kuroi neko

The happy king of a tiny hill, are we? As for the rest of the mighty active DiF squadrons, otetsuki tokubetsu kôkûtai hereby officially anounces that it will only take part in official scrap metal piling contests. If you're especially nice, we might consider chute folding events or friendly kills challenges, though . edit: of course not, wooden planes recycling would be the mission of a completely different air force! [ September 14, 2005, 10:36 AM: Message edited by: kuroi neko ]

Agreed, as it is now all the buttons are too similar and close to each other for comfort. In 45s turns it's not a big deal, but for blitz games that could cause lots of nail biting and hair pulling . If moving the buttons around proves to be difficult due to implementation constraints, a different color or style for each action would already be an improvement.

Sorry for being a bit dense, but when you say "same" cumulative and available XP, will that be subject to the 100 points limit, or the "true" total (like 1 or 2 K points for instance)? Along with a few players in the lobby, we were wondering if the appearance of "über" pilots after beta phase would not create some huge gap for the new players, who would not even be allowed to fly against the "aces", be restricted to puny planes while old brass would access the whole arsenal immediately, etc. I was thinking retaining experience, kills and skills up to the 100 points limit would be already not bad, and less intimidating for newcomers. Add maybe a free "escape death" for old pilots if you feel really generous and I'll be more than happy with that .

About AI, yes they are very predictable and get a bit funny in the head when it comes to discarding. Among other things like trying to attack whatever the cost, always matching altitude when possible, and not taking into account opponent status except for choosing between possible targets (attacking the one with less cards, but following blindly a plane having 3 times as much in hand). Luckily for them, the "draw at will" and "summon hot cards" special skills manage to even the odds a bit. Otherwise, pruning over-experienced demo players could become an issue . About pictures, you can actually change the both of them when you create a new wingman (even though the leader picture choosing slider appears greyed out, it works). Still a bit strange, but nothing prevents you from restoring original picture, or even change it (the transfiguration of promotion? ). [ September 10, 2005, 06:03 PM: Message edited by: kuroi neko ]

the idea behind intentional disconnect was rather to escape death by disconnecting at the critical moment (especially for an high scorer ace), or to scrap a game if the initial hand is was not good enough or whatever attack/defence sequence did not work out as intended. I would say there is little you can do to detect red hot cheating, but still a few clues may be gathered, for instance to allow an admin to look for the smoking gun if requested. By storing dated events like game start, moves and of course disconnect time over a few games, one can quite easily see if the disconnection date is tied to some repetitive circumstances (a few seconds after game start, just before a crippling blow is dealt, etc.). But of course this kind of recording could soon intrude into the privacy of players. Anyway, I personally would be happy to get rid of the bailout russian roulette for now .

I concur, actually NorthStar and I were forced to fight each other last night while we were expecting our first squadron team fight! How frustrating

If the server-side game keeps track of each move from each player, and is able to resolve the game alone (i.e. not needing the client to move any plane), I think a possible answer would be to let the game be played by bots on the server, taking only into account negative results for the disconnected player (i.e. fatigue and death, but neither kills nor experience). Players still connected could deal with the bot replacing the player as usual. This way, an intentionally disconnecting player would still have to face the consequences of his cowardice (i.e. could not escape an IMS2:D by pulling the plug ), while an accidentally disconnected player would stand a better chance of survival. Maybe the negative results could be mitigated somehow: ignoring death if it did not occur in the current turn, or allowing a given number of "suspicious" disconnections over a sliding time period (say 3 disco without sudden death, decremented after each disco, incremented after 2 days without disconnections).

Well frankly speaking the local game is quite limited as it is, more like a starting practice for new players or something. It seems to me all the "interresting" events (learder board, squadrons, campaigns) are designed to be played exclusively online, so I doubt much will be done to improve offline play.

Actually I just lost the escape death of a leader with about 30 kills due to connection timeout, AND IT WAS NOT EVEN ON MY SIDE OF THE LINE !!! My DSL connection was perfectly fine, I was using it for upload at the same time, and there was ample bandwith remaining (something like 8kbytes/s up and maybe 50 kbytes/s down). This thing has to be fixed pretty quickly, it is simply a game killer IMHO. Besides, getting your pilots extra fatigued will soon force you to fly with fatigue, due to the 4 elements limit. That is more than enough for a penalty, and should at least be less of a problem for regular players suffering from unreliable connections than for those who disconnect on purpose.

