Thanks for the clarification. I only use video when trying to capture behavior for a hunter education class, but never for data collection. It is not practical. A couple rapid stills have higher resolution and can capture all the data I need and require much less analysis time for data extraction and require much less storage.
Perhaps the reason you can't see it difficult to identify individual bucks is the short range PIR. The closer the subject is to the camera and the more light available, the less resolution that is required. (Keep in mind that there is a difference between what most companies advertise as resolution which is often interpolated and the native resolution of the image sensor used). With a camera with a long-range PIR, I often get triggers at 40 yards and sometimes even at 60. At low light and depending on position, it can even be difficult to distinguish small bucks from does.
As for flash avoidance, I have the data to show it statistically. This is one of those unknown unknowns until you look for it in the data. I first found this when black flash was just coming out. I found that with the same IR cameras, over bait, I would get repeated pictures of mature bucks, but over small fields, I'd get only one or two pictures of the buck and never again. Eventually, I started to dig up old pictures and scrub them to try to understand this. It turns out that if I was using bait, even mature bucks would accept the risk of the camera to get it. However in a small field, they learned not to trigger the camera. They did not spook or leave the area, they simply kept their distance and kept other deer between them and the perceived danger (flash). When I scrubbed old pictures, I found some of those bucks on the fringe of the flash when the camera was triggered by young deer. Next, I compared a BEC (wireless running unattended for months removing human scent as an issue) with Red Blob to the newer version with Black Flash. I eventually swapped the position of the cameras to remove that bias. I consistently got more repeat pictures of mature bucks using the Black Flash camera.
One more item of interest. When I looked at the data closely, I found mature bucks were closer to black flash cameras during the daytime as well. The physical form of the camera can be an issue when they are moved from place to place frequently, but for cameras in permanent positions, mature bucks seem to acclimate to the camera form but not the Red Blob flash. With a red blob flash, mature bucks seemed to remember and keep more distance from the camera even during the day. With the black flash version, they seem to ignore it day and night.
One note here, deer behavior has a lot of variation. And that variation may be greater from area to area. Here, there is a significant statistical difference between how deer react to red blob and true black flash.
One final note: When I say True Black Flash, I'm referring to a flash that is not detected by deer. Detection is a function of three factors, wavelength (frequency), intensity, and duration. Some cameras that advertise invisible flash are clearly visible and can be detected. At what point the level of visibility is ignored by mature bucks is not established. I would suspect that the less visible the more they ignore it but I don't know if it is linear or a step function.
Great discussion and keep reporting on your system! While you've already shared enough to make it clear it is not a fit for my application, it may be a great fit for others!
Thanks,
jack