keithLamothe

Join Date 11/2006

+21

Proposal for dealing with engine speed issues

(my apologies if I'm starting a new thread where I should have added to an existing thread; this seemed like a sufficiently distinct track of discussion)

Just finished my first beta DA game. Very stable, quite fun. Some bugs and the obvious incomplete stuff, but quite good for a beta. Thank you devs! Anyway, onto the main reason for my post:

I think I have an idea for dealing with the "engine cheese" problem. It's been mentioned before in a few forms but I thought I'd go a bit further.

Disclaimer: this may be a completely hare-brained idea (or programmatically non-feasible, or already beaten to death, etc), if so please let me know so I can move on to more productive things.

The problem, as I understand it:
- under normal (pre-engine-nerf) rules, ships could be built with 80+ speed (40+ for well-loaded warships)
- even after the engine-nerf, rather fast ships can be made (20+ for a huge hull transport at high tech levels, probably 30+ or even 40 with max miniaturization but I haven't checked that)
- due to the nature of GC2 movement and standard computer performance limitations, the AI can't intelligently deal with enemy ships moving much over 8. As a result, players who utilize the higher speeds can beat the AI like a red-headed step-process (less so with the nerfing).
- further reduction in engine speed, however it was accomplished, would make it take a LONG, LONG time to move across a gigantic galaxy. This may or may not be a true "problem", but it is worrisome to some players.

Proposal:
- add a "warpgate" starbase ability (conferred by a new line of modules) that allows an allied ship on the same square as the starbase with at least 1 movement point remaining to instantly "warp" to another allied warpgate-equipped starbase within X parsecs (where X depends on the StarbaseAbilityValue of the module or modules). This action would consume all remaining movement points for that ship.
- nerf engines like crazy so that speeds don't get above 8-10 (or whatever works out). This could be done by further engine space nerfing or by limiting ships to at most 2 engine components (which would probably require more programming).

Disadvantages/Costs:
- presumably requires changes to code, so that the parsing routine for StarbaseModules.xml could make sense of:

[StarbaseAbility]Warpgate[/StarbaseAbility]
[StarbaseAbilityValue]15[/StarbaseAbilityValue]

(replace ['s and ]'s with <'s and >'s, I couldn't get them to display due to them looking like tags)

Not to mention the probable need to add another command to the ship command window for "warp" that is only active in the above circumstances; and of course the code to implement the "warp" command (I imagine setting the x,y location of a ship isn't too bad, but I'm armchair programming here and the code to determine allowed warps would be a bit tougher).

- requires addition of a number of entries in StarbaseModules.xml and TechTree.xml, but that probably won't be anything compared to the coding.

- would either add to micromanagement of long distance travel (including new-production rally points from across the empire) or would require additional code in the auto-pilot plotting routines to take warpgates into account.

- will probably tick off people who like really fast ships and whipping the AI with inter-stellar-ballistic-missiles in the form of building a transport and "firing" a chunk of metal containing 500 billion angry marines at an enemy world in the same turn.

- could allow low-speed cheese in the form of minimal-to-no-engine ships stuffed with weapons and armor that could make it across the galaxy by gates when it would have been untenable to cross the distance on normal drives

- may not scale well to galaxy size

- would either make freighter trade routes look a bit silly ("why are they not using the gates?") or would complicate their code (I am NOT suggesting routing them through the gates).

- inevitably will add more complexity to the game. Whether that's good or bad I can't tell right now.

Advantages:
- AI wouldn't have to deal with possible-move trees much past the normal depth for an 8 move king (which is still not fun, but better than a 30 move king), since engine speeds wouldn't exceed that and any warped ship has to stay on the starbase square until the next turn (allowing processing of it's possible moves from that point on normal speeds).

- Players could still move ships across a gigantic galaxy in a few turns IF they had sufficient control of the space to keep the constructors/starbases alive and had the industry, money, and will to build the infrastructure. Players could even construct a gate-base in the middle of enemy territory and start warping in large fleets, but it would be risky to do so without having already established a degree of space superiority (and, of course, the square would have to be reached by normal engines first).

