Ask HN: IP cameras that don't require an app or internet?
The models I've been using have an ethernet port and wifi. Setup was connecting to the ethernet port and then accessing the web ui in a browser to configure settings (most importantly turning on RTSP or ONVIF feeds). The cameras I have are starting to show their age and a couple of them are starting to fail (PTZ slow or require reboots every few weeks).
I picked up newer models from Amcrest and foscam assuming they would have the same set up procedure (i made sure to get ethernet+wifi models and did research on being RSTP capable) but they all require downloading an app and creating an account to set them up, even if the end configuration is without internet for local video)
The foscam cameras have a web ui that just has links to the app stores and the amcrest cameras don't have any web ui available. I tried directly accessing the RTSP URLs and still no luck. both apps require account creation in order to use.
I've also tried some tp link, wyze and aqara cameras in the past but they all required an app/account. They also had the worst reliability, both in connection stability and physical failure rates.
Does anyone have specific model numbers of currently purchasable (US) IP cameras that genuinely don't require an app and account to set up?
203 comments
[ 3.3 ms ] story [ 231 ms ] threadTheir basic range all have a web GUI, RTSP and ethernet/wifi. Have a look at the M11 range, or the M30 for a cheap dome camera.
I heard they are the go to brand for any goverment organisation that cares about security and privacy.
Check out the UniFi subreddit for details.
Very disappointed, as that was my go to brand for standalone IP cameras.
It's just impossible for a business not to screw customers, isn't it?
There is a max number of cameras though.
You could spin up your own ShinobiCCTV if NVRs are your problem.
I can only conclude that they must be selling access to feeds, because even the most incompetent agile product delivery manager type person isn’t going to simply suggest every camera needs an account along with all the extra engineering requirements involved for that for simply no reason - there surely must be a financial motive.
Furthermore what good is the feed if they can’t also sell the associated metadata, such as the account holders details, nearby WiFi access points, the list of devices on the network.
This is yet another angle they can use to try track every aspect of a home including the infamous example of Samsung “analysing” what type of content you’re watching and selling that to ad companies.
OP is probably better off looking at enterprise/industrial manufacturers.
https://openwrt.org/toh/netgear/r7000
Once that is done you can use the browser for a few things, but for WiFi password and all I think app is still needed. Couldn't find a way to do it other way.
Worst thing is that this stupid Xfinity app eats phone battery like crazy. So I had to disable background activity and bluetooth permissions once I was done with the activation and initial setup.
- my most recent Comcast/Xfinity experience
Most products like this have some kind of "harder way" to set up without an account, while the smartphone way uses an account. For a router usually means that if you want the old-fashioned manual way you need to visit the router's gateway IP and set it up from there.
Sometimes "the smartphone way" even has a way around having to make an account, it's just that it's hard to find or discouraged. E.g., I set up my HP printer without an account or any ink subscriptions (and I even think it's a great printer, shocking, I know).
(I had a Nighthawk device relatively recently and it had no need for an account, but I don't actually know if this has changed recently)
Eh, probably they're just targeting users who want to view their CCTV on their phone when away from home, without their camera ending up visible to the whole world.
I mean, if you're selling a retail product to consumers, very few of them know WTF things like ONVIF are.
You haven't noticed many of the companion apps for IoT(of Shit) have those "store"/"discover'/"savvy user center" sections where they try their best to get you to buy more of their devices? Plus a barrage of notifications, in-app banners or emails how you're eligible for a "discount" or "promo"?
This - plus online services (re-broadcasting outside of home network, recording storage, face/object/sound recognition and notifications, etc - any software features that can be pay-gated, especially those that can't be done on-device because the device is ultra-cheap) is how they make money.
> I can only conclude that they must be selling access to feeds
I'm skeptical. Who's the buyer and what are they going to do with those feeds to make them useful? And if it's for something remotely legal - how they're going to untaint this data?
This works in cyberpunk novels - cameras sweeping data, AIs detecting that a neighborhood is $brand turf, classifying all the individuals, ..., massive profit! But reality is messier and way less logical than any fiction (and we don't have any AIs yet, while fiction has them abundant and dirt-cheap). Out of curiosity, I've just had a long session with GPT-4 (which is a really uneconomical way to do advertising, but maybe in 5-10 years?), telling it all I see in my room in extreme detail, down to all the scuff marks and pet hairs. It wasn't exactly bad - it managed to realize the obvious (which is much cheaper to deduce from my search and purchase history, huh), but let's say I haven't made any new records on my purchase list.
coulda just gone with Internet of Turds (IoT)
Axis cameras are high end and expensive, but they will, in my experience, do anything an IP camera could reasonably be expected to do, and they will do it well. They are European in origin and are available from various retail outlets to ship this week.
