Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Fonze

Doom Battle Chess

Would you like to see a Doom-based Battle Chess game?  

48 members have voted

  1. 1. Would you like to see a Doom-based Battle Chess game?

    • Yes! I'd even be willing to pledge some cash, too.
      7
    • Yes, but I dont want to pay for it.
      31
    • Not worth it/Don't care.
      5
    • No.
      5


Recommended Posts

Time to enter the night like a Doomguy enters an arena with a plasma rifle.

Well to start there was a problem with the knight not the pawn. I spent a while trying to fix the pawn. The knight had 6 moves 2 were duplicates when it should just be 8 moves.

The game now prioritizes double pawn moves to be five times the priority to any other move.

90 minutes in and the death wish glitch been fixed that would dare allow pieces to make a move into a capture point or think they're safe when they're not. Of course if a pawn is stuck and will die anyway and can't move no one steps into place to avenge them if they were to be captured.

Share this post


Link to post

This reminds me of the time I did actually make a kind of chess in ACS just for fun a few years ago.
https://drive.google.com/file/d/0B4gm20-64GnibER1bTFNUzFkeEk/view?usp=sharing

It's not quite fully done, the method I had for calculating possible moves (involving moving sectors 1 pixel and tagwait()ing them, there was some reason I had to do that) takes ages and I lost interest after a while anyway.
You move a piece by 'using' inside its square, then in the square you move it to or another square to cancel. You play both sides (red and blue).
So far you can move and capture pieces, you can't move a king into check, and if your king is in check you can't do any move that still leaves your king in check.

I was basically adapting code from a chess program I'd already written in Visual Basic for a high school project, so I could continue it I guess. The only AI I could be bothered to make for the VB chess was one that would just make random moves.

Spoiler

Share this post


Link to post

2.5 hours in, I've fixed another giant death wish mess. Such as the system thinking something was blocking a piece when no it wasn't. It affected the safety check from bishops, rooks and queen attacks.

I have also fixed an issue with the bishop and queen diagonals, where they were only checking northwest to south east, instead of north east and south west as well.

The logic seems far better tonight at now 3.5 hours. There's probably still a pawn - knight death wish, but if there is I haven't seen it for a while. It does seem like there are more pawn mishaps, like they do not dodge when they should, of course a logical pawn would stay in place if it meant defending a piece in jeopardy if it moved... but I haven't programmed it, and it doesn't just happen by magic, so its an issue.

Looks like a few times things will get in the way of a queen's diagonal or stay in the way. Mainly knights... always the knights.

I still have the game not attacking, same as last night, to improve the whole movement and safety logic. The games that I've played end up getting sort of gridlocked where the computer just shifts a few pieces back and forth to maintain not attacking, but not getting captured.

Here's the .gif turned into a video. I've been getting better at dealing with imgur.

http://i.imgur.com/ZZ4QbZF.gifv

Share this post


Link to post

Sweet, it's cool that there are no movement errors.

So we bow to barrel now? So much of my life has changed...

You know, when you think about it, maybe the barrels are Doom's toughest enemies. Cause they're always sittin there, taunting the player to just punch em a little; then repaying the compulsion with a gory death.

Share this post


Link to post

Well I am back. This time I've brought a barrel punching video with me:



http://i.imgur.com/HNs2wlv.gifv

I suppose the starting point is to get pieces to dodge diagonals from queens and bishops.

After almost 2 hours of no progress at all, I think I've solved the problem of not dodging. The game felt that if it moved, it would get picked off by a bishop or queen, because it was not checking if those pieces are being blocked.

Ba-da-bing. It works perfectly now.

If the other player has a pawn moved up 2 spaces, there is now a 10x priority for the computer to match that move. It helps gridlock the board pretty quick when no one is attacking.

So I think with this level of quick gridlock, its time to start making more advanced AI that values more movement. Let's say if a bishop has 3 squares of movement. It will evaluate and prioritize if it moves to square 3 will it have more or less squares of movement? More squares, the higher the priority. It also helps to avoid "dumb moves." Why did it move a rook back to its starting point when that bishop could have moved to a better position? This way it will also prioritize everything over pawns.

While I am at it, I will also add something that determines if the computer makes that move, can it capture anyone else from that move and then prioritize that accordingly based on what the captured piece will be.

