Short answer: Neither Ring nor Google documents a way to export a camera archive in bulk. Every retrieval path either company publishes is per-clip and hand-driven, and each one carries a ceiling: Google's own help page says you can only download clips from detected events, with a maximum clip length that changes depending on which of its four clients you happen to be using, and the Nest app holds only about three hours of saved clips before you have to download and delete to make room for more. Ring's help page says the phone app downloads one clip at a time, and that the number you can take from the web "will vary based on the length of your videos." None of that is hidden and none of it is malicious. It is simply not a number anybody puts in a comparison table, which means the cost of leaving a camera service is real, is measured in evenings, and is invisible until you are already inside it.
The column that is missing from every comparison
Camera services get compared on the things that are easy to tabulate. Monthly price. Retention window. Number of cameras on the plan. Resolution. Whether there is a free tier and what it takes away.
There is one property nobody tabulates, and it is the one that decides what happens to everything you recorded: how long does it take to get your material out.
Not can you get it out. That question almost always has a yes attached to it, and the yes is honest. The question is how many hours of your own attention stand between the archive and a folder on your own disk — because that is what you will actually be spending, on some evening you did not plan, usually because a price changed or a household moved.
I went and read what the two largest vendors publish about this. Not their marketing; their support documentation, which is written for people who already own the thing and is generally accurate because it has to be. Here is what it says.
What Google documents
Google's help page on creating, saving, and downloading video clips is the relevant one, and it is worth reading closely because three separate constraints sit in it.
First: not all of your footage is downloadable. The page states plainly that you can only download clips from detected events in the video history, and that you cannot download clips from footage with no activity. If you are paying for continuous recording, the continuous part is reviewable but the downloadable part is the event list. This is the single most under-appreciated line in the document. A year of recording was never going to come back as a year. It was always going to come back as a list of moments the motion detector agreed were moments.
Second: the per-clip ceiling depends on which client you are standing in. The same page carries a comparison table, and the numbers do not agree with each other:
| Client | Documented download limit |
|---|---|
| Google Home app | Bounded by the camera's max clip length setting |
| Google Home for web | Up to 5 minutes per clip |
| Nest app | Up to 2 minutes 30 seconds per clip |
| Nest app for web | Up to 1 hour per clip |
One account, one archive, four ceilings that differ by a factor of twenty-four. If you start the job on your phone because that is where the app is, you are working in the smallest window available to you and there is nothing on the screen to tell you so.
Third: the Nest app has a working-set limit. The page says you can save up to about three hours of clips in the Nest app, and that once it is full the app will tell you that you are out of storage space — at which point you must download and delete clips to make room before you can save a new one.
Read that as an algorithm rather than a limitation and it becomes clear what the evening looks like: create clip, save clip, download clip, delete clip, repeat. There is no queue, no batch, no background job. The loop is you.
What Ring documents
Ring's Managing your Ring videos page is shorter and equally direct.
The phone app downloads one video clip at a time. The web account will let you select several at once and delivers them as a timestamped .zip of .mp4 files — genuinely the better path, and if you are ever doing this, do it there. But the page attaches a caveat to the batch: the number of videos you can download at one time "will vary based on the length of your videos." There is no published figure. You find out what the ceiling is by hitting it.
Two other lines matter for anyone thinking about the exit specifically. The page says reviewing recorded video requires a subscription — the review surface is part of the plan, not part of the device. And it notes that past events "might appear in your history" while no longer being viewable or recoverable, which is a useful and slightly chilling piece of honesty about the difference between an index and the thing it indexes.
The arithmetic, done honestly
I am not going to hand you a total, because the total depends on your event count, your clip lengths, your uplink, and which client you use — and a confident figure from somebody who has not measured your setup is worth nothing.
What I will say is that the shape of the calculation is fixed, and it is multiplicative rather than additive:
evening_cost = event_count
× (select + create + save + download + delete)
÷ (whatever the batch size turns out to be)
Every term in that expression is documented above except event_count, and event_count is the one you already know. Open the app, go to a month you still have, and count the events in it. Multiply by twelve. That is the length of the list you would be going down.
The reason this never shows up as a number is that it is not denominated in money. It is denominated in your time, and a price in evenings has nowhere to appear on a page that has a column for dollars.
This is not a villain story
It would be easy to write the previous section as an accusation, and it would be wrong.
Serving thousands of individually addressed clips out of cold storage to an account that is on its way out is a real cost to somebody, and rate limits and per-clip ceilings are the ordinary engineering answer to that cost. A service that stops serving footage to people who have stopped paying is doing exactly what it published. Every constraint quoted above is in a public help page written in plain language, not buried in terms nobody reads.
The problem is not concealment. The problem is that this is a property of a purchase that the purchase decision has no slot for. You choose a service on price and retention, and you pay for it in a currency that was never quoted.
The one thing worth actually doing
Turn it into a number while nothing is at stake.
Pick an evening this week when you are not leaving anything and not angry at anyone. Pull down one week of your own material. Time it — actually time it, on a clock, including the part where you decide which events are worth keeping. Then multiply by however many weeks you are holding.
You will get one of two results. Either the number is small, in which case you have bought genuine peace of mind for the cost of one evening and you can stop thinking about it. Or the number is large, in which case you have learned it on a day of your choosing rather than on a day chosen for you — and the fix is available immediately and costs nothing: keep less, and decide what is worth holding on the day it is recorded rather than on the day you leave.
That second habit is the whole of the mitigation. An archive you curated as it arrived is an archive you can walk away from.
Where I sit, and why it is not a clean win
I build Background Camera RemoteStream, which records on the handset to the handset, so there is no retrieval step in the sense this article is about. Nothing has to be fetched back from anywhere, because it is already sitting on a device in the room.
I should be honest that this is a trade and not a victory, because the mirror image has its own failure and it is not obviously the smaller one. What the retrieval step gives you is a deadline — a renewal notice, a cancellation date, a moment when somebody makes you look at your own archive and decide. Remove it and you get a job with no deadline, and a job with no deadline is a job that stays undone until the device holding all of it settles the question on your behalf, without asking.
So the habit above is not advice for people leaving a subscription. It is advice for anyone keeping recordings at all, including on my own setup: decide what is worth holding on the day it is recorded. That is the part that works in both architectures, and it is the only part that does not depend on which one you picked.
More on the trade-offs of running a spare handset as a camera at superfunicular.com. If you have measured your own retrieval time on any service, I would like to hear the number — that is the figure the whole category is missing.
Sources, both read on 2026-09-02:
- Google Home & Nest Help — How to create, save, and download video clips
- Ring Help — Managing your Ring videos
Top comments (0)