen" above). For documentation of all settings used, employ Settings>Export Setting.
• If it’s an (ultra) low vision study: See this section on low vision in the FAQ.
• For every session, apply your preset, check luminance of the VDU, and verify size calibration (with "plastic card").
• Find an easy and reproducible way to ensure a well-controlled observation distance.
• For reliable documentation and 21 CRF Part 11 compliance: Use the Result→PDF button after every run. In Settings>Misc there are fields for ID and tested eye which are included in the PDF output (or any other export).
• Make use of automated results export / integration with your hospital documentation system. Never type a result manually.
• Make yourself independent from my website: Install a "frozen" version of FrACT10 as web app, or run from your own servers, as described in Long term support (at the end).
• FrACT in standard form might not be permitted by regulatory bodies because its settings are accessible to technicians and patients. Approaches to lock down FRACT include:
  • Drive FrACT solely with HTML messages. Implement a "Start test run" button in your software which sets all settings to your standard and then initiates the test run. This functionality is similar to the Click for demo run↑ button above and requires some programming expertise on your side.
  • I could add a setting to reload your specific Preset at the beginning of any run, also adding the preset name to the exported data. This would ensure that every result was obtained under standard settings. So far, there hasn’t been a need for this feature.
• Don’t hesitate to contact me: bach@uni-freiburg.de; open for consulting.

• Good sources on this topic
  • British Standards Institution (1987) Precision of Test Methods Part 1. Guide for the Determination of Repeatablity and Reproducibility for a Standard Test Method by Inter-laboratory Tests (BS 5497-1:1987 ≙ ISO 5725:1986), p 1
  • ISO 5725-1:2023, in Introduction (not behind a paywall).
• Precision, as defined above and in Wikipedia, refers to the variability observed in repeated tests and encompasses two aspects:
  1. Repeatability: quantifies precision in tests performed during a short interval of time in one laboratory by one operator using the same equipment.
     The repeatability value r according to the British Standards Institution is defined as the value below which the absolute difference between two single test results may be expected to lie with a probability of 95%. It can be quantified with the 95%-confidence interval (CI95) of the Bland-Altman Limits of Agreement (LoA). For example, a CI95 of ±0.2 LogMAR indicates that in 95% of the cases results will fall within a ±2 lines range.
  2. Reproducibility: quantifies precision in tests performed under different conditions, e.g. in different laboratories.
     Thus: reproducibility ≥ repeatability.
• Accuracy, also called "trueness", describes how close measurements are to their true value; a difficult concept in visual acuity. Thus we fall back to operational definitions: Acuity x is what test x’ measures.

FrACT’s variability has been documented across various studies, as outlined below. Conclusion: The 95% confidence interval of FrACT’s LoA is ≲ ±0.2 LogMAR in patients/participants with acuities of hand movement or better. If you want lower variability, increase the number of trials beyond 18 [roughly a 1/√n dependence, see Bach (2024)].

CF: counting fingers, HM: hand movement (→how to derive quantitive values for these).
The values for the 2006, 2007, 2009 & 2016 studies are based on re-analysis of the original data sets because formerly I used decimal acuity and also idiosyncratic measures of variability, obnoxious to convert.

Statistical Intricacies

1. To evaluate test-retest variability, a common approach involves calculating a correlation (e.g., ICC). However, this method is inappropriate because correlations are range-normalized (e.g. McAlinden C, Khadka J, Pesudovs K (2011) Statistical methods for conducting agreement (comparison of clinical tests) and precision (repeatability or reproducibility) studies in optometry and ophthalmology. OPO). Simply adding one very poor acuity and one very good acuity (extending the range) will automatically yield a misleadingly high correlation coefficient, see: Holopigian K, Bach M (2010) A primer on common statistical errors in clinical ophthalmology. DOOP [→PDF]. Therefore, I employ Bland-Altman’s (1986) limits of agreement (LoA).
2. The British Standards Institution’s / ISO’s repeatability value r (the "repeatability coefficient" in Bland-Altman 1999) happens to have the same value as LoA when there are two measurements per eye and no bias (thanks UG! This deserves a brief paper in its own right because the Bland-Altman papers are so arcane) (Bland-Altman 1996).
3. Most of the time I find a tiny bias in vision assessment: The second run’s acuity is a wee bit better, probably due to learning. Since the bias is not significant, I average the upper and lower LoAs, arriving at the single number for LoA as given in the table above.

