Author
|
Topic: SMAC Multi-Player How ???
|
DJ RRebel |
posted 11-07-98 02:24 AM ET
OK ... could someone please help me here !!! I'm excessively confused as to how Multiplayer will work in SMAC !!!Fact >>> Multiplay will have each player moving at the same time to cut down on waiting by about 99.9999999% I understand that !!! Fact >>> SMAC is not a real time game !!! Fact >>> I am confused !!! How could you have a game that proceeds on a turn by turn basis without having movement conflict ??? Example >>> Two players are not in a given map position (hereafter refered to as MP because I'm sure it'll be used alot) ... ok so the two players move into that one MP in the same turn ... because their turns are simultaneous, in theory they both occupy the same space .. so what happens ??? does one of the get chosen at random to get there first ??? But that would mess things up for the other person because he would have to re-move his unit, but what if he had planned a multi-unit advancement that relied on all the units being able to move ... then what ??? Would he be able to re-move all his units because the first one wasn't able to move where he moved it orriginally ??? This would end up making the game even longer once the factions started to meet up with each other on the map !!! What if the two units are enemies ??? Will a combat automatically be started ??? hmmm .. maybe if such a thing happens a diplomatic menu will pop up letting both sides negotiate for the MP .. but this could happen many times ... and then several times a turn near the end of the game and could be quite annoying !!! Maybe it's a feature that is turn on at first, but you can set it to an automatic responce depending on the faction you encounter !!! What happens if two factions setup cities in MPs next to each other without knowing it till the end of the turn ??? This could be very confusing for people who are not logically inclined !!! How is it posible to have a non-turn based non real time based game while keeping it simple enough for a majority of people to understand and enjoy ??? Even worse .. lets say player 1 has a unit that he sends to attack a city, then he sends other units to attack the city from a different route only because the first unit managed to do alot of damage otherwise you would not have send in those other units ??? But then you find out that the first unit met another unit of player 2 on the way to the city and got destroyed ... then what happens ??? Can you unattack with the other units because the first attack essentially never happened ??? There is so much room for turn-progression conflicts by having everyone have their turns at the same time I just don't understand how it is feasable !!! For those of you who refer to Firaxis programmers as gods ... if they can work around all the loopholes involved in programming same time turns, then creating a living being in 7 days is as simple as .. ummm .. well even simple than baking a cake !!! Same holds true for Secret Projects (Civ2's Wonders of the world) ... if 2 Factions develop a given Project on the same turn, what happens ??? This would cause total chaos once the equivalent of railroads came into effect !!! I'm not a professional programmer, but I took Pascal in school 10 years ago to know that writting SMAC is easy ... developing the logic behind it must be so complexly mind-numbing I don't think anyone at Firaxis behind the logic will be certified sane for the rest of their lives ... Hats off to Firaxis if you actually implement this flawlessly while taking into account the problems I've mentioned here among the many others involved in doing multi the way you intend on doing it !!! I'm truly curious !!! Does anyone else here have any ideas ??? I can think of a few other conflicts in this approach to multi, I'll post them as this discussion progresses ... do any of you understand what I'm trying to say ??? Can any of you see any other potential conflicts in the logic of multi-play SMAC ???
|
RyanR
|
posted 11-07-98 02:44 AM ET
The PCGamer article which SMAC is in is mostly about the multi-play of SMAC. I don't have the mag with me but it talks about "screen lock" which totally confuses me. Somehow it focuses on one screen or something. Not sure. |
DJ RRebel
|
posted 11-07-98 03:45 AM ET
Wow ?? Sounds weird .. I started a thread like this in the last forum, but no-one really responded to it ... I think it is a critical issue in SMAC and hope it is done properly !!!I'll have to go buy the mag tommorow ... I have the article in Computer games strategy plus ... and it doesn't talk about it at all !!! |
Old_Guy
|
posted 11-07-98 04:06 AM ET
I just read that PC Gamer article today and I still don't get it. We just know that if anyone can do it right, it's Sid, Brian, & co. |
DJ RRebel
|
posted 11-07-98 04:28 AM ET
Is the article on-line ???So anyways .. how do all of you think multi should be implemented ??? |
Bannor
|
posted 11-07-98 06:17 AM ET
ok screen lock means that when ever a unit is moved all units within a certain number of spaces is temporarliy unmovable. The chances of 2 people truly moving at the same time especially considering internet lag is really not a concern. I am sure there will be a host/server/creator of the game and that will make sure this doen't happen.If you send a unit to attack a city and it hasn't gotten there yet you can send it back movement take place during the turn. Of course I don't fully understand but it ought to be a lot of fun to play a truely new type of multiplayer game.
