next »

Well, I certanly don't know how this will impact on your project, but I've been wondering about implementing the following ideas into a strategy game:

• (Like in Commandos) To have a field of view for each human unit. This will improve the posibilities of strategies, also this will also improve the game (imagine a vehicle controlled by only one person.. -pause- now imagine a truck with some soldiers at the back -pause- don't you think the second vehicle will have an improved view than the first? Now imagine the SAME truck but with only one-quarter of the people and the driver -pause- The field of view will be quite minus expanded... That is what I mean. Soon I will post here some images to make you focus better on my idea. That, in my oppinion, is great for an RTS.
• (like in a game I like much) To have huan units with different names and surnames, different voices and faces.. This will make this game not just more real, also more inmmersive for the player and it will expand its stability.
• (Like what you get with endorphin) This will sound a bit abstract, but should be implemented to make an ideal good RTS: To have a Dinamic Motion Synthesis based engine, where intelligence and physics and also materials meet together, creating a flexible and wide set of posibilities for the player to get.
• Make soldiers be able to interact and modify their enviroment (for example, like in a game I know, human units are needed in a building to create vehicles.. And vehicles need humans to run and move and shoot.. As they also need fuel to also do that..
  Say, a lab needs humans from one proffesion, Scientistic, to get a research done, and so to let the army have acess to more technologies.
  ).
• • As said before, to have different proffesions for humans in game, so that these humans gain also experience in them (or have some "innat" experience), and therefore operate certain elements defined from the source by the programmer (for example, a vehicle  or a turret, by anyone.. etc) and to improve those elements when the character has the "tools" (say, a character that has changed his proffesion will wear, for example if he is a soldier, a different cloth, say a Kevlar vest and a camouflaged hat, or if he or SHE is a scientist, then he'll wear a shirt and will have a briefcase and pants adecuated to his enviroment, the lab.)
  • • Dynamic motion Synthesis should be present if is the case, as this would be good for anything! Imagine the process of animating everything? This will take months, and in a extreme case a year! If we have a human unit that has a different proffesion, would we use a different model (more animation procces) or we'll attach them a new cloth (with cloths physics also take part)? In this case D.M.S takes care of all this process, and it also adds the VERY NEEDED implementation of materials to physics, so that an object, the same one, could interact with different surfaces and react differently.
    Once I have more time I'll post more. But By the way, PLEASE, Think on this implementations and post oppinions.. I know it isn't an easy work, but this will expand even more the posibilities for this. And if you know, in real life a person do many jobs at the same time  , imagine in the military life..
    Think on it, it will be worth for it.

Your ideas are nice, only I wonder if would be able to use this Dynamic motion Synthesis. It sounds like patented stuff to me. :\

Solinx

dynamic motion synthesis is just a difficult word for dynamic animations
so even if that particular technology/name is patented, there is nothing to stop us from dynamically generating our own animations

however, i do fear that it would be hell for performance, and that some of these ideas may be too complex for the average player. it's not like in real life, where commanders have subordinates that do a part of their job. in an RTS, you have to manage everything, and quite quickly at that 

dynamic motion synthesis is just a difficult word for dynamic animations
so even if that particular technology/name is patented, there is nothing to stop us from dynamically generating our own animations

however, i do fear that it would be hell for performance, and that some of these ideas may be too complex for the average player. it's not like in real life, where commanders have subordinates that do a part of their job. in an RTS, you have to manage everything, and quite quickly at that 