Geovision cameras are low end and not expensive. They are Taiwanese in origin and are pretty easy to find.
I have personally configured a wide range of cameras from both of these manufacturers and I have never needed an app or internet connectivity. It's been a few years since I looked at Geovision's product lineup though, my information is not 100% current. I don't have any specific camera recommendations. If I were setting up a home NVR today, I would buy Geovision cameras and put them on an isolated network.
Both of these manufacturers are nominally ONVIF compliant (ONVIF compliance is a mixed bag and can't be fully trusted from any manufacturer IMO) and have readily accessible RTSP streams
Would recommend both of them. Notably, I had one exterior camera fail (wouldn’t run off PoE), and the Axis support/RMA process was smooth and fast.
It also, relevantly for this thread, will do standard video transmission w/o any cloud/app components.
Most camera you get that are cheaper are trading off in one or more of these areas to recoup cost.
Otherwise its just a single ethernet line from the camera to your switch, Axis cameras (and any others that support PoE and outdoor installs) have a nice shielded bit that the cable goes into at the camera end so its protected from the elements.
But they're expensive. They do have cheaper models nowadays, though. In addition to explosion-proof ones: https://www.axis.com/en-us/products/explosion-protected-came...
I was given a tour of an industrial facility where conditions required explosion-proof equipment. The tour guide made a point of calling out the explosion-proof motors in one of the rooms.
One of the other tour group members asked the question "Why would it matter if these motors survived an explosion?"
I visualized the facility exploding in a fiery blaze and a motor, fully intact, tracing a high parabola into the sky and landing, unscathed, in the front yard of a house.
The tour guide had to awkwardly answer: "Explosion-proof means that it won't cause an explosion."
Another industry term for this is Intrinsically Safe.
[1] https://blog.pruftechnik.com/intrinsically-safe-vs-explosion...
Especially for cameras in potentially hazardous or critical areas, being able to capture details of their final moments could be important (obviously great to have cameras which [also] don't cause explosions in the first place).
I can also think of many situations in which you would want motors which are rated to survive for some time through an explosion, for example something which is powering a fire sprinkler or ventilation system to give people time to escape. The military has "battle switches" on lots of equipment for this reason, better to remove safeties and destroy the equipment than have the equipment protect itself and stop working in a critical situation, potentially costing a soldier their life.
It was a formative experience for me, too. I try really, really hard to explain terms-of-art to my audience if I'm unsure of their familiarity. "Reading the room" can be difficult sometimes so I try to always preface w/ "Stop me if you know this already...".
I also learned not to be afraid to speak up when terminology isn't familiar to me. I've no doubt that's made me look ill-informed in some situations. I'll always take the judgement of others re: my knowledge vs. acting on a bad assumption later.
There are consequences (sometimes significant ones) from assuming somebody knows what you're talking about. People who act like they understand something when they don't (either because they're afraid to ask, or because they don't know they don't understand) are dangerous.
That bit made me think otherwise.
> I also learned not to be afraid to speak up when terminology isn't familiar to me.
100% agreed. I don't care much if others think I'm stupid. I care about actually being educated and understanding the situation. If someone thinks me asking questions makes me stupid, I'm probably better off with them thinking that and avoiding me for it :)
> That bit made me think otherwise.
I could have phrased that better. It was clearly awkward for the tour guide. I think the guide could have handled it in a less awkward way— maybe with some humor.
I found it amusing by way of the image it created in my mind, but that shouldn’t be taken as me discouraging others asking questions.
Wait, what? The implication is that their other cameras regularly cause explosions?
Explosion proof / flame proof enclosures don't prevent explosions. Their sole purpose is to contain the explosion and to ensure that the gases escaping from the enclosure do not ignite the surrounding atmosphere. The beauty of a flameproof enclosure is that I can put standard, non-protected electronics into an enclosure and use them in a hazardous area. It's a much more affordable way of deploying electronics into hazardous areas, rather than designing from scratch and certifying.
Some electronics use a spark gap as a form of surge protection:
https://en.wikipedia.org/wiki/Spark_gap#/media/File:PCB_Spar...
None of that would be acceptable in an environment that requires intrinsically safe equipment.