The main code is now 2,300 lines and tonight was 3.5 hours.

Well here is tonight's video minus advanced AI for Queens, rooks and bishops.

http://i.imgur.com/0mUloqx.gifv

Share this post


Link to post

The barrel video was funny ^^
You should've seen me on Visplane's map icannar where a chainsaw was provided in a secret, but the only things around were a wall of 5 barrels. So I did the only logical thing to do and chainsawwed the barrels, because I'm full of good ideas ;)

Once again, barrels 1, Fonze 0 :\


Man it's cool to hear the inner workings of a chess AI. Once again I really appreciate the work you're putting into this. I'm sure many others do/will as well.

Share this post


Link to post

I decided to Google "Doom Chess" and I found this 3 year old thread about a physical Doom chess board that never came to be: http://www.doomworld.com/vb/doom-general/59436-doom-chess/

I think tonight I can finish up the advanced AI, polish it, test it for a while, enable attacking, then add the blocking process to attacking, so a Queen can't kill another queen as the first move. That should eat up 4 hours.

Then tomorrow I will probably hit up the liberate feature where a pawn makes it to the other side and liberates a captured piece along with the logic that tells it... there are no captured not pawns, so why keep going?

There is also safety blocks too, such as if a queen is about to take a king, why not block with a bishop? After reading about Chess on the wiki page, there's also castling or whatever its called where a king and rook trade places, once per game per side. I'm not looking forward to that.

Then I suppose it will be time sometime next week to have the ability to actually play the game! Select how many players. Restrict movement, highlight squares, tell you where you can and shouldn't go. A save and load feature, where you type out the file to save as, and of course auto save and continue (which is different than load.)

After that I'll add the frills / reason for this project like the auto import from folders feature to animate the game. Pick your graphics for your characters (imp = pawn or batman = pawn) and liberate your captured pieces.

So there's the road map now that I'm almost done with the tough stuff. I think it'll be another 10 - 14 days for an alpha.

Once the game reaches that point, I'll add extras like undo / rewind, place your pieces, skill / difficulty levels. The game needs to be murderous first before it can ever have an easy skill level. Finally, email a file capabilities.

I'll probably stick it on a site somewhere, along with an online / HTML5 functionality for whatever "vanilla" game it turns out to be. Why download when you can just play it online? Well if you download, then you can add your own sprites and so on.

As for online player versus player functionality in real time, I'm not planning on it, unless there's enough interest for it.

There you have a road map for the next 3 weeks 4 hours at a time.

If its rare for a human to beat a computer at a game of chess, then I wonder how a computer can defeat itself?

Share this post


Link to post

Couple things man.

First, the pawns don't technically "liberate" a captured piece. I'd call it more of a "promotion" because getting pawns to the opposite end can lead to multiple queens on one team. No pawn is held back by a lack of captured pieces.

For special moves you will need a couple.
Castling involves the king moving over 2 squares towards either rook and putting the rook on the far side of the king. It must have certain check features, tho. For example, once a king moves he can no longer castle. The same applies to each rook. In addition, the king cannot castle if it: moves him out of/through/into check.

There's also another pawn special move called the "en passant" in which a pawn can diagonally take a pawn that moves 2 squares (through the attackers attack-square), even though they end up beside one another.

Everything else sounds really sweet! The frills you mentioned hit all my buttons and I don't think anyone's gonna care about internet chess in real-time, so no worries there.

Share this post


Link to post

I was hoping no one would notice en peasant. I forgot it and I had just been on wiki.

Promotion is good.

Share this post


Link to post

The advanced AI system can now do rooks, queens and bishops. It evaluates the directional movement... Then for every space that it can move it evaluates who it can capture and if its being blocked from capturing. The rook took a full hour, which I kind of expected. It also added 200 lines of code. Not a positive or negative, more like that's just how many lines of code.

30 minutes later, the first attempt for the bishops / queen diagonal are done. I say first attempt, because these systems are so 200 lines of code that it takes a while to construct them and there's no sort of half way point to test. Its either all or nothing since its relying on a complete knowledge to set a priority to move.

The diaginals use a slightly different system to analyze if I move here, I can attack this piece WITHOUT being blocked. In fact it might just be a better system overall. The current / usual system checks every grid space between the two of them and every piece for every square.

