Discrepancy Between 4x4 and 8x8 Resolution Distance Output

[JahKMC]

I am currently testing the VL53L7CX sensor(w/ vl53l5cx_api) with the STM32F7 MCU. The default resolution of 4x4, when running, would consistently return distances around 2.5 -3.0 meters with a target status code of 5 for multiple runs. 

I then tested 8x8 resolution and it began to return distances no higher than 1 meter with numerous error codes within the 64 zones target status (0,13,255 if I recall). Though, when I would hold my hand over the device it would return the ranges under a meter(this makes sense) with a target status 5 for all zones even when repeated. Please note that the room conditions of the 4x4 measurements and the 8x8 measurements were the same.  

Is there any reason for the readings to be present incorrectly when 8x8 resolution is used but 4x4 seems to work as expected? 

 

 

[John E KVAM]

In order to work, the sensor needs a number of photons returned from the target. Generally, we would like to have 20Million per second. But the sensor works very acceptably down to about 0.5M per second. At 2.5-3.0 meters you are getting to the limit of what you might expect. With a reasonably reflective wall (white) you can probably get to 3.5, but most normal paints are about 50% reflective, meaning you cannot range as far. 

In 4x4 mode we have 4X the number of photon detectors for each zone. And that is why it works to a longer distance.

I'll bet if you double check, the error status you get is 'low signal'. It just means the sensor did not get enough photons to get a result for that zone. 

the VL53L7 spreads the light to cover about 4 times the area that the VL53L5 or VL53L8 does. If you need to see longer you can switch chips, but you do give up the wider field of view. 

 

[JahKMC]

This makes sense.Thanks for the help. 

What you’re seeing is actually pretty common with the VL53L7CX when switching from 4×4 to 8×8.
At 8×8, each zone has a much smaller field of view and lower photon budget, so signal strength drops significantly. That makes the sensor more sensitive to ambient light, reflectivity, and distance, hence the shorter reliable range (~1 m) and mixed status codes (0, 13, 255). In 4×4, zones are effectively “merged,” giving stronger returns and more stable readings at longer distances.
Also worth checking your timing budget and integration time, 8×8 usually needs higher values to compensate.
This kind of resolution-vs-accuracy tradeoff reminds me of how image detail affects processing in tools like Paint 3D; higher resolution can introduce noise if not handled properly. If you’re exploring similar concepts visually, this guide might help: How to use paint 3d like a pro.