I understand your oppinion. I respect it.
When I wrote about Dynamic Motion Synthesis I referred to Natural Motion. Maybe you can deal with them, they have great stuff.
The idea about proffesions, names, surnames, individual characteristics(face, voices) is based on experience, the gameplay will be expanded significantly if these things are implemented.-
The way victory is get is through sacrifice..
About the Field of view, this would expand the AI autonoumus capacities, as the player can see what do their units are viewing (you can create a button to toggle on/of this real field of view..) Remember the examples about vehicles I've given you?
So in conclusion, if you want to have more massive control, I stay with the provided ideas by me: the technollogy provided by Natural Motion serves to improve the AI of our units to react with the enviroment in an intelligent way; The proffesion, skills, human unit unique characteristics will bring more flexible control, leting the player know what units are more experienced or more specialized in their carrer, and whom has to be took out of fight; Also the implementation of human-controlled units if you like to call it like that, will bring also even more control to the full aspect, as it is all managed by the user, say, the experienced scientists go to the lab (if the person isn't an stupid he will choose that), the unexperienced soldiers could go to the factory and develop vehicles there, some mechanics could take care of vehicles, and further more.. So in this way, the player has a severe control over the massive behaviour of his army, and HE creates his own strategy, HE says how the things are done, and HE makes the things even more easier or more difficult to be solved..
In my oppinion, Massive control isn't limited by my ideas, even more, it is expanded considerably.
Please, tell me if my acclaration is wrong, otherwise, give an oppinion about it, as if you think they're useful, we can take a look at them..
PNGS and reference images will be provided soon. If you're more interested on my ideas contact me on MSN today at GMT -3 5:00 PM to 6:30 PM
MSN adress: wren206@hotmail.com
Best regards to all the dev. team,
Aleph Wren.
PD: This Dynamic Motion Synthesis technology uses not much resources, there are plenty of users testing endorphin.. Where do I fundament my hipothesis? On Google Video. So from that I can say that this is a low-performance-cost technology, oriented to memory.

It will take some time to find and perhaps type too, but I'll try to dig up an idea I wrote about just a few weeks ago. An idea, for the larger part about training units, that would by large complement or coincide with what you have suggested here.

As for the animations... shows that I still have a whole lot to learn about these things

Solinx

(Like in Commandos) To have a field of view for each human unit. This will improve the posibilities of strategies, also this will also improve the game (imagine a vehicle controlled by only one person.. -pause- now imagine a truck with some soldiers at the back -pause- don't you think the second vehicle will have an improved view than the first? Now imagine the SAME truck but with only one-quarter of the people and the driver -pause- The field of view will be quite minus expanded... That is what I mean. Soon I will post here some images to make you focus better on my idea. That, in my oppinion, is great for an RTS.

I think the best way to implement this is to limit field of view while the unit is fighting and marching.  If it's standing still, the units facing angle is arbitrary anyways.  When marching, a penalty to line of sight on the rear should be implemented, and in combat the flanks should be penalized.  In neither case should sight be eliminated, just penalized.  I'm not sure if the game mechanics will work out in favour of this system, but in any case, the engine should be able to support it.

(like in a game I like much) To have huan units with different names and surnames, different voices and faces.. This will make this game not just more real, also more inmmersive for the player and it will expand its stability.

This all depends on scale, of course.  If we have hundreds of units (or even as few as 50) commanded at a time, it really only makes sense to do this for commanding officers and heroes, not individuals.

Make soldiers be able to interact and modify their enviroment (for example, like in a game I know, human units are needed in a building to create vehicles.. And vehicles need humans to run and move and shoot.. As they also need fuel to also do that..
Say, a lab needs humans from one proffesion, Scientistic, to get a research done, and so to let the army have acess to more technologies.

The problem here is management; it's taxing on a player to handle the logistics of making sure he has someone doing something everywhere.  The question is whether the added depth will merit the added management.  While having the capacity for fuel and "operator-required" objects would be great, I do think that in the vast majority of cases this is more taxing on the player than anything.  Modifiable environment kind of goes without saying; I think flammable forests and terraforming options for workers are already taken for granted, but beyond that there isn't much we can do that doesn't really reek of "prescripted".

As said before, to have different proffesions for humans in game, so that these humans gain also experience in them (or have some "innat" experience), and therefore operate certain elements defined from the source by the programmer (for example, a vehicle  or a turret, by anyone.. etc) and to improve those elements when the character has the "tools" (say, a character that has changed his proffesion will wear, for example if he is a soldier, a different cloth, say a Kevlar vest and a camouflaged hat, or if he or SHE is a scientist, then he'll wear a shirt and will have a briefcase and pants adecuated to his enviroment, the lab.)