I felt it would be easier rather than checking every space for the diaginols to just check does the direction of the piece I'm attacking equal the direction of the piece I'm checking if its between us. Then if they match the direction, is the piece that's blocking closer than the piece that I'm attacking.

That might be a much quicker way of doing things than checking each grid square, since each grid square counts as a process. Check if there's a piece on A:1, check B:2 when it can just be directions and distances.

I say that, but the kicker is when the game can check a usual 900 times per second does doing the same thing 2 - 8 times really matter? The game moves at 60 frames per second, but calculates 900 times on my PC of course.

After that long wall of geo'ish text. I've tested the game and the advanced AI / priority moves sure does work, to a predictable level.

I think that I do need to add a system for "is the place I am at better than the place I could move to?" It gets to the point where its just the queen shuffling around.

Now that I'm thinking about it, there should be a release priority, like should it move that specific pawn, because it will release a bishop or a queen that will then have greater range. So does that pawn become a higher priority?

Those might all just be frills for later after the animations / folder import.

Share this post


Link to post

That's up to you when you wanna put em in. You could release a beta when it's playable to have others figure out its flaws and work from there.

I'm a bit confused as to why the distance matters of any piece, unless your referring to a distance of time (moves) required for retaliation.

I agree that it's prolly better to just check directions and who can go where, rather than every single square. Prolly be easier on you if you could have the most amount of systems relying on each other rather than a million individual checks going on.

Quick question, is it able to prioritize moves based on what's going on? For a very common example, if my knight is 2 moves away from putting the king and rook in check together, does the AI know that and have the ability to reliably block it in time?
As another example, if I'm up in pieces, is it smart enough to know not to trade 1 for 1 unless it gives itself a great position or takes a great position away from me?
Or pinning a piece down so it can no longer move.
I guess that it could be called "situational awareness."

Share this post


Link to post

Well I spent 2 hours toiling in the code mine adding attack along with adding blocking conditions so the queen doesn't capture the other queen on the first move. With that being said, it seems like there is a flaw with the queen where she will not capture an obvious capture. other pieces go for captures. Of course its pretty rare when any pieces allow themselves to be in a position to be captured. Its usually a knight that can capture 2 and 1 has to move or a queen that can capture a pawn.

Anyway, work in progress. Its another half step in the right direction.

Share this post


Link to post

That's interesting, could the flaw with the queen relate to her priority level, assuming you've put those in for the pieces. Also, when you say obvious captures, what is the result from the queen not taking the piece? Is it possible that you could be missing a crucial strategical point that would be given up by the queen or something like that? It's hard to notice everything all the time, one reasons computers typically whoop us good, heh. I suppose if you've been setting up the board for specific encounters for testing then this point is likely moot, but it may be worth mentioning.

Anyway, I like the sound of progress and as always learning how stuff works is a plus :)

Share this post


Link to post

Why not bulls as rooks since they mostly run in straight lines and a combination of imps/zombiemen as pawns? The manc seems out of place to me

Share this post


Link to post

So to expedite problem solving, I've taken away all the pawns to watch the top tier killing begin. Without pawns, queen takes queen instantly, which should not be, because then king takes queen... another man's queen. You know how that goes. Actually no, rook takes queen, because somehow rook can just float through its own pieces.

Again, when there are no pawns its good to check for errors.

So almost 90 minutes in and I've stripped away more so its just rooks that attack. There was some sort of error where everything attacking from the 2nd player side thought it was attacking a different square. I've improved / enhanced the selection of attacks.

Since only the rooks attack, they plow through everything until there is only one piece left on the board. There's nothing yet that says you captured the king, good game.

Now I have only queens able to attack. Even though the queen and rook use the same method / code / command to check if they're being blocked from attacking, the queen is blocked mysteriously when she's not. Blocked by a rook at A:8.

20 minutes later, I've found the problem. Missing variable, so the queen and rook check for blocking pieces in the first column. Since the rook in the first column would always check the first column it would go. Anyway, this fix should be system wide, because the block checking is used for attacking, and checking for safe spots when dodging and moving.

Now the queen takes out the other queen on the first turn and the other king spends his time dodging the queen while she takes pieces to make up for his dodging... this is only because ONLY the queen can attack.

with that done, I've set everyone to be able to attack. I know... chaos. No pawns still.

