Faster recording

[AskMcGowan]

This is as it states in the subject,

Based on my test experience there is a lag between motion detection and recording start. 

So according to tests (and this is one example)

There is a street in the camera view. A car drives by at about 15mph - Camera detects it at the beginning but by the time it records I only see the back of the vehicle. So there is a 2-3 second lag.

I would like that lag to turn to 0-1 seconds.

[paintguy]

I agree. This is the second biggest issue besides the battery one.

[StuBee]

My two-cents.....

Sure we would all want instant-one..... but I would think there is a current technical limit that is between motion detected to be of such a magnitude to kick off a recording...and the recording actually starting.   Remember it's not an always on type of recording so there has to be lag of some size.

I try to keep in mind the intended purpose of the device is as well.  I don't think a car driving by your house is an intended target of a recording.  Yet someone trying to get into your door is, which probably lasts more than 1-2 seconds.

To help somewhat in the 1 sec lag, I position the cameras to capture a bit more than just the front of the door.  So the person approaching the door within at least 7 feet are in the motion zone.

[RRUGA]

I agree with the original poster. We have one camera that is in the driveway/backdoor area. I can walk out the door and get into the car by the time the camera starts recording.I know there is a fine line between performance and battery life that the engineers have to consider, but cutting this lag time down could be of immensive value. Even better is the fact that it can more than likely be addressed via software updates for this generations equipment.

[beachbumdb]

I agree the lag time is a problem if you want to catch potential problems on video.

You really do miss getting what you need on the video.  

I would be really disappointed to not get video of what is needed to give to the police.

I would defeat the purpose of this as a security camera system.

Just using the system, i find that i miss getting the license number of a vehicle entering our driveway due to a longer lag time.

I dont expect instant on, but if motion could trigger it in just a 2-3 second time that would help.

[retrorobot]

Completely agree. The activity I'm trying to capture involves a moving vehicle, and often the camera only gets the back end of passing cars due to the lag, which means I miss the action. Although I have the sensitivity turned up to 100%, the capture still starts late and sometimes there is no clear motion in the clip indicating the car had already passed by before the recording started. This lag time needs to be improved. 

[R_Lowe]

This would put Arlo in the same class as my Dropcam.

[StuBee]

But this isn't the same as dropcam.  Arlo only starts recording when a threshold breach is detected.  The treshold is breached when a significant movement has occurred (depending on what you set the % at).  It cannot go back in time and provide data that occurred prior to the arlo decision that significant movement had occurred.

Meanwhile devices that are ALWAYS recording (like dropcam) can go back a few seconds prior to the motion occuring and provide that in the sequence.  Arlo would never be able to do this (for good reason)

So it comes down to any lag that may exist between "Action Threshold detected" and "Start Recording"  I have found that to be 1-3 seconds, which is ok for me.

====
So regarding the cases above....when do you need to catch a license plate of a moving vehicle?  When you do catch one can you actually read it considering the quality of the video and the distance?

Can the Camera be position so that a Greater movement occurs in the middle of the camera detection and starts the recording earlier?

If someone was attempting to break-in to something, wouldn't the activity be more than 1-2 secs that you maybe missing?

[erikprem]

I am trying to use one camera, set closer to the top of my driveway, to trigger the other camera at my door.  So far, NSH using the rules but I'll keep on trying to get it right.

[WadeShuler]

I put "from one dev to another" because I think this message should be passed to the dev team.

I think there should be a public change log so the public can see a history of updates/fixes, see what day they were implemented, and what they took care of. There should also be a public page showing what features/issues/projects the dev team is currently working on.

I am a programmer and web developer. I use git (via GitHub, and git locally). You can create issues, solve them, etc. You have changelogs for a release history to see what updates took place when, and what they addressed. While I am not saying to use GitHub, we just need to see a change log to see when updates are rolled out and the issues they addressed. Also, it would be nice, if the community could see what the team is currently working on. I am sure they are working on stuff, but the community probably would like to know they spent $300+ and whether or not a feature they need/want is currently being taken seriously.

All we have is a bunch of people posting on the forums, requesting features. Does Netgear (Arlo) even hear them? Is there a system in place for putting this requests in, to get reviewed and addressed? What I mentioned above, would take care of that. I am optimistic and have faith in a large established company like Netgear, that they are working on it. However, a public display showing people a list of past updates, and current progress, would prove to the community what is going on behind the scenes a bit. It would allow them to stop posting 100s of times requesting a feature, because they know it has been heard and is currently being whipped up by the team. If I knew the team already is in the progress of addressing the featues I laid out in this post, I wouldn't have even bothered writing this.

