FAQ & Support
Support
VEET 2.1 Troubleshooting Guide
Common VEET errors and how to address them.
Release Date: 2/3/2026
VEET 2.1 Using the Command-line Interface
Instructions for how to access the VEET’s firmware through virtual serial port interface for troubleshooting or for performing bulk operations.
Release Date: 1/23/2026
FAQ Section 1: What the VEET Is
The VEET (Visual Environment Evaluation Tool) is a pair of spectacle temple arms that measure a wearer’s visual environment continuously during normal daily activity. Each temple arm is an independent instrument recording illuminance, spectral composition, optical distance to near objects, and head motion at the plane of the eye. The VEET is a research instrument operated under the open-science initiative Project VEET; it is not a consumer product and not a medical device.
The VEET is developed and supported by SIXTY D LLC, based in Bend, Oregon, founded by David Sullivan, former Meta Hardware Systems Lead and co-inventor of the VEET platform. The VEET was originally incubated at Meta Reality Labs Research, which assigned all Project VEET assets to SIXTY D effective June 4th, 2026. To reach SIXTY D about the VEET, use the contact form at https://projectveet.com/contact-us/.
No. Meta ended its support for Project VEET in 2026 and transferred the entire program — hardware designs, software, documentation, and this website — to SIXTY D LLC. projectveet.com is maintained by SIXTY D and is no longer affiliated with Meta. Support questions go to the Project VEET contact form at https://projectveet.com/contact-us/.
Light measured at the wrist or chest differs substantially from light entering the eye, because body-worn sensors are shaded, angled away from gaze direction, and covered by clothing. The VEET integrates its sensors directly into spectacle temple arms, placing them as close to the eye as possible with consistent placement across wear sessions.
Project VEET is the open-science research initiative: device documentation, data standards, analysis tools, and the researcher community, hosted here. SIXTY D LLC is the company that operates Project VEET and separately develops commercial instrumentation. Research support under Project VEET is provided at no cost and is not contingent on any commercial relationship with SIXTY D. Inquiries about either go through https://projectveet.com/contact-us/.
FAQ Section 2: Sensors and Measurement
Each VEET temple arm carries four solid-state sensors: a time-of-flight distance sensor (ams OSRAM TMF8828), an ambient light sensor (ams OSRAM TSL2585), a spectral light sensor (ams OSRAM AS7341), and a 6-axis inertial measurement unit (Bosch BMI270). Both temple arms are independent instruments, giving bilateral measurement.
The two temple arms are aimed differently on purpose, to cover a range of gaze behaviors. From the wearer’s perspective, the right arm’s sensors are oriented 20 degrees downward and 4 degrees inward toward the sagittal plane, approximating near-work gaze. The left arm’s sensors sit perpendicular to the glasses frame. Left and right channels are therefore not interchangeable and should not be averaged without accounting for this difference in sampling geometry.
The VEET ambient light sensor reports photometric data directly in lux, covering dim indoor conditions through direct sunlight — over 100,000 lux at the upper end. It also includes UVA sensing and light flicker detection.
The device will report values down to 0.001 lux, but accuracy below 0.1 lux is not well characterized, because the calibration instrumentation available at very low light levels is itself limited. Treat sub-0.1 lux readings as indicative rather than quantitative.
This matters most for chronobiology and sleep research, where nighttime light exposure is often the variable of interest and the relevant levels fall in exactly this range. Studies depending on precise values below 0.1 lux should validate against a reference instrument for their specific conditions, and should note the limitation in methods.
The VEET spectral sensor measures 8 narrow optical bands across the visible spectrum plus full-spectrum clear and near-infrared channels — 11 channels total, spanning approximately 350 to 1000 nm. Full spectral power distribution curves and correlated color temperature can be reconstructed in post-processing using the published calibration matrix and SPD Reconstruction Guide.
Less well, and this should be checked rather than assumed. Below approximately 10 lux the spectral sensor produces very low counts, and data from that regime should be inspected before it is trusted. Low counts propagate into anything derived from them, so spectral power distribution reconstructions and any alpha-opic quantities computed from them — including mEDI — inherit the same uncertainty.
