• What are you working on? May 2012
    2,222 replies, posted
[QUOTE=ShaunOfTheLive;36036048]Make a game called Super 3 or Mental Blox [url]http://boardgamegeek.com/boardgame/10578/super-3[/url][/QUOTE] This is very do-able. It's going to be dropbox hosted multiplayer. idk lol
[QUOTE=Ziks;36034194][img]http://puu.sh/vRTC[/img] Surprisingly I'm not having any frame rate issues yet, although I'm mostly only drawing LOD models. It took quite a while to load the models and textures though, but I think I know where I'm being inefficient.[/QUOTE] Did you cut around LS yourself or were there already "zones" in the game data that you used?
[QUOTE=Ziks;36033801][IMG]http://puu.sh/vRAb[/IMG] Blargh alpha blending issues Edit: Again, source is [URL="https://github.com/Metapyziks/SA-World-Viewer"]here[/URL] if you want to browse / help[/QUOTE] OpenTk right? This would be awesome for helping ,e learn opengl and the gta sa formats! I'll look into it tomorrow! Awesome!
[QUOTE=Map in a box;36036915]OpenTk right? This would be awesome for helping ,e learn opengl and the gta sa formats! I'll look into it tomorrow! Awesome![/QUOTE] No offense to Ziks, but I had a look at his code and I really wouldn't if I were you. Usual suggestions: [url=http://www.opengl-tutorial.org/]one[/url] [url=http://arcsynthesis.org/gltut/]two[/url] [url=http://ogldev.atspace.co.uk/index.html]three[/url] [url=http://openglbook.com/]four?[/url]
[QUOTE=Lexic;36037044]No offense to Ziks, but I had a look at his code and I really wouldn't if I were you. Usual suggestions: [url=http://www.opengl-tutorial.org/]one[/url] [url=http://arcsynthesis.org/gltut/]two[/url] [url=http://ogldev.atspace.co.uk/index.html]three[/url] [url=http://openglbook.com/]four?[/url][/QUOTE] Someone forgot [URL="http://open.gl/"]open.gl[/URL].
Perhaps it is just my imagination/self-serving opinion, but I find this terrain peculiarly fun to drive on. [img]http://eagle.undo.it:8083/img/kintel_90.jpg[/img] [img]http://eagle.undo.it:8083/img/kintel_91.jpg[/img] It offers a good balance between close quarter battles and long-range scouting without splitting the map up into too many isolated areas. I could see some interesting strategies develop with this pattern...I may need to revise the final terrain algorithm with this in mind. I'm kind of tempted to just leave it as-is and work on other (more difficult) parts of the game, coming back to terrain generation later. Maybe I'll just kill the camera roll/tilt and keep it. I'd need to pull out my bullshit fabricator to fit an explanation for these land features into the canon, though.
Your game is reminiscent of [url=http://www.youtube.com/watch?v=hwMOsDrbfd4]Thunder Brigade[/url].
[QUOTE=HiredK;36032463]Am I missing something or are you guys rating him down because of the look of the GUI? Seriously I'm starting to get tired of the majority of progs here thinking there's so good that they can just rate people down because they don't like [b]ONE[/b] thing about the project. In the list of people that rated him disagree, half of those don't even have a project on their own. Really I don't understand how someone can post 20 lines of code and get 20 programming king but when someone post a real project he get that kind of rating. Criticism is constructive only when done right, rating dumb/disagree without giving other reason then is very bad criticism and he got what... 64 agrees? I know ratings are not that important but seriously can't you guys see the amount of work but in this IDE? CountNoobula I think that your project is fantastic and keep up the good work :v:[/QUOTE] Sorry for brining this up again, but the argument you gave sounds like a typical "DeviantArt" argument to me. I don't have to have made my own IDE to say someone else's looks like shit. Besides, I think it's safe to assume most people here will have had their fair share of experience when it comes to programming. And even then, I don't see why people can't just rate someone down for not liking something about the project, even if it is just one thing.
