Welcome, Guest. Please login or register.

Pages: 1 [2] 3 4
Print
Author Topic: Map layout etc...  (Read 34645 times)
egalor
Community member

Posts: 94


View Profile Email
« Reply #15 on: April 30, 2009, 10:45:31 AM »

Actually, as for now, I'm not prepared yet to tell how many tiles per map would be best. However, in order to start doing map sketches I will need some sort of scale as a first thing.

And secondly, will the maps be multi-levelled, like in Fallout?

Logged
mvBarracuda
Admin
Community member

Posts: 1308



View Profile Email
« Reply #16 on: April 30, 2009, 02:30:36 PM »

There is no elevation concept like in Fallout egalor. Elevations were just little hacks to get multiple maps into one map file and they don't really provide any advantages. For now we can assume that different "elevations" of a map (e.g. different levels of a skyscraper) reside in different map files and are magically connected to each other via some C++ or Python code.

So when you change the elevation, a new map will be loaded, similar to how it was handled in Fallout (though the elevations resided in the same map file in Fallout, but it does not really make a difference in this context).
Logged
maximinus
Community member

Posts: 694



View Profile Email
« Reply #17 on: April 30, 2009, 02:37:44 PM »

Yep, that's right, maps are flat.

It MAY be possible to crowbar in some neat tricks to make it *appear* that we have some elevations, but I wouldn't rely on it right now. On the other hand, we should be able to have 'cover' with different heights. But the PC will always be at the same height.
Logged

Science is open-source religion
zenbitz
Moderator
Community member

Posts: 1164



View Profile
« Reply #18 on: May 01, 2009, 01:36:54 AM »

I don't see how you are going to have a passibly realistic game in SWEDEN without hills.  I suppose it doesn't literally matter whot he engine handles it - but maybe it's a exercise for the graphics dept.
Logged

We are not denying them an ending...
We are denying them a DISNEY ending - Icelus
maximinus
Community member

Posts: 694



View Profile Email
« Reply #19 on: May 01, 2009, 02:52:28 AM »

I don't see how you are going to have a passibly realistic game in SWEDEN without hills.

Agreed. But this is a limitation of the FIFE engine I'm afraid. To be dreadfully honest, I'm not that impressed with the feature set of FIFE. I mean it works, it's stable, it seems reliable but for building a game it still leaves a lot of work to be done (I'm not saying I could do better though - I appreciate it's a lot of work and too easy to complain!)
Logged

Science is open-source religion
Gaspard
Community member

Posts: 479


The Blackest Heart


View Profile
« Reply #20 on: May 01, 2009, 08:21:06 AM »

indeed more stress to the graphic department I'd think:



this is a map shot of a location in Baldur's Gate 2. It's the Umar Hills. As much info as I have conjured up over time about the Infinity Engine then that didn't use any real elevation either - it was all the visual tricks of the graphic departments.

So - in this map you're basically "climbing" onto a huge overlook in the NW corner of the map on top of which is a building (Valygar's Cabin). You can approach this directly from the south (note the stone slabs like steps) or from the east (also some stone slabs that look like steps). And well - this map has a very concrete 'hill' plus the high rocky outcroppings which mainly just act as visual candy.

Some more realistic impact could essencially be coded into the game in the fashion of blocking certain tiles etc. Like label tiles in a certain way that when you shoot arrows from a tile that's labeled 'low terrain' to a tile that's labeled 'high terrain' or 'hilltop' you do less damage or the % of a critical miss would be higher. Although that's more into the mechanics' dep. alley
Logged
maximinus
Community member

Posts: 694



View Profile Email
« Reply #21 on: May 01, 2009, 01:26:07 PM »

We could store a z value for each tile easily, so I'm happy to leave the rest of the problem to another dept!  Tongue
Logged

Science is open-source religion
Lamoot
Community member

Posts: 161


View Profile
« Reply #22 on: May 02, 2009, 02:25:58 PM »

As far as elevation goes, it can be faked rather well, even using the Disciples 2 method (not exactly called that, but the best example at our wiki. The approach in this game is not developed to the fullest, since the cliffs are too straight to really feel natural.

http://wiki.parpg.net/Disciples_2

Another game which used the same method is Out of This World and here the illusion is rather convincing.

http://www.elitesoft.de/uploads/images/39.jpg

http://www.elitesoft.de/uploads/images/25.jpg
Logged
Gaspard
Community member

Posts: 479


The Blackest Heart


View Profile
« Reply #23 on: May 08, 2009, 07:34:59 PM »

quoting barra from http://forums.parpg.net/index.php?topic=161.0

We're still puzzled if you should be able to enter buildings without a map switch or if the interior of buildings should reside in a separate map that gets loaded once you enter the building. Both approach got pros and cons: separate buildings are easier to handle when it comes to graphics creation and programming; in case of buildings that you can enter without loading a new map, you will need to worry about z-ordering and blocking as well as proper splitting of buildings (in several one tile-width wall pieces) to work around that.