|
DJ RRebel
|
posted 11-07-98 06:47 AM ET
ummm ... I didn't quite fully understand that .. could you explain it again .. a little slower please !!!  Or how about a link to the article anyone ??? |
Heckler
|
posted 11-07-98 02:34 PM ET
Well one way of solving the two unit problem (assuming that they are not enemies in which case simply begin combat) is compare movement rates i.e. the faster unit gets there first. However, this could lead to a problem with a faster unit coming from further away and getting there first which is illogical. So what I would suggest would be a quick comparison between movement rates and the distance traveled to get to the Map Unit in question in the case of a true tie then you would be given the choice of attack, retreat, and negotiate which would allow you to talk to the other player and decide between yourselves (ok I will send you 1000 energy units and my next tech advance just dont tell the spartans about the sneak attack!)Heckler |
DJ RRebel
|
posted 11-07-98 10:45 PM ET
hmmmn ... yes ... but my point is that the time this type of in sequence would save would end up complicating things on the diplomatic front in a huge way !!!I guess maybe your faster movement theory would be good one ... but what if it was a tie like it probably would be !!! Or what if both parties didn't want to fight !!! And do you know the chaos it would cause if you had railroad type movement !!! ugh !!! |
HermitTwig
|
posted 11-08-98 01:40 AM ET
All I can think of is that this "screen-lock" thingy could be a real problem in the later part of the game when you have most of the land "developed" with the equivalent of roads & railroads from Civ 2. Take playing on Earth in Civ 2. If you had the railroads to do it, you could move a unit from Siberia to South Africa in a single turn. This could be a problem when you run into a unit that someone else is trying to move in the Middle East.Also a question I have: In Age of Empires, two or more people could be not only allied, but they could actually control the same civ at the same time. Will there be this option in SMAC? |
DJ RRebel
|
posted 11-08-98 07:33 AM ET
That;s exactly what I meant by railroad type movement !!! How could they possibly take into effect a zero point movement ???It just seems like alot of anarchy to me ... I hope they do it well, because the whole essance of the game will depend on this factor !!! I hope they do it well, if not, our wait was for nothing !!! |
Heckler
|
posted 11-08-98 11:23 AM ET
Good point about railroads. However they could bypass the problem somewhat by having things moving from (point a) to (point b) via railroad not go through the interveening space (the shortest distance between two points is none at all.)Heckler |
Steel_Dragon
|
posted 11-08-98 11:49 AM ET
I always thought that decisions should be during the pause and movement made at th same time for all units( shorta like bewteen turns), Like MOO2. Have any idea if this is how their doing it. |
DJ RRebel
|
posted 11-10-98 02:18 PM ET
Apparently there's an article in PC Gamer that doesn't describe it well !!!  Heckler, your solution is probably the easiest, but also the most unrealistic .. in fact I never liked the idea of railroads in CIV2 (although I used them alot) .. I think they were too powerful .. at most railroads should have tripled or quadrupled road movement .. but not 0 movement ... that was just too farfetched !!! |
Titan
|
posted 11-10-98 04:56 PM ET
I think that they should allow many units to move through the same square in the same turn, to consider only the destination square. In fact, as Steel Dragon said it, it could be as in MOO2 where many ennemy units could be at the same place, and that the battle only starts at the order of the ennemy. A square, I think is larger than a square meter, so i dont see why there could not be more then one factions units on one square. The two units would be facing each other waiting the order of the general to engage fire. |
Titan
|
posted 11-10-98 05:19 PM ET
One other thing I dont understand in the working of the multiplayer game is the lenght of the game. A civ2 game could take over ten hours. Alpha Centauri Will surely be as long. So how are they are going to make a net version of SMAC. Not many persons like to spend ten hours straights on the computer without eating or going to the bathroom or things like that. There could be a save option, but if you live in Montreal, and make a game with someone in Texas, how could you be sure that he will be there the continue the game when you would. Unless you play only with your friends, I think that the internet games would have to be shorter one way or the other. The single player games should still take as much time. |
DJ RRebel
|
posted 11-11-98 10:58 AM ET
Yes .. about the length of games, Firaxis has set it up so everyone takes their turn at the same time ... they also said that a game could be as short as 30 minutes !!! (I have no idea how ... I guess alot of computer macros etc) .. Actually, I'm almost as curious about how the game will work as about the game itself !!!Time will tell !!! There's also supossed to be a time limit option for the turns !!! As for units sharing the same square, yes, I do believe that in theory that should be posible, but that would take away from the whole point of the game !!! As for what each square represents, i think you're way off with your one meter !!! I think it would be more like in between 5 to 100 KM each !!! Depending on the size of the map !!! |
GOdSMurf
|
posted 11-13-98 01:02 AM ET
DJ I've often been wondering the exact same thing. But, there's been a CivNet (which I know nothing about) - didn't they have to solve the exact same problem there? How did they do it?