[QUOTE=q3k;36034007]I don't think I have to be schooled about [url=http://code.hackerspace.pl/q3k/Cucumber/log/?h=feature-x86-64]the difference between a kernel and an OS[/url]. Or about [url=http://www.facepunch.com/threads/1181032?p=35943541&viewfull=1#post35943541]embedded system development[/url], for that matter. And if we want to get anal about definitions (you sure seem to want to!), then the Linux kernel itself doesn't offer any shell (or 'console', as you like to call it) whatsoever - it's the userland (usually via /bin/*sh on full-grown systems and /bin/sh from busybox on small systems) that does. And no, embedded devices usually refrain from using a full-blown GUI not because they run Linux - they do so because they don't require a GUI. It's that simple.[/QUOTE] Hi again! Any reason you moved off github for cucumber?
[QUOTE=Lexic;36037044]No offense to Ziks, but I had a look at his code and I really wouldn't if I were you.[/QUOTE] Is it something that I can improve on? Do you have any specific things I should look into fixing? [editline]21st May 2012[/editline] [QUOTE=Xeon06;36036491]Did you cut around LS yourself or were there already "zones" in the game data that you used?[/QUOTE] I'm only looking at item placement files (.ipl files) that involve LS.
[QUOTE=Ziks;36037886]Is it something that I can improve on? Do you have any specific things I should look into fixing?[/QUOTE] Well I'm not [i]entirely[/i] sure since your code is kinda confusing, but you seem to be using deprecated OpenGL. So you could...not do that? I'm mainly confused because you seem to be using shaders, VBOs and glBegin all at the same time, which would be another reason not to use your project as a learning resource: It's quite hard to tell what's going on.
[QUOTE=Lexic;36038138]Well I'm not [i]entirely[/i] sure since your code is kinda confusing, but you seem to be using deprecated OpenGL. So you could...not do that? I'm mainly confused because you seem to be using shaders, VBOs and glBegin all at the same time, which would be another reason not to use your project as a learning resource: It's quite hard to tell what's going on.[/QUOTE] I don't think I use glBegin anywhere, I only draw from VBOs.
Non-Deprecated OpenGL vs Unicode Support
[QUOTE=Ziks;36038194]I don't think I use glBegin anywhere, I only draw from VBOs.[/QUOTE] [url=https://github.com/Metapyziks/SA-World-Viewer/blob/master/GTAMapViewer/Graphics/ShaderProgram.cs#L286]GTAMapViewer/Graphics/ShaderProgram.cs[/url]?
[QUOTE=Lexic;36038206][url=https://github.com/Metapyziks/SA-World-Viewer/blob/master/GTAMapViewer/Graphics/ShaderProgram.cs#L286]GTAMapViewer/Graphics/ShaderProgram.cs[/url]?[/QUOTE] I never use ShaderProgram.Begin() and ShaderProgram.End(), just ShaderProgram.StartBatch() and ShaderProgram.EndBatch(). Begin and End are mostly for debugging so I can draw vertex data without a VBO. [editline]21st May 2012[/editline] The program spends 0.9 seconds loading models and 115 seconds loading textures. I think I can spot where I can optimise.
[QUOTE=Lexic;36038206][url=https://github.com/Metapyziks/SA-World-Viewer/blob/master/GTAMapViewer/Graphics/ShaderProgram.cs#L286]GTAMapViewer/Graphics/ShaderProgram.cs[/url]?[/QUOTE] It's not used anywhere.
Ah. Sorry for the misunderstanding. As I said, the structure is kind of hard to follow when just looking at the repo.
[QUOTE=Ziks;36038236]The program spends 0.9 seconds loading models and 115 seconds loading textures. I think I can spot where I can optimise.[/QUOTE] [url]http://www.opengl.org/registry/specs/EXT/texture_compression_s3tc.txt[/url]
[QUOTE=Philly c;36038543][url]http://www.opengl.org/registry/specs/EXT/texture_compression_s3tc.txt[/url][/QUOTE] Yeah, when I take out my own texture decompression it takes 4 seconds. I'll try offloading the work to OpenGL like you suggest. [editline]21st May 2012[/editline] [img]http://puu.sh/vWvP[/img] Mirrors edge mode
[QUOTE=Philly c;36038543][url]http://www.opengl.org/registry/specs/EXT/texture_compression_s3tc.txt[/url][/QUOTE] I wonder how many other specs there are referencing things from DirectX.
