Spring / game feature status? - Page 3

Spring / game feature status?

Discuss game development here, from a distinct game project to an accessible third-party mutator, down to the interaction and design of individual units if you like.

Moderator: Moderators

Pako
Posts: 174
Joined: 12 Jul 2009, 18:57

Re: Spring / game feature status?

Post by Pako »

Scout Helper lite

-draws a map of LOS where gradient for out of LOS is constantly darkened
-it should be distinguishable whether a position was scouted a minute, 5 minutes or 20 minutes ago
-L button changes between 4 or 6 states of LOS display
-L toggles specfullview when spectating
-I think the overall efficiency is way better than the engine LOS view(still both are always ON)

Bugs:
-LOS is not accurately in place
-some bug at area edges
-some cliffs shows through(y is 50 above ground because of this)
-resolution is shitty and update rate is barely fluent
-needs some more optimisation although performance is not bad at all
-two dimensonal arrays are used, one dimensional could be a lot faster?
-many minor bugs
-only optimized for my usage and shitty compruter
-maybe very hard to get working decently for all cases
-engine LOS contrast is set too moderately

Newer version will show blocked radar coverage as blue dimishing to green, but it's full of bugs and won't be ready anytime soon.

Image
Attachments
gui_scout_helper_d.lua
Extreme darkness
(7.8 KiB) Downloaded 119 times
gui_scout_helper_l.lua
Normal version
(7.79 KiB) Downloaded 125 times
Last edited by Pako on 24 Mar 2010, 05:21, edited 2 times in total.
User avatar
manolo_
Posts: 1370
Joined: 01 Jul 2008, 00:08

Re: Spring / game feature status?

Post by manolo_ »

used with los handler (handels the colors of radar, jammer, los)
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

What is not addressed is your desire for unscouted areas are all black. This is a bad idea. It's only good if the map is randomly generated (a possible future feature) so that no player knows what the map looks like before playing. But right now, players DO know what the map looks like before playing so it's annoying to make it all black.
Maybe it only piss off those who are not ready to more strategic gameplay. Where you can not just in the start give fast build orders for VTOL builders etc. You can not just blind shoot to specific points without actually scouting. But no one would be forced to have that. It would be just option to enable.
It makes the game play harder. Thats why FOW and LOS is used on multiple multiplayer RTS games, to bring a strategic element, not to piss off players. And it would be a _optional_ feature for games and mods, just like configuring the game speed or units speed, wind amount and other features effecting game play.
Even that player would know 100% the whole map. It still would force player to use more stratetic ways in the start. And it would allow a new player better changes in the start aagainst fast skilled player who knows the units etc.

It is like moving in your home in the dark. You believe you know your house totally but first few times you notice that you are pretty lost to find anything. You can find out stairways, doorways but to find specific things right away is much harder. It would take even more practice to actually walk around from point A to point B without touching any walls or other objects in the way.
Thats why we do have LOS and units that can shoot by the radar information and accuracy is depending about the other features. It is a strategic element same way as FOW.
zwz answered this but I will clarify in case you didn't understand. The graphic is not smooth when a unit is moving because the engine's LoS calculation is slow.
Maybe I was not so clear. I Know that graph is not smooth. Thats why the topic was created. The current LOS is ugly and it does not work. If anyone else has ever played any other RTS game where the LOS is done, it is done mostly on these days in real time, without lags, without hyper fancy coloring (I have not pressed the ; as you suggest, yet). Currently the spring LOS is more like fancy technical feature and not the feature to make gameplay more tactical.
The calculations can be easily accurate, but the graphics does not need to be tied to LOS calculation process but being a separated and visually impressive strategic element. Thats why LOS and FOW would be smooth edged. Not everything need to be a 100% accuracy by calculations but they can be balanced for the game play, without making a jerky, ugly and almost useless LOS feature like current version.
It is possible to make the graphic smooth but it would update faster than the engine's LoS calculations, so we prefer not-smooth-graphic because it's correct.


No it would not update faster. It would always be following the LOS calculation, unless the LOS calculations are not accurate and move exactly the same way as the unit. Here is said that the LOS calculations are done exactly and smoothly as the unit moves. But the graphics are not updated real time.

Even now we have real time range ring. And LOS calculations are done as well, right? But the graphical showing of that is not updating nearly as well. Think about the LOS effect being draw as smoothly and well as the range ring. Check out the starting post mockup. You can see the LOS line is blurry. Now image the LOS line to black edge of that blurry circle. And image that LOS would be smootly curved like it would be water what reflects the surface where it is flowing. The units already has acceleration and braking speeds. Units does not start moving and stop moving instantly. Their movement is smooth. And when the LOS is drawn as graphical element, like range circle. without fancy coloring or adding a more complex calculations. It is a graphical thing what is drawn by GPU, used the calculations of the current LOS (what is there because units appears and disappears from their view). But with this graphical element, that would be just nicely grayed and fainted away. The units would not disappear and appears suddendly or change to/from radar dots/icons.