It doesn't seem that you're talking about a RTS anymore.  I think this, more than anything, sounds like the "create-a-hero" system of Battle for Middle Earth, except with a wider list of aesthetic options.  I'm not against that, but I also think the focusses you talk about aren't geared towards the kind of gameplay the genre has.

Thanks for spending the time to write up this post, we appreciate your input.
 

This all depends on scale, of course.  If we have hundreds of units (or even as few as 50) commanded at a time, it really only makes sense to do this for commanding officers and heroes, not individuals.

the player doesn't have to name them himself. just giving a unit a name by combining a random first name and a random surname and a random face will be extremely easy to do, and is a nice touch

YES! Depending on the nation of the soldier!
You see? My ideas aren't an utopy
Also keep processing the skills and experience of the soldier with them also, as I am a faithful defender on that.

I think this idea is to bring some reality in the game through randomness.
If two persons take the exact same training, it doesn't mean they are equally adapt at what they learned during that training. To apply it to the game: Each individual unit would differ in combact effectiveness, resulting in one unit being better than another, even tho he is of the same type.

I'd love to go beyond that and think it should be an option to mod the game beyond that, but I'll be realistic and say we shouldn't over do it at the main game, which should also be accessable by casual gamers.

Solinx

Each individual unit would differ in combact effectiveness, resulting in one unit being better than another, even tho he is of the same type.

that is ok, but players should not lose or win by luck. as such, the average of a battalion is always as good as another one of the same type and training

Due to the randomness, we cannot assure that every battalion would be the exact same strenght as each other battalion of the same type, but the overall average should indeed be the same. The individual changes should not play too much of a role.

They do in my writing, but that is still unfound

Solinx

we cannot assure that every battalion would be the exact same strenght as each other battalion of the same type

with proper algorithms you can

we cannot assure that every battalion would be the exact same strenght as each other battalion of the same type

with proper algorithms you can

The experience on each skill will do the difference Look:
While you multiple skills improved, the armor of the person will be increased, as he knows more, and thats something I've taken from the game Original War..
If you tell a soldier to go and fight, and he survives, then He'll be better than the rest of his buddies; Also, you can establish a global parameter as a limit, for example, any soldier could not pass fron the skill level 3 from X ability (for example X=Mechanical skill)..

the player doesn't have to name them himself. just giving a unit a name by combining a random first name and a random surname and a random face will be extremely easy to do, and is a nice touch

It's not a matter of the player naming them, it's a matter of the player actually caring that they have names.  If we have 100 soldiers on the field at a time, it is highly unlikely that the player will select only a single soldier at a time.  If he selects any more than one, we really need to make decisions about what information is necessary, and what is clutter, and "proper names" fall into the clutter category.  This makes absolutely no sense if units are commanded in a "battallion" type system, and still is only noticeable in a practical sense if you are commanding two dozen troops (tops).

I think this idea is to bring some reality in the game through randomness.
If two persons take the exact same training, it doesn't mean they are equally adapt at what they learned during that training. To apply it to the game: Each individual unit would differ in combact effectiveness, resulting in one unit being better than another, even tho he is of the same type.

I'd love to go beyond that and think it should be an option to mod the game beyond that, but I'll be realistic and say we shouldn't over do it at the main game, which should also be accessable by casual gamers.

This has come up before, and I'll repeat my arguement earlier: if it has a noticeable effect on combat, then it can be unfair.  If it doesn't have a noticeable effect on combat, then it's irrelivent to begin with (and some people will still claim it could potentially be unfair!).  I will always remember one day in warcraft III where my blademaster got five consecutive critical strikes.

I can see your point about information overload, but

and still is only noticeable in a practical sense if you are commanding two dozen troops (tops).

Even with my trouble of remembering names, more than 24 are possible, as long as you don't constantly use them in one large group.

{rant mode}

This has come up before, and I'll repeat my arguement earlier: if it has a noticeable effect on combat, then it can be unfair.  If it doesn't have a noticeable effect on combat, then it's irrelivent to begin with (and some people will still claim it could potentially be unfair!).  I will always remember one day in warcraft III where my blademaster got five consecutive critical strikes.