|
Mortis
|
posted 11-21-98 07:33 PM ET
****!!, an on-the-topic thread, KILl IT!!!  Okay, now to the seriouse part... I think that the screen lock featre would be a pain in the ass when your at war, wouldn't it make more sense if it only the square the unit is moving into get's locked, after all, it takes 0.5 sec max to get the info over to the other computer 0.5 sec is a realy slow ping. And the rest is just animation. There is no point in locking up 8 squares when you only need 1. The other thing is, I seriously doubt the 30 min a game estimate, I'll put it around 1 hour. You don't always find 7 people willing to sit an hour playing a game. If one of them decides to leave would the AI take over?? Thirdly, I see alot of sense in electing someone a winner, although I doubt a lot of people would. I'll call it "being a gentleman" saying, "congratulations you;ve won". It's better to admit defete than to be slaughterd by a someone. Fouthly, I would like to know about the diplomacy, I would assume that in a MP game quite a lot of diplomacy would take place a "chat" rather than pre made statments. What i'm trying to say is, it's hard to govern a faction, wage a war and sign a new trade deal all in 30 seconds. |
Larry Boy
|
posted 11-13-98 02:53 AM ET
Until I read this little thread, I had it all logically laid out in my mind. Now I'm confused! THANKS REBEL!!! The way I had it impressioned is this: Every unit has so far that it can be moved in one turn (except a 'railraod' situation). How to describe this? It would be both people actually moving their units at the same time. For instance, let's play a quick game. Me and Bob are playing against one another. He has 1 base, and I have 1 base. I have a hovertank, he has a gravity-copter (I think I made that one up). Anyway, the turn begins, Bob looks at his base screen to see what to build, I move my hover tank. Instantly it is updated to his computer that I moved my hovertank. After he exits his city screen, and I move my tank, I look at my base, and he moves his helicopter. When we are both done, we hit the 'done' button or whatever. He saw me moving my tank, I saw him moving his helicopter, it happened in the same turn at different times. Am I being clear here? OK, let's say my tank and his copter are both planning on moving to the same square. He is quicker to hit the arrow keys than I am, so his copter moves, and both our screens are updated. I can no longer move my tank to that square because in that turn, he moved his copter there, so I fortify my tank to recive his onslaught the next turn. This is just my guess, I have no reason to back it up. Say, does anyone remember back in the days when Gettysburg was just a dream and Brian Reynolds, Mike Ely, and Sid all posted as much as anyone else? Those were the days... Anyways, there's my theory, hope I didn't confuse more people than I helped. |
Octopus
|
posted 11-30-98 03:50 AM ET
Well, with all the recent talk about everyone ganging up on the Hive, I've started thinking more about multiplayer, which I may have to try out, to teach some people a lesson . Reading this thread, it seems like some people think that the first person to move will have a clear advantage. Personally, when I play games, I prefer a reactive rather than proactive strategy (it is far easier to find flaws in someone else's plan than it is to construct a flawless plan of your own ), so I don't think it will be a disadvantage to let my opponent frantically move all of his units as soon as the turn starts. He'll just make more mistakes for me to exploit .The locking system described in this thread seems to make sense, but the problem is that bugs (if they exist) will probably manifest themselves as livelocks or deadlocks, with each machine waiting for the other guy to do something. Kind of a scary failure mode.