Notes:
- I would suggest that the starbase modules come from a new line of techs off the root (or near the root) of the propulsion tree. Technobabble-wise it could be explained as a revisit of stargate technology using the new insights from hyperdrive-style propulsion research.
- The numbers could obviously be tested and tuned, but right now I'm thinking of a progression like:

Warpgate I module : 15 parsecs
Warpgate II module : +10 parsecs (25 total)
Warpgate III module : +10 parsecs (35 total)
Warpgate IV module : +15 parsecs (50 total)
Warpgate V module : +20 parsecs (70 total)

So early on they could help a fair bit (while normal speed is still kicking around in the 3-6 range), but later on they could be used to cross a Gigantic galaxy in time comparable to that of the pre-nerf engines.

Alternatives:
- have the gates be planet based instead of starbase-based (improvements instead of modules). This would make the gate locations considerably more fixed, but I tend to prefer the starbase idea.
- make warpgate travel non-instant (instead moves at much faster rate determined by module level), where the ship disappears and shows up at the destination however many turns later. This seems like it would be a lot more complicated code-wise.

I could ramble on quite a bit longer, but I'll stop there and see if there's any point in my saying more.

Thanks for your time,
Keith LaMothe

Locked Post

Advanced Search

Watch this post

Do not email me updates for this post Email me updates for this post

Successfully updated karma reason!

Something along these lines has already been discussed, most recently in the "Yor Super Ability has to change" thread IIRC.

Successfully updated karma reason!

Seems like it would be a lot simpler to just say,
"More than 2 engines destructively interfere with each other. Can't go there."

Successfully updated karma reason!

My suggestion would be to first of all, prohibit planetary attack by forces which were not within a few squares of the planet at the beginning of the turn. I feel that if you're going to gain air superiority for a ground assault, it should take at least one turn. If you have a fully-escorted fleet of transports and you're not simply cherry-picking the planet, then you have nothing to fear from in-system fleets attempting to defend the planet.

Thusly, the AI could always make a proper defence by stationing a suitable defensive fleet or two within range of the planet. If it can't do this - perhaps because you have drawn its attention elsewhere, or because it simply doesn't have a military budget big enough to cover everything - then it was never going to hold onto the planet anyway.

Successfully updated karma reason!

A 2 engine limit might be harder to code than you might think. The AI builds ships with more than 2 engines all the time, and speed 16 AI ships is not uncommon. The problem with fast ships is the AI can't determine potential threats from ships that move faster than 8 or so. So if you hard code a two engine limit, the AI will have to be worked on to deal with those limits (since it exceeds 2 engines itself). And that still won't help the "8 move limitations" as just 2 advanced engines will pass that mark easily. Although, of all the ideas floating around, I like the two engine hard cap the most.

The biggest gripe seems to be blitz invasions, and not so much fast warships, as the AI "does not have a chance to respond to potential threats". Want an easy remedy? Have planetary invasions require full movement points (yes, you would have to park your transports next to the enemy planet for invasion the next turn). It would not matter how fast they got there if they had to wait a turn. This would give the AI a chance to respond to threats, while being able to determine more easily non-threats. I have had a number of times I was only moving ships through one AI's territory to another and triggered the threat warning and had them declare war. This would not happen if that trigger was only tripped with transports being right next to a planet. It's also one of the concerns I have with slower trasports, as you can't traverse their territory in a turn and reduces the game to a "you have to attack your neighbor".

I am not playing the beta, so I can't comment on the size changes made already. My thoughts are from a 1.4x perspective.

Successfully updated karma reason!

Well, not to spoil it for you too much but....

Ion engines in DA are (in beta 2A) an early cost-effective propulsion system, even if they're not as space-efficient as impulse engines which come after them. All engines are now considerably larger and more costly than they were in DL. With sufficient miniaturisation you can still stuff a fair number of engines onto a ship but it will cost more to make than it used to, and of course if it isn't exempted from maintenance, that will be higher too.

The upshot is that it is not necessarily pointless now to mix engine types to achieve a mixture of cost and space savings.

Successfully updated karma reason!

Mystikmind

Join Date 06/2004

+20

My suggestion would be to first of all, prohibit planetary attack by forces which were not within a few squares of the planet at the beginning of the turn

Even better... prohibit 'ANY' combat for a ship traveling beyond a certain distance.

Successfully updated karma reason!

Alstein

Join Date 07/2004

+55

Also, those mega engine ships will have a tough time taking out defenders.