The electronics also need to be carefully current-limited so there's no chance they overheat and cause an explosion that way too.
Non-explosion protected electronics... sure. All you need is an atmosphere of sufficient concentration and a spark with sufficient ignition energy OR a hot surface that is above the atmosphere's auto ignition temperature (think a hot resistor on a PCB). Some chemicals have absurdly low auto ignition temperatures or ignition energies. Engineers must spec equipment for chemicals they expect to encounter within the hazardous area. Engineers also have to factor in how often the flammable atmosphere will be present. All of this is spelled in out IEC60079 for Europe and RoW. US has their own standards.
Surely that should be "non-explosive".
Otherwise rain-proof and fire-proof could have interesting overloaded definitions...
XPQ1785
Starts the washer. When the sequence starts, the camera moves to the configured position to receive the wash spray. When the whole wash sequence is completed, the camera returns to its previous position. This icon is only visible when the washer is connected and configured.
Starts the wiper.*
https://help.axis.com/en-us/axis-xpq1785
My Geovision experience is also not current either. I've got three Customers using their cameras in isolated VLANs w/o Internet access. Both hardware and software reliability have been very good. Some of the outdoor units are coming up on 7 years old and still working fine. I think, in total, there are around 300 (mostly indoor) cameras in all their systems combined.
I work at a site w/ >100 Axis cameras, also w/o Internet access. They're phenomenal devices-- built like tanks-- but very expensive. I particularly like that you get root on the cameras (which are running Linux). There are third-party applications that can run directly on the cameras.
Such a rarity nowadays--great to hear.
https://www.kernel.org/doc/Documentation/cris/README
Edit: Official Linux support was removed with 4.17 in 2018
https://lkml.org/lkml/2018/6/3/142
Edit: added missing word.
[1]: https://en.m.wikipedia.org/wiki/Axis_Communications
(not to fear-monger, just something to note)
I think they’re still the industry leader too. The catch is that they are expensive and they’re mainly enterprise solutions
If an independent shop wants 2 cameras you charge them enough to make a profit on the overhead of dealing with them (or point them to a retailer if they want less than a pallet).
If Walmart want 2 cameras to try out you give them (or loan them) to them for free knowing if they like them you'll be shipping them by the pallet for years.
In larger installs it’s a VRF dedicated to the camera subnets and a similar dedicated NIC arrangement on the server.
Nothing should talk to the cameras except the recording server they’re streaming to.
Annoyingly, few TVs have multicast receivers, it's all Chromecast etc, so you need your own computers to decode and display.
You can find older used HD resolution Axis cams fairly inexpensively on eBay. They have a lot of different chassis/lens configurations and a lot model numbers so some research might be required to find the right camera configuration for your desired use case, especially if searching through the used market for models that have been discontinued by Axis.
I think a lot of the problems with IP cams aren't with the cameras themselves, but the poor state of open-source media players. Getting mplayer, ffplay, or vlc to play nicely with any ipcam has been a Labor of Hercules. If you're using the versions in repos--which are usually quite stale--all bets are off. I've had much better luck using the latest installers, direct from the project websites. So far, VLC seems to work best for me[1].
[0] https://github.com/BourgeoisBear/amdacli
[1] https://www.videolan.org/vlc/
The cloud thing can be disabled from their web UI or the HTTP API, but between all of the knock-offs out there (of any brand), and the eventual end of firmware updates, it's best not to trust any of this stuff.
I use a Debian Linux router with nftables to achieve this in my homelab.
If so: on my desk right now is a Raspberry Pi Zero with a NoIR camera, configured so that when it gets power a small Python script starts up as a web server and begins hosting the camera output.
Here's the code for both the server and a webpage to auto-refresh the camera view. MIT license, so have fun:
https://github.com/BenWheatley/PiWebcam
There are different maintainers for different models, but as far as I can tell they're all pretty similar: https://github.com/search?q=yi%20hack&type=repositories
Edit to add they work pretty well with Home Assistant with this HACS integration: https://github.com/roleoroleo/yi-hack_ha_integration
I am not affiliated with them, other than that I've bought equipment from them in the past.
I personally prefer Bosch and Hanwha cameras. Great optics, low light performance and solid firmware. Axis tends to be expensive and low light performance is not as good, for the price. UniFi Protect cameras are decent, but the standalone firmware is rather limited.
I connect these cameras to Frigate[1] locally.
