What are you working on?

[hogsy]

would be nice for an ide to have funtion tabs so u can see in one window a flow of a function.

I thought Visual Studio had something like that?

Seeing as I've started feeling better I decided to write a more lengthy post here, and recently I decided to give myself a bit of a challenge. I've switched over to Linux (Ubuntu) as my main operating system since I want to get away from Windows as I'm not personally happy about the direction that Microsoft is taking it and I want to push myself into supporting Linux with both OpenKatana and future projects. I originally intended to switch over to use Ubuntu 13.04 which I had been testing on my laptop but it appears there are issues with certain libraries in the 64-bit version of 13.04 which I had been trying to set-up on my desktop so for now I've decided to use 12.04 until there are less issues.
So far the switch to Ubuntu has been a little bumpy but that's been mainly because of the bugs in 12.04 though they're somewhat bearable, and after some short time I had OpenKatana compiling and running in a virtual machine with Windows XP. Now I just need to work on getting it also compiling directly on Linux...

I regret to say that currently I am somewhat the only person working full-time on OpenKatana, and even then my main focus as of late has been pretty much purely on the engine and less so on the game itself as there has been some uncertainty regarding the path of which the project will take. Our level designer has recently vanished, both our assigned music artists are not interested enough for me to feel the need to waste time on them, our sound artist has stopped speaking to us due to his personal situation and Eukos has lately been more focused on his own personal project which is based on Katana (though I suppose this gives me a motive to do more work on the engine).
OpenKatana certainly isn't dead, but let's just say that because of how its development has played out I don't think people should expect it to be quite "complete". I personally blame my poor judgement and huge ego for most of the issues that have effected development, because at the end of 2011 we actually had a playable game and now OpenKatana is far from playable in any way whatsoever. It's funny really... In 2011 I was laughing at all the mistakes John Romero and his team made while working on Daikatana but we've actually ended up making the same mistakes they did. This all said however, we did actually only reach our first milestone in December of last year so this might just be a perspective thing in-terms of what we want to do to allow us to make the game we want to make, personally I'm willing to dedicate another year to that but it's looking more and more like we''ll end up throwing OpenKatana out the door and move onto our next project before the end of this year.

Anyway looking past all of that, today I spent sometime going over a few different things. Firstly we had a few calls to some known Windows-specific functions which I set out to replace by adding a replacement function that has both the Windows and Linux equivalents to our Platform library. Initially the idea for the Platform library was to do exactly what SDL was intended to do but as development has continued we've actually ended up using SDL and other alternatives over our own code but I've continued to develop our Platform library for smaller things which is making it pretty useful for use with our other sub-projects too and it also means that even the Game and Menu projects can also use features from there which is kinda neat I guess. Secondly I also went over some old code and left myself some notes on things I want to go over and tweak tomorrow, there's also a lot of work currently going into lighting for models right now but I'll speak about that when the implementation of my changes is a little more solid. I had also playfully spent a short amount of time working on some basic vehicle stuff in the game-code but stopped working on it after a short while so I could focus on OpenKatana again.

There's a few sub-projects that are part of our work on OpenKatana such as our editor and other tools but I'm sure we'll speak more about all of that when we're closer to releasing something.

Also if anyone here is interested in assisting us with Katana (the engine and tools) and OpenKatana then feel free to let me know! Having another programmer or modeller on the team would be pretty helpful...

[Baker]

personally I'm willing to dedicate another year to that but it's looking more and more like we''ll end up throwing OpenKatana out the door and move onto our next project before the end of this year.

It was a very ambitious project, at least you did make it as far as you did.

[ceriux]

personally I'm willing to dedicate another year to that but it's looking more and more like we''ll end up throwing OpenKatana out the door and move onto our next project before the end of this year.

It was a very ambitious project, at least you did make it as far as you did.

=( that really sucks.

[taniwha]

Not a lot. I've been throwing little green men at the atmosphere at about 2000m/s for the last month or so (playing Kerbal Space Program: dangerously addictive).

[leileilol]

kernel filtering models. WHY DOESNT IT WORK

fyi - flQuake has bilinear filtering working on models, works there. but that's slow and not really the same

[mankrip]

Good point, I'll take a look.

[jitspoe]

Putting together a new release of Paintball2. Now with loading screens.

[hogsy]

Been working on getting our tools compiling on Linux while going over and implementing a few suggestions for the engine.