Main motto: Progredere, sed nihil frange1.

FrACT began 40 years ago and is available on the Internet in various forms for 30 years now. I am committed to providing unwavering support for decades. "Classic" FrACT was last (and finally) updated in 2020, and will continue to work in the two major operating systems due to its 64-bit architecture.

Initially, FrACT10 was exclusively running in a web browser. Subsequently, it was also released as an Electron-based application. If you currently use one of these applications, they will continue to function without gaining new features. In 2023, FrACT10 was replaced by a platform-independent Web App (details: blog). This development makes FrACT10 accessible on all platforms and allows for automatic updates, if so wished.

• If you want to run the "(b)leading edge" most recent FrACT10, install the Web App from here. Whenever I update FrACT10, your Web App automatically updates in the background: Give it 1⁄2minute, then reload, voilà!
• If you want a stable version for long-term usage (studies, clinical trials, student assignments, ESU, ...), install the Web App from the latest "frozen" version. Even if you clear your browser caches, specifically the cache which Chrome calls "Hosted app data", the Web App will reinstall the same version by itself. Frozen versions will remain on the server for years, even if I remove the direct links to them.
• FrACT10’s code is open source on GitHub (commit history). So even if I don’t, FrACT10 will live on 😎.

To make yourself completely independent from me (advised), install all files at an HTTP server of your own, like so:
Got to https://michaelbach.de/fract/FrACT10/FrACT/ (or a frozen version of your choosing). Download the entire contents of that directory (≈15 MB; e.g., with a "site sucker") to an appropriate location on your HTTP server. Then open the index.html therein and lo, FrACT10 runs from your server!

1Advance, but don’t break anything.

If you ran a previous version, all selections in Settings may automatically be defaulted in order to update internal structures (depends on the exact version).

How to calibrate?

Very easy: Enter the width of the calibration bar (easiest with plastic card) and the observation distance in the Settings>General interface. Be sure to have enough distance from the screen, so pixel resolution does not limit acuity. Also consider other entries in Settings and set them appropriately; the defaults can not be optimal for every situation. The Checklist is a must-read; in the FrACT10 Manual or classic FrACT3.x Manual [PDF] all settings are explained, probably in more detail than you wished...

For the Contrast Test, you also need to go to Settings>Gamma. There you can interactively linearise the luminance transfer function ("gamma correction") before testing. But see below for a cautionary note on the difficulties of contrast testing. Or use a screen calibrator (see What is special about the Contrast Test?).

All selections in Settings are automatically saved.

Starting the test

Apart from using the obvious buttons a test run can also conveniently be started from the numerical keypad (digit "5"). Enter the position of the appearing optotypes orientation via the numerical digit keys, or use the alternatives. The digit key "5", pressed twice, aborts an ongoing test run (and inhibits display of the premature result). In Settings you can select which test start on "5".

Number of trials?

• First, this depends on the number of choices with your chosen optotype: Sloan letters have 10 choices, Landolt Cs 8 or 4, gratings 4 or 2, Vernier lines 2.
• The default number of trials is set to 18 with 8/10 choices, 24 with 4 choices, and 32 with 2 choices. A paper on this: Bach M (2024) Freiburg Vision Test (FrACT): Optimal number of trials?
• For scientific applications, rather than increasing the number of trials, best repeat each condition (e.g. each eye at 18 trials, ideally in an ABBA scheme) and take the mean; this also allows test-retest quality quantification.
• Why are all the above trial numbers divisible by 6? Because every 6th trial is a "bonus" or "easy" trial (if enabled), where the optotype is presented at 3 times its current threshold estimate. This keeps your participants happy, especially since the last trial is thus always a ‘success’ experience.

4 or 8 gap positions, or 10 letters?

Sloan Letters have 10 alternatives, Tumbling E has 4 possible orientations, for Landolt C there are 4 or 8. Depending on your application, you may wish to use only the 4 primary Landolt gap positions or to supplement it with the four oblique directions; FrACT allows both. Briefly: 4 directions are less easily confused, but guessing probability is higher, making more trials necessary; 8 positions allow more rapid determination of visual acuity(less trials) as the guessing probability is lower. Final outcome should be identical, but the number of runs must be adequately chosen; see previous item. The letters are the standard Sloan letters, the Sloan letter "C" is physically identical to the Landolt C.