Now the LOS engine seems to be tryed to be 100% accurate with .00001 accuracy and perfect by every pixel by that way. Thats why there are smooth edges on RTS games for FOW/LOS so the calculations can be done more simpler way, real time and still in such accuracy that does not bring any prolems to game play. RTS games what use hex-mode maps. It is easy because you can just drawn the unseen areas as black. Games what has similar grid, like Age of Empires II, you can just paint the grid areas same way. 3D brings lots of problems if being 100% accurate, but everything can be work out when thinking what actually is needed to be a totally accurate (example the hit area effects for units need to be) and what does not need to be (graphical representation of LOS because using the fuzzy edge. The user can see that there is unit, the computer does not need to know because it can calculate it itself already.)
Fri13 you seem to be saying that games which do not use the black area view state have no intel warfare element. Correct me if I am wrong.
There is intel, but it is easier in the begin. When you can see where exactly to go, where you can sooner place build orders, the easier the game is. When there is need to advance first, scout and then react, there comes need for much more strategic element by the player.
Like now we can see right away when enemy is building big factories etc, without scouting or having a radar. The terrain gets morphed and anyone can see that right away.
There are different levels of game play, every level adds new strategic element.
1. No FOW/LOS
2. No FOW, only LOS
3. FOW and LOS

Now Spring has only level 2 on it. And it is not good looking. It is not functional so (and that is the visual).
Even the LOS could be mixed with FOW. By making it so that you do not see anything what is outside of FOW (not even terrain) but only the targets what radar gives. They can flow in the air or be placed some strange way. (Even now many can recognize the fighter from bomber or scout vehicle from assault vehicle by the speed and the movement, even that radar shows the dot.)
It would make the whole game play more strategic than just plain LOS where you can see the map like it would be just empty.
If you know where the mex spot is you can send a unit to it and give the mex order when they see the spot. All you have added is micro.
What you call "micro" is called a strategy. You need to send scout first, when you spot the location. Then you can add build order there.
When you do not know or you can not place order before you have found something, it makes it harder to expand as you have trained to do hundreds of times. There comes more possible unknown situations what ruins your plan because you did not count them. Just by adding a FOW.
Micromanagement would be that you need to start building every thing yourself. You could not do any kind build orders for factories or any terrain what is in your control. Micromanagement would be that you can not just build predefined unit, but you would need to select unit movement types, weaponry (tracks, wheels, lasers, cannons etc) and other features.
Why even skilled players are afraid the FOW? Because they have no clue what actually they should do first. They can not think so multiple different levels to actually first scout, then build. And in the start, every player can see where they have been. What areas they have not scouted and where is possible enemy.

That is the situation in the real war as well. Maps does not tell anything. You can find perfect place for defense by looking a map. But when going to place itself, it can be unsuitable for defense. Even that you would have satellite images. Thats why scouting is very important feature and intelligence is usually that what makes biggest mistakes using too old information.
And the defender always has the greater change to win than attacker, because they have the positions already on known place. In forest you have 3:1 for defenders. In urban fights it is even 6:1 to 12:1. Because attacker does not know the terrain in first place. In time you find out the terrain how to use. But every defense positions gives different results and always giving better change for defenders.
Maybe thats why so many are not so willing to use the "Random positions" feature either, because they can not use learn tactics right away. They need to scout on what direction the enemy is, check out the area where they actually are and make plans based that. It makes much more strategic game play right away.
I never understood the concept of FOW. You think army commanders went on a new battleground without a map? Well maybe some spaniards exploring South-America, but other than that, maps and pre-battle scouting were used to give a detailed description of the battlefield since culture developed. You think huge robots waging interstellar wars would be more primitive than ancient Mesopotamia?
That was the point of the TA. The commander is send to location. It can be even random place somewhere on the planet. It was harder to play. Even times coming after first time. If we are trying to think what there would be in the future. Why they still use radar towers or vehicles? Even today we can use satellites to show units. We can have satellite telemetry to give real time visual about map. So no need to LOS less than FOW. They have big cannons what can shoot less than 20km. Even 70 years ago were huge cannons what could shoot over 80km and sink battleships with one hit. TA has missiles what fly only few kilometers or on the whole map, while today we have missiles what flies hundreds and thousands of kilometers easily. In games we have nukes what are like todays 155mm artillery explosions... Not so good point to think about such thing.

