Jump to content
  • Advertisement


  • Content Count

  • Joined

  • Last visited

  • Days Won


MarkK. last won the day on November 2

MarkK. had the most liked content!

Community Reputation

2522 Excellent

About MarkK.

  • Rank
    Advanced Member

Personal Information


  • Twitter
  • Github

Recent Profile Visitors

17290 profile views
  1. I think we're trying to view unity source, or something like that.
  2. MarkK.


    I do read the one pixel under the cursor directly but I'm using that back buffer twice per frame. Once to render the gizmo with a much thicker line width (6.0) pick from that. Clear and render the scene and thin lines (1.0) and present that.. yeah, requires a flush...didn't see a reliable way around it. I thought about ray casting or similar but for me the better path is to keep my head on the shader side. My goal is to see how far I can push it in that space...I have some post process (or I guess it's called deferred) hopes and dreams for the point lights humble attempt in the current source branch.
  3. MarkK.


    One of the things I seemed to have ignored in my hobby career (besides shadow) were the rest of the off screen buffer glory. Learned a bit of magic this week. We're still approaching my comfort zone, but uncharted territory is on the horizon.
  4. MarkK.


    Half way there. The when works fairly well, still need to wire on the what.
  5. MarkK.


    currently playing with custom gizmos for a placement tool. Just passed the visual representation bit, moving into the picking realm hopefully tonight. Wish me luck...
  6. MarkK.

    Ray Tracing a Tiny Procedural Planet

    Oh yeah, here we go...
  7. Huh, that's 'kinda neat moments... When you start over from scratch, design aspects become more important. I can't help what I have yet to discover but for the first of I suspect to be many, a happy moment. Mouse input ponder. The usual choice is to constantly redirect the mouse cursor to the center and act from the deltas. I'm starting to hate that. Something about it. So I changed it up a bit where I'm still clipping to the window but wrapping on the left and right edge. This feels better at least to not have to hide the cursor and still not allow to click focus lost. This is what my front end input handling looks like. Now I don't care about mouse sensitivity or delta timing on the camera yaw movement and perhaps puts a skill modifier back on the user. In a sense. #define imp_bool(x) if(input.pAction_map->GetBool(x)) #define imp_float(x) if(input.pAction_map->GetFloatDelta(x) != 0.0f) void Game::update(float deltaTime) { // user input pump imp_bool(cancel) { action_this_frame = quit; } // camera imp_bool(forward) { camera.move(FORWARD, deltaTime); } imp_bool(backward) { camera.move(BACKWARD, deltaTime); } imp_bool(left) { camera.move(LEFT, deltaTime); } imp_bool(right) { camera.move(RIGHT, deltaTime); } imp_bool(up) { camera.move(UP, deltaTime); } imp_bool(down) { camera.move(DOWN, deltaTime); } imp_bool(jump) { } float x = input.pAction_map->GetFloat(mouseX); // 0.0--->1.0 (l--r) float y = input.pAction_map->GetFloatDelta(mouseY); // upper ndc space also camera.ProcessMouseMovement(x * 360.0f, y, true); // yaw rotation based on horz screen position // pitch is accumulated and clamped process side // // todo: take into account the physical velocity of the mouse movement // high velocity user movement skips a beat at edge transition // solution: track previous movement and apply // priority: barely notice-able // POINT pt; ::GetCursorPos(&pt); // keep y component if(x<0) { // left and right screen edge cursor wrapping POINT pt_desired = { display.dims.width, 0 }; ::ClientToScreen(display.hwnd, &pt_desired); ::SetCursorPos(pt_desired.x-2, pt.y); // todo: refactor out the magic number } if(x>0.9985f) { // partial cause of refactor above (but it's decent smooth at 800 x_res) POINT pt_desired = { 0, 0 }; ::ClientToScreen(display.hwnd, &pt_desired); ::SetCursorPos(pt_desired.x, pt.y); } } we'll see how it goes.... Next up... [ Fix your timestep ] yup, I'm a believer. I've heard said, 'If you're going to do multiplayer, do it early'. Hard to argue, set up you system to play nice with something hard. I've always partially ignored the previous state - this state tween concept when it came to timing, but this really, really makes sense to me now. Oh yeah, we're keeping that... And the closer.. I'll be spending the week cleaning up and minor miscellaneous tweaks. I understand the importance of tidying up and now is the point to decide how I branch out from here cleanly. I feel now is a good time to give the [ RIP Method ] a serious follow through. I agree, repetition is key (at least for me) Thanks @DavinCreed, that's helping. I've only started to browse. Have to go to work for now though.
  8. Action Shooter Genre appreciator and dabbler in, wait look at that moments... :) 

  9. MarkK.

    Khronos Releases Vulkan Unified Samples Repository

    Perfect. I indirectly found three glTF related projects of interest applicable to the previous spec. I wish we had public resources like this in the yester-year. Takes time to build it up I suppose. edit: we don't want anyone else handling the news lol...
  10. MarkK.

    The Story Begins...

    Progress Image(191031 Working on door handling makes me remember UDK. Today I push my project to be a little more visible. https://github.com/MarkKughler/NubDevice [dev.azure : NubDevice] project development board Over time, I've glued together more than a handful of starter frameworks. This time I've changed it up a bit and tried for a bit of consistency from the start so new objects feel natural to use. This is working well for shaders, textures, input and the display. I'm actually happy with it as a start. High-lights that may be interesting: start of custom gui (inspired by fpl::flatui) updated stack based state management (menu[n]/play/pause) push in pop out (or maybe old school, first in, last out) .dae loader (geometry and skinning info) (todo: keyframe) currently broken TBN path Hope to finally do some interesting shader work this time around and feel better prepared for the next game challenge. I'm a fan of openAL. That parallel to openGL way is a good thing in my book. Start the device, fill a buffer, attach to a source. Bounce a listener around. It's good. As my interface, that's all I need. A buffer load and a source to play. Fairly clean solution for home brew. Since I don't see a need for the buffer because we don't alter it, I'm thinking of making that init a one liner, except form maybe a reload need. Play from id is great. buffer_click = sound.add("buttonClick"); source_click = sound.attach(buffer_click); // and anywhere alSourcePlay(source_click); // sure why not :) The main character := reveal. nope...still not telling. I'll let you know when I do. heavy current inspiration is an first person arm mesh on the table. Still dreaming up what could go with it proper third person. Choices are fun.
  11. MarkK.


    Thar'll be your guy high tech group. Fun aside, lot of work to get there.
  12. MarkK.

    The Story Begins...

    I really want to do this bad. lol. (boat load of great art over here now. Few decent modular packages ready for tweaking) enough character icons to put front and center...just 'gotta make that code sing. I know...should use a upper tier engine...it's just not the same kind of puzzle.
  13. MarkK.


    @Rutin What a wild month. I knew I had an unattended comment in the queue. I've been at it pretty hard, Finally settled on some realistic moves, which are still a bit ambitious for one guy. I'm torn between Unity and home brew which is winning over here. 'Lov the Unity, fantastic tool. For me, home brew scratches that spot just right, so it's hard to resist. With that said, I'm fairly happy with a new raw ogl3 framework I'm putting together. The focus is on getting up and running quickly with the common elements (menu, game state, timing, display management, input, shader management, logging....) so I can get serious with the Challenge Group. I was leaning toward an upcoming code review and/or community source share. (deciding if it's worth it. been wanting to put something in that git thing for quite a few years now.)
  14. At most, a basic map explorer. pixel art is hard...
  • Advertisement

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!