Results Export

The test result is presented on the screen in formats as chosen in Settings>Acuity. Additionally, the result can be transferred to the clipboard (in a simple, final-result-only or full-history fashion); so just switching to a spreadsheet program running in the background allows pasting and archiving the result. More export options: FrACT10 Manual>Exporting.

What is special about the Contrast Test?

Experience taught me that contrast testing is technically more demanding than acuity testing.

• Avoid stray light falling on the screen; a "monitor hood" may be sufficient
• Be sure to linearise luminance ("gamma correction") before you test contrast (go to Settings>Gamma); the best approach is to employ a screen calibrator (I’m using the Spyder). Choose a gamma value of your liking (e.g. 1.8 or 2.0) and calibrate the screen/computer to that value; then set it thus in Settings>Gamma in the gamma value field.
• Normal contrast vision is better than normal screen luminance resolution (256 steps are not enough, especially after loosing a bit or two via linearizing ("gamma correction"). This is solved via dithering (Bach (1997) Anti-aliasing and dithering in the "Freiburg Visual Acuity Test’. Spatial Vision 11:85–89).

Mirroring, or more generally, "Display transformation"

In many environments the standard 4 m distance may be difficult to achieve. In such conditions, high-quality front surface mirrors can extend the effective viewing distance. Their image-flip can be compensated for by Settings>Display transformation.

Shortcuts

I am lazy and prefer to start actions without mousing, thus I added a number of shortcuts, see also FrACT10 Manual or classic FrACT3.x Manual [PDF]. Casing is irrelevant for shortcuts.

Avoiding more mousing: Use the tab key (⇥) to move between fields in Settings (shift-tab for backwards). The blue 🆗 button is activated by the return key.

• Jolly et al. (2019) A Randomized Crossover Study to Assess the Usability of Two New Vision Tests in Patients with Low Vision. Optomet Vision Sci 96(6): 443–452 [comparison of the Berkeley Rudimentary Vision Test (BRVT) & FrACT]
• Bach & Schäfer (2016) Visual acuity testing: feedback affects neither outcome nor reproducibility, but leaves participants happier. PLOS ONE 11(1):e0147803
• Tsai L, Thomas E, Tarrant J, Bentow S, Kasthurirangan S (2015) A Comparative Clinical Evaluation of Two Visual Acuity Testing Systems: ETDRS vs. FrACT. IOVS 56:3890 [meeting abstract]
• Tarrant J, Thomas E, Tsai L, Kasthurirangan S (2015) Evaluation of defocus curve performance with two visual acuity testing systems: ETDRS and FrACT. IOVS 56:3891 [meeting abstract]
• Kollbaum, Jansen, Kollbaum, Bullimore (2014) Validation of an iPad test of letter contrast sensitivity. Optom Vis Sci 91:291–296 [Contrast, comparison of Pelli-Robson, FrACT, and a new iPad test]
• Kurtenbach, Langrová, Messias, Zrenner, Jägle (2013) A comparison of the performance of three visual evoked potential-based methods to estimate visual acuity. Doc Ophthalmol 126:45–56 [Acuity, comparison of FrACT and ETDRS]
• Lange, Feltgen, Junker, Schulze-Bonsel, Bach (2009) Resolving the clinical acuity categories "hand motion" and "counting fingers" using the Freiburg Visual Acuity Test (FrACT). Graefe’s Arch Clin Exp Ophthalmol 247:137–142 [→PDF] [Acuity, ETDRS & FrACT in very low acuity]
• Bach, Schulze-Bonsel, Feltgen, Burau, Hansen (2007) Author Response: Numerical Imputation for Low Vision States. IOVS, eLetter Aug 2007 [Acuity]
• Bach (2007) The Freiburg Visual Acuity Test – Variability unchanged by post-hoc re-analysis. Graefe’s Arch Clin Exp Ophthalmol 245:965–971 [→PDF] [Acuity, methodology]
• Bühren, Terzi, Bach, Wesemann, Kohnen (2006) Measuring contrast sensitivity under different lighting conditions: Comparison of three tests. Optom Vision Sci 83:290–298 [Contrast]
• Schulze-Bonsel, Feltgen, Burau, Hansen, Bach (2006) Visual acuities "hand motion" and "counting fingers" can be quantified with the Freiburg visual acuity test. IOVS 47:1236–1240 [Acuity, ETDRS & FrACT in very low acuity]
• Dennis, Beer, Baldwin, Ivan, Lorusso, Thompson (2004) Using the Freiburg Acuity and Contrast Test to measure visual performance in USAF personnel after PRK. Optom Vis Sci 81:516–524 [Contrast]
• Terzi, Bühren, Wesemann, Kohnen (2004) Frankfurt-Freiburg Contrast and Acuity Test System (FF-CATS) – A new test to determine contrast sensitivity under variable ambient and glare luminance levels [Article in German]. Ophthalmologe 102:507–513 [Acuity]
• Loumann Knudsen L (2003) Visual acuity testing in diabetic subjects: the decimal progression chart versus the Freiburg visual acuity test. Graefes Arch Clin Exp Ophthalmol 241:615–618 [Acuity]
• Neargarder, Stone, Cronin-Golomb, Oross 3rd. (2003) The impact of acuity on performance of four clinical measures of contrast sensitivity in Alzheimer’s disease. J Gerontol B Psychol Sci Soc Sci 58:54–62 [Acuity]
• Wesemann (2002) Visual acuity measured via the Freiburg visual acuity test (FVT), Bailey Lovie chart and Landolt C chart [Article in German] Klin Monatsbl Augenheilkd 219:660–667 [Acuity]
• Bach (1997) Anti-aliasing and dithering in the "Freiburg Visual Acuity Test’. Spatial Vision 11:85–89 [Methodology]
• "Generic description" Bach (1996) The "Freiburg Visual Acuity Test" – Automatic measurement of visual acuity. Optometry and Vision Science 73:49–53 [PDF] Also consider the FrACT10 Manual or classic FrACT3.x Manual [PDF] which describes all options [PDF]