Two practical consequences. First, this threshold sits near the transition between photopic and scotopic visual response, so the regime where the instrument becomes least reliable is also one of genuine physiological interest. Second, many indoor evening environments fall at or below 10 lux, which places nighttime circadian research directly in this range. Studies whose primary measurements are made in dim conditions should inspect raw spectral counts, validate against a reference instrument for their specific conditions, and report the limitation in methods rather than treating derived values as equally reliable across the full illuminance range.
This is a configuration choice, not a hardware ceiling. The stock firmware sets spectral sensor sensitivity so the device does not saturate in full daylight, which necessarily sacrifices performance at the low end. SIXTY D can supply a custom firmware configuration biased toward low light levels instead, for studies whose measurements are concentrated in dim conditions. Automatic scaling across the full range was not feasible at the time the current firmware was written, so the two ends are a trade rather than something the device resolves on its own.
The implication for study design is that the sensitivity range should be chosen to match the primary measurement before data collection begins. A single configuration cannot serve both bright outdoor exposure and dim indoor evening measurement equally well, and data collected under different configurations is not directly comparable. Discuss requirements with SIXTY D at https://projectveet.com/contact-us/ during protocol design rather than after collection has started.
Not directly. The VEET spectral sensor records calibrated counts only; it does not compute circadian metrics on-device. Melanopic equivalent daylight illuminance (mEDI) and the other CIE S 026 alpha-opic quantities must be derived in post-processing, by reconstructing the spectral power distribution from the sensor’s channel outputs and applying the standard action spectra.
Two routes are available. The Spectral Power Distribution (SPD) Reconstruction Guide on the Sensors & Data page documents the method, and Project VEET publishes a Python reconstruction script and a simplified Excel example alongside it.
Alternatively, researchers working in R can use LightLogR, an open-source package developed by the Translational Sensory & Circadian Neuroscience Unit (MPS/TUM/TUMCREATE) that supports VEET data output directly. LightLogR handles import, gap detection, visualization, and the calculation of 62 light-exposure metrics across 17 families, including mEDI-based threshold, duration, and timing measures. It is available on CRAN and documented at https://tscnlab.github.io/LightLogR/. The group’s wider work is at https://www.tscnlab.org/.
The VEET’s maximum sampling rate is 0.5 Hz — one acquisition every 2 seconds — and each sensor is individually configurable to slower rates, or can be turned off by setting the sensing interval to 0.
No. The VEET does not require post-delivery calibration. Routine maintenance is limited to keeping the optical surfaces clean. The spectral reconstruction calibration matrix is published on the Sensors & Data page for researchers who want to verify or modify the processing chain.
Any VEET can be rechecked at the manufacturing facility for a small fee plus shipping. Contact SIXTY D at https://projectveet.com/contact-us/ to arrange it. Full recalibration requires specialized optical equipment and cannot be performed in a typical lab or clinical setting, so field recalibration is not offered or recommended.
Before arranging a check, read the next answer. Most suspected calibration discrepancies turn out to be a field-of-view mismatch rather than a calibration problem.
Because the two instruments measure different things. Most commercially available light meters use a cosine-corrected diffuser that integrates light across a full hemisphere. The VEET’s optical acceptance is deliberately narrower, chosen to approximate the field of view of the human eye rather than a hemispherical sensor.
As a result, the two are not directly comparable in either direction. A cosine-response meter will typically report higher values than a VEET in an environment with bright off-axis sources — a window or ceiling fixture outside the wearer’s line of sight — because the meter collects that light while the eye, and the VEET, largely do not. A discrepancy between a VEET and a handheld meter is therefore expected behavior, not evidence of a calibration fault.
This distinction matters when comparing VEET data against measurements from cosine-corrected loggers in the literature, and should be stated in methods sections rather than assumed away. Researchers validating a VEET against a reference instrument should match the measurement geometry, or characterize the difference explicitly, rather than treating the reference meter as ground truth.