Successfully updated karma reason!

not the defenders im familiar with.

Successfully updated karma reason!

Even better... prohibit 'ANY' combat for a ship traveling beyond a certain distance.

Or limit the number of attacks to the number of engines. That wouldn't be as drastic, and it wouldn't hurt the AI at all since at most it'll attack 2x in a turn.

Successfully updated karma reason!

I don't know, i think that in the current version of the game with the new engine sizes the problem is solved for warships. Engines are just too big for it to be worth putting more than one or two on war ships. In the current build you just loose too much firepower to compete with the slower defense ships.

I believe though that there still remains an issue for transport ships for which the sacrifice in terms of space is not so important. Making 20+ speed transports is still viable. A solution might be to increase the size of the engines for cargos more than on war vessels.

Even better... prohibit 'ANY' combat for a ship traveling beyond a certain distance.

I don't believe it is necessarily a good idea you still need a certain number of relativly fast ships to intercept troop transports that are generally in the current DA build much faster than warships.

Successfully updated karma reason!

Something along these lines has already been discussed, most recently in the "Yor Super Ability has to change" thread IIRC

Yes, I noticed that at the tail end of that thread, though it seemed more in terms of references to MOO stargates and other such implementations rather than a discussion of specifically how the concept would fit into DA. In any event, it doesn't surprise me that this sort of option has been brought up before and (apparently) found inadequate.

Seems like it would be a lot simpler to just say,
"More than 2 engines destructively interfere with each other. Can't go there."

Yes, it would be simpler, and I would prefer that exact rule for the normal-speed-nerfing part of what I was suggesting.

However, would that deal with the problem of only being able to go speed 10-12 (and that's near the end of the tech tree) or so in a gigantic galaxy? Or is that not really that much of a problem?

Have planetary invasions require full movement points (yes, you would have to park your transports next to the enemy planet for invasion the next turn).

That sounds like a good idea, actually. That way the transports at least have to be escorted if the enemy has any sort of defense fleet in the area.

Successfully updated karma reason!

The biggest gripe seems to be blitz invasions, and not so much fast warships, as the AI "does not have a chance to respond to potential threats". Want an easy remedy? Have planetary invasions require full movement points (yes, you would have to park your transports next to the enemy planet for invasion the next turn). It would not matter how fast they got there if they had to wait a turn. This would give the AI a chance to respond to threats, while being able to determine more easily non-threats.

This is the most brilliant idea to combat blitz invasion cheese yet --my hat's off to you, sir.

Brad, are you getting this? I think this idea would be way more effective than an engine nerf while not chastising people who like to build fast constructors and colony ships.

Successfully updated karma reason!

Hmm has somebody thought about slot system for ships?

Bit similiar that Mech Warrior games had. IE. Medium ship hull would have 2 engine slots while large could have 3 or so. It worked great for mech warrior, dunno if it would work with GC2.

Successfully updated karma reason!

Have planetary invasions require full movement points (yes, you would have to park your transports next to the enemy planet for invasion the next turn). It would not matter how fast they got there if they had to wait a turn. This would give the AI a chance to respond to threats, while being able to determine more easily non-threats.

This is the most brilliant idea to combat blitz invasion cheese yet --my hat's off to you, sir.

Not sure If I agree with that. Sure, I won't be able to launch a dozen troop transports and strike a dozen enemy worlds in the same turn. However, I can clear out the orbits of said planets, launch my troop transports, navigate the same distance, and in the next turn still take out the dozen worlds at the same time.

So basically all you have done is give the AI a single turn to repel a dozen invasion transports. I grant you, giving the AI that single turn is more than it has now, but I dare to say, it will not do the AI much good.

My opinion of course.

Successfully updated karma reason!

I grant you, giving the AI that single turn is more than it has now, but I dare to say, it will not do the AI much good.

I agree with this as well. Sure the one turn gift for the AI to respond is more than it has now, but what good would it do. Quixen states the logical scenario as to what would happen in such an event. In a way it even makes me question my original thoughts on having a ship power supply. I guess it will be what it will be in the finished product.

Successfully updated karma reason!

If transports have to park one turn and invade the next, it still allows the AI the chance to use any fleets it has stationed nearby to attack those transports. Without park-and-invade, there is no strategy it can formulate.

