If you have written a 3D engine (in asm, C, Pascal, Basic, whatever), I'd be curious to see it!

I am writing one now (in C), resuming my work from 15 years ago. I hope to share it soon, once it reaches demo state (working on texture mapping now and occlusion)

goood day comrade
do you plan using opengl, directx,.. or modern cpu extensions?
thank you for an open-source approach, it inspires/motivates/lifts from an abbys
goood luck

I plan specifically on NOT using opengl, directx, and perform all rendering in software.

Как дела в Украине? С какого города вы? У нас здесь футбольная комманда, много лет вместе играем, а в 2014 не скольких "подвинуло". Обидно.

I will be waiting.

У руля бл#яди, а я устал и хочу мира, спасибо. Чернигов...

macgub is the first to come to mind.
https://board.flatassembler.net/topic.php?t=15011
http://www.macgub.hekko.pl/

He too has been working in a similar manner for a long time. There are others, but I will need to look through my archives.

Everything just in GDI, no opengl, directx or external lib:
RayCasting
Textured cube
My first demo
My user's still working

Welcome. You simply need to come back home more often.

@avcaballero very nice! p.s. i can't locate your windows graphic tutorials the link in not working.

Yes, I'm sorry. I had to down my website because some av gave some false positives in some TinyC programs. Mainly in AOWC and in AOWG (this last one is what you are looking for). It seems that the key is giving information to the file such as icos, text info, etc, otherwise it seems to be suspicious. Until I have time to change them, I cannot upload.

Regards

You should not change your code to please the AVs. This is the way to the hell.
Better submit your programs on the AV web site with false positive report.
This way they will change their code, simply because they are wrong and you are right.
Surprisingly, this works most of the time for me.

I know, last week i had a lot of trouble trying to visit your pages.

My suggestion is to ignore the AVs. It is just more work for you and no benefit to the users or yourself. AVs don't work as advertised, as shown by the continuing reports of people with active AVs still getting caught with new stuff.

If you are worried about losing potential users because they trust their AVs so much then so be it. Such unquestioning blind faith in things needs to be removed and a dose of scepticism put in its place.

However, even so, AVs do serve a role in some situations where they can catch a lot of the old problem malware. Usually a few months to a few years after the malware has been released. So for old software you can run it through virustotal and check there. But still be aware of malware has has yet to be found even in old code. [/rant]

Yes, I think so, but I have no time to fight with them. I even had to change the folder from "AOW32" to "AOW32_". Fortunatelly I still can add some more underscores

They even flag my website as suspicious and google and others showed it as a red page saying something like "if you dare, get into", which is not very handy to show to anybody For some programs they extended the exclusion to the whole site. Fortunately at present is up again

I write a software voxel engine in a forth lang
https://github.com/phreda4/reda4

I not finish but i have and basic editor a viewer and a unfinished game.

some lib are common for a basic rasterizer, like zbuffer or polygon filling..
I do all in integer math, some fixed point.

The compiler use fasm for make standalone exe.

wait for see you implementation!

I just started looking into Spir-V https://www.khronos.org/spir. I've been using HLSL a lot lately. Not sure why, but switching between FASM and HLSL leaves something to be desired. I will be writing more and more compute shaders for mechanics simulations and probably writing some results to OpenEXR image files if it goes well. Not sure if every GPGPU language requires "views" to resources/buffers, and if the memory model is the same for each (still learning HLSL's), but I may look for an alternative to HLSL later.

I wonder if fasmg would provide conveniences for Spir-V? Maybe defining a 'struc' for an x-component Tensor would still require redefining it in the Spir-V shader/kernel, for instance. The message boards here have been very helpful also Was able to refer to someone's implementation of a Taylor Series sine function and then tried my own to a few decimal places. Would be interesting to use this in Spir-V with Quaternion rotations (got this in HLSL with cross products, etc, but used the sine intrinsics. I think Spir-V would need a new function or extension).

Good luck Comrade