mrvn wrote:I think they should not be able to fire while on the move. There could also be a delay before firing. Like the train stops, then it has to put down some supports and unfold itself before the turret can rotate and fire. Reverse that before driving away.
From the start of that paragraph to the end, I went from mild kneejerk aggravation to "I like!"
Deadly-Bagel wrote:so you now have to deal with the massive wave of biters
That's funny, because I've taken out quite a few spawners with bob's mods sniper rifles, and I don't seem to have any recollection of the "massive wave of biters" you speak of. Maybe I'm just getting old...
(Since bobingabout's in the thread, maybe he can speak to this with a bit fresher memory than me, though I don't know if his enemies have different AI than vanilla ones.)
Deadly-Bagel wrote:Oh well that would be with turrets then, no? Now you need to creep your train forward except now you need to build more turrets, provide them with ammo/power, then build more rails, enter the train and drive it forward. I thought you said this would be less tedious than turret creeping? You've added more steps.
I talked about three different scenarios. I even numbered them 1, 2 and 3 for you. You're conflating different scenarios. I said in #2 that using an artillery train + blueprints in map view with supporting lasers could be a less tedious way to expand one's base (rather than using traditional turret creeping). That's a different scenario than #3 which was about building a railway through enemy territory. And no, for #3 there are not really more steps because you already have to do all of the steps in question (laying rail and placing defenses for your rail - at least that's what I always do when placing rail outside my base walls, all of my rails have power, circuit wires, and lights, so power for lasers is already there, and I use blueprints for militarized rail that include everything so I'm not hand-placing anything) - the main change (between that and laying militarized rail when not so close to the enemies) is that you would have to proceed slow enough to allow the artillery train to do its job, and that small price should be much less than the price associated with traditional turret creeping (but who knows - we'll see).
With mrvn's idea of how artillery trains should work, I'm having additional ideas on this - *if* we ever get modular wagons in vanilla, it would be very nice to have such a wagon on an artillery train or in a train right behind the artillery train, far enough back so the construction zone barely reaches in front of the artillery train. Then you could drop a massive rail blueprint and have it auto-built in front of the artillery train without the bots getting too far out ahead and getting themselves killed. Any time the artillery train stops to fire the bots would then automatically stop building and wait for the attack to finish (because their construction zone has stopped moving forward).
Note that a big problem with traditional turret creeping is that it requires you get too close to spawners before you can get in range to take them out. The vast majority of the biters you end up fighting are not biters that happened to be there before you arrived, but biters that were spawned after you arrived. If artillery trains have the range to take spawners out before biters agro, none of those additional biters will ever be created so neither you nor your defenses ever have to fight them.
Deadly-Bagel wrote:Maybe you have to use them less often but I still fail to see how it is not a step backwards
A step backwards from what? (Since you've conflated multiple scenarios it's impossible to know what your basis is here, much less guess in what way it would be a step backwards.)
Deadly-Bagel wrote:why it would be superior to a static placement or non-rail-bound vehicle.
This is a game, not the real world. "Superior" for a game does not equate to "more powerful in the game world", it just means fun and interesting to play with. If you just want an "I win!" button, feel free to mod-in such a thing. I predict that getting good working defenses with an artillery train (assuming it's expensive enough so it can't just be spammed and used statically) would be more interesting than doing the same with static artillery. Of course, whether it really turns out that way depends on the details, which we do not know yet.
BTW, artillery trains should be
far more useful/convenient than both static placements and non-rail vehicles when it comes to keeping spawners away from your rail lines.
Deadly-Bagel wrote:Fun fact! We already have a vehicle with a mounted cannon. It's not that useful.
"Fun fact!" That vehicle does not provide ANY automated defense. You're just wasting time making arguments that ignore the automation aspect. I want my defenses, if I build them properly, to work without me having to intervene.
Deadly-Bagel wrote:Big picture, you say you can do all this from the map. Trains are bound by train stops, they go from point A to point B via the shortest available route, you can't tell them to patrol the perimeter. So expansion is actually a very tedious process of building the rails
Do you know what a blueprint is? Yes? You know you don't have to dirty your hands with the "processing of building the rails" when you use them, right? So how is it tedious? Are you using blueprints that are too small so you have to repeat them 100 times along each side of your base or something?
Deadly-Bagel wrote:and moving all the train stops
Um, you clearly don't know what you're talking about - you would never move the train stops as part of the process (at least not as Factorio works now because "moving" train stops by constructing them with blueprints causes them to lose their names). Luckily, there's no need to move the train stops.