This system could dominate the market, but it isn't there yet due to lack of features. To be honest, the look, feel, and lack of features, makes this system look like it is still in beta stages. Did Netgear jump the gun and release this to the public too early? You can pick up a Dropcam for $199 that has zones and many other features that the Arlo system doesn't have. Since I paid $200 for the base and 1 camera, then another $160 for an additional camera (the 2 camera package wasn't in stock anywhere, it was 1 camera or I think 4), I would say this makes Dropcam direct competition to the Arlo system. There also are other respectable namebrand systems, like Westen Digitial offers one for I think $200 also. There is competition that is, and should, be used as a baseline as what to expect from this system.

While I am sure everyone has features they "think" should be added, not every feature request is the same. One person might want to be able to change the hue of the video so they can make it blue or red and look cool, or to balance out some weird colored light in their house. In reality, 99% of people have no need to change the hue of the video, so that feature, understandably, would probably never get added. I wouldn't waste my time on it either. However, the brightness feature could stand to go brighter, and maybe to adjust the white balance. 🙂

From a business standpoint, implementing a feature costs money and time. If your feature is only needed or wanted by a small percent of the customer base, it doesn't make sense to waste time and money on it. Why add a hue feature, if only 1 out of 100k people want it? Just for one person? No way.

Then you have features that are absolutely necessary. Those features are driven by competition and customer demand. When all of the competition has a feature, it should be expected that this product has it too. When a large percent of the customer base requests a feature (no matter if it's a need or want), it should be expected that it is implemented. Of course, needs come before wants. There is priority, between bug fixes and additional features. Bugs affecting majority of the customers comes before a feature. ie: motion tracking died on over 50% of the cameras becomes top #1 priority.

I am mentioning that because I understand business and I understand the development and programming side. Also, maybe it will curb some of the feature requests that nobody but that person even wants. However, a publicly viewable ticket type system for requests would be a lot better than a forum. A moderator to approve a request if it isn't already requested, and to deny the request if it's a duplicate. Then let the community vote it up to get a feel of how many people think it's important. The current forum system, just doesn't work.

-------------------------------------------------

With all that said, here is the list of top features needing to be added soon!

I have lurked the forums a bit to see where the customers are at. That, combined with my own usage of the cameras for a few weeks, helped me put together a starting list of needs for this system. These are based on what the competition has, or what seems to be the most demanded features.

• Motion Zones - A zones feature is my biggest issue with this system! In my opinion, it should be top priority. You should be able to add more than 1 zone because just a square/rectangle box doesn't always cut it. Adding 2 or 3 zone boxes, would allow us to focus on areas of importance, and to eliminate false alerts. I have an alley that sets off the cameras every time a car goes down it. I have repositioned the cameras trying to cut out the alley. It literally barely is on the very very top of the camera. All I can see is the bottom 1/3 of the tires.. and it goes off every time. Also, the zones shouldn't all be the same. I think the ability to create zone rules just like you can create other rules. I could create a "Super" rule of say 150% to help detect a person a few feet out of the range. At 30 feet I can't pick up someone at my front door, but it picks up cars in the alley at over 60+ feet!! Another zone rule of "High" at 100%, "Medium" at 80%, "Low" at 50%.. Just examples of zones one could create. Then on the camera, I could add a "Super" rule to my front door area. Then add the other zones where I see fit. I do think that one should be able to add multiple zones to 1 camera, and other systems I have looked at support adding many zone boxes to it. Then, I could have an interstate in the back ground, and not pick it up because it's not in one of my zone boxes. This would dramatically reduce false alerts. We don't all have a perfect spot to place the cameras.

• Motion lag - I have noticed it, many people have complained about it. When a person walks across the camera, you only catch them as they are leaving. There is a delay from the time the camera detects motion and the time it starts recording. It is about 3 seconds. Not every intruder, robber, bad guy, will casually walk in front of the camera, and a blurry image of him running away with your TV isn't going to catch him.

• Record full time motion (with max override setting) - If the camera rule is set to record for 60 seconds, but there is a bad guy doing something for 3 minutes, you will only record him for 60 seconds. Then, so long as he is still triggering the motion, it will kick back on and record another video for 60 seconds. It should continue to record, so long as there is still motion. If its set to 60 seconds, it should record for at least 60 seconds, respecting the user's rule (or an option they can choose to stop recording early if motion has stopped). Then at 60 seconds, if there is still motion going on, it should keep recording until there is no more motion. There should be a failsafe time that the user can set, say 5 minutes.. So it would go past their 60 seconds defined by the rule, but keep recording up to 5 minutes. Not every bad guy is going to do his thing in under 120 seconds (max allowed recording time currently), and combined with the motion lag, adds a lapse in time that you could snag a clear shot of his face or other capture other evidence.