No. The VEET contains no camera and no image sensor of any kind. Distance is measured by infrared time-of-flight ranging, which returns depth values only and cannot reconstruct a scene or identify what the wearer is looking at.
FAQ Section 3: Distance Measurement
The VEET time-of-flight sensor illuminates the near field with infrared light and reads 64 points on an 8×8 angular grid across a 41° × 52° field of view, returning distance in millimeters. Effective range depends on target reflectance and ambient infrared, so it is best stated by condition:
- Manufacturer specification: up to 5000 mm against highly infrared-reflective targets under ideal conditions. SIXTY D has confirmed this in bench testing.
- Real-world indoor environments: objects at 2000 mm are reliably detected.
- Bright outdoor sunlight: solar infrared raises the noise floor and reliable detection may fall to approximately 1000 mm. This has limited practical effect on the measurement of interest, since near work largely occurs indoors and objects at arms length and closer will be resolved
A reported value of 0 indicates no return signal — no object was detected within range. It does not mean an object was measured at zero distance. For analysis, 0 should be interpreted as “far,” meaning beyond approximately 2000 mm. Treating zeros as literal zero-millimeter measurements biases computed viewing distances substantially toward near.
The VEET’s optical stack imposes a floor of approximately 10 mm; the sensor cannot report values between 0 and 10 mm. Readings at or near this floor generally indicate a fully occluded sensor — covered by a hand, hair, clothing, or stored in a case or pocket — rather than genuine viewing at that distance. Distinguish occlusion from near work before computing distance statistics.
Both bias in the same direction. Zeros read literally pull the mean toward near; occlusion readings at the 10 mm floor mistaken for genuine near viewing also pull toward near. A researcher averaging raw TOF values without handling either will produce viewing distances systematically shorter than reality. Handling both correctly is the single most important step in VEET distance analysis.
The time-of-flight sensor returns a confidence value alongside each distance measurement, and that value should be used to decide whether a reading is trustworthy. Critically, there is no single confidence threshold that works across the whole range. Returned signal strength falls off with distance, so a fixed cutoff would either discard valid far readings or admit noisy near ones.
Project VEET therefore defines the minimum acceptable confidence as a function of measured distance. The reference implementation is veet_tof_bands.py, which divides the range into eight bands, each with its own minimum raw confidence:
| Band | Range (mm) | Recommended min. raw confidence* | Anchor |
|---|---|---|---|
| artifact_occlusion | 0–100 | 243 | Sub-100 mm, treat as occlusion |
| extreme_near | 100–200 | 230 | IMI <20 cm risk threshold |
| close_near | 200–300 | 205 | Sydney Myopia Study; Gajjar 2022 |
| standard_near | 300–400 | 154 | Clinical near reference, 33/40 cm |
| near_intermediate | 400–600 | 103 | Upper edge of Ostrin near band |
| intermediate | 600–1000 | 77 | Ostrin intermediate band |
| room | 1000–2000 | 64 | — |
| far | 2000+ | 52 | Effective optical infinity |
*These values are subject to further refinement through ongoing lab measurements
Two things follow. Readings near the device require high confidence to be accepted, because a genuine object at that distance returns a strong signal and a weak return is more likely to be an artifact. Readings at distance are accepted at lower confidence, because weak return is expected there rather than suspicious.
The band boundaries are not arbitrary. They are anchored to distances that carry meaning in the myopia and visual behavior literature, so that binned results map onto quantities other studies already report rather than to an instrument-specific scale.
Note that the lowest band treats everything below 100 mm as occlusion artifact, a wider exclusion than the sensor’s approximately 10 mm optical floor alone would suggest. Sustained viewing closer than 100 mm is rare enough that a hand, hair, or clothing over the aperture is the more likely explanation.
