Advertisement Jump to content
  • Advertisement


  • Content Count

  • Joined

  • Last visited

Community Reputation

253 Neutral

About fyhuang

  • Rank
    Advanced Member
  1. fyhuang

    Network stability testing?

    Ah, sorry, I'm talking about clients connected solely on my LAN: no Internet players. All traffic therefore goes through the router; it's just that one computer is connected by ethernet to the router and the rest use the wireless. Basically I want to see if the router starts dropping packets on the wireless under high load, or multiple clients sending/receiving at the same time (which is what I suspect). I've found a number of Linux-based tools, but I have one Linux computer and multiple Windows computers connected to my network, so a cross-platform tool would be preferred. Thanks again!
  2. Hi all, I've been having problems recently with my wireless router (it's a D-Link) and gaming over the wireless. The games I've been playing (Call of Duty) I think use UDP broadcast to find games (which is not working), and also communicate in-game through UDP. I don't know if the problem is with my router or not, as clients are sometimes disconnected randomly or "lag" a lot. My question is: do you know of/is there any good network stability testing software I could use to get to the bottom of my problems? I've tried setting a ping from one computer (wired) to another on the wireless, but an entire night of pinging returned no dropped packets. I'm guessing that packets might start dropping when the load gets higher. Would, say, network throughput software also work for my purposes? I'm mostly interested in, by the way, free and/or open-source solutions; while annoying, this problem is not high-priority enough for me to spend real money on. Thanks in advance!
  3. All of this sounds very interesting, thanks guys! I'm actually a little bit inclined to agree with esrix, about the whole "caring about others' turns" thing. Considering the rest of the comments here, I'm not so sure "simultaneous turns" or "phase-based" movement is the best idea... but I still think it would be nice to allow players to do something while others' turns are proceeding... Here's another possibility: what if the player was allowed to move units during his turn, but could perform research, change build queues, and so on at any time, including during others' turns? Turns could be limited to being very short (like 3 minutes or something), and that would allow a higher degree of player interactivity. Even more extreme, how about this: players could "explore the map" (i.e. reduce fog of war) during non-turn times? perhaps using spy systems or something like that. Or there could be no fog of war at all: that might make it easier for players to care about other players' turns, as they could see everything the other players were doing. Thanks!
  4. Hmm, interesting. I had forgotten about Final Fantasy, although my idea was more along the lines of being "RTTB" on the "adventure map": think, say, Heroes of Might and Magic but in semi-real-time. So, one would command armies of some sort to move around on the world map in turns. Thanks!
  5. Hello, I've been sort of floating an idea around in my head recently about a turn-based, but still real-time, style of gameplay (I'll explain in a minute). I don't know when this will find its way into a real game, but oh well. I've been trying to find a sort of compromise multiplayer gameplay that will help introduce non-gamers into video gaming, and this is what I've come up with. I would like this to be workable in large multiplayer situations, as it is easier to enjoy a game (IMHO) when you're playing with friends. So first, I looked at turn-based strategy games, and found that they were perhaps too slow for big-group multiplayer play. Everybody has to wait for everybody else, and that's perhaps not enough engagement for someone to get "hooked." I then looked at real-time strategy games, and decided that a truly fun RTS (StarCraft) needs to be relatively fast-paced... but that comes with the danger that the beginner could simply be rushed and have to sit out for an entire game, having done almost nothing. My idea is this: what if there was a game, modeled perhaps after FreeCiv (and the like), that was turn-based, but also real-time at the same time? Explanation: the game plays by turns. Each turn ends when all the players have ended their turns (perhaps at different times, perhaps with a time limit), then the next turn begins simultaneously for everyone. During each turn, however, all players can make moves at once; in other words, each turn is "shared" by all the players. Units will have a maximum moves-per-turn value, of course. I think the advantage to this compromise of sorts is that it doesn't require lightning-fast reflexes and keyboard shortcuts and extreme skill to be able to play effectively, but rather favors more careful strategic planning, due to the turn-based nature; however, one can still gain an advantage by moving before another player in the same turn. What I'd appreciate feedback on is this: is this a feasible game mechanic (i.e. any conflicts due to the mixed nature can be reasonably handled), and more importantly, do you think it would actually be easier for beginners to pick up (dependent on other aspects of the design, of course), and would they enjoy the game more because of its slightly faster-pace? Thanks in advance for your thoughts!
  6. I set the clamp values you suggested, and it seems to be fixed now! Thanks!
  7. Hi, I'm in the process of coding/debugging a skybox, and I've run into a small problem: there are thin lines between the faces of the cube I'm using. Here's a screenshot to illustrate: And a magnified crop: This is the code I'm using the draw the skybox (C++): double hel = m_EdgeLength / 2.0; RenderDevice::get().disable( GL_DEPTH_TEST ); // Top m_Textures[0]->apply(); glBegin( GL_QUADS ); glTexCoord2f( 0.0, 0.0 ); glVertex3d( -hel, hel, -hel ); glTexCoord2f( 0.0, 1.0 ); glVertex3d( hel, hel, -hel ); glTexCoord2f( 1.0, 1.0 ); glVertex3d( hel, hel, hel ); glTexCoord2f( 1.0, 0.0 ); glVertex3d( -hel, hel, hel ); glEnd(); // Left m_Textures[1]->apply(); glBegin( GL_QUADS ); glTexCoord2f( 0.0, 0.0 ); glVertex3d( -hel, hel, hel ); glTexCoord2f( 0.0, 1.0 ); glVertex3d( -hel, -hel, hel ); glTexCoord2f( 1.0, 1.0 ); glVertex3d( -hel, -hel, -hel ); glTexCoord2f( 1.0, 0.0 ); glVertex3d( -hel, hel, -hel ); glEnd(); // Front m_Textures[2]->apply(); glBegin( GL_QUADS ); glTexCoord2f( 0.0, 0.0 ); glVertex3d( -hel, hel, -hel ); glTexCoord2f( 0.0, 1.0 ); glVertex3d( -hel, -hel, -hel ); glTexCoord2f( 1.0, 1.0 ); glVertex3d( hel, -hel, -hel ); glTexCoord2f( 1.0, 0.0 ); glVertex3d( hel, hel, -hel ); glEnd(); // Right m_Textures[3]->apply(); glBegin( GL_QUADS ); glTexCoord2f( 0.0, 0.0 ); glVertex3d( hel, hel, -hel ); glTexCoord2f( 0.0, 1.0 ); glVertex3d( hel, -hel, -hel ); glTexCoord2f( 1.0, 1.0 ); glVertex3d( hel, -hel, hel ); glTexCoord2f( 1.0, 0.0 ); glVertex3d( hel, hel, hel ); glEnd(); // Back m_Textures[4]->apply(); glBegin( GL_QUADS ); glTexCoord2f( 0.0, 0.0 ); glVertex3d( hel, hel, hel ); glTexCoord2f( 0.0, 1.0 ); glVertex3d( hel, -hel, hel ); glTexCoord2f( 1.0, 1.0 ); glVertex3d( -hel, -hel, hel ); glTexCoord2f( 1.0, 0.0 ); glVertex3d( -hel, hel, hel ); glEnd(); // Bottom m_Textures[5]->apply(); glBegin( GL_QUADS ); glTexCoord2f( 0.0, 0.0 ); glVertex3d( -hel, -hel, hel ); glTexCoord2f( 0.0, 1.0 ); glVertex3d( hel, -hel, hel ); glTexCoord2f( 1.0, 1.0 ); glVertex3d( hel, -hel, -hel ); glTexCoord2f( 1.0, 0.0 ); glVertex3d( -hel, -hel, -hel ); glEnd(); Z buffer testing is turned off, and blending is also turned off. I know it's probably not a terribly complex problem, but any help that you guys can give me would be greatly appreciated. Thanks in advance!
  8. fyhuang

