I am interested in others results, my test setup and results are below:
Test Setup:
(1) Note the Wyze app version (under Account>About) and FW version (under Account>FW Update) [e.g. IOS App version 2.38.4 (3), FW 4.58.0.3210]
(2) Using livestream 2K setting, move your vehicle (with license plate aiming toward the cam) both toward the cam and away from the cam until you can read the letters/numbers on the license plate (while zoomed in to see), then park the vehicle
(3) Measure the distance in feet from the base of the cam (on the ground) to the license plate and record the distance
(4) In the 2K livestream setting, select the Wyze “Take Photo” vs phone screenshot and then go to your photo library and look at the info for the photo and provide the file info (e.g. 3MP 2560x1440 2.7 MB Jan 26, 2023, and file type and size if available)
(5) Change the livestream setting from 2K to SD and try to read the license plate from the same distance – SD should be blurry. Again, take photo as in step (4) for comparison purposes
(6) Photos optional to include in the thread
Test Results:
IOS App version 2.38.4 (3), FW 4.58.0.3210
42 feet max for 2K setting to read license plate (overcast); 3MP 2560x1440 3.0 MB Jan 26, 2023, 2:43 pm; jpg 861 KB
blurry (cannot read) on SD setting; 3MP 2560x1440 2.6 MB Jan 26,2023, 2:41 pm; jpg 776 KB
Note: Other forum topics for the Wyze cam v3 Pro, that are related, challenge whether the cam is really streaming/recording at 2K or less; why the 2K setting is not significantly better in resolution than the SD setting; and what factors determine image quality. Also, previously the v3 Pro was not streaming in 2K with the IOS app (not an issue with Android) and Wyze updated the IOS app which seems to fix the 2K streaming issue, but did the fix reduce the bit rate (which contributes to image quality) for the 2K setting to make it work? More data, like what I am asking, may shine some light on the above.
Update: I added some pictures to support the test results: - 1st pic is from the Wyze “Take Photo” option in livestream - stationary vehicle is 42 feet from the cam - 2nd pic is a zoom of the 1st pic to show the license plate - this is an iPhone screenshot while zoomed
Oh, I totally understand. I am sure this thread will be referenced by others in the future, so I thought some people might want to see a video example of a license plate at various distances since other people will be drawn to your thread because you are asking some great questions and some of them will be interested in this related test from lifehackster.
As for your question on the V3Pro resolution, it is definitely in 2K, this has been verified in numerous ways, including using digital analysis. I think the main issue with the quality concerns has to do with the amount of compression causing pincushion distortion, and there are smaller issues related to artifacts, and the wide field of view without post-processing measures to fix it. Most people won’t really notice it unless using specific testing materials like @TheNetGuyDotCom did in his awesome tests and review on the camera.
Anyway, I look forward to seeing the results from anyone else who also takes the time to replicate the test setup and post their results. I am going to follow this thread and see what else others find too.
Liked “thenetguy” review of the v3Pro! Regarding compression, since my tests only addresses a non-moving object (license plate), the compression/artifacts should be minimal?
I had a V2 recording of a car license plate about 20 feet away and the readability was like 50% because of artifacts. With OG it’s more like 90%.
Is there anyone who has both V3 pro and OG? I’d like to know if their license plate readability is comparable. If so I’ll probably stick with OG when I upgrade.
Hi,
If you are trying to capture license plates of moving vehicles on the street OG-Tele might be better bet. I have this setup and it’s approx 40 feet from camera to cars. Also you can capture in daylight, but not near sunset or at night.
I’ll see if @TheNetGuyDotCom will grace us with an opinion on that. He understands all that a lot better than me, and he’s done a lot more testing on it, so I’d like to hear what he has to say related to your question.
Thanks. I’m just trying to capture stationary cars parked in the driveway so the OG is sufficient for that for now. I think I’ll run a test during night time as well and see if the license plate is readable.
I am however, interested in the telephoto for facial recognition, to use it to supplement my floodlight with V3. Would telephoto’s night vision or starlight sensor be sufficient for facial recognition during night time? Especially with the floodlight on I would think there’s enough light.
OK I did a test last night and OG is rather horrendous when it comes to capturing license plate during night time. Given Telephoto also cannot handle license plate during night time, I doubt V3 Pro can since I read its night vision is not as good as V3. If someone parks in my driveway in the middle of the night, I would like to capture the license plate. Has anyone else done any license plate tests on other Cams?
I have not done any v3 Pro tests at night, just during the day. Reading a stationary license plate at 42 feet meets my goal. I have very bright motion detector lights in my driveway (not using spotlight on v3 Pro) that should provide enough light for seeing details at night. I guess that others are not very interested performing the test that I did and supplying their own data so that the forum can confirm what the max distance that the Wyze cam v3 Pro can see a license plate in the 2K setting?
I also have motion lights mounted above my garage and front porch, and Christmas lights etc. around the yard right now, but the lights actually made the license plate overexposed on the OG so all I see is a white rectangle.
I think not everyone has a long driveway that would allow for this test. In my case my camera is mounted inside the front porch, and since my driveway goes up to the street, it doesn’t catch the whole driveway anyway. So the max I can go is 20-30 feet.
The resolution of the Wyze cam v3 Pro is 2560x1440p. The resolution of the cam v3 and OG (and OG Tel) is 1920x1080p. Therefore, the image quality (which is made up of resolution, frame rate, bit rate, and compression) will be different among these cams with the v3 Pro being 2K resolution and the others being only Full HD. So one would expect the max distance to see a license plate would be less for the Full HD cams.
Unless Wyze provides users with the ability to adjust frame rate and shutter speed, you’re not going to have much luck doing nighttime LPR with a v3 Pro or an OG Telephoto. Nighttime LPR requires very fast shutter speeds to eliminate glare and motion blur. With the proper setting the plate can be read, but all you will see is the license plate and the (dim) headlights / taillights of the vehicle. LPR also requires sufficient illumination (infrared or visible light) reflected from the plate to make it visible to the camera sensor.
I’ve got an OG Telephoto arriving today, and I plan to do some daytime LPR tests to see how it works. A daytime LPR camera is certainly better than no LPR camera at all, and will at least help catch some porch pirates, mail box thieves, and the like. I’ve already got my own LPR cameras, so I don’t need it myself, but I’m hoping the OG Telephoto will be something I can recommend to neighbors.
I am going to follow this thread and see what else others find too.