does anyone here have an opinion about the nebula engine?
Can somebody tell me if this has fixed the CPU burning bug?
[url]http://www.upurload.com/files/b9f93ed6b74d89cfcfb9c90a4e2dc445.zip[/url]
[QUOTE=Eleventeen;18982915]Can somebody tell me if this has fixed the CPU burning bug?
[url]http://www.upurload.com/files/b9f93ed6b74d89cfcfb9c90a4e2dc445.zip[/url][/QUOTE]
The font doesn't render correctly for me:
[media]http://i50.tinypic.com/xp4w9t.jpg[/media]
Also my CPU usage shoots up to 100% when it's running, so I guess that means I can confirm that it's not fixed.
[QUOTE=Eleventeen;18982915]Can somebody tell me if this has fixed the CPU burning bug?
[url]http://www.upurload.com/files/b9f93ed6b74d89cfcfb9c90a4e2dc445.zip[/url][/QUOTE]
no
So I [i]just[/i] got shadowmapping implemented in some rudimentary form. (This is purely a free-time project, so development is slow)
[img]http://img51.imageshack.us/img51/9567/screenshotsm.jpg[/img]
As you can see, I highlighted in red prominent artifacts where depth-texture is fighting with the recomputed values in lightspace. I'm not entirely sure how to fix this issue. Any ideas from those of you who have done similar things?
Also, not marked are a bunch of texture filtering artifacts. I need to make texture loading and filtering way better than it is right now. I believe it still uses bilinear filters and shitty compression on normal maps.
[QUOTE=Cathbadh;18985044]So I [i]just[/i] got shadowmapping implemented in some rudimentary form. (This is purely a free-time project, so development is slow)
[img]http://img51.imageshack.us/img51/9567/screenshotsm.jpg[/img]
As you can see, I highlighted in red prominent artifacts where depth-texture is fighting with the recomputed values in lightspace. I'm not entirely sure how to fix this issue. Any ideas from those of you who have done similar things?
Also, not marked are a bunch of texture filtering artifacts. I need to make texture loading and filtering way better than it is right now. I believe it still uses bilinear filters and shitty compression on normal maps.[/QUOTE]
If you slapped cool effects on everything and ditched the crappy textures (e.g. the grass) it would look a lot more impressive with minimal effort.
Still, p. damn cool.
Just enable anisotropic filtering, set the anisotropic filter to 16x, and add a small bias to the values in your shadow map (or, if you want, subtract a small bias from the values you compute when comparing with your shadow map). And disable any texture filtering on the shadow map.
[editline]06:43AM[/editline]
Or, if all your objects are closed, you can instead render their backfaces into the shadow map (instead of the front faces), and then you get automatic biasing without having to do anything.
Spent the morning re-coding animated doors so when you mouse over, they open and stay open, then when you mouse off they close. Just realised I forgot to make it not play when the door is locked, oops.
Just started working on a portal style game, if you have ever played ASCII portal/ seen the video then it is basically the same idea but done with normal graphics. Today I wrote a draft of the rendering code and gave it a test. to my surprise the code handled depth itself pretty well, though there are a few things I need to fix before it looks good.
I'm using SDL to get the 'view' at the entrances of the paired portal, then drawing those pixels behind the portal.
I'll get a screenshot up in the next few days once I fix some bugs.
ASCII Portal:
[media]http://www.youtube.com/watch?v=MnvxDk2sAZg[/media]
[QUOTE=Mattz333;18986493]Just started working on a portal style game, if you have ever played ASCII portal/ seen the video then it is basically the same idea but done with normal graphics. Today I wrote a draft of the rendering code and gave it a test. to my surprise the code handled depth itself pretty well, though there are a few things I need to fix before it looks good.
I'm using SDL to get the 'view' at the entrances of the paired portal, then drawing those pixels behind the portal.
I'll get a screenshot up in the next few days once I fix some bugs.
ASCII Portal:
[media]http://www.youtube.com/watch?v=MnvxDk2sAZg[/media][/QUOTE]
Sounds like an awesome project, can we see a screenshot of this draft?
Guys check this thing out, found it on gamdev
[hd]http://www.youtube.com/watch?v=NjUuAuBcoqs[/hd]
Isn't that... beautiful ?
[QUOTE=Xerios3;18987832]Guys check this thing out, found it on gamdev
[hd]http://www.youtube.com/watch?v=NjUuAuBcoqs[/hd]
Isn't that... beautiful ?[/QUOTE]
Wow. O_O
[QUOTE=Xerios3;18987832]Guys check this thing out, found it on gamdev
[hd]http://www.youtube.com/watch?v=NjUuAuBcoqs[/hd]
Isn't that... beautiful ?[/QUOTE]
I think it's made by somebody from this Forum.
[QUOTE=Pirate Ninja;18987974]I think it's made by somebody from this Forum.[/QUOTE]
-snip-
I actually think it's from gparent but im not sure.
[QUOTE=iPope;18987329]Sounds like an awesome project, can we see a screenshot of this draft?[/QUOTE]
[img]http://img514.imageshack.us/img514/1632/50698530.png[/img]
Still very basic, right now I'm going to get a larger viewspan.