There are game play elements just for game play. It is better when we can make games harder or easier depending the players. Having a good FOV/LOS is nothing what would force everyone to be using when playing. It would be up to games and game hosts to enable such features if there would be such possible. FOV/LOS feature is about engine, not about games. If the spring does not allow FOV/LOS to being there well done, then it is harder for game makers or modders to implement one.
Titan-V2.jpeg
Only parts mockup, basic is using LOS calculations.
(166.08 KiB) Downloaded 2 times
And this is how it could look. Smooth edges, real time update (same way as the range ring is updated. The dark/gray area just get smoothly expanded/hided. When the FOW/LOS edge is blurry, it gives player still the advantage to actually see the partially units (other than just appearing/disappearing) and computer still calculates the LOS correctly for shooting etc.
Last edited by Fri13 on 24 Mar 2010, 00:19, edited 1 time in total.
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

JohannesH wrote:Obviously if you change micro you change strategies too. But it mostly just promotes map memorization, which kinda sucks.
How many currently remembers correctly where are every metal spot, every hill, shore and the routes, even on the maps what they play daily?

It is very easy to plan something for such map when you see it right away. When such map suddenly gets FOW, the gameplay change. You might not remember right away where is every thing in placed, only that what direction and what area.

But FOW is not there for whole game. It is just for beginning to make the game start more difficult. The LOS is the real key in the game, it even has great advantage when you can see very easily where every unit can see and where not. It lets player to focus to the area what there is. Knowing what areas in map are what are not never scouted and what are out of LOS or even radar (if there would be somekind nice overlay ev-en for that top of FOW/LOS, more like what the minimap has now).
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

This is the example of how FOW/LOS affects the gameplay in the start.
Attachments
high_and_low.gif
1600Kb GIF animation with 3 frames (4 second each)

1. FOW + LOS, gunships sended to patrol
2. FOW + LOS scouted littlebit more.
3. LOS (100% scouted area) only.
(1.56 MiB) Downloaded 2 times
Pako
Posts: 174
Joined: 12 Jul 2009, 18:57

Re: Spring / game feature status?

Post by Pako »

Fri13 wrote:This is the example of how FOW/LOS affects the gameplay in the start.
With the rate you type bullshit you could make a working system in no-time.
luckywaldo7
Posts: 1398
Joined: 17 Sep 2008, 04:36

Re: Spring / game feature status?

Post by luckywaldo7 »

Pako wrote:Scout Helper lite

-draws a map of LOS where gradient for out of LOS is constantly darkened
-it should be distinguishable whether a position was scouted a minute, 5 minutes or 20 minutes ago
-L button changes between 4 or 6 states of LOS display
-L toggles specfullview when spectating
-I think the overall efficiency is way better than the engine LOS view(still both are always ON)

Bugs:
-LOS is not accurately in place
-some bug at area edges
-some cliffs shows through(y is 50 above ground because of this)
-resolution is shitty and update rate is barely fluent
-needs some more optimisation although performance is not bad at all
-two dimensonal arrays are used, one dimensional could be a lot faster?
-many minor bugs
-only optimized for my usage and shitty compruter
-maybe very hard to get working decently for all cases
-engine LOS contrast is set too moderately

Newer version will show blocked radar coverage as blue dimishing to green, but it's full of bugs and won't be ready anytime soon.

Image
This widget is extremely cool btw.
Google_Frog
Moderator
Posts: 2464
Joined: 12 Oct 2007, 09:24

Re: Spring / game feature status?

Post by Google_Frog »

That was quite a long post. Argh better respond quickly or risk losing his title of "long postmaster extraordinaire".

You have stated that the current LOS display is ugly. I agree it is a little ugly but I would not like a better looking LOS display at the cost of less (or less accurate) information shown by the display.

