Funny you call it magic, what actually does the conversion is Imagick.

In my project I have it integrated in the upload process. You upload a PNG/JPG and it does its thing. Since it's written in PHP (my project), and PHP has an extension to call Imagick, I didn't need to write any complicated code.

You can see on this page if your programming language of choice has any integration with Imagick.

But there's always the command line interface. Depending on your process it may be easier to create a script to "convert all images in a folder", for example.

but 2KB vs 200KB is paltry on even a terrible connection in the 2000s).

You still need to resize the images and choose the right ones (even if only for the device's performance).

So we might as well do that small extra step and add conversion to the process.

What I really wish is that we could get more browsers, sites, and apps to universally support more modern formats to replace the overly bloated terribly performing and never correctly pronounced animated formats like GIF with something else like AVIF, webm, webp (this was a roughly ~60MB GIF, and becomes a 1MB WEBP with better performance), or even something like APNG…

Isn't that the users' fault? And of the websites for allowing those huge GIFs.

Apparently browsers have supported MP4 for a long time.

https://caniuse.com/mpeg4

For most of the images that I tried you can only see differences with the images side by side. It's really subtle.

I do have one example for which my config must be bad, compresses a lot but introduces a lot of noise

MultiViewer (which is an unofficial program, mind you) does support Linux, but you need to download the installer manually to install and update.

Other than that it works great

Ok, I understand what you meant, thanks.

Basically, after I’ve read all of that, it’s clear as day that security is not a priority on Testing. And while band-aid solutions do exist, it’s simply not designed to be secure.

Yeah, I wouldn't run it in a production environment.

I remember having some issue like that, but I'm not sure if this was the fix.

Try unchecking "Show desktop notifications when the song changes" on Spotify's settings (right now it's under the Display section).

Makes sense, thanks.

The band played 2 early albums + their latest one in sequence. Listening to whole albums in one go was great for many reasons.

Fellow PT-PT ISO user here. And although I use PT-PT in the OS, both my mechanical keyboards' physical layout is DE ISO, which has most special symbols in the same place. (finding DE keyboards is easier)

I've considered switching to UK ISO before. Typing brackets "[] {}" and a semicolon ";" is harder in PT-PT. Especially the curly brackets {}, which are really awkward to type with my small hands.

I'm running a 6700XT and weirdly enough it pre-compiled in Linux but not in Windows.

It's really stuttery for a while in Windows, with low GPU usage and erratic frequency, until it normalizes.

I'm getting none of that in Linux, smooth from the start in-game. Only getting some weird fps fluctuation in the start menu.

From the F1 channel, here it is in action (shakedown test).

I don't know if we're discussing semantics. A performance score is attributed, and before the fix their scores were all 166. It doesn't work, as you said. So the consequence is the preferred core being "random", isn't it?