Where do the band boundaries come from?
| Boundary | Basis |
|---|---|
| 100 mm | Device-level. Below this, occlusion is far more likely than genuine viewing. |
| 200 mm | The <20 cm close-reading risk threshold. The IMI Clinical Management Guidelines Report identifies reading at very close distances (<20 cm) and for continuous periods (>45 min) as significantly associated with myopia development and progression, rather than total near-work time. Objective support comes from Clouclip-based work in Chinese children, where time at a working distance <20 cm was an independent risk factor for myopia. |
| 300 mm | The <30 cm close-reading threshold. In the Sydney Myopia Study, close reading distance (<30 cm) and continuous reading (>30 min) independently increased the odds of myopia in 12-year-old Australian schoolchildren, with the association holding after adjustment for age, sex, ethnicity, and school type. Gajjar & Ostrin’s 2022 systematic review reaches the same conclusion across the wider literature: working distances <30 cm and continuous near work >30 min are risk factors for onset and progression. |
| 400 mm | Standard clinical near-testing references of 33 cm and 40 cm (3.00 D and 2.50 D accommodative demand). |
| 600 mm | Upper edge of the near band used in Ostrin’s Clouclip work, where near viewing is defined as 10 cm to <60 cm. |
| 1000 mm | Upper edge of the same intermediate band, defined as 60 cm to <100 cm. |
| 2000 mm | Effective optical infinity for this application, and the practical limit of reliable indoor detection. |
Primary sources
- Gifford KL, et al. IMI – Clinical Management Guidelines Report. Invest Ophthalmol Vis Sci 2019;60(3):M184–M203.
- Huang H-M, et al. Objectively measured near work, outdoor exposure and myopia in children. Br J Ophthalmol 2020. PMID 32075819.
- Ip JM, Saw S-M, Rose KA, Morgan IG, Kifley A, Wang JJ, Mitchell P. Role of near work in myopia: findings in a sample of Australian school children. Invest Ophthalmol Vis Sci 2008;49(7):2903–2910. doi:10.1167/iovs.07-0804
- Gajjar S, Ostrin LA. A systematic review of near work and myopia: measurement, relationships, mechanisms and clinical corollaries. Acta Ophthalmol 2022. doi:10.1111/aos.15043
- Bhandari KR, Ostrin LA. Validation of the Clouclip and utility in measuring viewing distance in adults. Ophthalmic Physiol Opt 2020;40:801–814. doi:10.1111/opo.12735
- Ostrin LA, et al. Wearable sensors for measurement of viewing behavior, light exposure, and sleep. Sensors 2021;21(21):7096. doi:10.3390/s21217096
A whitepaper documenting the confidence model in full is in preparation.
The time-of-flight sensor can report up to two detected returns per cell of the 8×8 grid, recorded as Object 1 and Object 2. This is among the most frequently misinterpreted features of VEET data, and more than one treatment is defensible. What matters is choosing an approach deliberately and reporting it, rather than assuming the second return is either redundant or automatically meaningful.
SIXTY D recommends researchers to take one of two approaches.
Approach 1 — Object 1 only. Analyze Object 1 and discard Object 2 entirely. This is simpler, easier to describe in methods, and adequate for many study designs.
Approach 2 — Object 1 with Object 2 fallback. Use the second return only where the first is unusable, applying the following logic per cell:
- If Object 1 has high confidence, report Object 1 as the distance for that cell. No Object 2 can meaningfully exist in the same region, so the Object 2 value for that cell is ignored.
- If Object 1 reports 0 or falls below the confidence threshold for its distance band, check Object 2. If Object 2 has viable confidence, report Object 2.
- If Object 2 also reports 0 or low confidence, interpret the cell as far, meaning beyond approximately 2000 mm.
Confidence in both steps should be assessed against the distance-dependent thresholds described in the previous answer, not a single global cutoff.
Whichever approach is used, state it explicitly in methods. Two studies applying different treatments to the same raw data will produce different distance distributions, and without a stated convention the difference is invisible to anyone reading the results.
A whitepaper documenting the TOF interpretation in full is in preparation.
