Tech tree - other dimension (probability)

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#16 Post by Bastian-Bux »

HoI = Hearts of Iron

Many of us are heavily influenced by games from Paradox Entertainment, a Stockholm based comapny which developes awesome hardcore strategy games.

You'll find infos about their games on http://www.paradoxplaza.com/

The greatest jewels are the Europa Universalis games (to be found in the low cost bin), Hearts of Iron, Victoria and afaik also Crusader Kings (dont own that one myself).

If you browse through this forum you'll find many references to Paradox games, cause afaik no other company has developed such a complex and still great strategy gaming system yet.

solartrix
Space Floater
Posts: 46
Joined: Thu Nov 11, 2004 3:57 am
Location: San Francisco

#17 Post by solartrix »

Looks cool. Never played it, tho. May have to fix that...

Back to the thread - any way you can send me a screen shot of their tech development screen?

Tieom
Space Krill
Posts: 5
Joined: Thu Jun 10, 2004 9:09 pm

#18 Post by Tieom »

Just to toss in $0.02:
The three sliders could be replaced with 1 bar that you move two dividers around on.
It's neater, and can be fit into a smaller space, but might require more programming.

Mock-up:
Image

Doesn't require any 'lock slider' buttons, either... I find those annoying.

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#19 Post by utilae »

Hey, that's pretty sweet. Simple and takes up little space. Nice to have the different colors too.

Ablaze
Creative Contributor
Posts: 314
Joined: Tue Aug 26, 2003 6:10 pm
Location: Amidst the Inferno.

#20 Post by Ablaze »

Just a heads up: One or more of the programmers are strongly opposed to anything slider. I'm not really sure why, but a while ago one of my ideas was dismissed out of hand in one of the design threads because I used the word "slider" when describing the UI.

That was the reason given too.. "We've avoided sliders so far, I don't see any reason to start using them now." Nothing about the actual content of the idea.

I would avoid putting a slider on your UI if I were you. Perhaps you can obfuscate the fact that it's a slider by making it look like a button or something.
Time flies like the wind, fruit flies like bananas.

Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#21 Post by Bastian-Bux »

:twisted:

Ablaze, yeah, you happnd to put your hand in an hornet nest there :). To sad our old Forii are gone, you could read pretty interessting stuff there (and I would be much higher up in the posting list). :)

noelte
Juggernaut
Posts: 872
Joined: Fri Dec 26, 2003 12:42 pm
Location: Germany, Berlin

#22 Post by noelte »

Ablaze wrote:Just a heads up: One or more of the programmers are strongly opposed to anything slider. I'm not really sure why, but a while ago one of my ideas was dismissed out of hand in one of the design threads because I used the word "slider" when describing the UI.
hehe, that was not a programmers decision. Blame Aq. or Tyreth .... :P

Honestly, i believe it was a agreement between the fo designers ....
Press any key to continue or any other key to cancel.
Can COWs fly?

Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#23 Post by Bastian-Bux »

:). Its a pure design decision we made in the design team. So don't blame programmers.

Actually there are severall keywords that will bring one or more designers to jump to you throat. *whisper* slider is one of them, snowball effect another one.

/me indulges in remembrances of that heated discussions. :twisted:

User avatar
skdiw
Creative Contributor
Posts: 643
Joined: Mon Sep 01, 2003 2:17 am

#24 Post by skdiw »

We decided to use HOI style UI because HOI is simpler, there gonna be too many sliders otherwise, and micro is required for sliders. I originally thought about how to improve the slider system, but I found the HOI is better in the end. You just click on the tech you want and everything else is taken care of. Each tech have a fixed number of turns that it will be researched and a cost/turn. You can click on the tech and cost/turn will be deducted from your rp pool then you can your tech after X turns.

The sad part of tech tree design is that there won't be random factor because of balancing purpose. Every race gonna get the whole tech tree like every other race. There won't be much unique techs or randomness.
:mrgreen:

Bastian-Bux
Creative Contributor
Posts: 215
Joined: Fri Jun 27, 2003 6:32 am
Location: Kassel / Germany

#25 Post by Bastian-Bux »

skdiw, that might be true, but who says that every race is equally adept at every branch of the tech tree? Or who says that there won't be boni or mali for a partuclar research area (theory, application, refinement)?
Or who says that there won't be specific boni to severall techs for a race (like boni on algae farms and other ocean based thingies for aquatic races)?

There are many ways to form the research of a race without "capping" techs out completely. If my boni for shields add up to 100%, and my mali for armor are as high as 50%, well then I'll logically be 4 times better at shield techs then with armor.

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#26 Post by utilae »

Ablaze wrote: Just a heads up: One or more of the programmers are strongly opposed to anything slider.
It was Drek. I think it is rediculus really. A slider has advantages that other control don't. I think in certain cases a slider is better, especially for distributing X among A, B & C (like Tieom's slider mock up shows).

Name a control that would do what Tieom's slider mock up does, but better.

iamrobk
Space Dragon
Posts: 289
Joined: Sun Sep 07, 2003 12:27 pm

#27 Post by iamrobk »

I must agree with utilae. I believe that a "no sliders" attitude is definately bad. It really restricts how much we can do.