Queen takes queen, king takes queen as it should be. Time to put in a safety check when attacking so both queens won't just be sacrificed. You get my queen, I get yours.

Logically the queen should move to a position to take the king.

I think at 2 hours in I'm calling it a night. I'm going to attempt to use the dodge's safety checker for the attack safety checker and its a process.

3,2000 lines of code in the main code.

Share this post


Link to post

This has become one of my favorite times of day :) I always look forward to hearing (reading) what's going on.

I like the idea of simplifying the board to test the AI. You could also look at the endings to some famous games to compare the AI's logic to what actually works(ed). I'm referring mostly to ones with most pieces taken off the board, but the others would be useful farther down the line.

Share this post


Link to post

Too bad CoolKids didn't stick around to see all of this game development talk. Even if its not real game development so much as AI / strategy development... which can be endless. I can tell its one of your favorite parts, that's why I try to keep up with it.

And yes simplifying the problem areas is always best, otherwise there's just too much going on and you waste time with things you don't need. I've watched game development Twitch feeds and Youtube videos where developers waste so much time getting to a problem area. Like its 5 minutes into a level, so they play all 5 minutes when... you're a programmer, just program a quick warp.

I can also remove or arrange any piece to recreate anything. Its easier to program than making something to delete and move around the pieces. It will be in there somewhere at some point.

Last night I also made the "R" button reset the board and the debug.txt file print out. I also added Alt + F4 support to auto exit the game. It was driving me nuts having to move my mouse up to close the game. Then having to alt + tab to the debug.txt file. There's so much being output to the .txt file that every move takes a second or two. That's only because its being output to a .txt file, the game without debug mode still moves 60 frames per second with 900 calculations per second, but I haven't checked that lately. It might be down to 800 calculations.

By calculations I mean it CAN through the entire code 800 times, but since its limited to 60 frames per second, it only runs through them 60 times per second. If I remember correctly, when there's nothing at all, no long code it goes at 1,500 calculations per second... on my computer. Well running just the board makes 1,200 calculations per second without the main code.

Also, I'm pretty shocked that now 3 people would actually pay for this project.

Share this post


Link to post

Cool concept. But for the board, I would use the greenish checkered marble hell flat.

SYS said:

Star Wars Battle Chest had the best animations


I remember seeing this at my cousin's house, because he had Sega CD, and thinking how awesome it was. Sega CD was probably the coolest system (add-on) back in the day. CDs themselves seemed pretty high-tech back then.

Share this post


Link to post
Kontra Kommando said:

Cool concept. But for the board, I would use the greenish checkered marble hell flat.


I intend to just make that selectable, there will a directory of images and the player chose the image from the directory.

Share this post


Link to post

What the Hell ever happened to him? One day he had that "any good Android games" the next he was like "screw 65% of the people here." it was kinda sad, but I laughed because of the apparent randomness of it. On second thought, maybe we shouldn't derail this thread for that ;p

geo said:

Also, I'm pretty shocked that now 3 people would actually pay for this project.


Me too, it's a pretty unique thing and the support for customization is unprecedented in the Chess world so far. At least to my knowledge.

More reason to either sell it or also put out a free version with a fuck ton of ads ;p
You gotta make money on this to pay for your time somehow.

I think support for mobile will severely boost sales as well, but idk how much of a bitch itd be to program for. Stuff like this is ideal for a gamer on-the-go. Never mind that it's easier to get 2 people to crowd around a phone than a desktop :)

Share this post


Link to post

Chances are two Battle Chess fans that haven't heard of Doom stumbled into the forum and it resulted in 2 extra votes to pay.

Here's the game I made in 30 minutes for CoolKid's birds discussion and Doomworld's joke pictures:

Share this post


Link to post

Ha ha we're up to 4 now.

geo said:

Chances are two Battle Chess fans that haven't heard of Doom stumbled into the forum and it resulted in 2 extra votes to pay.


Chances are they all 3 were between the ages of 25 and 45. I cant really see many people younger than that who'd prefer to spend money and older folk prolly would prefer a regular chess board, or a series of hand-written letters ;p

It takes those who play many free games to discover the importance of actually buying a game; you get what you pay for ^^

One drawback to the developer is that people will expect more out of a game with the more it costs, but if all the features I'm reading about are implemented (especially in a user-friendly way) and the AI is either solid or will be later updated to be smarter, I could see this going for more than 10 bucks a pop.

