Fractal Softworks Forum

Please login or register.

Login with username, password and session length
Advanced search  

News:

Starsector 0.97a is out! (02/02/24); New blog post: Simulator Enhancements (03/13/24)

Pages: 1 ... 17 18 [19] 20 21 ... 71

Author Topic: Starsector 0.6a (Released) Patch Notes  (Read 553304 times)

Silver Silence

  • Admiral
  • *****
  • Posts: 980
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #270 on: June 01, 2013, 01:25:39 PM »

I do gear for it. Heavy Maulers, Gauss Cannons, HVDs, Railguns, HILS, Tach Lances, all weapons typically mounted on my ships. Though right now, in the Fairy mod, I'm just using Guardian PD with 750 range, but a terrifying 2000 DPS and 600 maintained DPS after the charges are spent. And I have 3 of them. Hounds get annihilated so fast that their wrecks often smash into me at full speed because they died without a chance to maneuver. I thought the Autopulses with 1200 range in this mod were cool, HA. -HA-.
Logged

flashydragon

  • Ensign
  • *
  • Posts: 25
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #271 on: June 01, 2013, 02:28:08 PM »

I do gear for it. Heavy Maulers, Gauss Cannons, HVDs, Railguns, HILS, Tach Lances, all weapons typically mounted on my ships. Though right now, in the Fairy mod, I'm just using Guardian PD with 750 range, but a terrifying 2000 DPS and 600 maintained DPS after the charges are spent. And I have 3 of them. Hounds get annihilated so fast that their wrecks often smash into me at full speed because they died without a chance to maneuver. I thought the Autopulses with 1200 range in this mod were cool, HA. -HA-.

This^ 8)

Quick question...what are capital ships supposed to be used for, if not death-star-style extermination campaigns, then? Is this the first step in removing the ship type? Where is the line between "this is too strong" and "this is how you use the kit"? :-\

I'm still not convinced that removing auto-resolve is a good idea. It just sounds like a move to dramatically lengthen my already late late late night gaming sessions  :'(
Logged

ValkyriaL

  • Admiral
  • *****
  • Posts: 2145
  • The Guru of Capital Ships.
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #272 on: June 01, 2013, 05:24:31 PM »

I hate the fact that auto resolve is removed so much that i'm simply just not going to update... ever... if there is a logical reason behind removing it, i'd sure as hell would like to hear it.
Logged

Gothars

  • Global Moderator
  • Admiral
  • *****
  • Posts: 4403
  • Eschewing obfuscatory verbosity.
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #273 on: June 01, 2013, 05:55:29 PM »

I hate the fact that auto resolve is removed so much that i'm simply just not going to update... ever... if there is a logical reason behind removing it, i'd sure as hell would like to hear it.

Here:

The idea is that even when you have a big fleet size advantage, you still don't want to just deploy all and steamroll, because of associated CR loss and supply costs. It's hard to envision the role autoresolve can play in that environment; you'd possibly need to tell it what you want to use, what you don't, how many ships it's ok to use, etc. At that point you might as well play it out. If you're "grinding" low-level fleets for money/XP, you wouldn't want to use it anyway. Either you'd end up with too many losses, or too many expenses.

You exaggerate Valky, but I'm not too sure either I like the complete removal of autoresolve. It has indeed no in-game use, but sometimes a player is just not in the mood to switch from the campaign to the combat layer. That will probably be more true if there are important things to plan and think about for campaign progress. I see no reason to not allow the option of autoresolving at high CR cost.

Don't think it will matter all that much though, it's not like the mechanic itself could provide you with any fun.
Logged
The game was completed 8 years ago and we get a free expansion every year.

Arranging holidays in an embrace with the Starsector is priceless.

Alex

  • Administrator
  • Admiral
  • *****
  • Posts: 23986
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #274 on: June 01, 2013, 07:48:15 PM »

Right; it's more a question of me not seeing how it fits into things. I can see the "I can't be bothered right now" use case when there's other campaign stuff it's distracting you from (and when there's that campaign stuff to be distracted from in the first place.) I'd like to see how CR shakes out first, though, and if that really ever comes up with the new battle mechanics.

(Side note: I really dislike the idea of autoresolve being better than fighting it out in any circumstances.)