Impaler
Creative Contributor
Posts: 1060
Joined: Sun Jun 29, 2003 12:40 am
Location: Tucson, Arizona USA

#28 Post by Impaler »

Ok lets not digress into the Slider debate, its already a done deal and as Ablaze stated any use of sliders is dismissed out of hand by senior development team members. Just accept it and move one.

Now as for the original topic, the predictability of reserach and how to make it less predictable yet still fun. I am oposed to any system inwhich my choices to reaserch one thing will perminently reduce my chances to aquire something else, its simply not fun and is its use is a poor substite for a balanced tec tree.

I like the Moo1/2 "Spend X points then get a % chance to discover" system but I am worried that combining it with HoI might be too complex, its definatly worth looking into though and we should brainstorm on how to implement it well.

My personal preference is to use the Moo1 system inwhich theirs is a TEMPORARY inability to reserch cutting edge tecnologies, after sufficent advancment all of the older tecnologies open up and can be aquired. This will avoid situations inwhich a low level tecnology is realy nessary/critical for the player to have in the mid/late game and they are screwed if they dont have it. Rather all tecnology restrictions are short term, you have to do with out Hyper missles for longer then normal but you eventualy get the option after some other weapon system have been developed. Hyper Missles by this point arn't going to be cutting edge but could still be a major improvment over your realy obsolete missles.

Some method of desiding what constitutes High/low tech is needed and again I would steal from Moo1 here and go with a simple "Overall Field Level" for each of tec Catagories. Level 36 Growth, Level 21 Production ect ect. This tec level number is produced by adding up values for each Theory, Aplication and Refinment you possess in the field. The addition is not strictly adding up a sum, rather you get the Raw value of your highest Theory, half the Value of your second highest, third for your third ect ect and 1 for every Aplication and Refinment.

This tec level can be used throughout the whole tec tree as a pre-reqisite for Theories, Aplications and especialy Refinment. Every 10 levels of Field Theory alow the devices of that field to be refined to the next level, so in the example above you can Refine your Growth Devices to level 3 and your Production devices to level 2. Their would ofcorse be a level 0 devices which is what you first recive when you do the Aplied Reserch afterwards each Refinment adds one level to the device (think of this like getting Upgrades in StarCraft, you start at 0 bonus and get "Bonus level 1", "Bonus level 2" ect ect), Think of Aplication as simply the 0th level of Refinment.

Another idea along the same lines. When a new Theory is Discoverd the player dose not imediatly have the option of researching all the the Aplications that fall under it. Rather only a small random selection is avalible (possibly just one choice). Only after this Aplication is completed dosoe another open up, eventualy all of them can be reserched given enough patience.
Fear is the Mind Killer - Frank Herbert -Dune

Tieom
Space Krill
Posts: 5
Joined: Thu Jun 10, 2004 9:09 pm

#29 Post by Tieom »

Ah well... Too bad about the slider(s).

...

*throws his $0.02 down the drain*

solartrix
Space Floater
Posts: 46
Joined: Thu Nov 11, 2004 3:57 am
Location: San Francisco

#30 Post by solartrix »

I have a real wild tech-tree idea to throw out there. This will sound crazy, and it's kinda tough to explain without a whiteboard, but hear me out...

What if the first level of theory (setting aside applications and refinements for a moment) had 5 possible breakthroughs, and as those breakthroughs were unlocked (discovered), they were used to populate the row and colum headings of a 5x5 matrix. The 5x5 matrix becomes the basis for the 2nd level theories (essentially, all of the 2nd level theories have 2 prerequisites).

If your level one theories were A, B, C, D, E then your level 2 theories are now AB, AC, AD, AE.... (there would 10 of them, if I did my math right). Some of the combinations might be duds or dead-ends, but others would be useful and would be used to populate the level 3 theory matrix. Again, you could end up with 10 level 2's that get passed on up as the basis for the level 3 theories. And some of these level 3 combos would be passed on up to the level 4 matrix. And so on and so forth, but maybe only for 6 or 7 levels until we literally had 100's of theories to play with.

You still wouldn't have much control over which theory was next in your que. In fact, your "theory points" might work something like a big tank with different marks for the various levels. For example, a level 1 breakthrough costs 100 research points, level 2 is 150, level 3 200, so that once your research point "tank" fills up to 100 points, you have an X% chance of getting a a level 1 breakthrough. But if you get to 150 points (no level 1 breakthrough yet), you now have an X% chance of the level 1 breakthrough and an X% chance of having a level 2 breakthrough, and so on. Once you get the breakthrough, 100 or 150 points are taken out of the tank and the game of chance starts over. So, you could get a level 2 combo breakthrough before you got all your level 1 theories. Or maybe even a level 3 breakthrough before you had very many of your level 2's.

What's interesting about this (to me, anyway) is that there could be so many combinations that your tech development would no longer be linear and might actually be quite different from game to game. All the same theories would still be in each game, but their order of appearence would be dramatically different because the "tech tree" would now be much broader than it is deep, and your scientists could be "working on" very different levels of theory all at the same time.

If we tried something like this, we could still keep the applications and refinements pretty much linear, or use Impaler's idea about keeping some of the applications hidden at first (which might be another nice twist).

Does any of this make sense to anyone else? Or am I totally nuts?

Post Reply