Yes, under specific conditions. Each VEET time-of-flight sensor emits its own infrared VCSEL flashes, and each device runs on an independent clock with its own sampling phase. Devices stored close together, or several devices aimed at the same near target — as in a validation session, calibration setup, or shared charging tray — may register one another’s emissions and return spurious distance values. This is not a concern in normal single-participant field use, where devices are separated in space. When testing multiple VEETs simultaneously, separate them physically or offset their measurements in time, and treat distance data from multi-device settings as requiring verification.
Potentially. Infrared face-detection systems in smartphones and computers emit in a band that overlaps the VEET time-of-flight sensor’s operating wavelength, and may produce spurious readings when active. Rigorous characterization of this effect has not been completed as of this writing, so its magnitude and frequency are not yet quantified. Researchers whose protocols involve substantial device use should be aware of the possibility. SIXTY D welcomes contact from groups observing anomalous near-distance readings during screen use, via https://projectveet.com/contact-us/.
FAQ Section 4: Data
The VEET records time-series data as non-proprietary comma-separated value (.csv) files, alongside log files recording device state and events. The VEET 2.1 Data Interpretation Guide, Log Interpretation Guide, and Data Processing Guide on the Sensors & Data page document every field.
Connect the temple arm to a Mac or PC over USB-C and it mounts as an ordinary disk drive. No account, login, or software is required to read the files. VEETManager is used for configuration and sensor preview, not for data retrieval.
Each temple arm has a minimum of 16 GB of onboard flash storage, holding months of logged data. For devices in continuous use, unload data weekly to keep individual file sizes manageable.
Each temple arm has a built-in 350 mAh rechargeable lithium-polymer battery supporting more than 24 hours of continuous data logging on a single charge. It charges over standard 5V USB at a maximum draw of 175 mA and reaches full charge in under 2 hours on wall power.
| State | Voltage |
|---|---|
| Fully charged | 4200 mV |
| Connected to wall power | 4080–4200 mV until disconnected |
| Deep Sleep threshold | 3500 mV |
The VEET charges to 4200 mV and then holds between 4080 and 4200 mV for as long as it remains connected to wall power. Full charge typically takes about 2 hours, and up to 3 hours in adverse conditions such as high ambient temperature. Battery voltage is recorded in the device log files, so it can be reviewed after a session.
Deep Sleep preserves the clock for more than a month, so a device that enters it resumes with correct time once charged. This is different from full depletion, where the clock is lost and resets to the sentinel value 1325376003, invalidating absolute time for data recorded afterward until the device is reconnected and resynchronized.
Below 3500 mV the VEET enters Deep Sleep. Logging stops, and the device remains in this protective state until it is recharged.
Recharging before a device fully depletes therefore preserves timestamp integrity. Verify the clock on the next connection for any device that has been left uncharged for an extended period.
The VEET clock synchronizes with the host computer when connected over USB. Time may differ between the two temple arms by several seconds, which matters when aligning bilateral data. The clock is maintained for more than a month after the device enters low-power mode.
Timestamps are recorded in Unix time (seconds since 1 January 1970 UTC). https://www.epochconverter.com/ is a convenient tool for converting to human-readable time when inspecting raw files.
Time is lost and the clock resets to 1325376003, corresponding to 1 January 2012 00:00:03 UTC. This value is deliberately nonsensical: no real VEET data was ever collected then, so its presence is an unambiguous marker that the device lost power and has not been resynchronized since.
Treat this as a data integrity check. Screen every dataset for timestamps at or near 1325376003 before analysis. Records carrying it are correctly ordered relative to one another but have no valid absolute time, so they cannot be aligned to time of day, to photoperiod, to the opposite temple arm, or to any other data source. Reconnecting the device to the VEETManager resynchronizes the clock, but does not recover correct timestamps for data already recorded in the reset state.
Not in practice. Current VEET devices will reach end of life well before 32-bit Unix timestamps overflow in January 2038.