Well, a Lasher, 1-2 Hounds and maybe a couple Talon or Piranha wings. FP-wise, the same weight as something like a Conquest, thus why I get the engagement. In terms of combat proficiency, not in remotely the same leagues.

This is true, but that's also the kind of situation you don't want to use autoresolve in because it's likely going to produce bad results. It can't not, it's basically impossible to tell apart situations where one side has overwhelming superiority due to how the fleets match up tactically. Oh, you could catch some cases, but not all, or even most. There's just too much customization possible.

(It's also not an obviously bad engagement for the other side to take, harassing your ship into a CR loss and an eventual defeat could be a viable option.)


So... I get it, autoresolve sounds like a good thing to have. But when would you really use it? When it has a chance of totally messing up the outcome, because the battle looks close on paper? Or when the enemy is so outgunned that they're trying to run away, and you're hitting "autoresolve" because it has a better chance to nab those pesky Hounds (and you already don't have to fight it out and can just let them go)? "Save/reload until autoresolve wins" is not a use case I want to design for, and that's about the only thing that comes to mind.

Hmm. Another case involves the player figuring out how autoresolve works and getting a fleet together that does will with it. I'm not keen on that as a core game mechanic, either.

It seems to me that it's mostly useful now either in the save/load case, or in the "over-aggressive weak fleets" one. The second should be addressed in the new release, they're much better about gauging their chances.

That said, I definitely want to hear what you guys think. Are there cases where you'd want to use it that I didn't cover in the above?
Logged

naufrago

  • Admiral
  • *****
  • Posts: 511
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #275 on: June 01, 2013, 08:15:32 PM »

Well, aside from farming exp, credits, and loot...

Sometimes I'll engage a fleet that I can easily steamroll, but autoresolve would end up killing off most of my fleet. In those cases, I'll manually go in, take out the biggest threats and whatever else gets in my way, then autoresolve since an easy victory is assured at that point. Being completely dominant can be fun for a while, but I'll get bored if there's no way to skip to the win when total victory is guaranteed.

What I'd like you to take from that is that maybe keep autoresolve if you want to end combat early, but stipulate that only the ships currently deployed are allowed to be used by the autoresolve. You could even take it one step further and, instead of getting rid of second-in-command, dictate before battle which ships you want deployed for autoresolve (ie. tell second-in-command which ships he can use) and keep the rest in reserve. It's fine if you make autoresolve universally worse than manually fighting if it lets you skip trivial battles (although I'll be annoyed if one of my ships gets heavily damaged or destroyed needlessly like that).

You could make it so that you can't autoresolve escape sequences if your primary concern is using autoresolve to catch speedy ships, and lower autoresolve's odds of catching speedy ships when they try to escape.
« Last Edit: June 01, 2013, 08:17:51 PM by naufrago »
Logged

Alex

  • Administrator
  • Admiral
  • *****
  • Posts: 23986
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #276 on: June 01, 2013, 08:31:22 PM »

What I'd like you to take from that is that maybe keep autoresolve if you want to end combat early

What happens now is you can end it early - with a win - once all the enemy ships are retreating. There's a message to let you know when this is happening.

It's fine if you make autoresolve universally worse than manually fighting if it lets you skip trivial battles (although I'll be annoyed if one of my ships gets heavily damaged or destroyed needlessly like that).

The problem I see there is, again, we're either in 1) autoresolve will handle it poorly because it looks more even on paper than it is or 2) they're trying to get away, so you can skip it already.


You could make it so that you can't autoresolve escape sequences if your primary concern is using autoresolve to catch speedy ships, and lower autoresolve's odds of catching speedy ships when they try to escape.

Hmm. I'm starting to think that "catch a bunch of stuff that's running away" might be the only case left where it's really useful. It's almost a given that you've got vastly superior forces, and it seems more feasible to tune it to be really bad at catching ships - though not utterly terrible - while disallowing a loss.

Though there's still the CR issue there. Yeah, you could have "what the 2nd in command gets to deploy" selection stage... hmm. I'd considered that and thought it a bit of a hassle, but maybe that idea is worth another look. Then you could base the amount of stuff captured on how much stuff you allowed to be deployed.