Glad you asked! If you used FrACT for your research, please cite it.

• Bach M (2024) Freiburg vision test (FrACT): optimal number of trials? [PDF] (most recent)
  or
• Bach (1996) The "Freiburg Visual Acuity Test" – Automatic measurement of visual acuity. Optometry and Vision Science 73:49–53 [PDF] (the "generic" paper)
  or
• Bach M (2007) The Freiburg Visual Acuity Test – variability unchanged by post-hoc re-analysis. Graefes Arch Clin Exp Ophthalmol 245:965–971 [→PDF]
  or
• Farassat N, Jehle V, Heinrich SP, Lagrèze WA, Bach M (2024) The Freiburg Acuity Test in preschool children: Testability, test–retest variability, and comparison with LEA symbols. Transl Vis Sci Technol 13:14.

Also please mention the version plus release date (top right, can be copied from there).

The briefest way to refer to the present page (its URL):
https://michaelbach.de/fract/

• Lieberman & Pentland (1982) Microcomputer-based estimation of psychophysical thresholds: The Best PEST. Behavior Research Methods & Instrumentation 14:21–25
• Newman & Sproull (1979) Principles of Interactive Computer Graphics. New York, McGraw-Hill, Inc.
• Metha, Vingrys, Badcock (1993) Calibration of a color monitor for visual psychophysics. Research Methods, Instruments, & Computers 25:371–383
• Mittelviefhaus, Bach, Jedynak, Kommerell (1993) Der Freiburger Visustest. Ophthalmologe 90:1320–135
• Pelli (1996) VideoToolbox
• Pentland (1980) Maximum likelihood estimation: The Best PEST. Perception & Psychophysics 28:377–379
• Petersen (1990) Zur Fehlerbreite der subjektiven Visusmessung. Fortschr Ophthalmol 87:604–608
• Treutwein (1995) Adaptive Psychophysical Procedures. Vision Res 35:2503–25
• Visual Standards – Aspects and Ranges of Vision Loss
• Bach (1997) Anti-aliasing and dithering in the "Freiburg Visual Acuity Test’. Spatial Vision 11:85–89

While FrACT began as a "hobby project" (see History below), its widespread adoption entailed rigorous quality control.