|
DJ RRebel
|
posted 11-15-98 04:39 AM ET
LarryBoy, your version would be the easiest, but it would NOT be a true turn based game !!! It would be a not-so-real-time game !!! With the only difference that there would be a forced time out (every turn) so people wouldn't get too far ahead !!!Your version would give the advantage purely to the person with the fastest connection or better mouse ... etc etc etc !!! Larry ... sorry for ruining you beleifs as of late !!!  GodSmurf ... (Where did you get that name ???) ... I never saw CivNet, I heard it was a flop though !!! I'm still waiting for Civ2Multi-player Gold !!! (Frigging MicroProse ... delays delays delays !!! Anyone else ever play CivNEt ??? |
DJ RRebel
|
posted 11-30-98 12:25 PM ET
I agree .. this simultaneous turn thing has got me completely baffled .. I still think that is the most interesting thing about SMAC !!! I really wonder how they'll piece it together !!! |
DJ RRebel
|
posted 11-15-98 04:40 AM ET
I'm guessing CivNet was just two people connected on the net and they had to wait for each person's turn !!! |
Golf
|
posted 11-30-98 08:39 PM ET
In one of many interviews i have read lately i remember reading Brian Reynolds saying that the game will have certain time limitations for each turn and that you had so much time to complete each turn. Also he stated that he felt that most games were going to have to end with the Faction Dominance chart being the deciding factor after an agreed to number of turns. Hope this helps. |
Jack_Schitt
|
posted 11-15-98 05:09 AM ET
so far as civnet, I LOVED the game.I loved the multiplayer aspect SO much, that I decided to wait for civ II to go multiplayer... so much for that idea  as I remember it, there were two modes to play in with civnet, turn or simutaneous... ... in turn (if I remember correctly) it was as you'd expect, each person in turn took his turn... SLOW... ... in simutaneous, it was still kinda turn based...but everyone could move units around. ie. I could see an enemy move while I was moving. thinking fast and moving -could- save your butt. there was a timelimit countdown for both, configureable at the start. it was possible to not have a time limit. you could NOT make these selections during the game, and it was common to "save game and go to turn" late in the game when there were ALOT of men to move. in simutaneous... the timelimit wouldn't start clicking down until only one person was still moving (or was it after one person hit "end" ?) *sigh* that was a long long while ago overall multiplayer gameplay was buggy, it crashed often. alot depended on the host system and host internet connection. when the host died, the "new" host (game auto selected) was often unable to let the game continue, due to his internet route or his CPU... there was also a bug in starting, that would allow you to have 3 settlers (but no city) within the first 5 or so turns... another common thing to starting the game "fast start?" which meant "ok to use this bug?" as far as how many players in a game, I'm sure it was at least 6... maybe 8 max... If I had it installed on my new system, I'd check 
Jack_Schitt aka OldGroo (if anyone from civnet remembers the "Grooites" ) sidenote, I always setup games via IRC, so that I could traceroute the players... as there was no internal ping option...or any internal game board to find games.
|
DJ RRebel
|
posted 11-16-98 11:40 AM ET
Well, at least for SMAC, we'll have this board to meet people for games ... I hope we don't get 50 million threads called: "Does anyone want to play ???" ... lol !!!By then Firaixs should have a section of their site devoted to helping players meet each other ... and even then, I'm sure lots of us will have web sites for that as well !!!  As for CivNet .... why wasn't it very popular ??? |
BigER
|
posted 11-16-98 11:45 AM ET
Well, I am going to go out on a limb here and say; Idon't think CIVII lends itself to multi play very well. Maybe for the first few centuries. But imagine you have 25 cities and the next guy has 25 cities. etc. If you are playing by modem over the internet, well, time is money. Over a lan is more like it, but how many of you have a LAN at home. I do, but no time to use it for games. Anyway, just my opinion. |
DJ RRebel
|
posted 11-16-98 12:36 PM ET
Who here would be interested in joining a league ???I'd be willing to set it up on my web-site if you guys wanted to !!! Nothing fancy though .. I would really want to keep the game fun !!! I'm the type who gets really competitive during a game, but would like to laugh about them when it's over !!! I'd probably set it up like the NFL, where ech team plays once (maybe twice) a week (it would be up to the players to decide where they play !!! This way, people would actually have time for other leagues or solo play !!! We could have some games of 2 players some of 3 and some of even more !!! I guess the number of point the winning player would get would depend on the number of opponents !!! EX: 2 Players (3 pts): Winner 3pts, loser 0 pts 3 players (6 pts): 1st 4pts, 2nd 2pt, 3rd 0pts 4 players (9 pts): 1st 5pts, 2nd 3pts, 3rd 1pt, 4th opts 5 players (12 pts): 1st 6pts, 2nd 3pts, 3rd 2pts, 4th 1pt, 5th 0pts or something to that effect !!! I'd also like to keep it relatively small ... definately way less then 20 people !!! If you guys are interested, let me know !!! I'd preffer people who are in the forum at this point or any point before the release of the demo !!! That way I'll know you're semi-serious at least !!! There could also be team play .. wow .. the ideas are really stirring now !!!  6 players (18pts) 2 teams: 3 winners 6pts, 3 losers 0pts !!! Of course, I'd be taking many suggestions before we started and we'd have some kind of a vote on the rules, point system etc ... So ??? What do you think ??? |