On the other side buildings that you can enter without a map change provide some additional gameplay depth. E.g. You could fire from the inside of a building through the windows of it at somebody who walks around outside. That would also work the other way round.

Anyway my thoughts concerning the 'open' buildings vs the separate-map buildings:

When you look at Baldur's Gates and Icewind Dales then you can definitely make some awesome eye-candy more easily with separate maps, but then then the possible tactical approaches to infiltrating buildings (sneaking under windows/listening behind closed doors) and combat (shooting through windows or sneakily sniping a tavern patron from a house on the opposite street during the fraction of a moment when someone else opens the door to enter/exit) etc are cut down in number.

But when the 'outside' map is flexible and interesting enough and actually offers various types of cover and obstacles for tactics then a separate map for buildings wouldn't be too bad, or what ?


---

Has it been decided what the scope of this 'PARPG demo' is going to be ? I remember from earlier discussions that it's not going to be a large thing consisting of various or too many different locations. But has it actually been decided on how big is it really going to be ? How large of a map, how many buildings, a range of how many NPCs etc ?
 
If not then here's a thought/minor proposition that came to me, oddly when I was reading that post by mvBarracuda:

How about loosely designing one minor location in some detail, a village or an urban suburb perhaps in which the demo took place ?

As people are trying out modeling buildings and characters already then this design(-document) might include a semi-detailed map of how and where buildings/trees/hills etc would be distributed including a few inhabitants of the place. As of yet the background story and mechanics are not wholly in place - so this would be a sort of playground for the programmers and graphic artists, but instead of being a hodgepodge of buildings and scenery it would be an orderly game map already.
As the story engine gets developed and scripts added the demo location would slowly start coming to life and various aspects like combat and sneaking/running/skiing can be tried out. When 'finished' it would be something like the demo for Van Buren was - not necessarily part of the final game, but still in the game world and showing off what the game's got to offer.

The location should be relatively independent from the game world (so the writing department could work on the storyline and world history etc entirely separately) and .. um.. flexible ? I could not come up with a better descriptive word :/

An area which has a shoreline or a river (to try out how water and ice could look and work), trees, buildings, debris, snow, a hill or two etc - various scenery that the player could theoretically interact with and use in combat situations/moving around the place which would all be used in the final game.
A character or two that inhabit the place that in due time would get personalities and roles and could give the player simple quests that involve the scenery of that specific map or maybe even another location that you would access by 'travelling to' on the World Map (to test and try that out, also). (dumb quest example: "Could you fetch me some dry coal from the coal-house a bit further away from here - you'll get to it by following that path over there which leads over the frozen river, watch out and try not to fall through the ice though...")

A separate thread for this kind of location's creation could be started in the General Discussion or Writing subforums.
Logged
maximinus
Community member

Posts: 694



View Profile Email
« Reply #24 on: May 09, 2009, 08:15:48 AM »

How about loosely designing one minor location in some detail, a village or an urban suburb perhaps in which the demo took place ?

That´s kinda what is happening right now. But you are right, it´s very loose.
Logged

Science is open-source religion
Dave Matney
Community member

Posts: 230


29640090 naveofhearts@hotmail.com naveking matneyx
View Profile Email
« Reply #25 on: May 09, 2009, 03:04:10 PM »

I would suggest a combination of both styles, if it's not too difficult to code.  The Baldur's Gate style allows for some huge-seeming levels without having to worry about actually programming in the rules for multiple level buildings and basements. (Or caves, mine shafts, whatever)  I can actually see the shipwrecked submarine scenario benefiting a lot from this method.

On the flip side, I always thought it was annoying in BG and IWD to have to load a map just to enter a single-level shop, then reload the city map when I left.  Especially when playing BG-Tutu, which threatened to crash every time it loaded the Beregost map.