Then again, pricing has to be thought through because if reducing the price by slightly less than half leads to double the sales then that would generate more profit.

Perhaps a beta version with stripped-down features (visible so people can get an idea, but not usable) could be a good way to get a feel of pricing through a somewhat small focus group, similar to what businesses do for new products.

Share this post


Link to post

Its that time again to have a 2 person conversation!

I've decided menus. Gotta have menus to setup a texture selection system for the floors. It works pretty smooth, escape to get in, escape to get out, enter to go further or back out if you're on the return to game menu item.

When started, the game generates 2 folders / directories if they do not exist. The sprites/ and the tx_floors/.

30 minutes later, I've started making the select a texture menu, because it will be easier to start here and go to the select a pawn graphic than select a pawn graphic, then make the select a texture. Its virtually the same system though... just one more folder I guess. Well the floor textures is you picking a .png while selecting who gets to be pawn will be selecting from folders. I assume each sprite/directory/ will have an index.png or a default.png as an avatar the game will read and show the player who you're selecting.

Anyway, it detects all the .pngs in the tx_floors/ directory. Then you select one with the arrows. No mouse support yet. 30 minutes more and now I have the textures showing up in a grid along with the arrow keys to select which one.

To close out the 4 hours tonight, it changes the textures. It saves and loads the setting to the settings.txt.

The code in the settings.txt looks exactly like this:
tileA = "tx_floors\SLIME14.png"

It only loads textures from that directory, but I suppose in the future if someone wants to have tileA = "..\..\hardcore_texxturepax\joe.png" it could load that one texture into the game. Again, a future idea if anyone wants it.

If anyone is crazy enough to want white and black tiles to be selectable, let me know. Personally, I think it gets confusing, but I'm not you. So let me know.

Here's the video / gif of the floor selection in action:
http://i.imgur.com/K4R6sgh

Share this post


Link to post

My favorite floor was the one that Jaxxoon would refer to as

Jaxxoon R said:

Skull clusterfuck. Skullterfuck.


Hahaha, that shit still cracks me up.

I like what I'm seeing and reading. It's a cool feature to change the tiles and whatnot. I'd kinda like to see em both changable, but after seeing it with the skulls that'll prolly be the setup I use, lol.

Itd be nice to hear what more people have to say. Speaking of which, I'll update the OP to reflect what's going on so people don't have to sift through all the replies.

Share this post


Link to post

Yeah I'll just make the other tile selectable. Someone out there might need it to make their fire pro wrestling mod where there is no board only mat. 30 minutes later, its in along with a few more error preventions when loading the settings.txt

As for the new person willing to pay; now I think the Doommunity is just messing with us.

Share this post


Link to post

I was 30 minutes into last night when I thought I'd just start rewriting stuff. 3 hours later, I have rewritten a lot of code with my new queen system that checks directions rather than tile spaces in every direction. I rewrote a blocking system that should be better to include the directional system.

The old system went something like this... queen / rook / bishop = 0 Then it went out 8 tiles one by one in each of its directions. Then figured out which 8 tiles are still even on the board. Then with the block, it calculated the same way 1 by 1 and one by one meaning that all of this was still calculated 900 times per second. So no human would ever notice. Well that just makes the code long.

As I've explained before with the directional system, queen / rook / bishop look in the directions the can move. Is something I want there? Is something in the same direction closer meaning I'm blocked? It used to work like that for just one function in many functions I think just for the queen.

Now it works like that for all three of the major pieces and it doesn't need to check if its gone off the board, because when it looks it only sees pieces. No pieces exist off the board, so it never finds a piece, rather than checking tiles would check beyond the board.

The directional look system also only needs to check is a piece in this direction? Rather than is a piece on this tile? Then each system checks a second time. I think the numbers turn out to be a queen's tile checking old system makes for 94,000 checks and the new system is 1,680.

Anyway, this is all 400 lines of code rather than 1,000 lines.

So now with attacking, the pieces check who they can attack, check if they are blocked, check if they are safe and check if that unsafe position is blocked. Currently only the queen checks all of these things, but with the new system in place, I'll make the rest do the same.

