Canardia

handbone = FindChild( model, "hand_right" ) handpos = EntityPosition( handbone, 1 )

These lines look at bit wierd: gx=Curve(MouseX() - GraphicsWidth() /2, gx, 10) gx=Curve(MouseX() - GraphicsWidth() /2, gx, 10) gy=Curve(MouseY() - GraphicsHeight() /2, gy, 10) It should be: gx=Curve(MouseX() - GraphicsWidth ()/2, gx, 10) gy=Curve(MouseY() - GraphicsHeight()/2, gy, 10)

It works also in Lua.

Well, first I want to use all possible techniques to get the maximum speed out of C++, and if it's still too slow, then I rewrite it in Fortran, or at least parts of it

Yeah but when it's ready, then I will have lots of fun with trying to optimize it I found an interesting new command in C++: alloca(), which allows you to use always stack memory, so that could be one thing to try out for maximum speed. Also compiling it with MinGW or GNU C++ should bring some more speed, and changing pointers to references (although that is not always faster, but quite often). And now I also learned that 64-bit can be faster, and not to forget that I could try to put some heavy and/or blocking CPU loops on other cores. And moving contants in front of expression. And redefining the typedef for float to double, although this increases speed only when memory is not moved around too much. And finally compiling it with Intel C++ compiler.

I think you're lying, you don't have any proof of that

A good real life example is making a verlet fluid simulation. It seems it needs a lot of power, and the fastest languages can barely handle 3000 particles at 60 FPS. Interesting is that 64-bit C++ is faster in many tests than 32-bit C++, maybe it's time to start testing 64-bit Windows? With LE3 I will finally have an engine which utilizes my GeForce 8800 fully, but does it utilize my Intel quadcore CPU also fully? it seems I would need to compile LE3 as 64-bit version to find out how much faster it would be than the 32-bit version.

I heard there will be also a new demo out soon, which allows limited programming with C++. That way you could already start writing your game, but to finish it you need of course to buy the full version, for technical and legal reasons. Usually a decent part of the game is also written in Lua, while the main language does the CPU intensive stuff and controls the game flow. In Lua for example often little model logics are done, like a spinning propeller, radar, or advertisement barrel/plate, but also switches, doors and physics joints are often done in Lua, because they need realtime programming to adjust the joints and pieces millimeter accurately.

The tutorials are in C++, and the demo has only Lua enabled. With the full version you can choose between 350 different languages, but of course only 5 have headers written by the community. The official SDK has only BlitzMax, C/C++ and Lua, but C#, Java, Pascal have been made and Fortran is coming end of this month (written by me), and Monkey might also come later on (also written by me). Basically everyone could write their own headers, by looking how the C headers are done. It took me only 2 days to learn Fortran, so a new language is also not hard to learn, especially if you have experience with various other langauges.

Just fresh out from the press: IBM will drop the Lotus part from the name, because it misleads many people who remember Lotus 1-2-3, Lotus Smartsuite, and yes even Lotus Notes which is often thought as ancient and dead. This is very cool, because it removes the long name nonsense and makes it clear that it's a real IBM product, and everyone knows that IBM is the leader in computer technology, mostly on the hardware side, but since the aquisition of Lotus also on the software side, and of course since the invention of Fortran. Also a SharePoint killer app will be included out-of-the-box with IBM Notes/Domino at some point. I don't have any details on this, but I would assume it would integrate IBM Quickr plugins in it, without needing a seperate IBM Quickr server. As I mentioned earlier on Vowe's blog, I converted from C/C++ to IBM Fortran. I never liked Microsoft C#.NET, because it's slow and bloated (and I just can't understand why someone would use Garbage Collection, since for me it means only that the programmer has no skills and discipline and should not program in first place), and IBM Fortran beats even C/C++ in speed and of course it beats both C#.NET and C/C++ in easyness to write code, thus increasing productivity a lot. IBM Fortran for IBM Notes/Domino is indeed a double RAD solution: radical performance combined with radically easy development. Usually they work reciprocally, but not with this approach. I have a plan to make Fortran 2008 a sexy and modern language, and it involves several steps: 1) Establish web site for Fortran 2008: http://www.fortran2008.com [DONE] 2) Establish IBM Domino based custom forum for Fortran 2008: http://www.fortran2008.com/forum [DONE] 3) Make a OpenGL and FreeGLUT demo for Fortran 2008, because it seems that nobody has done it before: http://www.fortran2008.com/posts/3D46BEECFB7212EDC22578C200420D3E [DONE] 4) Implement the most advanced realtime 3D engine for Fortran 2008 [Coming in 2011-07-30] 5) Implement IBM Fortran API for IBM Notes/Domino [Coming in 2011-09-30] 6) Continue implementing and supporting several other essential high quality libraries for Fortran, like sqlite3, sdl_net, etc... and make easy to follow tutorials so that even beginners can understand them. Source

Don't forget that Android is also Linux, so you can sum Linux+Android votes. Yes, I have also Android for PC in my deca-boot And vice versa: You can run Debian on an Android phone too (phone calls still work!).

Make a small buffer and render it to it.