Project VEET publishes a Python-based VEET Data Charting Tool with an interactive interface, a Validate VEET Python script for file analysis and integrity checking, and a Python SPD reconstruction script with a simplified Excel equivalent. All are free to download from the Sensors & Data page.
For researchers working in R, LightLogR supports VEET data output directly and provides import, gap detection, visualization, and 62 light-exposure metrics across 17 metric families. It is developed by the Translational Sensory & Circadian Neuroscience Unit (MPS/TUM/TUMCREATE), available on CRAN, and published in the Journal of Open Source Software. Documentation is at https://tscnlab.github.io/LightLogR/ and source at https://github.com/tscnlab/LightLogR.
The researcher and their institution own the data collected with their VEET. Project VEET does not receive, access, or claim rights to study data. Researchers are encouraged, not required, to publish datasets under FAIR principles.
FAQ Section 5: Fit, Wear, and Study Design
The VEET is compatible with commercially available glasses frames and ships with four hinge designs to accommodate different frame types. Installing a VEET temple arm requires one screw. Exchangeable temple arm extensions cover lengths typical of standard glasses, roughly 125 mm to 145 mm.
Contact SIXTY D at https://projectveet.com/contact-us/ for the current list of compatible frames. The commercial eyeglass market changes constantly as models are discontinued and introduced, so the list is maintained and updated rather than fixed. Confirm frame compatibility before purchasing frames in quantity for a study.
A VEET pair adds approximately 40 grams. The VEET temple arms replace the frame’s existing arms rather than mounting alongside them, so true added weight is slightly less than 40 grams once the removed arms are accounted for; approximately 40 grams is a reasonable planning figure. Each arm is about 20 grams and measures 2.0 cm high by 1.4 cm wide by 8.0 cm deep, excluding the extension. The frame front and lenses are unmodified.
There is no manufacturer-specified minimum age, and suitability should be evaluated by each institution’s IRB in the context of the specific protocol. Studies have successfully deployed the VEET with participants as young as 7. Success at younger ages depends heavily on proper fitting and may require limiting daily wear duration.
The practical constraint is weight. Approximately 40 grams of added mass proves too much for many younger participants to tolerate over a full day. Researchers planning pediatric studies should account for this in wear-protocol design and compliance expectations, and should consider shorter wear windows rather than assuming full-day tolerance. The Optician’s Guide documents fitting procedure.
Yes, and some researchers have done so deliberately. A single arm halves the added weight to approximately 20 grams, and participants have shown higher tolerance to the resulting asymmetry than to the full bilateral load — a worthwhile trade in pediatric protocols or extended-wear studies where compliance is the limiting factor.
The choice of side is consequential. The right arm’s sensors are oriented 20 degrees downward and 4 degrees inward, approximating near-work gaze; the left arm’s sit perpendicular to the frame. A single-arm study therefore measures a different sampling geometry depending on the side selected, and its data is not directly comparable to the opposite arm in bilateral datasets. Researchers should specify and report which arm was used, and select it to match the study’s primary measurement question — the right arm for near-work distance, the left for a frame-referenced view.
No. Frames can be fitted with plano (non-corrective) lenses, or used with no lenses at all, for participants who do not require correction. The VEET measures independently of the lenses, so corrected and uncorrected participants can be enrolled in the same study.
Key facts for an IRB submission: the VEET has no camera, no microphone, and no Bluetooth or Wi-Fi radio; it records no personally identifying information; all data remains on the device until the researcher retrieves it over USB; no account or cloud service is involved; and the device is non-invasive, worn externally on spectacles, with no external buttons a participant could use to interfere with logging. Added weight of approximately 40 grams for a pair should be disclosed as part of the burden assessment, particularly in pediatric protocols.
Yes. The VEET is designed for the rigors of daily wear and each device can be reused by different participants across studies. It can be cleaned with typical optical cleaning solutions and isopropyl alcohol wipes.
In non-charging use the VEET operates from -10°C to 37°C (98°F) at up to 95% relative humidity, non-condensing. When charging or connected to a computer, ambient temperature should stay below 27°C (80°F).