Since a piece will not attack if it is unsafe to do so, I wonder if I can make a protection system where my bishop can get your bishop and then your queen can get my bishop... so my bishop would never attack, unless there is a rook that will take your queen if she took my bishop.... therefore my bishop will indeed take your bishop knowing my rook will get your queen. If a pawn can then take my bishop... to not take your bishop.

Share this post


Link to post
Quote

 

That seems to be my impression too, but I'd at least try this one out because it's Doom. I play chess at a reasonable level but I can hardly visualize anything when I look at these boards. Well any chess moves, at least -- what I can visualize is that the game will be over by the early middlegame, because the opening will feature a ton of infighting.

 


What? What infighting are you talking about; it's chess, remember?

Idk, I've played chess on many different boards. In fact, my personal board is one of those where the pawns look like the bishops which look like the queens. Many people find it to be very confusing. I haven't played a full game of chess on my board in ages, lol. Games always move towards other boards or are just moved to the phone chess apps. I suppose after playing chess on over a hundred boards I no longer care what the pieces look like, I'll figure it out. I could see things getting a bit confusing if both teams had completely different monster sets, but even that wouldn't bother me one bit. Part of chess is taking your time, double-checking yourself, or finding out that you don't really need an opponent to lose because you done did it to yourself, ;p

Romero's head would be a good king too, but that's all somewhat of a moot point here as Geo will not have anything to do with that. This program will be able to make anything you want any piece you want, so long as you have the animations for them and put em in the right place in the right format.



To geo: thats pretty cool that you essentially stream-lined it and cut at least that part of the code by more than half (damn).
The directional look system sounds like it'll be very efficient.

But the pieces may need some kind of additional check system to note more general things about the game. For example, one of my favorite openings is moving the pawn in front of the knight up one square on the side opposite of the typical king/queen pawn up 2. Afterwards, I move a bishop up to the edge of the board, so it puts the opponents bishop in check, while being protected by my knight. There are few moves the opponent can do to save his bishop. He can move a pawn up to block, but that also blocks in his bishop. He could take my bishop, which is his best move. Of course I'm ok with a straight-up bishop trade early on when they are important, especially since I get to develope a knight in the process, which is also good in the early game. The last option is to do something else, which that early on won't typically need to be dealt with immediately, as pressure is different in the early game. If the opponent does nothing, when I get a chance I will immediately trade bishops, bonus points to me if I take the bishop king-side, as it will cost the opponent his castle right off the bat.

The AI needs to be able to handle these types of situations. As I said, it's best move would likely be to take my bishop of pawn up to block. Taking my bishop puts the onus of revenge on me, which puts me a move behind and gives it a chance to attack again since my move was spent on getting even.

Also, as I stated before, a check system for piece count. I always found that the old point system made the most sense. Pawns are 1 point, knights/bishops are 3, rooks are 5, and the queen is 9. Using this system gives me a very good understanding of how I'm doing after each trade. The computer could use it to realize that it's either up in pieces and needs to keep trading 1 for 1, or that it's down in pieces and should avoid 1 for 1 trades at all costs. Depending of course on positioning(idk if that's possible) and point values of all pieces involved.

A straight-up protection system might not be good because it might cause the comp to never trade, or never initialize the trade.

Of course this is all stuff about what is the icing on the cake for me, I'm very likely gonna be happy no matter what.

Share this post


Link to post
Quote

Yes! I finally get to use this gif. :)

The thing is that there are certain atomic-level pattern recognition things that are done automatically -- spotting legal moves, forcing moves, basic tactics, basic positional patterns, the basic things your opponent's last move changes, etc. And complex analysis builds off of that. So if it takes you 3x longer to spot the simple patterns that exist in the position because you have to "read" what the pieces are, you're going to take a lot longer to calculate all the variations you need to. So for any given time control, both sides' play will be weaker than with a normal board. (The exception is if the player is weak enough that they don't do much of anything automatically even with normal pieces.)

That's typically why serious chess players don't like playing on weird-looking boards, and why chess pieces used for competitive play are standardized (Staunton).

One good analog would be how fast you read normal English vs. how fast you read a cypher. You're going to be a lot slower if you have to consciously interpret every letter, and this will be true until you spend a crapload of time with that cypher.

So yeah, we can fool around with it just because it's DOOM, but probably no one will be playing even "serious casual games" with it (except maybe correspondence chess where a normal-looking analysis board can be used on the side).