Deadly-Bagel wrote:ensuring they're in the right places (easiest would be with a square factory, a unique stop in each corner, good luck if it's not a square)
Since you don't have to move the train stops, you don't have to do anything to ensure they are in the right places when expanding - they already are.
(And you act like people aren't using blueprints for giant rail systems already. Many of us can handle train paths that are not squares just fine. If someone is using the scheme I talked about, it is likely that before they get too far they are going to make themselves a modular set of blueprints that makes it easy to construct whatever set of angles and corners they want to use. E.g., if they just want to use vertical stretches, horizontal stretches and 90 degree corners then they will make themselves a set of blueprints with such features in it, along with long versions of straight sections to save time when placing blueprints. People are of course already doing this now for their defenses.)
Deadly-Bagel wrote:and this is ON TOP OF walking up the requisite wall defences. Not exactly a shining example of automation.
As Factorio stands now, the process would involve issuing construction and deconstruction orders in map view, and then your robots do the rest - not just in some small area where you're killing spawners, but around your whole base - that's a nontrivial amount of automation. If you wanted to fully automate base expansion, performing multiple expansion steps with one command (and who wouldn't want that?), you'd need some functionality like what the
Recursive Blueprints mod provides (which a lot of people would really like to see in vanilla, including at least one of the Factorio devs).
Deadly-Bagel wrote:As I said there is something of a need for this but there are far better implementations.
Better how? Like I said, this a game, so fun and interesting is more important than highly effective. And being different and being more complex makes things interesting. We already have static weapons, and we already have direct player controlled mobile weapons. An automatic train-mounted weapon (that is not just a "train install" of some other weapon) would be something quite different than what we have now, and different is good.
Deadly-Bagel wrote:you're failing to take into account many factors. It's still a train, if it hits a behemoth it's going to stop dead and be killed, yeah they could make it absurdly powerful but they could have also done that to the tank and existing trains and yet they didn't.
You're failing to take into account the fact that you can already kill behemoths with appropriately constructed trains - you just need a number of locomotives on it to do so. I would guess it's more likely than not that they make artillery trains tougher than normal trains, but even if they don't the multiple locomotive solution exists.
Deadly-Bagel wrote:It's therefore going to have to go behind your defences.
Hell no! Out front for sure!
Deadly-Bagel wrote:If it's going to be there anyway why not just use static artillery?
Wait... what? That was a total non sequitur. What does "behind the wall" vs "in front of the wall" have to do with "moving" vs "static"? (Hint: Nothing.) You could make a case for how close the lasers are - there's some advantage to having it so lasers are in range of any spitters that are in range of whatever artillery you are using. Since lasers out-range spitters by 10 tiles IIRC, that still gives quite a few usable tiles in front of the lasers. You could have "cowardly artillery" and actually put it behind the lasers, but that's not how I roll - e.g. when playing bob's sniper turrets go up front, regular turrets/lasers go in the rear. Anything else reduces the useful range of your best-ranged weapon. Ditto for lasers and gun turrets in vanilla -- lasers up front, gun turrets in the back (or at best, side-by-side if there's only one turret row).
Deadly-Bagel wrote:they wouldn't need to be placed close to each other
You're making assumptions about range which may not be true. Sure, if they have a range which is much, much larger than biter agro range (which I'm not expecting to be the case as it would be totally OP), then you could space them out a lot and still keep spawners back far enough. But if their range is just a bit more than agro range, then that doesn't work - spacing them out would very quickly result in defensive holes that allow spawners to get too close.
Deadly-Bagel wrote:the train, stops, etc, realistic costs could be comparable
Unless the range is massive, I don't think so. You only need two train stops total, regardless of base size. You only need about twice as many rail signals as you have trains on patrol. One train could patrol a very large section of perimeter because you don't need instantaneous response - if a spawner shows up and starts spewing biters, it's OK if it takes a couple of minutes for the next train to arrive and blast it. (That's probably still faster than I would have gotten there anyways, and now I don't have to.) Trains can travel a *long* distance in a couple of minutes.
Deadly-Bagel wrote:Actually I think an even better approach would be a massive cannon you plant in the middle of your factory that increases range somehow
"better approach" how? It sounds insanely dull to me! What interesting design challenges might I run into while placing this gun in the middle of my base? It sounds like "none" to me. With trains, on the other hand, you have just tons more possibilities. For example you could use accumulators+switches+circuitry to measure power usage in major sections of your base's defenses, thereby detecting when lasers have fired, and automatically send extra artillery trains to any section which is seeing high amounts of enemy activity (though we don't know enough yet to know how useful that would be - we'll have to wait and see).