Of course Josh can add these functions to the official DLL too, but I don't see any reason why we should have to wait when people need it now. I've suggested earlier too that the community could do some testing releases, and if they work well, Josh can add them to the official release. That's actually what he has done already many times: he added hnphan's SSAO, and in 2.43 he added Aily's SSAA. I'm sure there are also more examples which were added from the community to the official engine. It's kinda nice, so we have registered member's open source and commercial software working together. Others could learn something from this collaboration model So, I offer to support the Community Edition C/C++/Fortran headers metaofficially. The DLL can be done by anyone, we just need a changelog so we can keep the headers updated. klepto2 or Lazlo will be probably interested in supporting the C# headers.

I was just thinking the same, that I could release a custom engine.dll which has also TMesh GetWaterPatch(), but maybe you could add that too since you're already on it? Then I would need to make custom C/C++ headers too which go along with the custom engine.dll. Maybe the custom headers and custom engine.dll should be named a bit differently, to avoid accidental mixing up if they are in common system paths and include directories? For example: enginece.dll, enginece.h, enginece.cpp, where "ce" stands for Community Edition (and it's still only 8 characters, so Windows can handle it).

I have done a very simple routine for raising and lowering terrain. It can affect either a single point or a square area. Circular area would be easy to code too, just need to add a sin() and cos() function which call the point raising function.

I think I read somewhere that LE2 will be supported for a quite long time even after LE3 comes out, because there are already big game projects in development, and I think LE2 will be even still sold for some time after LE3 comes out. How ever it will go, I can always trust that Josh won't drop support for customers who are seriously developing with any Leadwerks product. And even in the worst case scenario, if Leadwerks for some reason didn't exist anymore, I have a backup plan: the source code. After checking legal rights and stuff, a source developers' committee could be formed who could then try to fix possible bugs or add new features, including a possible help of hired professional 3D developers.

Yeah Android should use Fortran (actually one guy got Fortran already running on Android, and he got like 25MFLOPS perfspec, which beats any phone today: most have like 7-8MFLOPS), and not Java or C++. However, the Java in Android is fortunately not the real slow and bloated and buggy Java from Oracle, but it's Dalvik, which is a binary compiled Java, so it's not too bad. And meanwhile the Android NDK is offering more and more C++ access with each update, so soon you don't even need Dalvik at all.

That's nice to hear, so the investment in Android development is secured

Yeah that trend is exactly how I see things going. Android is growing a lot, Symbian keeps growing but at slower pace until it starts to shrink rapidly, camp jumpers from Nokia have joined the also sinking boat of Apple, and WP7 has not really started to grow yet, if it ever will, depends on the camp jumpers.

Symbian is a dying platform. Shareholders have demanded that Nokia keeps supporting Symbian until WP7 is fully working on their phones. The history doesn't matter, but we have to look how the future will be, else we work for the wrong thing which we know to be obsolete. WP7 will have a increasing market share, and it's up to the developers and users how popular it will be. So you should not support WP7 if you don't want it to compromise Android's future market share. The market is not doing things itself, but it's the people's behaviour which does it.

I don't see Apple anywhere near 2nd place, not OSX, nor iOS. Linux and Android are the most popular platforms after Windows, as this Poll also clearly shows. Just let Apple die in peace, it was a nice platform (after Amiga of course) when it had a Motorola CPU, but those times are gone. I have already today a 1920x1080 video recording with HDMI connector Android phone with 48GB SSD, and end of this year HTC will bring a 3D video recording and playback (without glasses) Android on the market. I don't think Apple and Nokia will ever be able to do that.

Why would you guess that it is my guess? I'm usually not guessing anything, but gathering information. http://www.game-debate.com/games/index.php?g_id=1235&game=Minecraft

Minecraft is not a low end game, it's a mid-end game. It needs at least: Minecraft PREDICTED System Requirements Intel Processor - Celeron M 1.5GHz AMD Processor - Sempron 2200+ Nvidia Graphics Card - GeForce 8400 Series ATI Graphics Card - Radeon HD 2400 Series RAM (Memory) - 512 MB Hard Disk Space - 1 GB And it does not run on Intel GMA 950 and many other low price laptops, on which a low-end game would run. So basically the requirements for MineCraft are the same as for a LE2 game, except that it would run on LE2, but be too slow for decent gameplay. Here LE3 brings mid-end gaming to the missing features of LE2, since it can run on mid-end machines with decent speed.

But the problem is, it's not guaranteed that people with basic PCs would buy your game at all, indeed I know many casual computer users, who use only free software, and the maximum they want to spend on a PC is a 256KBit line. So in the end you would have two audiences: #1: People who want to buy your game, even if it's too low end for their PC #2: People who want to buy your game, and your game matches their PC features So I would say that the audience #2 would buy more copies of your game than audience #1. Of course you can make also 2 versions of the game, but that would be almost the same effort as making 2 games.

To make big profit with selling games, you don't need to hit the most machines, but the most people who are willing to buy your game. Even if you make a very low end game which runs on all machines, it might be that nobody wants to buy it, because it's not worth the money since both the gameplay and the visual quality is horrible. But if you make a hightech game which looks awesome and has some creative gameplay, but maybe not so much different assets, you might find a lot of people who want to buy it. I think Portal was that kind of game, very simple graphics, but very high tech and very creative gameplay.