    GAIM Issues need help asap

    Well, I wouldn't think that MSN is the most well-documented protocol, first of all, so I'm already pretty impressed that the gaim people have pulled off a reverse-engineer of MSN (if that is actually what they did, I'm not sure about the details). Have you filed a bug-report? If you run the debug version through a debugger, figure out which function it's crashing in, and (possibly) provide a core dump to the gaim developers, you could potentially apply a quick fix yourself or help the gaim developers fix the problem for future releases. About your problem specifically, I don't know, but once I had a crashing problem with proxies, and I was able to apply a quick fix to the source code (by checking which function it was crashing in through a debugger) until the gaim developers had fixed the problem. Cheers!
  9. fyhuang

    How to pause a game (Part 2)?

    Well, this is how I handle such a situation: in my Input class's getKey() function (or input handling function, or whatever), which returns a boolean specifying whether a certain key is pressed: bool getKey( SDLKey key ) { return m_Keys[(int)key] && !m_Lock[(int)key]; } So a key is only 'pressed' if the corresponding index in the lock array is not. Both arrays are boolean arrays. When a key is pressed down: if ( event.type == SDL_KEYDOWN ) { m_Keys[(int)event.key.keysym.sym] = true; } And when a key is released: if ( event.type == SDL_KEYUP ) { m_Keys[(int)event.key.keysym.sym] = false; m_Lock[(int)event.key.keysym.sym] = false; } The final step in this is, in your game loop (or wherever you put this): if ( Input.getKey( SDLK_p ) == true ) { // Pause the game Input.setLock( SDLK_p ); } So basically, keys can be "locked", so that once you've processed the key once, you won't have to process it again until it is pressed again (because releasing a key also releases its lock automatically). The lock must be activated from the function that tests for the keypress and acts upon it. There's probably a better way of doing it, but mine works, so whatever :). Hope it helps!
  10. fyhuang

    PCI bus compatibility?

    Quote:Just making sure here, you do know that PCI-X and PCI-Express are entirely different things, right? Yes. My question is of compatibility between PCI and PCI-X devices/buses, not between PCI and PCI-express. Thanks!
  11. fyhuang

    PCI bus compatibility?

    Quote:A "normal" 32-bit PCI card will work with a PCI-X slot as long as the card to be put in utilizes 3.3v level logic. PCI-X version 2.0 is also compatible -- 3.0 may be as well. Would I be correct in stating that most off-the-shelf PCI cards use 5v logic (i.e. a Creative sound card)? If that is the case, then I will not be able to plug that card into a PCI-X slot? Thanks!
  12. fyhuang

    PCI bus compatibility?

    Hey all, For those of you who are more attuned to your hardware than I am... I've been wondering about the compatibility of the different PCI standards/bus variations that there seem to be. For example, 64-bit PCI-X on server motherboards, 32-bit PCI, 66 MHz and all that kind of stuff. A couple of basic questions: Is it possible to plug a "normal PCI" card into a PCI-X slot and have it work? i.e. is PCI-X backwards-compatible with PCI? Similarly, will a 32-bit PCI card work in a 64-bit PCI slot? Will a 66MHz (PCI?) card work in a 133MHz (PCI-X?) slot and/or vice versa? I really don't understand what the difference is between these PCI variants and whether they will all work together fine. Thanks in advance!
  13. fyhuang

    Access control in Python?

    If not Python for configuration files, then what do you suggest? These will be somewhat like Portage ebuilds, in that they will also define instructions for compiling/installing the package (i.e. some packages have weird compilation instructions, and some binaries require post-install commands). Portage "jails" the executable code it runs somehow, so that writing outside a special "work/temp" folder results in an access violation... which is what I want to achieve also.
  14. Hello, I'm trying to write a Python program that uses Python scripts for its configuration. This program is intended to be a package management tool for Linux boxes, and thus will be running with super-user privelages most of the time. However, since Python configuration scripts will be distributed with all packages, I need a way to restrict the power of the Python configuration scripts. Thus, I'm wondering if it's possible to implement some sort of access control in Python. I did a quick web search for Python and access control, and nothing really interesting came up. If anyone has any suggestions, they would be greatly appreciated. Thanks in advance!
  15. fyhuang

    Gah! I can't think of a good title for a game!

    Quote:They came from the top right/left corner Yeah... unfortunately, it is actually a vertical scroller, not a *diagonal* shooter [smile]. So enemies still come out from the top, but the bg (and ground enemies) are "iso-diagonal". I'm liking "Zero Point" ATM (although it makes absolutely no sense at all). Cheers!
  • Advertisement

Important Information

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