Yeah, I think that could work. You take out the unpleasant randomness of losing a battle because of it, and have an inefficient shortcut to a potentially bothersome task of chasing down a bunch of stuff. Though those battles actually tend to be a fun change of pace, especially when the enemy has a decent amount of escorts and they use them to delay/escort the fleeing ships.

You'd probably still end up with the occasional "how'd that lone Atlas get away from 3 Tempests", but I suppose that's the breaks when you trust the 2nd in command.
Logged

Dri

  • Admiral
  • *****
  • Posts: 1403
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #277 on: June 01, 2013, 08:53:36 PM »

Don't load in the art assets and just have the autopilot simulation run greatly accelerated in the background? I know this may still take some seconds to run its course but it sure beats out coming up with some funky RNG system and other formulas that govern autoresolve.

Sword of the Stars 2 uses its combat engine for autoresolve, just greatly speed up, and that game has full on newtonian physics and full 3d battlefields and while it takes ~5-10 seconds to play it out the results are much more accurate and believable because its an actual, fully played out battle. So why can't Starsector use its own combat engine just speed up?
« Last Edit: June 01, 2013, 09:04:07 PM by Dri »
Logged

naufrago

  • Admiral
  • *****
  • Posts: 511
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #278 on: June 01, 2013, 09:13:08 PM »

What happens now is you can end it early - with a win - once all the enemy ships are retreating. There's a message to let you know when this is happening.

Ah, I missed that part in the patch notes. Good to know.

Well, autoresolve seems like a tricky subject, so I'll trust your opinion. But I'm glad if any perspective I can offer helps at all =)
Logged

Alex

  • Administrator
  • Admiral
  • *****
  • Posts: 23986
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #279 on: June 01, 2013, 09:31:56 PM »

... and that game has full on newtonian physics and full 3d battlefields ...

I can't really speak about SotS since I'm not familiar with the game, but neither of those are indicative of the complexity of the simulation. Other stuff (like the AI, and managing the state of all the entities/weapons/etc) is much more CPU-intensive.

At this point - in Starsector - the graphics and the simulation tend to take roughly the same amount of time per frame, unless there's a serious imbalance in the hardware. So, at worst, assuming the game was just barely running at 30 fps, you would speed the battle up by a factor of 2 or so by ditching the graphics. That's clearly not good enough.

Why look at the worst case? Well, we're talking about hardware that can handle the game. If *autoresolve* is what makes the game have higher performance requirements, something is very wrong.

(Even if we decide to increase the simulation step size, I don't think it'd do enough. At the point where it would, we're talking about missing collisions and seriously affecting the fidelity of the simulation in other ways. Really, it's not even practical with high-end hardware - at best we'd be talking something like 30 seconds for a decent-sized battle.)



Well, autoresolve seems like a tricky subject, so I'll trust your opinion. But I'm glad if any perspective I can offer helps at all =)

I appreciate the feedback/ideas. It is a tricky subject, so seeing others' perspective really does help.
« Last Edit: June 01, 2013, 09:43:34 PM by Alex »
Logged

Talkie Toaster

  • Captain
  • ****
  • Posts: 256
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #280 on: June 02, 2013, 03:19:09 AM »

... and that game has full on newtonian physics and full 3d battlefields ...
(Even if we decide to increase the simulation step size, I don't think it'd do enough. At the point where it would, we're talking about missing collisions and seriously affecting the fidelity of the simulation in other ways. Really, it's not even practical with high-end hardware - at best we'd be talking something like 30 seconds for a decent-sized battle.)
This is how SotS does it, which leads to problems like missiles being more effective in autoresolve as PD has fewer chances to hit them.

Plus... SotS2 has suffered from huge performance and stability issues since day 1, and is no exemplar of game design either. Not really something to use as a target.
Logged

Gothars

  • Global Moderator
  • Admiral
  • *****
  • Posts: 4403
  • Eschewing obfuscatory verbosity.
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #281 on: June 02, 2013, 03:43:06 AM »

That said, I definitely want to hear what you guys think. Are there cases where you'd want to use it that I didn't cover in the above?

To expand on what I said before, for me the only real reason to miss AR would be the forced switch out of the campaign layer. The only thing I can really compare it with is the Total War series. First I have to say that I absolutely love the battle system in those games, I would buy them just for it (same with Sector). But they take time and a lot of concentration. The same goes for the campaign layer, where I have to plan much more long-term. So when I'm busy planning ahead (invasion of Spain, marriage of my daughter, assassination of the pope), the last thing I need is to be completely interrupted in my thoughts by a skirmish against some rebels. I autoresolve in these cases, even at the price of losses.