It's called luck. Happens in real life often enough. So why should games be statistical models in which the outcome of every conflict could actually be predicted by just comparing a few numbers that are set in stone and never vary one bit?

I find it illogical to leave out randomness. How can it be unfair if the odds are equal? Sure, people will percieve it as unfair when they loose, but you won't hear anyone about it if they win. It's an easy thing to bitch about, but if chances are equal, there is nothing unfair about it.

In the past randomness was probably left out because of the added complexity in the calculations, meaning computers couldn't handle it, but that is certainly no issue today. Games could be so much more complex and realistic in their calculations today, but instead of enhancing gameplay, the visual appearance seems the only part that noticably improves.
{/rant mode}

Solinx

It's not a matter of the player naming them, it's a matter of the player actually caring that they have names.  If we have 100 soldiers on the field at a time, it is highly unlikely that the player will select only a single soldier at a time.  If he selects any more than one, we really need to make decisions about what information is necessary, and what is clutter, and "proper names" fall into the clutter category.  This makes absolutely no sense if units are commanded in a "battallion" type system, and still is only noticeable in a practical sense if you are commanding two dozen troops

well it's just a matter of detail. the player hovers over a soldier and sees "Hank Pierce" or something in the tooltip. he doesn't have to remember or use that in any way, but it's a nice touch

There's no problem with having the capacity for such a naming convention in the engine (in fact, we should have the capacity).  The question is whether it's going to be noticeable or just clutter.  I think if we are going with a battallion-type system where units are created and managed in groups, this will be clutter no matter what.  There is just so much information on such a large scale battlefield that a name isn't meaningful.  If you cursor over a (group of) soldier(s), what should be seen is important information like his (their) health or fatigue status.  Soldiers will fight and die nameless because that is the reality of warfare.

As for randomness, the point in fairness is this: if the statistical odds can vary enough that the players find it noticeable then something is wrong.  If after strategy, tactics, micromanagement, and all that junk is applied we actually notice that a group of units is performing well above average, then the benefit has gone beyond random; it is downright unfair.  If it is not noticeable above other factors, then it was pointless to begin with.

I believe randomness is best modelled internally instead of externally.  An example of external random would be the modular division by 11 of how many seconds have passed since midnight on September 20th, 1978.  It gives us a random number (not necessarily a well distributed random number, but it's an example of how randomness is generated) between 0 and 10, but the number had nothing to do with what was going on in the game.  Alternately, internal random would be the way two formations met during a charge, resulting in different units being caught in very different situations.  It's still randomness, because a player can't exactly control formations with total finesse, but it was something that happened within the game that caused this uneven distribution of units in battle.  The difference is it is within the ability of players to control that randomness to a certain extent, but not totally (so it's still random).  This is, in my opinion, the only fair random.  In almost any situation, I believe that it will produce uneven distributions well enough to render the approach you recommend pointless.

Again, I'll reiterate: if it's noticeable on an everday basis, then it will be very unfair.  It's true that life isn't unfair, so why don't we give each player a random resource boost?  That's not fair, right?  It's the same thing from a gameplay perspective.  If it's not noticeable on an everday basis, then players won't even notice that it's there.  I think experience values and upgrades model differences in unit strengths more than well enough.

Stupid thing must have logged out when I was typing <_<

Stupid thing must have logged out when I was typing <_<

I don't understand.. :S Are you making an apology?

the post above that is his, but was posted as a guest. Darvin means that he somehow got logged out when posting

so why don't we give each player a random resource boost?

Be it a one time event for each game and with a great scale of possible resource boost, yes, it would be unfair. But given that we are talking about 100+ units, with a limited variation, randomisation would hardly ever result in big differences when looking at the complete span of a game.

Im bothered with the fact that I still haven't found the stuff I wrote, because I don't recall everything of it

One of the points, and I have posted about it before, is that units should be trainable for a scalable time, improving their skill the longer they train. (This would of course have diminishing returns and provide less of a benefit than actual combat experience) When the basic recruits are less adapt, they can be trained for a longer period, turning the randomisation of stats into internal randomisation. Not completely controlable, but not completely out of control either.