Lmfao! That gif was funny ^^
No offense though, but if it was supposed to be a joke maybe you should remember that text doesn't convey tones very well. Maybe next time don't fail ;p

By atomic-level actions I assume you mean autonomic-level and yes that's true. Of course most serious tournaments are done with standard boards. Why do you think I mentioned that I haven't played a full game on my board in ages? What I have found through playing different people and also talking with them as we play, is that the older people get the less they like non-standard boards. I'm still in my 20's, and most people that I play with who are also in their 20's dont yet mind boards like that.

One other thing about pieces then, in order to somewhat standarize the board perhaps each piece could have a height multiplier to differentiate itself somewhat. The rest would be up to the player's choice of characters for pieces.

Idk who this would appeal to, but I can tell you that it likely won't be an online thing. I would just want it for 2 player with someone right next to me, practicing against the AI, or using it for an email-chess interface, so I don't have to have a board or multiple boards set up at all times. If it later gets the capability to email out progress then the other player wouldn't even have to use the same program.

It is somewhat of a novelty thing. All battle chess games are, though. But if I play this once with every person I play chess with irl, I'll have gotten probably days or weeks worth of total time enjoyment out of it. But there's something else. This will do more than Doom. It will do any character you have sprites for. If Geo also includes a 2D "classic" view than it could even take over as a regular chess app.

I respect your views, they're not wrong, but I think there's more that you're not taking into consideration.

Share this post


Link to post

After a ten day banishment to loser land, the Doomworld birdman powered skelecopter came to rescue me. Anyway, now that I'm back I have decided to continue making chess and keep bumping this thread.

Tonight I put in 4 hours on camera controls that took maybe 30 minutes and more importantly automatic import functionality for graphics.

So to tease you, I'll talk about camera controls. Before it was always arrow keys to move the camera and the mouse wheel to zoom in although I should probably have keys to zoom in as well. Now there are mouse controls. Hold center mouse button to move the camera. Holding the right mouse button will make the camera pan. Still no mouse controls for the menu or anything else. I probably don't even need more than one button with the mouse to be closer to touch screen controls.

Now for the better part of the night. I have it so the game looks in the sprites\ folder for any directory. Then it looks for an index.png. It looks for index.png because it only shows a 64x64 image, so the index is more like a profile / avatar of the folder.

If it cannot find a index.png, it looks for any png to load as an image. Then .jpgs and then .gifs. If it doesn't find anything, it just has an empty gray square with "no index.png" and the name of the folder. The game will not let you select the empty square, but any other square sets for the designated piece. I've got all 12 classes setup in the menu and the menu is broken up into 3 sections.

I've got around 35 sprites to chose from. I even stuck in some Doomworld avatars, purple enemies, Romero's head, flaming barrels, a tree, arachnatrons, cyberdemons, masterminds and eyes. The game saves and loads the selected sprites like the tiles.

Where it is now is at a glitch where the selected sprite hovers over the board. Its one of those situations that I'll fix tomorrow and then I'll make a video and post it.

This will solve the great Romero Head = King debate. Zombiemen can now be pawns, just like chain gunners.

Share this post


Link to post

10. Here is the new video demonstrating selecting the characters / sprites as pieces, along with the folders they're all kept in to show that they're loaded externally. It has to be the most visual update yet.

http://i.imgur.com/tzyEyqK.gifv



Here are the details in typical Geo wall of text fashion:

I've fixed giant memory leaks that made the memory usage bloat. It wasn't a priority last night since I try to give 4 hours each day. I've added background behind the index.pngs and the folder label over them like the no images already had as of last night.

I've come up with a more efficient way of loading the images rather than its a new milisecond, time to load the image again and again and again. Anyway, that stuff just adds to the memory usage / memory leak that bloats any program. Now the game stays somewhere around 5 MB with 6.5 MB at the most.

The game now saves and loads the settings file properly. It wasn't a priority last night and no one would ever know unless I mentioned it here and now. The "image selection" menu is much better. Right and left arrow keys now function, up and down moves you between the columns instead of to the next one in line. Its much more efficient when going through 35.

I have added a current piece graphic to the piece selection. There is now more of a highlight when selecting the graphic. Same with the texture selection.

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×