Hardware
How machines behave in the real world. Heat, wear, power draw, and the parts that give out before the rest.
Research
We build the machine, so we study the machine.
Calyraen makes the hardware, the operating system and the platforms that run on top. Owning the whole stack means we can measure it end to end. We publish what we find, including the parts that did not go our way.
Imagery in production
Our standard
A claim you cannot check is just marketing.
Every study here comes with its method and its numbers. Change our minds later and the old work stays up, with a note that says so.
What we look into.
One stack, studied from the metal up.
Operating systems
What people do with CalyOS when there is no IT department in the room, and where the software gets in their way.
Society and safety
The effect of what we ship on the people who use it, from parental controls to how readable our own words are.
From a question to a published answer.
-
It starts with something we cannot explain
A return rate that does not match the lab. A feature nobody uses. The interesting work begins where the data stops agreeing with the story.
-
We measure before we theorise
Because we own the stack, we can instrument it properly instead of guessing from the outside. The measurement comes first. The explanation comes second.
-
Then it goes out in full
Method, data, and the parts that surprised us. Reviewed inside the company, then published so anyone can take it apart.
“The studies that make us look worse are the ones worth trusting. So we keep them up.”
The record so far.
40+
Studies published
3
Fields, one stack
7 yr
Oldest work still up
0
Findings pulled after the fact
Asked and answered.
What is in each study?
Its method and its numbers. If you cannot check a claim, we do not treat it as one.
Do you publish work that makes Calyraen look bad?
Yes. Nothing here has been pulled after the fact, including the findings that did not go our way.
What happens when you change your mind?
The old study stays up with a note saying what changed and why. We do not quietly delete our past work.
Why can you measure the whole stack?
We make the hardware, the operating system and the platforms on top. That lets us instrument the whole thing instead of guessing from the outside.
The index.
Everything we have published, newest first. Filter by field, or search.
Imagery in production
Research Where laptops actually fail first.
We read three years of returns and sorted every machine by the part that brought it back. What we found changed how… Jun 14, 2026
Imagery in production
Research What families do with the controls, not what they say.
We sat with real households while they set up parental controls, and watched the gap open between the settings people ask for… Jun 3, 2026
Imagery in production
Research What a cold data centre costs you.
Most server rooms are kept colder than they need to be. We spent a month measuring what that habit actually buys you. May 23, 2026
Imagery in production
Research Reading levels in our own help pages.
We read our support articles the way a first-time customer does, then scored how hard they were. Some were working against the… May 12, 2026
Imagery in production
Research Joining a network without an IT department.
We watched small firms set up CalyOS Server on their own. Here is where they got stuck, and what the setup does… May 1, 2026
Work with us
Have a question worth measuring?
If you think we got something wrong, or you want the raw data behind a study, tell us. A person reads every note.