Deadly-Bagel wrote:I just don't see any advantages over static artillery
That seems to be because you simply don't understand good game design and interesting game play, and you just want stuff that's powerful and easy. If you want to offer a viable alternative to artillery trains for keeping spawners at bay, than suggest something that's actually interesting and (at least potentially) complex, not simplistic and boring.
Deadly-Bagel wrote:I can guarantee it won't be nearly as awesome as you think it will be
You've already failed then - your guarantee is worthless. I've never thought it would be "awesome". I do think it will more likely than not be a positive addition to the game, and I will most definitely be playing around with it once (if!) it comes out, and I find your arguments about "why it will suck" to be extremely weak. But none of that means I hold the belief that it will be "awesome".
Deadly-Bagel wrote:the production of the shells requires quite a lot of setup
Do you also forgo nukes? Because if you're making nukes, you already have everything but one last step to make regular rockets. (I'm not quite sure why you're into Factorio if even one step of automation is "quite a lot of setup" and deters you from even doing it.)
Deadly-Bagel wrote:Look at the Rocket Launcher. They "fixed" it right? Do you use it? No.
Actually, I'm in my first major 0.15.x game and I have been specifically getting ready (tech'ing up) to give rockets a shot again. (I probably last tried them in the 0.12.x time frame - not a good impression at the time, but I don't know that I had that many upgrades either - and my fighting style was quite lacking back then.) And... I'm back - just tried rockets (and nukes, and uranium rounds, and combat shotgun, and combat robots). Rockets don't seem that bad. For small nests combat robots by themselves and/or uranium rounds is easy/quick, but for somewhat larger nests the rocket (with destroyers, I generally always have destroyers active when fighting) lets you stay back a bit farther so you're not getting caught in biters as easily, and it's slightly quicker at killing spawners as well (with the only time consumed being the slight pause of your character while he fires). Combat shotgun seems to have been nerfed significantly. Nukes - well, I don't
need nukes to take out large bases, and making nukes is a bit expensive, but they sure are a time-saver! (And time is a valuable resource.) I just happen to be at the point where I have all of these weapons researched right up to (but not including) space science so it should be a good comparison between them. I used to (0.14.x) mostly go out using bots and combat shotgun, but now I think I might use bots + uranium ammo + rocket + nukes as my weapons configuration and switch between weapons as suits the situation. (I still have to try the flame thrower - I don't have that one tech'ed up yet however so I'll wait before comparing it. I don't like the idea of setting forests on fire while fighting, though, but maybe I'll get over it.)
Anyways, now that I've tried the 0.15.x rocket and found some use for it, I'd have to tentatively say you're claim of "Do you use it? No." is going to be false.
Deadly-Bagel wrote:going off the combat revamp of 0.15 I can guarantee it won't be nearly as awesome as you think it will be. {...} All this for literally no advantage over AP rounds which are cheaper, easier to make, more versatile, higher DPS, research boosts SMG + Tank + Turret damage, and everything flows nicely into uranium rounds for even more damage lategame.
That whole paragraph boils down to "Hey, the factorio devs suck at this combat stuff so whatever combat thing they're adding is going to suck." Well, yeah, they kind of suck at the combat stuff, but I figure it's a dice roll and we've got a better than 50/50 chance of getting something fun to play with out of it (and probably better still if they improve it based on feedback after it comes out). And I notice you don't mention nukes even though they were part of the 0.15.x changes - do you think nukes suck too? Or do you concede that the devs don't always totally suck at combat stuff? Frankly, nukes turned out better than I had expected because I figured they would just be too overpowered and boring, and while they sort of are overpowered, they're not exactly boring - they are actually rather dangerous to use (and there's some nice dynamic positioning aspects in terms of getting the most bang out of each shot). Anyways, it's the same devs whether they make artillery trains or static artillery, so that whole paragraph constitutes no argument for one or the other. Actually, I take that back - I believe your paragraph is an argument FOR artillery trains. Because even if artillery trains did end up kind of sucking, some of us will still have some fun coming up with and trying out creative designs/schemes aimed at making them not suck (whereas with a static defense, there's a whole lot less you can try to do with it).
Anyways, if they give us artillery trains, you can have your static artillery - just put down some track and plop an artillery train on it. So why are you complaining? (OK, maybe a little complaint about inability to blueprint trains.) The reverse does not work - if some of us want to play with automated mobile artillery and they give us non-automated and/or static artillery, there's nothing (short of modding) that we can do with that.