I have posted two updates about RADDBG on Linux, and all I can say is that, while the majority of responses have been very positive, the rest make it obvious why it has taken so long for the Linux ecosystem to provide a ~viable consumer desktop experience (for Windows to decay so much that there is not really any alternative).
There is no coherency in the design of the platform. There is no sane team calling the shots, so the user base spends their time bikeshedding over what should be trivialities. Everybody has an opinion about everything. Window managers (X11? Wayland? X11 on Wayland?), to package managers, to libc versions, to the right folder to use for X/Y/Z, to which shell interpreter to use, to how to ask the user for a file path (libportal? zenity?), to how to show the user a trivial message box, and so on. And this is just the beginning—the trivial stuff that most people reading this will get, without needing a ton of extra context.
If your users or developers are debating about any of this, YOU FAILED. The platform SUCKS. It will only succeed when centralized teams (e.g. shipping hardware using the Linux ecosystem) steer the ship and make decisions. Then, people write to that, and that becomes the actual standard.
This week:
— third-person camera
— shader hot reload (instant iteration!)
— learned sRGB vs linear + tonemapping the hard way
— procedural sky + fog
— per-voxel color jitter, no more flat dirt
Next: offscreen HDR pipeline
#gamedev#voxel#indie#vulkan
Dear @OpenAI
I would make a great candidate for CEO of OpenAI 501 3(c).
- i am currently a CEO of a multi-billion dollar company
- i have a mustache most men only dream about
- i can bench my weight
get your people in contact with my people, we will make it happen