all device IDs for push notifications COULD be stored locally on the cameras, and pushed directly from there upon detections. that design wouldn’t require their server or cam plus services for motion or person detection (using the built in person ‘edge ai’). And since CamPlus Lite isn’t available, that’s another reason a server connection isn’t needed.
the site also says ‘insert a microSD card for 24/7 recording’. so, I wonder if it still detects/records shorter motion events on the microSD card. if so, do those events use the built in ‘edge ai’ logic as well, or is that just for CamPlus?
Of course they sure would like more CamPlus subscriptions. The AI saves them computing on cloud servers, but is that savings passed on to the users for those offline features? Either way CamPlus can still offer plenty other features. Hopefully they don’t lose focus on what makes a great SD only cameras too…
If someone doesn’t have Cam Plus, then yes, it saves on computing.
If someone has Cam Plus, then I don’t believe it saves on computing…the Cloud AI will still end up processing or verifying the entire video, I think it’s just that the local AI can immediately check frames in realtime without any latency and the Cloud AI can verify more thoroughly if that is correct as soon as those frames are uploaded. At least that is how I assume it happens since they said both work together when you have cam plus for improved recognition. In that sense it might use more computing overall since it does it locally, then verifies it all in the cloud again a second time.
I know the VP of product mentioned that they were working toward moving Wyze away from Cloud-only paradigms to have things work more locally. I am guessing that is a long term move and will take time to make such a drastic change…but I do think/hope they will eventually move things to be able to give notifications more than every 5 minutes. This camera is a good step in that direction by having some local AI processing. I’d love to see it further updated to be able to be more locally controlled and without cooldowns.
I can’t imagine that Wyze would move away from the Continuous \ Motion Activated recording to the SD that is already in place on all their wired cams w\SD capability. I can only assume that the SD is going to record the same way it does on their cams now. SD recording has never been linked to any subscription level in the past for these cams. I don’t see that changing.
So the field of view of the V3 Pro is 116-degrees, compared to the V3’s 130-degree.
The V3 Pro is appealing because I am using Person Detection to turn on my LIFX Flood Lights. The Trigger Time-to-Powering On the lights using a V3 is much better than the Outdoor Cam it replaced, but if I can improve upon that time, that would be great. But since it still has to pass through the cloud to IFTTT or the Alexa App, it might not make much of a difference.
Best way to improve on that is use a Sense motion sensor (V1 or V2) to sense the person. It will also sense animals, lawn mowers, and car engines, but positioning will help much of that.
For instance, mine is above my front door where a porch overhang blinds it to the street and people on the sidewalk out front, and I turn it about 30 degrees to blind it to the kids, lawnmowers, and pets next door.
It is so fast at responding to the people coming up my driveway that my signal light indoors goes on long before they can reach the doorbell.
Wait, what? They’re imposing a 5 minute cooldown on a local AI feature of hardware you’ve paid for? IBM wants its 1973 business model back.
Is it the same Person Detection you get with Cam Plus?
No, it is not the same Person Detection. It is not as powerful and does not include features like no cooldown, unlimited video, or extra verification included with Cam Plus and Cam Plus Pro.
While it has the PD Edge AI on the cam, there is no way for you to get a notifications or an event in the event tab without the server. The Edge AI is useless without it. Any non-subscription cam on the server is subject to the 5 min timeout cooldown just as it has always been.
With CamPlus, the Cam PD Edge AI and the Server PD AI will be used for greater accuracy.
They would have to be manufactured with multicolored adjustable LED’s, controller circuits, and software to do that. More expensive than single output.
The V3Pro is a brand new cam offering that Wyze just announced. No cams have shipped yet so no one has seen what the reported “Updated User Interface” has in it.
How is it that you are reporting it doesn’t have a feature when no one has seen it yet?
If you are talking about general access to the SD card, then I suspect it will work similar to the way it works now – you choose a date from a calendar, then you move a timeline to get to the time, expanding the timeline to get better resolution when you get close.
If you are talking about an event, then the event on the Event tab should take you exactly to the time of the recording on the card (press ‘Playback’ there).
If you are talking about the time stamp, you can hit ‘Sync’ under Advanced Settings.
But surely you see that is disingenuous and makes little logical sense. The “old”, current service model limits notifications because remotely analyzing and storing video sequences costs money. With “edge” AI there is no cloud AI processing or storage - the local CPU handles person “AI” and they say the video is recorded only on local SD. The only cost is a miniscule few bytes of network transit to route the notification back to your app. (This is dwarfed by the millions of hours of video they transmit and store for free containing unwanted non-person videos with trees waving because their old cameras could not do that AI pre-processing; the new model saves this waste.)
It’s of course their call and obviously in their business interest to cajole people to pay for CamPlus, but unfortunately they are apparently artificially and needlessly kicking their nifty new baby 2K "Edge AI !"camera in the teeth to do it. Ah well, still a nice new model. Anybody know why they didn’t just call it V4?