Edit:
Spent quite sometime fixing some bugs and now I'm currently rewriting some of the weapon system so that it's more streamlined. It's going to be a long night...

Edit:
And it's now five in the morning, I think I'm done for today heh!

[revelator]

Dont Work yourself to hard Things take time.
Taking a break myself atm cause else ill burn out, but sometimes hard when you'r muse goes on a powertrip

[hogsy]

http://www.indiedb.com/games/openkatana/videos/walking-around#imagebox

A video demonstrating some small things, some of which are obvious and others not.
Also I suppose it might be a good idea to create my own thread for OpenKatana related updates?

[revelator]

Wouldnt hurt Things tend to go unnoticed in the bigger threads like this one.

[mankrip]

kernel filtering models. WHY DOESNT IT WORK

fyi - flQuake has bilinear filtering working on models, works there. but that's slow and not really the same

I got flQuake's filtering working in Makaqu, although the colors are wrong.

From what I've understood, there's no way to adapt that code to use kernel-based dithering without a significant slowdown, due to the way it calculate the texels' coordinates. The best way would be to rewrite all of the texture mapping code for MDL models, which is what I want to do. The original code looks quite messy, I'm sure it can be improved a lot.

I could just shoehorn the filter in, but I don't feel comfortable with the idea.

[leileilol]

I got flQuake's filtering working in Makaqu, although the colors are wrong.

That's because it's written for a 32-bit renderer - you'd have to use a big 18to8 or 15to8 lookup table for it to look right. I only really mentioned it for reference on how it worked with the spans, for potential clues on getting Kernel to work. I might use it for dithered spans mode though, since it's fun seeing true color data surviving in a palette.

[mankrip]

That's because it's written for a 32-bit renderer

It actually seems to be coded for a 24-bit one (or 32-bit, with the first 8 bits being unused or reserved for alpha), since it uses the 8 to 24 table and the LRP macro is run with 8-bit and 16-bit offsets. Here's my crappy code (by the way, you'll have to disable the x86 ASM codepath for it to work):

Code: Select all

// flQuake - begin
#define LRP(X,Y,L,IL,B) ((((((X>>B)&0xff)*IL)+(((Y>>B)&0xff)*L))>>16)<<B)
#define LRPM(X,Y,L,IL,M) (((((X&M)*IL)+((Y&M)*L))>>16)&M)
// flQuake - end
// mankrip - begin
extern int dither_kernel[2][2][2];
static int
// integers for dithering
   u
,   v
,   idiths
,   iditht
,   idiths2
,   iditht2
,   X
,   Y
,   XY0a
,   XY1a
,   XY0b
,   XY1b
   ;
// mankrip - end
void D_PolysetDrawSpans8_ColorKeyed (void)
{
   #if   0//id386
   D_PolysetDrawSpans8 (pstart);
   #else
   spanpackage_t
      * pspanpackage = pstart
      ;
   int      lcount;
   byte   *lpdest;
   byte   *lptex;
   int      lsfrac, ltfrac;
   int      llight;
   int      lzi;
   short   *lpz;

   do
   {
      lcount = d_aspancount - pspanpackage->count;

      errorterm += erroradjustup;
      if (errorterm >= 0)
      {
         d_aspancount += d_countextrastep;
         errorterm -= erroradjustdown;
      }
      else
      {
         d_aspancount += ubasestep;
      }

      if (lcount)
      {
         lpdest = pspanpackage->pdest;
         lptex = pspanpackage->ptex;
         lpz = pspanpackage->pz;
         lsfrac = pspanpackage->sfrac;
         ltfrac = pspanpackage->tfrac;
         llight = pspanpackage->light;
         lzi = pspanpackage->zi;
         // flQuake - begin
         #if 1
            if (lsfrac < 0x8000)
            {
                lptex -= 1;
            }
            lsfrac -= 0x8000;
            lsfrac &= 0xFFFF;
            if (ltfrac < 0x8000)
            {
                lptex -= r_affinetridesc.skinwidth;
            }
            ltfrac -= 0x8000;
            ltfrac &= 0xFFFF;
            #endif
         // flQuake - end

         do
         {
            if ((lzi >> 16) >= *lpz)
            if (*lptex != TRANSPARENT_COLOR) // mankrip - transparent pixels in alias models
            {
               // flQuake - begin
               #if 1