• Alternative storage (Dropbox, Google, FTP, NAS) - with auto sync - While you provide a subscription option, not everyone wants, or can, pay for it. Storage is a dime a dozen in today's world. We have dropbox, network storage drives, and many cloud storage options (ie: Google and Microsoft both offer it, along with dropbox, and 100s of others). We should be able to have our videos sync with at least Dropbox and our own NAS (I have a 2TB WD MyCloud). That would give us the option of a cloud storage service, and our own storage system. FTP would be super simple to implement, so easy it shouldn't be excluded. Google and Microsoft cloud services would be icing on the cake for giving user's flexibility. I just dropped $300 on this system, I think you will be alright without my cloud subscription. I am sure I will add another camera or two in the future (if you add the needed features). Not providing options, for the sole reason of profit, is bad business. I am sure they are not available just because the system is immature, and hope you will add them soon. - We should be able to sync the videos (either every x hours, or before they get deleted after 7 days) automatically.

• Alarm - This is a home security system (according to the title of your arlo.com homepage). Home security systems have alarms. We should be able to add (to the rules) an alarm. I don't want an alarm on my driveway, but I do want one on my front door. The alarm should be a separate notification sound on our phones. You guys could make an alarm box that we mount on the wall, and the system would have a rule to sound that alarm. For example: You place an alarm box in an appropriate area in your house. Then you have your frontdoor and backdoor cameras set to sound the alarm. If someone broke in and triggered the camera, they would be recorded, and the alarm would go nuts. You could even go as far as to compete with ADT, have an operator come over the intercom and ask if they need to notify the police. Again, thats just future product ideas. However, a simple alarm feature that sounds an alarm on our phone is something that can be done with the current system.

• Custom Notification Tones (for each rule) - We should be able to select (or add our own) sound for each rule. My frontdoor could have a different tone so I know to check it out, while I could keep the driveway notification the default one. Right now, when I get notifications, I don't know if it's Arlo, which camera, if it's paypal, or any other app on my phone. They all sound the same. This is possible as other apps allow you to customise your notification sound for their app.

• Delete Trashed Videos - When we delete a video, it goes to the trash bin to live forever until it gets purged. This is our system, our houses, our lives, and our videos. We should be able to completely delete the video from your system. Adding a delete and purge/delete all on the trash bin would take 20 minutes for one of your programmers.

While I am sure I could come up with a handful more, this list is what I think should be top priority. I compiled this list from my own usage, what the competition offers, and what the customers here on the forums are complaining about the most. This list is extremely reasonable, and could/should be able to be added over the next couple of months. Some features would only take a few days to implement for the dev team (delete trash videos, add custom notification tones, add an alarm, alternate storage). I am sure the zones feature, and fixing the motion lag, would take a bit longer. Honestly, all this could be done, fixed, and pushed to consumers within 3 months... So we will see 😉

Feel free to add your own requests, or weigh in on how important you think the features I listed are. Contact support, give them this link, and make sure they have read this thread 🙂

PS: Please take note to how important a feature really is. This isn't a list of all and every feature, but features that SHOULD be have been available the day it hit the shelves, and what the competition already has. Let's keep this a targeted list of features based on importance and priority, that majority of the consumers would use and need.

PPS: To the devs, I am sure there are important bug fixes that probably are a priority. We understand. I just want to make sure the absolutely necessary feature requests have been heard, and get implemented. I like this system, but would love it if it can actually compete with the competition by having the features they offer at least.

Thank You!

[cnsheets]

New user. Have had a 4 camera system for less than a week now. Overall i love it. But like many others, the delay between the motion detection and when recording starts is frustrating. I receive a push alert that motion is detected, and when I look at the video, I either get the tail end of someone walking/driving away, or nothing. And every camera is fixed in a position to where the motion would be captured left to right or right to left across the camera view as suggested. 

Can the buffer of the video be increased so when motion is detected, the recording picks up at the beginning of the buffer.

So normally, motion detected at 9:10:25, the recording lag causes it to start recording at 9:20:33   Well now I missed the 8 seconds of the person walking up and only capture them walking away.

If it had like a 10 second buffer, motion detected at 9:10:25, with the buffer, the recording actually starts and picks up the buffer at 9:20:15, now I capture the motion either right at or slightly before the motion occurs and I dont miss anything.....

Please  please do something to increase the reliability and capture of motion detection!! I want to capture what triggered the motion, not after the motion already occured and missed.