I have another crazy idea for the GOTM

[Moonsinger]

I waked up this morning and had this idea. It would be so cool if we could compare the result of all the players in any GOTM ressults in a graphical format from 4000 BC to 2050 AD.


Now, I know that this is possible for the following reason:

• 1. After we finish the our game, there is an option to 'replay' our game and there is a power/score/culture map display at the bottom of the screen for each year. That means that those information are saved with the game for each game turn.
  
  2. Now, if we can extract those information for each game, we should be able to generate a graphical map of culture/score/power for all players (each player can be represent by a color line or something on the combine power/culture/score map). Now we know how each of us are doing in each game turn in comparing to our peers. Isn't that great?

Since I don't have the time and the talent to decode the savegame file, this project is beyond me. However, I'm hopping that SpiderZord, ainwood, or you will take over.

What do you think?

 

[Jurimax]

I love the idea, I know that my line will be towards the bottom of the graph, but that way, I can see myself progressing

I would however add one 'small' thing. The time when someone goes into another era. That would make the comparison even more profound.

With only the graph you can play the best game of your life and still be to the bottom. I mean, the graph doesn't represent everything. If you play for example an OCC, 5CC, ... the graphic representation won't represent the greatness of your game.

Anyway, I like the idea

Greets Jurimax

 

[ainwood]

Moonsinger,

Great idea

I have a half-written utility for showing the replay info already (on the map). I'll add in the histograph bit if I can.....


I guess a second utility option for direct comparisons between multiple players wouldn't be too arduous either.

can do


P.S. - I don't think the 'eras' are saved anywhere. If anyone knows elsewhere, please PM me.

 

[Aeson]

I put this little program together for those who can't wait for Ainwood's more stable, less error prone, more feature packed utility. Also I was just a bit bored tonight.

I probably won't be updating it, but it does give a basic impression of how a game's score progresses. The scores aren't scaled quite right. Seems to work with both PtW and 1.29f saves, but I haven't tried many. The curve used for the Jason score is included, but isn't exact.

Just type in the proper MaxScore (you can get it from the calculator) and then open the saves you want to look at or compare.

CivScoreChart.zip (179KB)

 

[tao]

Originally posted by Aeson
Just type in the proper MaxScore (you can get it from the calculator) and then open the saves you want to look at or compare. (179KB)

Aeson: you might add a note on the calculator page, that as a Macintosh user, you neither see the Jason Score nor the values used. It may reduce some confusion.

Or better yet: make it working on Macintosh.

 

[Moonsinger]

Originally posted by Aeson
I put this little program together for those who can't wait for Ainwood's more stable, less error prone, more feature packed utility. Also I was just a bit bored tonight.

I just checked out your little program and I'm totally impressed. I wish I have just a little bit of your programming skill.

PS: I'm very happy that Ainwood is going to take over. If there is anything you guys think that I can help, please let me know.

 

[Aeson]

Or better yet: make it working on Macintosh.

Thanks for the heads up...

I think I figured out at least part of the problem. A function name I was using isn't supported by IE5 Mac. Can you see the Jason score and best dates in this one?

 

[Dianthus]

Aeson/ainwood (or anyone else ),

I guess that means you have some info about the file format? I would be interested in having a bash at this, but the only promising links I find using google are for Apolyton (which is currently down). Would you mind sharing some info?

 

[Jurimax]

Great work so far...

only one little problem, it doesn't work with 1.29f.
I really hope someone can fix it, thanks

Jurimax

 

[Muchembled]

There is something at http://jmuchemb.no-ip.com/JDM/ that should answer your question. Just look at the results ("résultats") of a GOTM ("JDM"), select the players you want to compare and submit by clicking on the small image.

Score : After the year of the victory, score evolves according to the bonus. It's all the more interesting since we can see if it was better to prolong the game or not. So you can see discontinuities at the victory dates and at 1750 AD : there would be other ones at 1950 AD or else. Jason score isn't used.

Power and culture : After the year of the victory, it's ... quite obvious

 

[Aeson]

After looking at the turn/stat format more closely, I've made some major fixes to the program. It actually works now! There is a color prompt with each game opened added as well for visability's sake.

civscorechartv02.zip (Win, 180KB)

Looks like the curve needs to be changed to something more along the lines of a sine wave.

 

[Aeson]

EDIT: DP

Since this post is here anyways...

I guess that means you have some info about the file format? I would be interested in having a bash at this, but the only promising links I find using google are for Apolyton (which is currently down). Would you mind sharing some info?

4 BYTES "HIST"
4 BYTES TurnsPlayed (0-541)
4 BYTES ?
for each turn:
...4 BYTES turnnumber
...4 BYTES date
...4 BYTES civs
...4 BYTES civnumbers[civs]
...4 BYTES turnpower[civs]
...4 BYTES turnscore[civs]
...4 BYTES turnculture[civs]

 

[Dianthus]

Thanks Aeson. Thats enough for a start. Have you any idea about the dates cities were founded and wonders created? I guess this would be in there because these are shown in the play back when you finish the game. I think it would be pretty cool to be able to play back the mini-maps side by side for multiple save games to get an idea of the expansion rate/technique.

 

[Aeson]

Ainwood is working on that sort of utility.

As for the format of that part of the save, I'm not sure the GOTM Staff wants to make it public at this time.

 

[Aeson]

Hows this for a new curve? I think a change of the sort is in order... any math wiz's who can polish this up a bit?

f(x)=(MaxScore*520/1000) * sin((x-310)/196) + (MaxScore*520/1000)

f(540) = 99.949% of max score

 

[Bamspeedy]

I'm no math wiz, but here is a how my game went, if you want a comparison. I imagine I was below the Jason score at the start, because of territory points I was giving up early in the game due to ICS, but the settler flood quickly surpassed it. Wouldn't which particular victory condition you go for affect the curve? I was looking for a place to put more info in (like victory condition) and there isn't any dates listed.


Score curve will be posted on May 1st.

 

[Aeson]

Thanks for showing me the graph Bamspeedy, but it might be best to edit that out until the submissions are closed. It's not much of a spoiler, but still in the general area. If you want to post it in the game submitted spoiler thread that would probably be best.

 

[Aeson]

Here's the graph of the function I posted above in case anyone is wondering f(0 thru 540). It's closer but still not right.

 

[Aeson]

...at the start

I'm not too worried about the curve up till about turn 80-90... when winning by conquest starts being possible on some maps. People just won't be winning before then, and so that part of the curve is never used.

Wouldn't which particular victory condition you go for affect the curve?

Yes, but the curve is mainly to simulate the score passed up that a best milked game could have gotten, and milked games generally follow the same progression. The curve will vary from map to map, but trying to account for that is quite a bit beyond the scope of this formula.

I just want a decent median, and if players want to crunch the numbers to find their 'best' finish for their game, it's something that basically any system is going to allow for.

 

[Muchembled]

My previous post was useless since the server crashed the same day. All is ok now and http://jmuchemb.no-ip.com/JDM/ should work.