DJ RRebel
|
posted 11-16-98 12:39 PM ET
ummm .. that point system didn't make any sense did it ???should be more like 2 players = 2 * X pts 3 players = 3 * X pts etc etc ... Well, as I said, we'd all have to figure out the rules together !!!  |
jonesEv
|
posted 11-16-98 05:30 PM ET
I'd like to make a suggestion and provide an explanation.It is mentioned in the gamespot preview that a planetary governor election may be held and that anyone with a 3/4 majority wins the game. This is all very well and good in a SP game, as it provides a method of ending the game early once you have established supreme dominance so that you don't have to spend 500 years looking for that one last size 2 city. However, the question I ask is: Why would anyone want to vote for another player in multi player? The answer (at least my opinion) is: Because you get extra victory points for voting for the winning side, lose some for voting against and are unchanged if you abstain.This seems like the only smart way to implement this feature. If anyone has any suggestions, please post! OK, now for the explanation. I posted this already in the GameSpot preview topic so this may be familiar to some folks. I think this is a decent explanation of what they intend: Turns are timed (time limit optional and variable) and simultaneous. Simultaneity of turns is handled through what was described as a "classic record-locking system". For those non-db folks out there, this is a method of handling read or write requests on a multi-user database. Basically, if one guy is looking at some data, and another guy comes along and changes it, the data that guy 1 is looking at differs from the data in the db. This is a bad thing because if the first guy decides to change and save his copy of the data afterwards, guy 2's changes are lost. So, what happens is when guy 1 requests data for write, a "lock" is placed on the data that he's modifying. If guy 2 requests that chunk of data for write, the locking system will refuse. What this boils down to is: only one person can move in one area at one time. The one big question here is: How big is that area? Actually I lied, there's another big question: how long does the locking player keep the lock? The area is probably a "circle" of squares with a radius equal to the maximum movement of the selected unit. The duration of the lock is probably from the time a player selects (highlights) a unit to the time that unit moves. They probably update the locking radius each square, so that it decreases throughout the move. They might make the locking radius one, and resolve locking on a square by square basis. However this might result in moving half your move and being stopped by an enemy. Another question comes to mind as I post this for the second time: Can enemies you can't see lock you out? This seems necessary, but it allows you to be aware of an enemy at a distance equal to your move plus the enemy units move, with greater precision as more of your squares overlap. What I really hope they can avoid (FIRAXIS are you listening?) is making the beginning of a turn a rush to move units into advantageous (sp?) positions. This probably would only be a concern with a small number of important units (presumably you can only lock one area at a time), but it would be extremely annoying if this induced manic clicking contests.
|
jonesEv
|
posted 11-16-98 05:37 PM ET
Regarding the railroad issue:Firaxis could say that you have to move units along railroads in steps; you can only move your maximum movement at one time, but you could repeat this for as many moves as you have time for. Not only would this reduce the maximum aforementioned locking radius to the max move points of a unit, it would address the unrealistic infinite move situation in Civ2. (One spy viewing all cities in one turn costing 0 move) |
Ultra SupremePaco
|
posted 11-16-98 06:03 PM ET
DJJ, go ahead and throw me into your league. I'll make a good teammate. Send additional info to my e-mail account.~Paco |
DJ RRebel
|
posted 11-17-98 11:33 AM ET
JonesEV .. thank you thank you thank you thank you thank you for explaining that in English !!!  Finally I understand the principles ... now, I only hope they are applied well to the game, you point about knowing an enemy is near is a very good point !!! Anyways ... thanks again !!!  So basically, this is not a true turn based game if you have to race against someone else for map positions !!! It's a merger of turn and real-time !!! Paco ... great ... I started a thread about it for more details, and I'll get back to you in a week or so with more plans !!!  (Post your ICQ and stuff in the ICQ thread if you can) |
jonesEv
|
posted 11-17-98 06:48 PM ET
I think I should clarify what I meant about 'resolving the infinite railroad move'.Basically, the system I mentioned would take significantly more real-time to move units long distances over railroads, thus limiting the total number of moves some in timed turn games. It should be noted that this would have no effect (other than being annoying) in sp games and that the move (even without this system) is no longer infinite in any timed turn game. |