There was a lot more to the writing than just this, but I can't remember a few important pieces of the puzzle, so what more I do remember makes little sense for now. It's really frustrating to know you had the answer to the problem, had actually written it down and eventually forgot both about the answer and the location of the writing :S

Solinx

Be it a one time event for each game and with a great scale of possible resource boost, yes, it would be unfair. But given that we are talking about 100+ units, with a limited variation, randomisation would hardly ever result in big differences when looking at the complete span of a game.

Ah, but here is the crucial arguement; if due to a statistical anomalie early in the game one player never survives to the mid-game (or, more likely, the other player gets an early advantage which is carried throughout the game) then statistical odds won't "even it out" later on.  Again, I repeat my fundamental arguement: if it's strong enough to be noticeable, then it can potentially tilt the game to one player's advantage, which just isn't fair to begin with.  If it's not noticeable, you're mucking around with stuff that isn't adding anything to what the player can perceive.

One of the points, and I have posted about it before, is that units should be trainable for a scalable time, improving their skill the longer they train. (This would of course have diminishing returns and provide less of a benefit than actual combat experience) When the basic recruits are less adapt, they can be trained for a longer period, turning the randomisation of stats into internal randomisation. Not completely controlable, but not completely out of control either.

The problem I have with that isn't necessarily a realism or fairness one, but a gameplay one.  This strengthens the turtle strategy and encourages people not to go out, take territory, and fight over resources.  At very least it makes this much more difficult, risky, and less rewarding.  Ultimately, that's the more interesting gameplay than sitting around waiting for your units to reach maximum veterency.  In other words, this has profound gameplay implications.

Ah, but here is the crucial arguement; if due to a statistical anomalie early in the game one player never survives to the mid-game (or, more likely, the other player gets an early advantage which is carried throughout the game) then statistical odds won't "even it out" later on.  Again, I repeat my fundamental arguement: if it's strong enough to be noticeable, then it can potentially tilt the game to one player's advantage, which just isn't fair to begin with.  If it's not noticeable, you're mucking around with stuff that isn't adding anything to what the player can perceive.

I figured you'd notice that flaw. I have been thinking about this for some time, but for now, the best I can come up with is to take away the randomness for the first x minutes and then slowly increase the randomness range.

The problem I have with that isn't necessarily a realism or fairness one, but a gameplay one.  This strengthens the turtle strategy and encourages people not to go out, take territory, and fight over resources.  At very least it makes this much more difficult, risky, and less rewarding.  Ultimately, that's the more interesting gameplay than sitting around waiting for your units to reach maximum veterency.  In other words, this has profound gameplay implications.

Aye, it's not the complete picture I had. I know your argument, and yes, there needs to be something to balance against the turtle strategy. I know I had something for that, I just don't recall :S The notes should be in the notebook I have for Openwar and mods, but that notebook is unfindable. 

oh, and just to be clear. I do not think any of this should be put in the main game. Things like this are too abstract for casual players. However, I do hope the engine will be able to handle such modifications, making it possible to create a game that does work with more realistic and complex mechanisms.

Solinx

I have to say I agree with Darvin:

if it's strong enough to be noticeable, then it can potentially tilt the game to one player's advantage, which just isn't fair to begin with.  If it's not noticeable, you're mucking around with stuff that isn't adding anything to what the player can perceive.

and I also agee with Solinx:

However, I do hope the engine will be able to handle such modifications, making it possible to create a game that does work with more realistic and complex mechanisms.

as long as it is not too much work, the engine should support all these things

I figured you'd notice that flaw. I have been thinking about this for some time, but for now, the best I can come up with is to take away the randomness for the first x minutes and then slowly increase the randomness range.

That still doesn't eliminate the problem; a run of luck can still tilt what would have otherwise been an indecisive battle in one player's favour.  There are many times when games come down to a hair, and in these cases a small amount of luck can have a profound impact.  I still hold that there isn't a middle ground where it's actually noticeable to the players, but still won't tilt the results of the game on in all but freak incidents.


I do agree with allowing the engine to support these things, however.  I don't think having statistical variation would be hard to implement, I just don't think it's very good for gameplay.  However, people who are designing mods or entirely new games can make that decision for themselves.

By the way: I encourage anyone who is interested in test how does some of my ideas work in reality to try Original war, even if it's the demo. It's worth it.

While reading the character creation an idea popped into my mind:
 Why not to add on creating the general or commander in chief who commands all the soldiers in the RTS? or how about making your commander? Choose a face or make your own by parts, nose, eyeballs, mouth, all in 3d.. And choose some characteristics used in YOUR game. This could improve also to create random faces for your soldiers or employees!!

About this incredible way of animation, the webpage for the interested people is http://www.naturalmotion.com/euphoria.htm
I recommend this to bring the best results and reduce the tedius animation process, also a game played "will never be the same" with this.

I'd recommend it as well, if it were free, or cheap...

Endorphin 2.7
$9,495 / £5,495 / €8,995 / ¥1,195,000

Maintenance (12 Months)
$2,395 / £1,395 / €2,295 / ¥295,000

Check the prices

Solinx

OK.. Forget it..

Lol, that's a price we can't afford

---------------------------------------------------------------------------------------------------------
To conclude: What ideas you like or accept to use along with the engine? Post them here, if you don't have much time you can quote them.-

I'd recommend it as well, if it were free, or cheap...

Endorphin 2.7
$9,495 / £5,495 / €8,995 / ¥1,195,000

Maintenance (12 Months)
$2,395 / £1,395 / €2,295 / ¥295,000

Check the prices

Solinx

I know a way you can get it for free.. Send me a PM and I'll tell you wich one.

By the way: It will be interesting if units could also have an inventory to have items (like guns, etc.). This could be global for any strategy game (Commandos,etc.).

---------------------------------------------------------------------------------------------------------
To conclude: What ideas you like or accept to use along with the engine? Post them here, if you don't have much time you can quote them.-

Ouch Man! This hurts! Nobody seems to want to give an idea about my ideas.. Didn't anyone find these propositions useful?
  I personally love the thing to use A field of view like in commandos, IE...

Well, things looks boring, also "nobody is at home" so I'll start with what do I think should be good for any RTS to improve it:
-That Field of view, of course (for any human unit), and Alert states, that of course, they influde considerably in the way things are.(If a unit is in alert state, its field of view will be more stretch, the unit will turn more violently, his voice when speaking will hear more excited, he or she will turn and focus up to where he/she hears any sound in his/her short range of hearing; for example if he/she listens to the sound of an engine that does not belong to a vehicle of its own army).-
-The existence of a variety of skills, trainable by doing work (when working in a workshop if your unit is a soldier that has picked up tools as a mechanic, or when this mechanic is driving a tank, he is improving his experience if in combat, if he also drives the main weapon; or if a soldier is ordered to enter and drive a jeep, as he uses the gun of the vehicle when in combat, then his mechanic skills will be increased.) and the experience level will influence the level of efficiency of the work that has a relation with this skill (vehicles can be developed more faster, when attacking an enemy the unit will inflict more damage, the vehicle will move more fastly).-
As I said before, I consider myself these two things the ones that could be used to improve ANY real time strategy game made with the future engine.-

Right now a lot of us have school-related commitments.  I myself have a killer final exam tomorrow, the last of them for this semester.  With that said, I do drop by every other day to check for new posts.  It's not that we don't care, it's that we just don't have time to work on these projects.

After I'm done my final, I intend to put a lot of effort in the pre-projects and a few other commitments.  While I'd love to dive right into a RTS engine, I do not feel we have the resources to make anything that would do justice to the ideas discussed on these boards.

As for me, I'm still learning software design, and I'd like to learn some more of it before starting anything around here

www.neoaxisgroup.com

Here is an interesting link of some other guys who are using the .Net framework for their engine project. Looks they've done in a demo an RTS..!
I'll take a look at those demos..
The features this engine shows are great.. I think you could take this as an example..
Good luck!

It could be useful if you can implement the AGEIA SDK to the engine. I'm sure physics will be an important part to take care of also.