Some example model numbers: Bosch Flexidome IP Starlight 6000 Dome Security Camera - NIN63023A3; Hanwha Techwin XNO-8080R WiseNet X Series Network Bullet Camera 5MP 3.7-9.4mm; AXIS P1468-LE Bullet Camera
I usually find these on eBay.
[1] https://github.com/blakeblackshear/frigate
I know UBNT have had a few missteps recently, but in general I like their kit and am pretty happy with it.
Do you happen to know a way to use the audio-backchannels of the to broadcast audio out of the cameras? And/or use them as "speakerphone" with SIP/xy? Somehow integrate them with pipewire?
In general, you probably want to look more towards the commercial CCTV market, even at the cheap end, rather than at any cameras aimed directly at home consumers.
They are available on Amazon.
Even if you don't use Synology, this is a handy list. https://www.synology.com/en-us/compatibility/camera
It is known to have major vulnerabilities https://www.bleepingcomputer.com/news/security/over-80-000-e...
So, Hikvision seems to not be a great answer in 2023 after all (an apparent 100% shift from just several years ago). I wonder about the current status of the ban, and what the rebranders and their distributors are doing; a cursory look at web material still shows Hikvision-derived products AFAICT (though I haven't checked in much depth).
As for management software; a dedicated NVR appliance is most convenient IMO in terms of hands-off just-works, but if one wanted to put more time and resources into it, Blue Iris looked interesting the last time I looked several years ago; again, I wonder how that landscape may have updated since then.
As long as you take a pretty hard stance to wall off the Hikvision stuff from the outside internet, it seems fairly harmless. Obviously this requires setting up your internet to manage that, but that's not impossible.
https://www.hikvision.com/en/support/cybersecurity/security-...
Precondition:
The attacker has access to the device network or the device has direct interface with the internet
I looked at DIY with Raspberry Pi high quality cameras, but the outdoor requirements mean I really need a manufactured housing. (Camera will be in an area very hard to reach for servicing.)
I wonder if there is any project to hack the generic chinese camera makes and install custom firmware?
The only downside hardware-wise is I don't get any indoor IR night vision with these, which some of the nicer "smart home" account-locked ones do.
It's honestly not too bad to set up if you run [1] and [2] in Docker. I've done disaster recovery scenarios of my home infra where I straight up disconnect the modem's uplink and everything works without any issues.
[1] - https://github.com/bluenviron/mediamtx
[2] - https://github.com/seydx/camera.ui
As a Comcast customer in a region with no competing provider, I really respect and appreciate their proactive stance toward ensuring all their customers frequently have the opportunity to reaffirm this level of confidence in our local networking infrastructures.
I also have a few IR cameras that work ok. The biggest issue is finding decent ir LEDs.
The ones that come with the IR cameras aren't great. I purchased standalone IR LEDs, which are better for reliability and ir illumination.
Most IP cameras use 3W just creating a RTSP stream. add another 5W for IR LEDs that come on in the evening.
If i was shopping, i'd look for sensors with very high sensitivity that don't need 5W of IR to get good night images.
I have a few where i've physically disconnected the IR array and the shutter will adapt and still get very good static images and a little bit of smearing when something moves. It all depends of your use case though.
Smearing is bad. If you want to identify anything, it's worthless. If the goal is to just see if anything is there, then almost any camera works... But that's not useful for much.
I'm also curious if anyone has found cameras available through Amazon or Ali which guarantee support.
edit: I found this -- https://github.com/OpenIPC/wiki/blob/master/en/guide-support...
But above all, it's exactly this that's the biggest problem. I'd wager like 75% of devices have an OpenIPC capable chipset. But there's only rudimentary tracking of what devices have what chipsets.
And it's unknown what other wrinkles folks might run into if they do get a device with a valid chipset. Most should just work, but a number will also require more embedded hacking.
Such an amazing project though. Stunningly wide support for a huge array of chipsets.
I've recently worked with some of D-Links consumer IP cameras, DCS-8302LH and DCS-8000LHV2. They offer an app and optionally cloud storage but they ship with ONVIF support. No web interface afaik, but what I found was that I just needed to start interacting through ONVIF to get going with a local-only setup (this would automatically disconnect the camera from the app as well).
These cameras are some of the cheapest available, but they have been solid.
1: https://a.co/d/8k1BWBY 2: https://a.co/d/4a0xnPP
For Wifi ones, like my doorbell camera, I had to use their app to connect them to Wifi but subsequently block their internet access with no issues. I can still react to events and control settings using Home Assistant and the Dahua HACS plugin.