jonesEv
|
posted 11-23-98 06:26 PM ET
Yes, you probably could lock someone out for one whole turn, but then you would not be able to do anything else for your turn. It is also possible that they'll code it so you don't put a lock out until you actually start the movement process ( as opposed to when you select a unit). There are obvious benefits to this approach, but there are issues that it might raise, depending on the specifics of the networking implementation(upon that I am not qualified to speculate). |
Gord McLeod
|
posted 11-17-98 07:16 PM ET
Regarding the multiplayer voting options - in CivNet, it was quite possible and indeed customary if I remember correctly to have a mixture of human and computer controlled nations, except of course for larger games where all slots are filled by human players. In a game where you're playing against 2 or 3 friends, it might well become important to hold planetary governor elections - if you can get enough computer factions to vote for you, you could pull off an early coup. |
jonesEv
|
posted 11-30-98 09:01 AM ET
In reply to Octopus' message (posted way up there): Christ, I hope not! It should be pretty simple for one end to figure out that the other isn't all there. I think, however, that this issue is being slightly overblown. They've had a while to think this through and the locking schema shouldn't interfere with gameplay if its done right. They've been playtesting a while... Here's hoping this gets posted at the bottom... |
DJ RRebel
|
posted 11-18-98 12:11 PM ET
Does anyone actually still play CivNet ??? |
DJ RRebel
|
posted 11-30-98 12:25 PM ET
I agree .. this simultaneous turn thing has got me completely baffled .. I still think that is the most interesting thing about SMAC !!! I really wonder how they'll piece it together !!! |
Old_Guy
|
posted 12-01-98 04:05 PM ET
Dushan--I was also thinking that maybe attack/movement orders wouldn't be executed until the end of the turn. I think this is pretty common in wargames. The problem is that in SMAC you also have diplomacy and this complicates things. What if I gave an order for my unit to attack yours, but before the end of the turn we call a cease fire or make peace? Sure the game could then say 'You have signed a cease fire, are you sure you still want to attack?', but even if you cancel the attack, that unit has essentially lost the turn.I have a feeling that multiplayer will heavily rely on automation features. City governors, building queues, and unit pathing/behavior orders will ease micromanagement and speed up the game. Maybe multiplayer won't be an RTS click-fest, but it could be pretty fast-paced. One bright spot is that Firaxis has probably thought about this since the development of Civ2. However, no matter what, I imagine that any bugs that pop up when SMAC is released will most likely be multiplayer issues. We'll just have to see. |
dushan
|
posted 11-18-98 03:04 PM ET
Hmm, locking sounds reasonable, except for this : you can stop an enemy from getting to critical fields simply by spending your whole turn with your unit ready to move, but not moving. So you can wait with your settler in the town, while you're building a defensive unit, and no one can approach the squares around the city...I think the only true multiplayer turn based system where all players move at once, is where you give orders to each unit during the turn and the orders are only executed all at once at the end of the turn. I don't think this would work very well in SMAC, but I've seen a game once that used it quite nicely. Dushan |
DJ RRebel
|
posted 11-25-98 10:57 AM ET
I agree .. there are soooo many potential loop-holes, I really wonder if even the Firaxis gods can fill them all !!! |
Octopus
|
posted 11-30-98 12:17 PM ET
"Here's hoping this gets posted at the bottom... "No such luck, jonesEv  "It should be pretty simple for one end to figure out that the other isn't all there." You'd think so, but if one end "forgets" that it has the lock, or thinks the other end has the lock, or if both ends think they have the lock simultaneously, then both will end up waiting for the other guy in some way or other. I'm not saying this is definitely going to happen, I'm just saying if bugs are in the code, this is one particularly nasty way they can manifest themselves. This sort of bug might be very hard to catch in playtesting, because it is very timing dependent (i.e. the two multi-players have to do a particular thing at a particular time, etc.). Even if the bug could be generated once, it might not be so easy to reproduce and debug. I've got a lot of confidence in Firaxis, but nobody's infallible, so this is something I'm concerned about, since this is the kind of bug that might slip through testing.
|
DJ RRebel
|
posted 11-30-98 12:25 PM ET
=================== ggrrrrrrrrr ----------I agree .. this simultaneous turn thing has got me completely baffled .. I still think that is the most interesting thing about SMAC !!! I really wonder how they'll piece it together !!! |
DJ RRebel
|
posted 12-02-98 05:03 AM ET
Good point old guy !!!As I said, the number of loopholes in programming this sort of thing is virtually infinate, I'm truly curious as to how it will be done !!! |