Well, it is probably not as drastic with Starsector since you only have one fleet instead of several armies you have to worry about. Still, even now, this is almost the only thing I use AR for. I don't grind, but if I have the idea that I absolutely need an Omen now, I will set course to the Tri-Tach station and autoresolve every smaller battle on the way there, because fighting is not what I want right now. Honestly, there's a good chance I don't even know anymore what my plan was after a battle.

... which, thinking about it, may be the core of my problem. Mh. Had the same problem in Mount and Blade, with trading runs where I forgot destination and price of my goods during battles (which I didn't like btw).

Mhhh. Now I want a memo function. :)
Logged
The game was completed 8 years ago and we get a free expansion every year.

Arranging holidays in an embrace with the Starsector is priceless.

RawCode

  • Admiral
  • *****
  • Posts: 511
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #282 on: June 02, 2013, 06:54:43 AM »

autoresolve usefull only for grinding and nothing else.

this game not about grinding, there is no reason to allow player autoresolve.
Logged

ValkyriaL

  • Admiral
  • *****
  • Posts: 2145
  • The Guru of Capital Ships.
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #283 on: June 02, 2013, 07:01:46 AM »

At the current stage of development, SS is about grinding since the other features haven't been added yet, autoresolve is a way to skip boring battles when your fleet far outmatches the enemy to the point that you have 100% chance of winning anyway, but i suppose you knew that right?
Logged

Alex

  • Administrator
  • Admiral
  • *****
  • Posts: 23986
    • View Profile
Re: Starsector 0.6a (In Development) Patch Notes
« Reply #284 on: June 02, 2013, 09:46:31 AM »

... and that game has full on newtonian physics and full 3d battlefields ...
(Even if we decide to increase the simulation step size, I don't think it'd do enough. At the point where it would, we're talking about missing collisions and seriously affecting the fidelity of the simulation in other ways. Really, it's not even practical with high-end hardware - at best we'd be talking something like 30 seconds for a decent-sized battle.)
This is how SotS does it, which leads to problems like missiles being more effective in autoresolve as PD has fewer chances to hit them.

Plus... SotS2 has suffered from huge performance and stability issues since day 1, and is no exemplar of game design either. Not really something to use as a target.

Ah, good to know. In fairness, losing some simulation fidelity might be ok - I mean, ok, say the missiles are more effective, but the overall end result is probably still pretty accurate. Unless you got for dedicated missile boats, I suppose :)

To expand on what I said before, for me the only real reason to miss AR would be the forced switch out of the campaign layer. The only thing I can really compare it with is the Total War series. First I have to say that I absolutely love the battle system in those games, I would buy them just for it (same with Sector). But they take time and a lot of concentration. The same goes for the campaign layer, where I have to plan much more long-term. So when I'm busy planning ahead (invasion of Spain, marriage of my daughter, assassination of the pope), the last thing I need is to be completely interrupted in my thoughts by a skirmish against some rebels. I autoresolve in these cases, even at the price of losses.

Hmm. I'm a fan of the TW series, too. (Well, I picked up TW: Shogun 2 a while ago, and that didn't go so well, though I liked the earlier games a lot. But I digress.)

Thinking about how that worked, yeah, at some point you just want to focus on the grand strategy and your individual battles don't matter as much anymore. You've got the numbers/income to absorb extra losses. There are also lots of trivial battles where you do have an overwhelming army taking out some piddly province with a unit of spearmen or some such, and ensuring 0 losses is just not worth the loading time.

In those circumstances, autoresolve is great.

Moving on to Starsector, though: considering an inferior enemy should decide to flee, the only time that'll come into play is either in an escape-style battle or if the enemy fleet overestimates its capabilities *compared to the outcome autoresolve would create*. The latter is more of an AI problem, and is largely resolved in 0.6a.

There's the possibility of suicidal types attacking you, though. The "Cult of Lud" comes to mind here... right, ok. Need to think about that some more, but seems like a pretty good reason right there, even if not immediately relevant.
Logged
Pages: 1 ... 17 18 [19] 20 21 ... 71