You say a lot of things about FOW increasing stratergy as the map is not seen. This is all countered by the memorising map argument. Once you have players which have memorised the map FOW does not add any strategy (i.e. meaningful decisions).
And it would allow a new player better changes in the start against fast skilled player who knows the units etc.
I assume this means something along the lines of it gives new players a better chance. FOW is the exact opposite, new players do not know the maps.
Like now we can see right away when enemy is building big factories etc, without scouting or having a radar. The terrain gets morphed and anyone can see that right away.
I agree entirely. The updates to terrain which is outside LOS is an annoying bug. Spotting enemy buildings is an entirely different issue to start of game FOW as the FOW would mostly hide the start state of the terrain.
If you know where the mex spot is you can send a unit to it and give the mex order when they see the spot. All you have added is micro.
What you call "micro" is called a strategy. You need to send scout first, when you spot the location. Then you can add build order there.
Mere terminology. I will define my definition of micro and strategy (not always the same definitions but this is what they are here):
  • Micro: All actions that a player inputs. Everything. Mouse press, mouse move, keyboard etc..
  • Strategy: Decisions that a player must make a during a game. It can be high or low level. The decision must have at least 2 non-obvious things to decide between.
By "All you have added is micro" I meant that the FOW as not added and decisions that the player must make, all that has been added is more actions the player must input into the game to achieve the same medium level decision; "I want a mex here". The player has played the maps enough times to know where the mex to enough accuracy as to not matter. The extra actions are switching attention to the unit once it has reached the destination and telling it to make a mex.
When you do not know or you can not place order before you have found something, it makes it harder to expand as you have trained to do hundreds of times. There comes more possible unknown situations what ruins your plan because you did not count them. Just by adding a FOW.
The only added difficulty is micro. The possibility of unknown situations is unaffected by the addition of FOW.
Micromanagement would be that you can not just build predefined unit, but you would need to select unit movement types, weaponry (tracks, wheels, lasers, cannons etc) and other features.
I'd call this a decision but anyway... if the player has played the map before they will know which unit types are required.
That is the situation in the real war as well. Maps does not tell anything. You can find perfect place for defense by looking a map. But when going to place itself, it can be unsuitable for defense. Even that you would have satellite images. Thats why scouting is very important feature and intelligence is usually that what makes biggest mistakes using too old information.
But imagine that real war was fought on identical bits of land thousands of times. Generals would end up knowing exactly where to place their things. This is what RTS is like.
It is very easy to plan something for such map when you see it right away. When such map suddenly gets FOW, the gameplay change. You might not remember right away where is every thing in placed, only that what direction and what area.
Just no. Maybe you cannot remember things well but with my handy printout of the heightmap, metal map, feature map, pathing map and extensive playing of the map I'm sure I could remember enough to make FOW worthless.
User avatar
manolo_
Posts: 1370
Joined: 01 Jul 2008, 00:08

Re: Spring / game feature status?

Post by manolo_ »

bla bla long text bla bla

imagine that:

team game 5v5
one player starts with air and makes some airscouts and after 1min all player know the map --> buuu huuuuu
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

manolo_ wrote:bla bla long text bla bla

imagine that:

team game 5v5
one player starts with air and makes some airscouts and after 1min all player know the map --> buuu huuuuu
So you believe that FOW is forced to every player, every game and every time....
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

Google_Frog wrote:That was quite a long post. Argh better respond quickly or risk losing his title of "long postmaster extraordinaire".

You have stated that the current LOS display is ugly. I agree it is a little ugly but I would not like a better looking LOS display at the cost of less (or less accurate) information shown by the display.

You say a lot of things about FOW increasing stratergy as the map is not seen. This is all countered by the memorising map argument. Once you have players which have memorised the map FOW does not add any strategy (i.e. meaningful decisions).
You can not memorize all the actions. You can not just order out right away all the units movements and place buildings where ever you want without scouting.

When it makes even more difficult to give a build orders to builder to build mexes for far away. What you need first to do is scout, it takes your time away from another things and it makes it even move complex strategies to be needed used in the beginning.

It is about strategies and planning what FOW makes more important to be used correctly and every game is different. Even that there would be one map only.

Now you do not need to build so many scouts anywhere else than just to scout the areas where enemy is, to give a warning.
FOW would force to scout own side as well first.

And this is just a idea of feature what can be used on single player maps, any other game than just clone of TA's or multiplayers.

The audience is not just TA-clone multiplayers there. Audience is all Spring engine games there.
User avatar
CarRepairer
Cursed Zero-K Developer
Posts: 3359
Joined: 07 Nov 2007, 21:48

Re: Spring / game feature status?

Post by CarRepairer »

CarRepairer wrote:
zwzsg wrote:
Fri13 wrote:Terrain is black until you scout it.
TA had that in all the missions, but none of the multiplayer games (bar n00bs). Maybe there is a lesson to learn there, a lesson such as how terrain black until you scout is a bad annoyance that prevents you from queuing mexxes, or some other moral story.
In multiplayer this gives an advantage to players who memorized the map over those that haven't (or haven't even played it). Skewing an already imperfect balance in multiplayer games around here is very no-no. Not a popular suggestion. People prefer balanced games (as in both teams have an equal level of aggregate skill) and they have to settle for a "good enough" balance in spite of smurfing and other issues. There's no reason to make balance worse.
User avatar
triton
Lobby Moderator
Posts: 330
Joined: 18 Nov 2009, 14:27

Re: Spring / game feature status?