Oh, and the reason that those early Black Isle games were such eye candy is that every map was developed as a single tile (I actually seem to remember reading that some of them were even hand painted) first, then they laid their grid system over it and assigned where the player could and couldn't walk.  Then they worried about buildings and trees and stuff.  (Just to ask, is this an option in FIFE?)
« Last Edit: May 09, 2009, 03:09:15 PM by Dave Matney » Logged
eleazzaar
Community member

Posts: 255


attempting lucidity


View Profile
« Reply #26 on: May 09, 2009, 06:24:35 PM »

Hmm I would favour buildings that you can enter without loading a separate map. It's possible to implement it by splitting up the buildings (and larger objects in general) into tile-wide pieces.

I don't understand what function slicing the building graphic into strips serves.  Is that simply the method to remove obstructing walls so you can see what is behind, if so, surely there are other possible methods.


There are two distinct issues here that are getting confused.

1 * Weather indoors are part of the same contiguous space with outdoors.

2 * Weather the game must load the map when moving between the inside and outside.


Besides the two scenarios discussed, it should also be possible to load an outdoors area and all the interiors of buildings in that area in one chunk, so that while the interiors are presented to the player as separate "maps", there is no loading necessary while moving inside and outside.



Has it been decided what the scope of this 'PARPG demo' is going to be ? I remember from earlier discussions that it's not going to be a large thing consisting of various or too many different locations. But has it actually been decided on how big is it really going to be ? How large of a map, how many buildings, a range of how many NPCs etc

At this point any such decisions would be arbitrary.  IMHO the demo should be released when we have enough stuff done well enough to give an idea what the game is supposed to be like.  Nobody can predict exactly how well or how much different parts of the game will be done when everything passes the "good enough to be a demo" threshhold.
Logged
maximinus
Community member

Posts: 694



View Profile Email
« Reply #27 on: May 11, 2009, 04:00:13 PM »

Has it been decided what the scope of this 'PARPG demo' is going to be ? I remember from earlier discussions that it's not going to be a large thing consisting of various or too many different locations. But has it actually been decided on how big is it really going to be ? How large of a map, how many buildings, a range of how many NPCs etc

At this point any such decisions would be arbitrary.  IMHO the demo should be released when we have enough stuff done well enough to give an idea what the game is supposed to be like.  Nobody can predict exactly how well or how much different parts of the game will be done when everything passes the "good enough to be a demo" threshhold.

Mmmm..... there shouldn't be a demo release.

WHAT! Of course there should be a demo release, I hear you say. and of course I agree. But these are not contradictions. I believe in release early, release often, so we always have the newest and best code in SVN. Building a demo would just divert attention away from other jobs. We just update code slowly and carefully, all of a sudden we will have a little demo. Just don't concentrate on 'THE DEMO' as such. We're not a games company  Wink

When it looks good, we bump up the version number and let barra make more noise about it  Tongue
Logged

Science is open-source religion
mvBarracuda
Admin
Community member

Posts: 1308



View Profile Email
« Reply #28 on: May 11, 2009, 04:07:05 PM »

Oh, and the reason that those early Black Isle games were such eye candy is that every map was developed as a single tile (I actually seem to remember reading that some of them were even hand painted) first, then they laid their grid system over it and assigned where the player could and couldn't walk.  Then they worried about buildings and trees and stuff.  (Just to ask, is this an option in FIFE?)
You could get such large background images working with FIFE in a number of different ways but there are no plans to go down this route. That would mean abolishing the tile approach and more or less handpainting the whole background. You would need to have experts in the field to make it look good and to me it's simply not worth the effort.

Most important point: tiles are reusable so you're way more flexible this way.
Logged
maximinus
Community member

Posts: 694



View Profile Email
« Reply #29 on: May 11, 2009, 04:32:04 PM »

Oh, and the reason that those early Black Isle games were such eye candy is that every map was developed as a single tile (I actually seem to remember reading that some of them were even hand painted) first, then they laid their grid system over it and assigned where the player could and couldn't walk.  Then they worried about buildings and trees and stuff.  (Just to ask, is this an option in FIFE?)
You could get such large background images working with FIFE in a number of different ways but there are no plans to go down this route. That would mean abolishing the tile approach and more or less handpainting the whole background. You would need to have experts in the field to make it look good and to me it's simply not worth the effort.

Most important point: tiles are reusable so you're way more flexible this way.

Custom painted maps look cool but are much more work. And a bugger to edit later on. I actually thought of this a few weeks ago pro the cons outweigh the pros by quite a margin, especially for a project like ours.
Logged

Science is open-source religion
Pages: 1 [2] 3 4
Print
Jump to: