Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
DFD-R2: 1099 vs. 1074, 1147 vs. 1068
#1
DFD: Daily Feature Deathmatch

The Cruel Fight For Implementation

This is a Daily Feature Deathmatch post. If you are unfamiliar with the background of this event, please read the announcement, the adjustment and the schedule.

Fight 1

[0001099] Re-enable Lightning rod logic vs. [0001074] Multiplayer Games about AI

Fight 2

[0001147] Streak vs. [0001068] SubjectToUnits, -Infantry, -Buildings

After the fight is over, two of these issues will be suspended, while the other two move on to the next round.
Remember that the coders will not take part in the discussion, so make your arguments complete, concise and convincing - when it's over, it's over.

Part of that is clearly marking what outcome you support for which issue.
There should be no ambiguity in the issue you're talking about, and it should be clear what outcome you support. Feel free to put your stance in bold, and use simple terminology like "kill #69" or "I want #42 to survive".
This use of simple terminology should be part of a larger argumentation - if this is all your post consists of, it will be ignored. We are interested in argumentations and details to consider, not votes.

A decision will be made either way, a lack of discussion will not cause all issues to live.

Be friendly, be civil, be logical.
You are allowed to try to deconstruct the arguments of those arguing against your candidate, but remember that they don't make the call - there is really no point in getting personal.

The discussion should be contained in this thread, argumentations elsewhere will be ignored, but you are allowed to transfer and adapt points made elsewhere in the past.

We want a good, clean fight.
Let's get it on! Dual M16

These fights are largely automatically generated - if an issue turns out to be unfit for combat, it will be disqualified and the opponent will go into the queue.
Forum Rules

(01.06.2011, 05:43:25)kenosis Wrote: Oh damn don't be disgraced again!

(25.06.2011, 20:42:59)Nighthawk Wrote: The proverbial bearded omni-bug may be dead, but the containment campaign is still being waged in the desert.
#2
[0001074] would possibly be easier to implement, what with the logic appearing in network games and stuff, that and not many people seem to want lightning rods back....

Neither of the other 2 issues garner support from me, streak bonus' are dumb, and we have promotion bonus' anyway, and i'm struggling to envisage how subject to [TYPE] is any different from verses=0%
[Image: MRMIdAS2k.jpg]
MRMIdAS: No longer allowed to criticise Westwood on PPM
#3
First pairing I'm indifferent tbh, if had to pick then go for 1099 as 1074 doesn't really appeal to me - maybe seeing it in action would sway me but for now 1099 gets it

2nd I go for 1068 as would both provide cover and add a bit of realism in that bullets can't go through tanks/buildings
#4
Fight 1:
Support #1074
Kill #1099

Being able to form and break alliances with AIs like in RA1 sounds overall more interesting than re-enabling the LightningRod, as the latter has a very limited usage scenario.

Fight 2:
(29.07.2010, 00:46:13)MRMIdAS Wrote: and i'm struggling to envisage how subject to [TYPE] is any different from verses=0%
uh... then you should re-read the description and comments (assuming you have read them...) Tongue

As this is my issue, my vote cleary goes to support #1068.
It's simply unrealistic that a straight-shooter like a prism tank can fire over a battle fortress hitting a terror drone behind it.
For other usage scenarios and advantages this feature would bring, simply read the description and additional info.

And I don't like temporary bonuses in games in general, so kill #1147.
#5
Support #1074. This would make AI games a lot more interesting than picking them as allies before game start. Plus it would be fun to screw around and break alliances mid-game for lulz.

Kill #1147. It's just stupid. #1068, on the other hand, is a very AWESOME idea. It would be a better way to make weapons damage in a straight line without using railguns.
I'm what Willis was talkin' about.
#6
While I'm not hugely fussed about fight one, [0001068] caught my eye. I like this idea. This feature would actually have a noticeable affect on the balancing of certain armours, with the possibility of sending in specific unit types specifically to form a protective shield for other, more vital units.
I support it, for this reason.
#7
I'm with Orac here, 1068 seems pretty cool, I mean we have builings stopping projectiles, why not tanks and other solid things? Now compare that to 1147, which is a port of a terrible feature from an overrated FPS. Streaks are bad when some guy hiding in a corner with a sniper rifle or running round with overpowered weapon X is getting them, so how is adding them to an RTS where any unit could get them and they don't have enough individuality for any player to tell which of those GIs is about to unleash an airstrike against them or whether that one died 10 minutes ago any better?
#8
For fight one:
LightningRod logic is pretty useful... in Tiberian Sun. However, in YR, how many uses would it have? Weather Storms don't work in the same way as TS's Ion Storms did, you don't have random bolts all over the map, just within a small radius fired from specific clouds. If something like TS's Ion Storm events made a return, or map-wide weather effects were implemented that could benefit from this tag, I could see a use for it. However currently, I don't see much use for it.

The second issue is, admittedly, quite an interesting idea. It could allow for Skirmish players to mix things up a bit more in-game. However, I get the feeling something like this could be hard to program. Additionally, there would need to be a limit programmed so the player doesn't just ally with every single AI on the map and there's no-one left to fight, but that's just common sense. Mechanics like this are usually quite open to abuse in RTS games (ally with all but one AI, kill it, break one alliance, go back to step 2), but in comparison to the previous issue, it seems much more beneficial.

Therefore, my stance is support #1074, kill #1099.


Edit: and I forgot to post anything on fight two.

For fight two:
Kill streaks are a debatable mechanic in FPS games at the best of times, and I really don't see such a thing working that well in RTSs either. I mean, the veterancy system is already there to provide for this sort of thing and it works much better than what is being requested. I think time would be better spent improving it than making something like this. I mean, this is basically just the super weapons on infantry request again, except only for veteran and elite infantry.

I like the second issue. We already have solid buildings, solid units make sense. However, I'm just wondering what the performance hit would be on the game keeping track of all these moving solid blocks. Also, as was already mentioned by someone, given that several infantry can occupy one cell, that would need to be accounted for. However, it could make for some interesting gameplay changes if this were implemented.

My stance here should be pretty obvious given my rant on the first issue - support #1068, kill #1147.
Ares Project Manager.
[Image: t3wbanner.png]
[Image: cncgsigsb_sml.png]
Open Ares positions: Documentation Maintainer, Active Testers.
PM if interested.
#9
With this AI feature of 1074 - If possible I believe that it'd be useful for the AI to check whether an alliance is a reasonable thing to accept. Even if the AI was just comparing the strengths of your and its armies (iirc, the AI already has that capacity?) and then accepting an alliance only if you are strong enough to be worth avoiding conflict, ie: you are a strong player, this might make this feature less open to abuse.
#10

Administrative Notice:

Given that there have been no new comments for over four days, it is assumed this fight is over; we will proceed to judgement.
Forum Rules

(01.06.2011, 05:43:25)kenosis Wrote: Oh damn don't be disgraced again!

(25.06.2011, 20:42:59)Nighthawk Wrote: The proverbial bearded omni-bug may be dead, but the containment campaign is still being waged in the desert.
#11
Fight 1

I don't have anything against the LightingRod request, especially not considering requests like #957, in fact, I think actual lightning rod buildings or building upgrades could be a neat little tactical addition to the game - just like in TS, when you played against Nod, you could choose to invest into pavement to keep out subterranean units, in this case, if you played against Allies, you could consider building lightning rods, to deflect damage during a lightning storm away from your important buildings.
It would be a passive defense structure, just like walls or pavement.

However, as much as I think #1099 is a valid request with a good usage case, the fact of the matter is that #1047 helps making the AI more "human-like", and games with a number of AIs less predictable.
Ideally, we could also ensure that AIs form distinct alliances among each other, dependent on their strength, position, etc., to give their alliance-behavior a very "human" feel.

I don't know how feasible #1047 is just yet, but I do think if we can get it to work, it'd be appreciated more than lightning rods.

Kill: #1099
Support: #1047

Fight 2

This is one of those fights where I'd like to kill both issues.

The streaks request is one of those "yeah...no" requests. It's a valid request with an understandable desire, but it simply doesn't fit into YR's gameplay and it's way too much effort for way too little gain with way too much unwanted silliness.

The SubjectToEverythingUnderTheSun request is one of those "omg let's add more complexity!!" requests. Sure, ideally, we'd have 100% accurate physics, a pixel-perfect collision system, dynamic, gameplay-relevant lighting and unit-individual lines of sight.
But we don't. YR is vastly simplified, and not all of these simplifications are made to be removed.
Yes, it's questionable that machine gun bullets happily fly through tanks and hit their target. But for the vast majority of projectiles, the system makes perfect sense, players usually happily suspend their disbelief and just enjoy the game, and, last but not least, it has been this way for over a decade.
Seriously.
It's not like this is a sudden development, and before, all was 100% realistic.

The point is: I understand the request. I see where it's coming from, I admit it'd be nice if stuff behaved "realistically". But ultimately, the current system has been in place since at least Tiberian Sun, and this is the first time it's seriously being questioned.

Which, by the way, generates a whole new side to this: People could get confused when they expect their C&C to behave like C&C has always behaved, send their Sniper to attack a bunch of conscripts, and then wonder why the fuck the units don't get damaged - not realizing the bullet is caught by the Rhino in the middle.

There is, by the way, a reason I only use bullets as examples: The request exaggerates the amount of unrealism. Independent from the fact that weird beams and battle-grade, tank-mounted lasers don't exist just yet anyway, even if they did, you don't know how powerful they'd be.
I mean, seriously - if a Prism is powerful enough to significantly damage a tank, don't you think it could happily melt through a Conscript and still hit the target?
Don't you think that a Magnetron's "beam" is maybe just an illustration for the player, and that an actual anti-gravitation beam might just be invisible and depend on other factors than line of sight?

When arguing about things unrealistic in C&C, you always have to remember that C&C isn't realistic. Arguing about realistic bullet behavior is one thing. Arguing about realistic behavior of fantasy weapons is kind of ridiculous.
Not saying I've never been guilty of that, just saying it's something to remember when doing that.

But there's more!
This request, if implemented, would be the start of a slippery slope.
Today, it's "We already have SubjectToBuildings, we should also have SubjectToInfantry|Units|Aircraft!"
Tomorrow, it's "It's illogical a human can stop a bullet, but a rock can't - we should have SubjectToRocks!"
The day after, it's "We already have SubjectToRocks, could we also get SubjectToTrees?"
Next, it's SubjectToSlopes, then it's SubjectToClouds, then it's SubjectToBeams, that gets extended to SubjectToProjectiles, and ultimately, we'll end up at stupid shit like SubjectToAir because someone proposes a vacuum overlay type.

Point is: Sometimes, a simple extension is enough.
The gain of this request would be insignificantly little, and the number of additional development effort and ingame computation required stands in no comparison to it. It's just not worth it.

And even if all the above weren't true: It still wouldn't make a goddamn difference.
Why?
AI.
Because you just know, if we code it in a way that a trooper happily keeps firing against the obstacle instead of repositioning himself, we'll get a bug report demanding a fix within a day.
So with the next patch, units simply wouldn't fire against the obstacles anymore. They'd go "There's my target! Damn, there's a conscript in the way! *steps to the side* CRUSH KILL DESTROY!!!". The truth is, once properly implemented, the pathfinding of all units would simply ensure the units wouldn't shoot against obstacles anyway. So if there's a passing Conscript running into the line of fire, they'd just move to the next cell and fire as before, and if a line of Conscripts were built to catch the fire, they'd just treat them like they've treated walls for a decade: They'd look for a better position to shoot from.
(Which, of course, would draw new bug reports about how pathfinding/AI is broken, since units walk away instead of attacking, but hey, you don't have to deal with those, right?)

And the funny thing is, even all that, even the more realistic scenario of how SubjectToVictims would play out if it were implemented perfectly as imagine by its proponents, is still irrelevant.
Why?
Because the simple fact is, ingame, the damn difference would be exactly one shot.
You would shoot at the target.
You would hit the Conscript standing in the way.
Scatter= would kick in, the Conscript would move, because he doesn't like being hit.
The line of fire, from your current position, would be perfectly free and you could go on hitting your target without any change of your own behavior.

You built a wall of Conscripts to take the damage? Good for you! I hit each of them once and your damn "wall" looks like a swiss cheese running around in panic.

If the modder activated PlayerScatter=, this will even work on your own units.
Which, by the way, poses an entirely new issue: Has any of the proponents ever thought about how exactly groups of units are supposed to work with this?
YR doesn't have formations.
You would constantly have to manually keep your units in one straight line in order to keep everyone effective. Otherwise, the last line of your units will happily massacre the entire rest of your army, or only the first line will fire while the rest of your army spends the battle walking around looking for a better spot to fire from.

As said, I understand the thinking and the spirit behind this request, but as should be visible from the above, it is riddled with inevitable problems that simply make this feature completely unfeasible - even if it were implemented perfectly and as requested, it would damage gameplay, significantly obstruct the flow of the game, and dramatically reduce the effectiveness and intelligence of all units in the game.

As such, I have no other choice but going with the streaks request.

Kill: #1068
Support: #1147
Forum Rules

(01.06.2011, 05:43:25)kenosis Wrote: Oh damn don't be disgraced again!

(25.06.2011, 20:42:59)Nighthawk Wrote: The proverbial bearded omni-bug may be dead, but the containment campaign is still being waged in the desert.
#12
Fight 1
Forming alliances with the AI on the fly would be awesome. Just like in RA1 when you hear AIs battling each other, suddenly silence, and then they'll going after you instead.

Lightning rods are a good thing to counter or attract thunderstorms. Tesla coils could withstand lighning and act as lightning rods. Power plants could also act as lightning rods, but they'll be destroyed instead. Even though, it's still a border case without ion storms.

Fight 2
Uhm... what Ren said.

Seriously. Adding too much realism is unfun. Want everything to behave like it would in the real world? Probability suggests most of you didn't like Physics in school. And now you want it, nonetheless?

If you got a dozen of Conscripts shooting at one single GI, by the time the GI is dead, your Conscripts would have decimated themselves. Same would hold true for tanks. Yeah, that's realism. YR is not designed to support that.

Realism.

How the hell can a magnetron actually lift a tank? If real world physics apply, with magnetic forces or some kind of radiation, it could only move it collinear to the virtual line between them, but never orthogonal to that. (Except for when the magnetron actually shoots a stream of positrons through the tank, shooting west to east, therefore using the earth's magnetic field to induce Lorentz force to lift the tank, of course. But we all know that's impossible as a positron would not fly through the tank [roughly for the same reason you can't x-ray a tank], it would most likely hit one of its nuclei, thereby converting a neutron into a proton causing some decay, or anihilate, if it hits an electron, creating two mesons. But that's not important here, as it isn't decisive for the outcome of the war.)

Do we need SubjectToTinFoil for Mind Control? No known mind rays can pass though tin foil. Ask your local goverment.

I also don't like the streak thing, but hey.
#13
Result:

As above.
Forum Rules

(01.06.2011, 05:43:25)kenosis Wrote: Oh damn don't be disgraced again!

(25.06.2011, 20:42:59)Nighthawk Wrote: The proverbial bearded omni-bug may be dead, but the containment campaign is still being waged in the desert.




Users browsing this thread: 1 Guest(s)