If I wanted to write a parser for my own scripting language, what would be the best way to get started? Looking at an existing open-source parser, like the one for Ruby?
Wondering if anyone can comment on this. I am trying to figure out the best way to implement waiting in my program. What I mean is that I have a client and I want to do.
this->SendLogin();
LoginInfo li = this->WaitUntilLogin();
this->ConfirmLogin(li);
Basically it sends the login and waits until it receives the information about it.
I am just wondering what the best way to implement this. I am probably going to have to implement WaitForSingleObject and CreateMutex for the waiting. Not sure how I should implement it into the classes though.
I know I could do callback functions but I am going to end up with a lot of these that depend on one another and splitting them between functions will get messy very fast.
Right now the rough ideas are coming together, so any constructive comments would be appreciated.
Look at Squirrel or Lua.
-snip-
[QUOTE=high6;18990541]Wondering if anyone can comment on this. I am trying to figure out the best way to implement waiting in my program. What I mean is that I have a client and I want to do.
this->SendLogin();
LoginInfo li = this->WaitUntilLogin();
this->ConfirmLogin(li);
Basically it sends the login and waits until it receives the information about it.
I am just wondering what the best way to implement this. I am probably going to have to implement WaitForSingleObject and CreateMutex for the waiting. Not sure how I should implement it into the classes though.
I know I could do callback functions but I am going to end up with a lot of these that depend on one another and splitting them between functions will get messy very fast.
Right now the rough ideas are coming together, so any constructive comments would be appreciated.[/QUOTE]
this->SendLogin(this->callbackfunction);
this::callbackfunction, logininfo li:
this->ConfirmLogin(li);
something like that, although you'd probably have to dos some kind of universal callback system using void* and casts.
[QUOTE=Eleventeen;18990650]this->SendLogin(this->callbackfunction);
this::callbackfunction, logininfo li:
this->ConfirmLogin(li);
something like that, although you'd probably have to dos some kind of universal callback system using void* and casts.[/QUOTE]
But the login can end up taking time and the send login returns instantly.
[QUOTE=andersonmat;18990643]I'm not a 100% on this but what if you did a while loop when the login result is null?[/QUOTE]
Would be a waste of cpu cycles. That is what WaitForSingleObject is for.
[QUOTE=high6;18990688]But the login can end up taking time and the send login returns instantly.[/QUOTE]
no, end your login function with sendlogin and a pointer to the callback function and start the code again at the callback function
[QUOTE=Pirate Ninja;18988180]I actually think it's from gparent but im not sure.[/QUOTE]
Ahaha. Thanks, but if I made that, I would've plastered it all over the forum by now ;)
[url]http://www.gameprogrammer.com/fractal.html#ptI[/url]
I am trying to do : Midpoint Displacement in One Dimension with Cimg
Anyone care to elaborate in pseudocode/words/code how you would take care of calculating the points.
I thought of using a array like
0,0 - > 0,1,0 -> 0,0.5,1,0,5,0
Only how would I move everything in the array from a certain point 1 place further ?
[QUOTE=Eleventeen;18990992]no, end your login function with sendlogin and a pointer to the callback function and start the code again at the callback function[/QUOTE]
That is what I am trying to avoid because I have several functions that do this.
[QUOTE=high6;18992884]That is what I am trying to avoid because I have several functions that do this.[/QUOTE]
then create a modular callback system
itll be worth it in the long run
[QUOTE=nullsquared;18985913]Just enable anisotropic filtering, set the anisotropic filter to 16x, and add a small bias to the values in your shadow map (or, if you want, subtract a small bias from the values you compute when comparing with your shadow map). And disable any texture filtering on the shadow map.
[editline]06:43AM[/editline]
Or, if all your objects are closed, you can instead render their backfaces into the shadow map (instead of the front faces), and then you get automatic biasing without having to do anything.[/QUOTE]
Look closely at the image that I already provided. I am already only rendering backfaces to the shadowbuffer. That's why you see the z-fighting going on at the edge of the cube. I've also already been doing biasing. I do this by slightly altering the scale-bias transform matrix because it is a convenient constant in the shader.
I guess it isn't so much z-fighting in the shadow buffer as it is the simple granularity of the texels in the shadowbuffer. I could increase the bias, but then I get bad artifacts like the one you see around the top of the teapot, and the shadows don't quite line up right, or I could increase the resolution of the shadowmap. (which pretty much just increases memory usage. Rendering the shadowmap is really cheap as far as the pipeline is concerned.)
To prevent the Z-fighting, could you not just disable the Z-Buffer while rendering shadows?
I guess what I need to do is create stacked shadowbuffers so I don't have to draw the entire scene in one big high-resolution buffer. The objects far away in the scene (like the columns) don't require nearly as much detail, but the self-shading teapot in the foreground requires quite a lot of detail. I can draw the scene that's close to the camera with a zoomed-up shadowbuffer so I can get good resolution without sacrificing memory for shadows close to the camera.
Sorry, you need to Log In to post a reply to this thread.