               #if 1 //#ifdef BILINEAR
               unsigned int c00,c01,c10,c11,xm,ym,xmi,ymi;
               unsigned int r, g, b, bcolor; // mankrip
               xm   =   lsfrac;
               ym   =   ltfrac;
               xmi   =   0xffff-xm;
               ymi   =   0xffff-ym;

               //Pal[255]   =   0xdeadbeaf;   //transparent color, used by particles and sprites
               //d_8to24table
               // mankrip - begin
               bcolor = *(lptex    ) * 3;
               r = host_basepal[bcolor + 0];
               g = host_basepal[bcolor + 1];
               b = host_basepal[bcolor + 2];
               c00   =
                  ( (r << 16) & 0xFF0000)
               |   ( (g <<  8) & 0xFF00)
               |   b;

               bcolor = *(lptex + 1) * 3;
               r = host_basepal[bcolor + 0];
               g = host_basepal[bcolor + 1];
               b = host_basepal[bcolor + 2];
               c01   =
                  ( (r << 16) & 0xFF0000)
               |   ( (g <<  8) & 0xFF00)
               |   b;

               bcolor = *(lptex     + r_affinetridesc.skinwidth) * 3;
               r = host_basepal[bcolor + 0];
               g = host_basepal[bcolor + 1];
               b = host_basepal[bcolor + 2];
               c10   =
                  ( (r << 16) & 0xFF0000)
               |   ( (g <<  8) & 0xFF00)
               |   b;

               bcolor = *(lptex + 1 + r_affinetridesc.skinwidth) * 3;
               r = host_basepal[bcolor + 0];
               g = host_basepal[bcolor + 1];
               b = host_basepal[bcolor + 2];
               c11   =
                  ( (r << 16) & 0xFF0000)
               |   ( (g <<  8) & 0xFF00)
               |   b;
               // mankrip - end

            //   c10   =   host_basepal[3 * *(lptex + r_affinetridesc.skinwidth)];
            //   c11   =   host_basepal[3 * *(lptex + 1+ r_affinetridesc.skinwidth)];

               c00   =   LRP(c00,c01,xm,xmi,0)|LRP(c00,c01,xm,xmi,8)|LRP(c00,c01,xm,xmi,16);
               c10   =   LRP(c10,c11,xm,xmi,0)|LRP(c10,c11,xm,xmi,8)|LRP(c10,c11,xm,xmi,16);

               bcolor = LRP(c00,c10,ym,ymi,0)|LRP(c00,c10,ym,ymi,8)|LRP(c00,c10,ym,ymi,16); // mankrip - edited
               // mankrip - begin
               r = host_basepal[ (bcolor >> 16) & 0xFF];
               g = host_basepal[ (bcolor >>  8) & 0xFF];
               b = host_basepal[ (bcolor      ) & 0xFF];
               *lpdest = BestColor (r, g, b, 0, 254);
               // mankrip - end

               #else // KERNEL DITHERED

               int x,y,x1,y1,c00,c01,c10,c11,xm,ym,xmi,ymi;
               unsigned int r, g, b, bcolor; // mankrip
               xm   =   lsfrac;
               ym   =   ltfrac;
               xmi   =   0xffff-xm;
               ymi   =   0xffff-ym;



            //   (byte *)r_affinetridesc.pskin
            //   r_affinetridesc.skinwidth
            //   r_affinetridesc.skinheight
               Y = ( (lptex - r_affinetridesc.pskin) / r_affinetridesc.skinwidth); //(pspan->v)&1;
               X = (lptex - (r_affinetridesc.pskin + r_affinetridesc.skinwidth * Y)) & 1; //(pspan->u + lcount) & 1;
               Y &= 1;

               #endif

               #else
               *lpdest = *lptex;
            //   *lpdest = ((byte *)acolormap)[*lptex + (llight & 0xFF00)]; // mankrip
               #endif
               // flQuake - end
               *lpz = lzi >> 16;
            }
            lpdest++;
            lzi += r_zistepx;
            lpz++;
            llight += r_lstepx;
            lptex += a_ststepxwhole;
            lsfrac += a_sstepxfrac;
            lptex += lsfrac >> 16;
            lsfrac &= 0xFFFF;
            ltfrac += a_tstepxfrac;
            if (ltfrac & 0x10000)
            {
               lptex += r_affinetridesc.skinwidth;
               ltfrac &= 0xFFFF;
            }
         } while (--lcount);
      }

      pspanpackage++;
   } while (pspanpackage->count != -999999);
   #endif
}


[ceriux]

i didnt make the arms.