I didn't read it as dismissive. I think the "Why doesn't this work?" at the end is key, implying they know they're probably wrong, but don't know why. It's pretty common to form and present hypothesis like this, hoping that anyone who knows the actual theory can easily provide a counterexample showing why it's wrong. It's not at all intended to be dismissive.
Its not enabled by default, but enabling MGLRU setting /sys/kernel/mm/lru_gen/min_ttl_ms to 70 has entirely resolved the issue for me. It _does_ usually pick firefox or vscode to kill, but my system remains interactive, with no need for earlyoom.
Class D amplifiers use that same trick, just at much higher switching frequencies (pushing quantization noise ultrasonic, where it can be filtered). Since transistors are most efficient when fully on/off, very little power is wasted as heat. That's what made the modern revolution in tiny amps possible.
Interesting, could it be mitigated by the server doing its own prediction and defogging the peeker early? Or is lag prediction in CS2 not entirely client-sided.
It's more that momentum-based defogging means that the peeker has control over how to manipulate the server's prediction, whereas the victim who is already disadvantaged by network latency now gets an additional penalty of the movement initiation being not telegraphed.
To solve this, the fog of war would need to use purely positional near-edge tolerances, which defeats the entire purpose of fog of war to begin with, which is the pre-aiming reaction time advantage of tracking the peeker through walls in addition to having a farther lever point from the cover than the peeker.
That doesn't sound at all like any ORM I've ever used. I've struggled in the past because The ones I've used are actively hostile to laying out data in the database in a way not proscribed by the ORMs philosophy. Heck of the ORMs I've used, one didn't support parameterized joins and the other didn't support joins at all.
---
It's not usually a DB guy gatekeeping, it's that multiple apps use the same database so layout changes are costly.
Except for the "multiple ORMs" part which is a level above it, it applies to the only one I've used extensively: Django for python. It has standard defaults, but just about everything overridable, and because models are python objects you can add methods or properties for extra data. There's even ways to define your own field types (the "serialization/de-serialization of individual properties"), which a decade ago people were using to provide json fields through libraries long before it was officially supported.
...and Django was like this 15 years ago when I first started using it. The core design hasn't changed, it just sounds like most other ORMs don't really know what they're doing.
>and Django was like this 15 years ago when I first started using it. The core design hasn't changed, it just sounds like most other ORMs don't really know what they're doing.
Django is an opinionated web framework that uses an ORM, not just an ORM.
Django can by all means be a great way to make a web site (I have little experience with it) but if you have a db that is accessed by various systems written in Java, dotnet, erlang or whatever else I suspect the smooth sailing of Django can run into headwinds quickly and the python plumbing you have to deal with then quickly becomes an issue in itself.
> but if you have a db that is accessed by various systems written in Java, dotnet, erlang or whatever else I suspect the smooth sailing of Django can run into headwinds quickly
Only if those systems are constantly adding/removing tables and columns. And adding isn't a problem, Django just ignores what's not specified in the models.
Django does have default table and column names based on the models that it prefers, but all of it is overridable in officially-supported ways. We're using it with mysql databases originally made for VB6 and C++ with inconsistent naming schemes that aren't even close to Django's defaults, that nowadays are also accessed by perl, php, and python. Most of our python uses are daemons that only use the models and none of the rest of the web framework - the models are defined in a common library they all use.
reply