As I said way up above, if you properly escort your invading transport all the way in protecting it from interception, then the AI will have to concede that world. Always though it should be looking to avoid such a checkmate if possible.

Successfully updated karma reason!

I don't know, i think bigger and more expensive engines are the way to combat a blitz. This will not only make the ships slower, but take longer for these transports to build. When you're talking about a blitz, you're talking about what might be a 10-30 turn advantage on the AI to rush. If you don't hit that window, you're essentially screwed. It's also gonna make you think a lot more if you have to throw upwards of 1000 bc just for an invasion try... its just not efficient really. I also like this idea because at the end, when i feel it is a mop up exercise, I will be able to produce the faster ships. I think the biggest problem with the engines before is that they were as cheap as free

Successfully updated karma reason!

However, would that deal with the problem of only being able to go speed 10-12 (and that's near the end of the tech tree) or so in a gigantic galaxy? Or is that not really that much of a problem?

Galaxies could be resized, say medium to 5x5 large to 6x6 huge to 7x7 and gigantic to 8x8, so that speeds are lower but distances are also made shorter... gigantic would be 1/4 current size. Of course, that affects # of stars and such, but it would make those sizes playable I guess.

Successfully updated karma reason!

Galaxies could be resized, say medium to 5x5 large to 6x6 huge to 7x7 and gigantic to 8x8,

This would almost certainly cause a mutiny among avid GalCiv2 players. What are you thinking? Reduce the map size? Blasphemy! If anything make the map size bigger to compensate for the faster ships! (Just kidding here guys and gals)

If transports have to park one turn and invade the next, it still allows the AI the chance to use any fleets it has stationed nearby to attack those transports.

Let me put it this way. If I have to park my transports next to the planet for a turn before they can invade, don't you think the average player would completely obliterate any opposition in the area that could be perceived as a threat before moving said transports? Once again Giving the AI a snowballs chance in... well you know, at some sort of defense, but still just a fraction more than what they already have.

I think the Devs will do their best to fix this problem and keep the game enjoyable.

Successfully updated karma reason!

So basically all you have done is give the AI a single turn to repel a dozen invasion transports. I grant you, giving the AI that single turn is more than it has now, but I dare to say, it will not do the AI much good.

I disagree. If you can launch those dozen troop transports and defend them for one turn then I think you probably had enough military strength to fight a prolonged conflict anyway. That's not cheese at all. Currently, the reason why blitzing is cheese is because it allows the player to obliterate an enemy with a much better military in a single turn. Having transports need to park one turn would make that impossible. It also makes sense; an invasion is going to take time to prepare for.

Successfully updated karma reason!

don't you think the average player would completely obliterate any opposition in the area that could be perceived as a threat before moving said transports?

If they can do that, then don't you think they deserve the planet? Also, if there is a problem in that area it can be solved by having the AI put more engines on their ships.

Successfully updated karma reason!

I don't know, i think bigger and more expensive engines are the way to combat a blitz. This will not only make the ships slower, but take longer for these transports to build. When you're talking about a blitz, you're talking about what might be a 10-30 turn advantage on the AI to rush. If you don't hit that window, you're essentially screwed. It's also gonna make you think a lot more if you have to throw upwards of 1000 bc just for an invasion try... its just not efficient really. I also like this idea because at the end, when i feel it is a mop up exercise, I will be able to produce the faster ships. I think the biggest problem with the engines before is that they were as cheap as free .

Absolutely right. The remaining problem is that AI still fails to build a decent economy, and the player will be able to build those more expensive/faster ships. Hopefully Brads re-write for the planetary management will help as well.

Successfully updated karma reason!

Eh. Anything to do with nerfing engines, making them bigger, more expensive, or limiting the number of engines allowed has a fundamental problem: These strategies are more effective at combating blitzing with bigger maps than smaller maps. I think that having transports wait one turn is a *far* better solution that works across all map sizes equally.

Successfully updated karma reason!

I guess in that regard, you may be right Jeff. I have yet to play above tough and have found up till this point I am walking over the AI. Maybe at the higher levels you cannot do as I suggested.

I was just speaking from my own personal experience with the game, but with the levels tough and below, the above suggestion will not work. An experienced player will realize this and continue the mass assaults.

Successfully updated karma reason!