The standard VEET is built to be water resistant, but no formal IP rating was obtained for it, so it should not be treated as rated for immersion.
SIXTY D has manufactured IP67-rated VEET units for specific studies. If your protocol requires a formally rated device — for example, research involving swimming, water exposure, or humid outdoor environments — contact SIXTY D at https://projectveet.com/contact-us/ to discuss it.
Non-impact physical activity is fine, and the VEET is designed for the rigors of daily wear. SIXTY D recommends the VEET not be worn during watersports or impact sports.
This matters for study design in myopia research, where outdoor activity is often the primary exposure of interest. Protocols should anticipate that some outdoor time will go unmeasured if participants remove the device for sport, and account for that gap rather than treating recorded outdoor time as complete.
There is no indicator provided to the wearer. The VEET is always on, and collects data whenever it has sufficient battery charge and is not connected to a computer. It continues logging while charging from wall power; connecting it to a PC is what stops collection.
For a researcher or curious participant who wants to confirm the device is active, the time-of-flight sensor’s infrared flashes are visible through most cell phone cameras, which are sensitive to near-infrared. Point a phone camera at the sensor aperture and the flashes appear as faint pulsing light not visible to the naked eye.
Because there is no wear-time feedback, compliance cannot be confirmed by the participant in the moment. Verify data completeness at each download rather than relying on self-report.
VEETManager runs on Windows 10 or 11, or macOS Sonoma (v14) or later, and requires a USB 2.0 port. It provides a graphical interface for setting logging intervals and previewing live sensor data.
FAQ Section 6: Section 6 — Privacy
The VEET has no camera, no microphone, no Bluetooth, and no Wi-Fi, and records no personally identifying information. It logs only illuminance, spectral channel values, optical distance to near objects, and motion — none of which can reconstruct an image, capture speech, or identify an individual. Data remains on the instrument until the researcher retrieves it over a wired USB connection.
No. The VEET has no GPS, no camera, and no network connection. Its depth sensor returns distance values within a 41° × 52° cone but cannot resolve or identify objects, and its light sensors record intensity and spectrum but not scene content.
No. The VEET has no wireless connectivity and transmits nothing. Data is written to onboard flash and retrieved by the researcher over USB, with no account required. SIXTY D receives no study data unless a researcher chooses to share it.
FAQ Section 7: Support and Troubleshooting
Begin with the VEET 2.1 Troubleshooting Guide on this page, which lists common error conditions and resolutions. For issues not covered there, the VEET firmware can be accessed over a virtual serial port for diagnostics, documented in the VEET 2.1 Using the Command-Line Interface guide. If neither resolves it, contact SIXTY D at https://projectveet.com/contact-us/.
The VEET exposes a virtual serial port interface for troubleshooting and bulk operations. Connection procedure and available commands are documented in the VEET 2.1 Using the Command-Line Interface guide on this page.
VEETManager checks the device firmware version every time a VEET is connected. If a newer version is available, VEETManager prompts to update. No manual version tracking is required.
Because updating requires connecting to a computer, consider checking firmware at the start of a study rather than mid-collection, so that all devices in a cohort run the same version throughout.
Researchers should use the Project VEET contact form at https://projectveet.com/contact-us/. Study participants and parents should contact their study’s researcher directly rather than SIXTY D, as SIXTY D holds no participant records.
Yes. SIXTY D has committed to supporting the VEET device and user community through at least 2029, with continuity of technical leadership under David Sullivan, co-inventor of the VEET platform. Device support for researchers under Project VEET is provided at no cost, and is requested through https://projectveet.com/contact-us/.
FAQ Section 8: Getting a VEET
There are two routes. SIXTY D can provide a limited number of evaluation units to researchers new to the VEET, though current stock is quite low. Devices can also be purchased directly — see the next answer. Either way, start at https://projectveet.com/contact-us/ and describe your planned study, so that availability can be matched to your timeline.