Post by triton »

boooring, download this and use L view like me :D
I think you need to press F11 ingame to enable it.

http://springrts.com/phpbb/download/file.php?id=3958

I think that fri3 idea would be good with randomly generated map ether way forget it.

edit : pako's widget seems to be far better I'll try it soon
User avatar
Spawn_Retard
Posts: 1248
Joined: 21 Dec 2006, 14:36

Re: Spring / game feature status?

Post by Spawn_Retard »

random map generation isnt possible for quick games, if you want to sit around for an hour waiting in a battle room then it is.
User avatar
manolo_
Posts: 1370
Joined: 01 Jul 2008, 00:08

Re: Spring / game feature status?

Post by manolo_ »

Spawn_Retard wrote:random map generation isnt possible for quick games, if you want to sit around for an hour waiting in a battle room then it is.
sounds like a normal xta game, so implement it :D
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

Who/where could actually use this kind smooth FOW/LOS feature?

* Singleplayer games
* Random generated maps
* Players with new/random maps
* Players with random location -enabled
* Players who are normal living people who do not memorize every pixel on the map (take it as a joke!)
* Players who want add a new level of strategy to game.
* Players who want enjoy visually nice RTS game.

For repeat:

* If you (player to host games) do not like it. You do not need to enable it from host options.

* If you (player searching multiplayer games) do not like it. You are free to choose another game to join.

* If you (game developer) do not like it, you would not need to add the engine feature to your game.

No one is here to force YOU to play games with this kind engine feature enabled.
User avatar
manolo_
Posts: 1370
Joined: 01 Jul 2008, 00:08

Re: Spring / game feature status?

Post by manolo_ »

Fri13 wrote:Who/where could actually use this kind smooth FOW/LOS feature?

* Singleplayer games for missions yes, but how much real campaigns are out there? ask argh
* Random generated maps atmo not possible, as far as i know
* Players with new/random maps noobs will be confused and there head will explode
* Players with random location -enabled why? so somebody will cb me, coz he starts near me?
* Players who are normal living people who do not memorize every pixel on the map (take it as a joke!)
* Players who want add a new level of strategy to game. bla bla bls
* Players who want enjoy visually nice RTS game. use loscolor iwdget and modify it
btw who will make this gadget/widget? move this thread to feature request. but i would like some kind of shader to hide mapdeformation by enemy building that i didnt scouted yet
Google_Frog
Moderator
Posts: 2464
Joined: 12 Oct 2007, 09:24

Re: Spring / game feature status?

Post by Google_Frog »

TLDR:
2 separate feature requests:

Feature 1: LOS that looks nice
Answer: Lua

Feature 2: Hidden map terrain at start
Answer: Not useful at the current time (no SP, campaign). Your arguments for FOW adding strategy in multiplayer have no basis in reality.
but i would like some kind of shader to hide mapdeformation by enemy building that i didnt scouted yet
There is some very simple code to fix this issue.

Code: Select all

for name, ud in pairs(UnitDefs) do
  if (ud.yardmap)  then
    ud.levelGround = false
  end
end
Unfortunately it has to be in unitdefs_post.lua, game side. The obvious conclusion is game devs feel (or outvoted by those that feel) the slight, unnoticeable graphics 'ugliness' when large buildings are built on slopes is a good reason to not fix their games.
User avatar
manolo_
Posts: 1370
Joined: 01 Jul 2008, 00:08

Re: Spring / game feature status?

Post by manolo_ »

Google_Frog wrote: Unfortunately it has to be in unitdefs_post.lua, game side. The obvious conclusion is game devs feel (or outvoted by those that feel) the slight, unnoticeable graphics 'ugliness' when large buildings are built on slopes is a good reason to not fix their games.
what do u mean with ugliness?
Fri13
Posts: 54
Joined: 03 Feb 2007, 13:15

Re: Spring / game feature status?

Post by Fri13 »

Google_Frog wrote: Your arguments for FOW adding strategy in multiplayer have no basis in reality.
Well, then many RTS games just has not implented it for strategy element for multiplayer games, including the Supreme Commander (what I just got yesterday).

It just is a stratetic element what sets player to focus to such areas what are scouted. Bigger the map, bigger the affect is.

It is psychology as well, what terrain you can not see, you do not think it same way as you would see it. Even how familiar the terrain would be.

So, because we are having now the conclusions that engine does not actually need big modifications but it can be done with LUA. The questions in the first pos is answered and results can be moved as a wish.
Post Reply

Return to “Game Development”