• Quantitative correctness of the optotype geometry. This is easily verified on the screen with a ruler and a little trigononometry.
  • To make this easier, you can use this tip:
    At a distance of 57.3 cm, one degree of visual angle corresponds to 1 cm. Multiply this with 6 and set the resulting 344 cm as testing distance. Calibrate the ruler (e.g. with plastic card), then:
    → 1 millimeter on the screen corresponds to one minute of arc. Use, e.g., tumbling E, and if you have the Starting LogMAR at 1.0, the first optotype’s stroke width should be 10 mm.
  • For very low acuity, when converting between angle α and screen pixels x, be sure to use this formula
    x = 2 · distance · tan(α / 2), and not x = distance · tan(α).
    (Maybe not that important: the difference even at 50 cm observation distance is less than 1%.)
• Quantitative correctness of the threshold outcome. This was done by controlled comparisons to chart testing in my lab and in independent laboratories.
• Version / Change Control. The code for FrACT10 is open source. Issues and their solutions are tracked with git; see FrACT10’s commit history (history of classic FrACT below).
• Documentation Practices. The code is self-documenting through descriptive long variable / method names; a detailed manual describes everything; all changes are tracked and available on github.
• Data Management.
  • No confidential user-traceable data is recorded nor stored by FrACT itself.
  • Settings are reproducible and traceable via PRESETS.
  • Result data integrity is ensured via automated export mechanisms
• Low-level programming practices can only be mentioned in passing here
  • Robustness. The Settings dialog is augmented with plausibility checking of the input values. E.g., when a numerical entry has incorrect number formatting or is set to an implausible value, the parameter is reset to its default value.
  • Systematic code modularization
  • Testing suite, unit testing
  • Automated GUI end-to-end testing with Cypress

As is well known, any interesting computer program cannot be shown to be entirely correct, and usually isn’t. Thus the ongoing validation by independent and critical users is most important for the verification of FrACT. Thank you! All pertinent comments welcome. Also see Bugs? below.

In 1985 I programmed the first version, using a custom-built Z80 CP/M computer with a graphics board based on the NEC 7220 graphics chip, using serious digital differential algorithms (Newman & Sproull, 1979) in assembly language just to draw the Landolt C. It became more than a curio after porting to Pascal and incorporation of the Best PEST threshold algorithm (Lieberman & Pentland, 1982) [nowadays one would call this Bayes-based] and was routinely employed in my lab to screen subjects’ acuities before experiments.

The port to the Macintosh in 1992, translating from Pascal to C++, facilitated the use of anti-aliasing (Bach, 1997), which immediately made the measurements more precise (by allowing intermediate pixel values) and enabled a higher acuity limit at practical distances (a few meters).

In 1996 the first Internet version was made available after adding gamma correction for better quantitative definition. The rich feedback spurred many bug fixes and feature expansions, paving the way from a ‘hobby’ to a professional application, now validated in independent laboratories.

The pressure of requests finally motivated me in 2002 to port the program to an operating system-agnostic environment, for which I chose ActionScript/Flash (formerly Macromedia, acquired by Adobe). Consequently the program then ran in near identical fashion on the Macintosh, Linux and Windows operating systems for nearly two decades.

Anticipating the demise of Flash, in 2020 I ported the program to a JavaScript-based framework (Cappuccino); thus it runs on all platforms, including tablets. In 2021, FrACT10 achieved parity with the last "classic" FrACT3.10.5 and its code was open sourced. In 2022 the browser-only FrACT10 was complemented by apps for MacOS & Windows, based on Electron. In 2023 the install hassles of apps were solved by packaging FrACT10 as a Web App for all platforms. In 2024, I implemented HTML messages to control FrACT10 and read/set its Settings and more; this facilitates automation and embedding FrACT10 in larger systems.

Version history of classic FrACT

[→even more early history details]

• Bugs in FrACT10
  • Currently (2025-11) no known bugs (except with the grating’s circular mask, only on Safari, a Safari bug)
  • My issues section on GitHub lists bugs (if known) and also lists planned improvements and new ideas. Please alert me if you find problems, or file a bug report / issue. Usually I respond rapidly.
  • For historical interests: closed issues.
• Bugs in "classic" FrACT
  • At this time (2025-11), I am not aware of any bugs in classic FrACT. Please tell me should you find problems.
  • Resolved
    • 2020年03月06日 Bug: When the optotype contrast is reduced, the Vernier rendering is wrong, because the Vernier doesn’t take contrast into account, but the background does. ←Solved in 3.10.5
    • 2019-05 Enhancement: Indicate in the results whether a ceiling/floor effect occurred. ←Added in 3.10.2

Should you desire to contribute: Donations are gladly accepted via Buy Me a Coffee.