First of all I must say DiF had me hooked to my screen right from the start. I had practically never played such kind of card games before, but the mix of luck and thinking is excellent. One can jump into the cockpit and have fun 5 minutes after installing the demo, but mastering the subtelties of this system seem to require quite a bit more time. Most of the games are full of surprises and suspense. The part of luck allows for interresting games even between players of uneven skills. The computer does a very good job of relieving the players from all the tedious "manual" operations, resulting in a very simple, fluid and intuitive gameplay. About the implementation The general design is really good. The web-based application allows for simple (from the player's point of view ) and quick pilots management, and the leaderboard, awards system etc. combined with skill upgrades are just right to give a role-playing feel to the game. The penalty for dropping in mid-game is a bit harsh IMHO. I had a provider disconnect that costed me my leading ace once. I was not amused . About the interface, I would say it's a bit awkward in the present state: the main window is basically empty, but other popups like private chat, mission selection or pilot management are soon cluttering the screen. I would rather prefer these dialogs or requesters to be created as children of the main window. As noted before, a few keyboard shortcuts would be handy too (and as far as my Windows programming goes, they are not the thoughest thing to add to an application ). About bugs As I started to look at the campaign system, I frequently stopped games after a few seconds to retry another setup. Once I was playing the Brittish at Dunkirk and due to my quitting the game and restarting it, all the British pilots were gone. So there was just a pair of 109 escorting stukas, and the 109 started attacking their own bombers. The final score was 4 points for me, 1 dead stuka and a couple of other German wrecks . For some reason the 100 exp. points limitation system sometimes fails. I had a pilot with 85 available points while around 40 points were already spent. He was a "promoted" wingman, actually the one replacing the leader lost due to connection dropping. Unfortunately I did not pay much attention at the time (I did not even realize there was a limit then ), and the poor guy is gone now, so I can't exhibit the example. The variable action time limit is a bit awkward at the start. I understand some kind of timeout has to be put to avoid stalling the game if some player goes AFK for some reason, but this timeout should not start before every player has had time to look at the staring screen! For now, "ace" (10 seconds) games force you to simply skip the intro screen (unless you're *really* good at fast reading), going into the fight without having a chance to look at your adversary's configuration. Another timeout may be set for watching the startup screen (let's say something between 30 seconds and 1 minute), and the game would only start after each player either clicks the intro screen or hits the time limit. Also a minor thing, the game setup is surprisingly slow, given the extremely limited number of players online. With maybe 3 or 4 people in the lobby, it takes sometimes like 30 seconds between game creation by the host and apparition of the game in the "join online game" list. .NET is sure a resource hog, but still I hope the server will take it when (as we all hope here) the lobby will be overcrowded 24/24 ! Some whistful thinking Most has been said before and I know changes are underways, but here is my list of current (minor) gripes: - seeing the leader's cards and both planes upgrades (I mean expendable skills, hopefully a player can remember the permanent upgrades ) at all time during combat is a much awaited improvement - the delays between AI plane moves are really slowing down the game. A "next" button to skip to the next phase and/or a way to specify the delay would be good. In multiplayer, the "next" button could be set to work only if all the players clicked on it (assuming this does not eat up too much bandwidth). - having been a faithful IL2 player a couple of years ago, I also suggest having a look at the HyperLobby application. It is a really neat game lobby that can handle literally hundreds of players. The idea is to have "rooms" where the players can wait for the host to launch the game, instead of having the host create the game and the other players joining only afterwards. The joining screen is visible to all players in the lobby. A dedicated chat window allows only the players who occupy slots to chat (in order to agree on who goes on which team, asking someone to leave because you're waiting for a buddy, etc.). Players can withdraw by clicking again on their name in the slot they want to free. Once the desired number of slots are occupied by the desired persons, the host (i.e. the player occupying the first slot) launches the game. A timeout system and the possibility for the host to kick players prevents lock-ups. Also, a separate list of in-game and non currently playing people would be very useful. Plus a system of friend list (I guess this will come handy soon, when dozen or hundreds of people will be around). - if we assume an "hyperlobby-like" interface (i.e. when a 2v2 game launches, all 4 players have already chosen their side), having an intermediate screen for allowing players to agree about each other's altitudes and upgrades would be nice. Once the teams are set, another screen for each team allows teammates to chat before flight, buy upgrades and set altitude (a bit like the current "create mission" screen, with an added chat window and no possibility to swap pilots, of course). As an added bonus, they could decide who want to play first (i.e. the sequence of play is still chosen as usual, but the players decide who goes first inside each team). This would add a small tactical variable, for instance trading the davantage of initial height for a lower playing order. Maybe this is too much to ask for, but I think that would allow the same kind of enjoyable pre-flight briefings I experienced while playing IL2 online. - the "eye candies" are pretty nice, but I wonder why all this 3D has to end up in huge and static pre-computed GIF animations? A (relatively) simple OpenGL code could do the trick for a lot less work, and add pleasant variety to the animations (nothing easier than varying camera angles or even damage representation, smoke, etc.). Even a software renderer could handle these kind of graphics, no need for accelerated hardware. Just a wild speculation, of course . - the leader board could be refined a bit. Especially some stats on a per-mission basis (like kills/mission) could be filtered out to avoid unsignificant results like a 2 or 3 kills per mission with a single mission. Pruning out players with less than something like 5 or 10 missions would be enough to have more reliable stats. Also, adding an ELO-like ranking system or a ladder system (the winner goes up by an amount equal to the difference of positions if his adversary is higher) would add even more challenge and allow relatively new players to be ranked on a more equal basis.

Here is an action/reaction summary I made primarily for my own use, but now that it's done, why not put it online? I found useful to have the occurence percentages of each card at hand, it helps getting a better grasp of the various sequence probabilities. For instance, the choice between tight turn and barrel roll in defence, or what cards to discard to maximize the chances of getting some others. I would appreciate any comments, changes or additions.