For studies aimed at starting in 2027 and beyond, we recommend reaching out to discuss the next generation VEET device
Yes. Contact SIXTY D at https://projectveet.com/contact-us/ to discuss purchasing.
Quantities of the VEET 2 device are very limited. SIXTY D is also actively pursuing the next-generation device, and welcomes hearing from research groups who are interested — planned studies and stated requirements inform what gets built. Groups with upcoming protocols are encouraged to make contact early, whether or not they are ready to order.
VEET 1 and VEET 2 share identical sensor hardware, and SIXTY D supports existing VEET hardware in the field, although at this time all known VEET 1 hardware has been removed from the field. For questions about data comparability across VEET generations, contact SIXTY D at https://projectveet.com/contact-us/.
SIXTY D intends on developing the next-generation device, and welcomes hearing from research groups who are interested — planned studies and stated requirements inform what gets built. Please reach out if you may be interested in this! https://projectveet.com/contact-us/
FAQ Section 9: Scope and Limitations
No. The VEET does not assess myopia risk and does not prevent, slow, or treat myopia. It is not cleared or approved for clinical or medical use and is not intended for measuring patient biometrics for any medical purpose. It is exclusively a research instrument for collecting data on visual experience.
No. The VEET is a non-medical research instrument with no FDA clearance or approval. For questions about regulatory status in your region, contact SIXTY D at https://projectveet.com/contact-us/.
The principal known limitations are documented in detail above and summarized here.
Distance (TOF)
- Effective range varies with target reflectance and ambient infrared, falling to roughly 1000 mm in bright sunlight.
- A reading of 0 means no detection, not zero distance.
- The sensor cannot report below approximately 10 mm; readings at that floor usually indicate occlusion rather than genuine near viewing.
- Multiple VEETs in close proximity may interfere with one another through overlapping infrared emission.
- Infrared face-detection systems in phones and computers may cause spurious readings. This effect is not yet characterized.
Light
- The light sensors have a deliberately narrower field of view than cosine-corrected meters, so values are not directly comparable to hemispherical instruments.
- Illuminance accuracy below 0.1 lux is not well characterized.
- Below approximately 10 lux the spectral sensor produces very low counts under the stock firmware configuration; spectral data and anything derived from it, including mEDI, should be inspected before use. This is a deliberate trade against daylight saturation and can be reconfigured in firmware for low-light studies, but the two ends cannot be optimized simultaneously.
Timing
- Inter-arm clock offset may reach several seconds, affecting bilateral alignment.
- Full battery depletion resets the clock to a sentinel timestamp, invalidating absolute time for affected records until the device is resynchronized.
Wear
- Approximately 40 grams of added weight limits tolerable wear duration in younger participants.
FAQ Section 10: Background- Myopia
Myopia, or nearsightedness, occurs when the eyeball grows too long, causing light from distant objects to focus in front of the retina rather than on it. It affects approximately 1.5 billion people worldwide and raises lifetime risk of permanent vision loss from retinal detachment and macular degeneration.
Myopia results from a failure of the eye’s growth regulation, and the mechanism is not fully understood. Genetics contributes, but rising prevalence points to environmental drivers — reduced time outdoors, lower ambient light exposure, and sustained near work. Because ocular development continues until roughly age 20–21, establishing which factors matter requires longitudinal observational study, which is what the VEET is built to support.
Subjective assessment, typically parent-completed surveys, correlates poorly with objective measurement of both screen use and time outdoors. Questionnaires also cannot capture illuminance, spectral composition, or viewing distance in any quantitative form. The VEET replaces recall with continuous instrument measurement.
Individual visual-environment studies have historically used incompatible instruments and undocumented data formats, making aggregation across cohorts impossible. Project VEET publishes a common instrument, common data formats, and the metadata needed to interpret them, so datasets from different groups can be compared, harmonized, and analyzed jointly under FAIR principles.
Contact Us
Researchers:
Use this form to inquire about the VEET project or get support for issues with your VEET.
VEET Wearers:
Please contact your researcher directly for questions or support.