reepblue

I've tried this and that'll be a lot of translating you'll need to be doing but I would appreciate it. I recommend exposing the smart pointer as well just in-case something is missed. I'm curious on how things like the world and actor system would port over. That, and getting context drawing simple text, shapes, and textures to the screen might be a challenge, but you will most likely figure it out. Only brought that up since I seen a lot of people draw things to the screen that way over the GUI system. Would be really cool just to link Leadwerks code to Ultra and have everything or most things just work.

I have been getting as kick of your video thumbnails but I honestly don't think your exploiting the algorithm to the fullest. Maybe next time try adding red borders, arrows and circles to create a sense of urgency. Emojis are cute and everyone likes them so maybe add a few of those in for good measure. You can also try being creative by including meme references and/or anime girls in your thumbnails. Play stupid games, win stupid prizes. 🤡

That was Leadwerks 2, different technology.

You need to make a custom shader. I know this is possible because I was experimenting with light maps a while back. I recall just duplicating the UV variables and multiplying them together. It's been a while, and I think I've deleted the shader.

That's incredible.

I was also seeing this with the Text areas too. I agree that Voxel Raytracing should only before hardware that supports it. However, there should be things we can do to make sure customers with lower end cards can still get decent results. If I have to go around placing GI probes everywhere, I wouldn't really mind although it kind of goes against your design goal a bit.

I don't see the addition of reflection volumes a bad thing really. If it can work with the current system, I wouldn't mind the finer control for things like that. 100% automation can only get you so far imo. I'm not sure if you'd want to consider also doing a onetime GI raytrace within those volumes like Leadwerks, but I'm mostly concerned about the reflections at the moment.

Yeah, that's what I expect. One idea for that can happen older GPUs we can do a generic cubemap builder and set it to the world's skybox. This way the reflections are as good as the ones in Leadwerks, but without complicating your current reflection system. I did this with Leadwerks years ago, and hopefully now we can do this without talking to the graphics api. The question is if you want to have this as an engine feature or have this done by somebody like me. I really think something like a world->BakeGlobalCubemap() function would be nice.

Release. I don't have fraps for the framerate, but the Xbox Gamebar thing is reporting 27 frames per second with 98% GPU usage.

In the shader, you can delete the decal code block and it'll break the decal from working on the material.

That example really lags on my GTX 1050 with 4GB of VRAM. I also find the AO fading in and out very strange and distracting. Hopefully it just you didn't get that far. Also noticed you have doubled the default voxel resolution. Setting this back to 64 helps but makes the AO fading more obvious. Perhaps I should be running this on a stronger card, but this PC is just more convenient.

Last night my app stopped working as expected. I had a layout on how I wanted scene loading to work but now the "menu" doesn't disappear and I was too frustrated to figure out why. I think I'm just going to wait until everything settled before writing code I want to ship with.

The only thing is that UAK had a Win32 library which is something that you probably don't want to support in the new engine.

I'm getting a crash with this. The example you posted before works fine but IDK why this is throwing me to lists.h. void StageUI::ProcessEvent(const UltraEngine::Event& iEvent) { // This throws me to list.h //if (GetInterface() != NULL) GetInterface()->ProcessEvent(iEvent); if (iEvent.id == EVENT_KEYUP) { if (iEvent.data == KEY_ESCAPE) { auto stage = m_wpStage.lock(); stage->SetPaused(!stage->GetPaused()); } else if (iEvent.data == KEY_TILDE) { auto stage = m_wpStage.lock(); if (!stage->GetPaused()) stage->SetPaused(true); const bool bConsoleHidden = m_pPanelConsole->GetHidden(); m_pPanelConsole->SetHidden(!bConsoleHidden); } } else if (iEvent.id == EVENT_PRINT) { m_pTextAreaConsole->AddText(iEvent.text + "\n"); } } This is where it stops.. _Unchecked_iterator _Unchecked_begin() noexcept { return _Unchecked_iterator(_Mypair._Myval2._Myhead->_Next, nullptr); }

I'm trying to rebuild my in-render interface after the window rebuilds to a different mode, but this doesn't seem to update the new size. void StageUI::Redraw(const float fScale, const iVec4 v) { m_pCamera->SetPosition(v.width / 2, v.height / 2); m_pInterface->SetScale(fScale); m_pInterface->Redraw(0, 0, v.width, v.height); m_pInterface->UpdateLayout(); m_pInterface->Draw(); }

I use GitHub, a NAS, and blu-rays for really old things.

How will this work if I create my camera in code, but load a map with a light? How will the map know what camera to render the shadows to? What if someone wanted to create security cameras, 3D skyboxes, or even portals? How would they render shadows to the main camera and also the other cameras? I understand that this is the first pass on this but I'm not warm to this change.

I guess concept artist are a thing of the past. 🤷‍♀️

I'm waiting on a certain engine with tools I'll love, with the features I want, and performance I can't live without. 🙃

FYI, I tried to run the client on a fresh install, and I got the dreaded OpenAL32.dll was not found error.

I have only feedback I got were from those using a Cyrillic keyboard. A customer from Spain mentioned they just set their layout to the English QWERTY layout and everything is fine. There's also a disparity between lower case and upper case characters. For example, Leadwerks and Ultra always see the E key as 69, no matter if it's a capital e or lowercase E. SDL2 on the other hand only uses the 69 value for that key only if shift/capslock is active. If you have a chance, I guess you can take a gander at SDL's source code as a reference. I'm in the middle of reorganizing my setup but I plan to send time researching and experimenting this. I think I'll start with installing the Cyrillic keyboard layout and build a test app with SDL to see if it'll automatically fix this problem. Then it's just a matter of figuring out how it works within the library.

I have looked into this more. Apparently things like the Source Engine and SDL2 make a virtual keyboard and the end user or the input manager uses that value instead of the raw input. This includes arrow keys and all. I'm planning on installing more keyboard layouts to Windows and using the onscreen keyboard to test this and I would really hate to rely on another library for this functionality.