[QUOTE=swift and shift;36037765]Any reason you moved off github for cucumber?[/QUOTE] I just moved my development branches off there, as at that stage I mostly use a remote to be able ot share the code between many machines, and not as a publishing / collaboration mechanism. I will continue to push major updates to github, just not as often.
Garry, if you read this, could we have a GitHub account link added to user profiles on facepunch (like steam account, deviant art, xbl etc)? [editline]21st May 2012[/editline] I mean if something as small and obscure as Google+ has one, you may as well add GitHub
Why not BitBucket as well?
[QUOTE=Lexic;36037044]No offense to Ziks, but I had a look at his code and I really wouldn't if I were you. Usual suggestions: [url=http://www.opengl-tutorial.org/]one[/url] [url=http://arcsynthesis.org/gltut/]two[/url] [url=http://ogldev.atspace.co.uk/index.html]three[/url] [url=http://openglbook.com/]four?[/url][/QUOTE] What do you mean by that? :saddowns: [editline]21st May 2012[/editline] GitHub, BitBucket, and maybe google code.
Could someone give a few convincing arguments why I should use a version control? I have used git before (which was a pain in the ass mostly) and Hg (which was more pleasant) but I can't say I see the benefits. Yes I could potentially lose code but all the important stuff is on dropbox (requires no effort on my part). Also, I was never in a situation where I desperately needed to salvage code I deleted. It just doesn't happen, and if I actually need something I deleted I know I did it with a reason and write a new, better version. Is there something obvious that I'm missing? Because right now it only seems like one extra thing to keep in mind (committing and pushing) and extra work to do. Maybe I just haven't worked on a large group project before but I could definitely see how that could be done without too much trouble without Git, SVN or Hg
[QUOTE=Jookia;36037328]Your game is reminiscent of [url=http://www.youtube.com/watch?v=hwMOsDrbfd4]Thunder Brigade[/url].[/QUOTE] I thought of Fury3 when I first saw his screens. [url]http://www.youtube.com/watch?v=6_8tSpXLuK8[/url]
[QUOTE=Darwin226;36039477]Maybe I just haven't worked on a large group project before[/QUOTE] Yes. Being able to work independantely on a large code base with multiple people involved at the same time, without having to resort to sending out emails telling people to not touch stuff is hell without a (D)VCS. A (D)VCS lets you manage development branches more easily, lets you quickly revert changes, merge different versions effortlessly, maintain multiple versions and releases at once while letting you quickly move code commits between them...
[QUOTE=Darwin226;36039477]Could someone give a few convincing arguments why I should use a version control? I have used git before (which was a pain in the ass mostly) and Hg (which was more pleasant) but I can't say I see the benefits. Yes I could potentially lose code but all the important stuff is on dropbox (requires no effort on my part). Also, I was never in a situation where I desperately needed to salvage code I deleted. It just doesn't happen, and if I actually need something I deleted I know I did it with a reason and write a new, better version. Is there something obvious that I'm missing? Because right now it only seems like one extra thing to keep in mind (committing and pushing) and extra work to do. Maybe I just haven't worked on a large group project before but I could definitely see how that could be done without too much trouble without Git, SVN or Hg[/QUOTE] Branches. Being able to find out when a bug was introduced between commits. Teamwork. Revision control. Backups. As for not losing code, I hope you have a great backup strategy.
[QUOTE=Darwin226;36039477] Maybe I just haven't worked on a large group project before but I could definitely see how that could be done without too much trouble without Git, SVN or Hg[/QUOTE] You don't even need a large team to see the benefits of a version control system. I've been using Git for small team projects (~3 people on a school project for example) and it's very nice to be able to play with the code in a separate branch so you don't break anything in the process, merging the branch back if your experiments work. It also helps prevent people from overwriting each others' code because version control software is smarter than a blind copy-and-replace.
So far, none of those are really sticking to me. Branches? Revision? Meh... If I'm not working with other people the most convincing reason to use VCS would be to feel more professional TBH.
Sorry, you need to Log In to post a reply to this thread.