Fractal Softworks Forum

Please login or register.

Login with username, password and session length

Author Topic: Shield Stat Outliers  (Read 1814 times)

FooF

  • Admiral
  • *****
  • Posts: 1389
    • View Profile
Shield Stat Outliers
« on: December 19, 2020, 04:58:42 PM »

All the Conquest threads piqued my interest in shield stats (because the Conquest has a horrible shield), so I took a look at all Vanilla ships across Tech levels, hull sizes, and shield types to see what observations I could make and where the outliers were.

My initial investigation was in regards to shield upkeep vs. base dissipation rate. I was curious to see what the "average" was and then see what ships were significantly outside of that mean range and possibly why. This led to then looking at shield efficiencies, shield arcs, whether or not Front Shields are, on average, more efficient/offer more arc coverage than Omni, etc. It kind of became a rabbit hole that I've spent far too much time thinking about. :p

Here are my observations (wall o' text ahead...):

1. Shield Upkeep relative to Base Dissipation Rate

Your average Starsector ship has a shield upkeep rate somewhere between 40-50% of its base dissipation rate. I always knew it was high, but I didn't put two-and-two together until now. Only 7 ships (barring phase and shield-less ships) have below-average upkeep:dissipation ratios, while only 6 ships have higher-than-average ratios (i.e. shield upkeep is disproportionately high). The vast majority of ships fall between the .4 and .5 ratio, making this stat fairly universal across the board. In fact, these upkeep ratios are fairly universal even across Tech level averages (Low = .46, Midline= .42, High = .44) so that distinction doesn't really come into play.

The outliers are interesting and somewhat surprising.

The Tempest is the king of shield upkeep:dissipation ratio at .2, with the Odyssey not far behind at .25. Tied for 3rd are the High Tech ships, Shrike, Omen, and Medusa at .3. Basically, all the over-performers are High Tech. They're spending less relative to their base dissipation rates to keep their shields up. This should not be surprising considering that High Tech relies heavily on shield use for defense.

Regarding the less-than-great ratios, this outliers here are also interesting.

The worst ratio in the game belongs to the Astral, as its shield upkeep is 70% (.7) of its base dissipation! From an absolute upkeep perspective (just the upkeep cost itself), the Astral is tied for 5th at 420 flux/sec but its dissipation is only 600. Next is the poor Tarsus, coming in at .66. The Tarsus has a terrible shield profile, among the worst in the game (for additional reasons to be mentioned later.) The next worst ships are the Paragon, Apogee, and Hyperion at .6.

I find it interesting that High Tech basically has a monopoly on both the best and worst ratios in the game. Why the Astral has the worst is fairly straightforward to me: it doesn't have a lot of flux generation to begin with and as a carrier, its dissipation isn't the greatest. It really doesn't hurt the Astral, though, so I don't think a change is needed. The other high-tech ships surprise me a little but they also have among the best efficiencies in the game (see below), along with 300+ degree shield coverage.

2. Shield Efficiency

Shield efficiency is the ratio of damage the shield takes per 1 point of damage (lower is better). The vast majority of shields in the game have a efficiency between .8 and 1. Anything above (bad) or below (good) that range are the outliers. There is a split between the tech levels on efficiency. The average Low Tech ship is 1.07, the average Midline is .89 and the average High Tech is .73. This jives with the general perception of the Tech doctrines as we know them.

However, while High Tech has the best efficiencies, only High Tech has efficiencies below .8. There are High Tech ships that fall into the average range but no High Tech ships falls below that range. Likewise, only one Low Tech ship, the Kite gets to .8, the rest are 1 or below. Midline, naturally falls in the middle with a few egregious outliers, as most are .8.

The best shield efficiencies are High Tech, as mentioned. There is a 6-way tie between the Paragon, Astral, Hyperion, Scarab, Medusa, and Omen, all having efficiencies of .6. The Apogee and Shrike are just behind at .7. Like the shield upkeep values, none of these should come as a surprise: all these ships use their shields as their primary defense as part of the High Tech "school."

The worst shield efficiency, by a fair margin, is the Conquest coming in at 1.4. It takes sole possession of "worst." There are 9 ships with an efficiency of 1.2, and most are civilian craft but the warships include the Sunder, Enforcer, Condor, and Prometheus Mk. II. The Enforcer, I believe is going to 1 in the next patch. However, the Sunder and Conquest, both "glass cannons," have notoriously bad shield efficiency. Likewise, all Low-Tech ships, minus the aforementioned Kite, are around 1 for efficiency.

***From an efficiency standpoint, the Conquest's shield is truly terrible. With 20,000 capacity, but modified negatively by the 1.4 efficiency rating, the Conquest's effective shield HP is only 14,285. Only the Legion and "junk" capitals have less. Add to this that it is also "average" in its upkeep but has below-average arc coverage (90) and it is in the top 3 worst shields in the game. This appears to be by design, though, because any ship with good firepower and good maneuverability has to have a deficiency somewhere for balance reasons. Personally, I think 1.4 is still a tad high. "Raised" to 1.2, its effective shield HP would be 16,667, a little less than that of the Onslaught. Still bad but not egregiously so. I don't think the Conquest has *that* much firepower or maneuverability to justify such a poor shield efficiency, to be honest, but I don't think it's in a terrible spot as it is.

***The real outlier here, from my perspective, is the Odyssey at an efficiency of 1. It ties with the Buffalo for having the worst efficiency among High Tech ships. For perspective, the Apogee at .7 efficiency and 12,000 capacity has an effective hard flux pool of ~17,000 (barring any other modifiers). The Odyssey has 15,000 due to its efficiency equaling its capacity. Even the Aurora, with .8 and 11,000 capacity, comes close at 13,750. What I'm saying is that the Odyssey, in terms of shield tanking, is actually not any better than its High Tech cruiser cousins or even an Onslaught. Again, like the Conquest, this appears to be by design. Both the Conquest and Odyssey are fairly nimble for being such big ships and allowing them to take hits on shields and retreat via ship system would make them difficult to pin down to deliver real damage. However, purely from an internal consistency standpoint, they both seem wildly out-of-line with their respective fleet doctrines. If the Odyssey was buffed to a .8 efficiency (still low by High Tech standards), it would have 18,750 effective shield HP. I don't think it needs it but from an internal consistency standpoint, one could make the argument.

3. Front vs. Omni Shields

I wanted to see if there was a quantifiable difference between base Front shields and base Omni shields since the Front Shield Conversion explicitly states that Front shields are more efficient, have wider arcs, and open faster. As such, the conversion halves upkeep, doubles the coverage arc, and opens the shields faster. I consider the Front Shield Conversion to be an almost universally good use of OP for most ships (broadside ships being an exception) due to all the shield perks it gives. In general, more ships have Omni shields than Front (30 to 22) and there are a disproportionate number of Frigates that have Omni over Front (11 to 5). I don't consider one to be "better" than other, though an Omni-shielded ship has the option of getting the Front Shield Conversion, which does a lot of good things.

In practice, the data shows that Front-shielded ships do not, in fact, have better average upkeep:dissipation ratios than Omni-shielded ships because they are virtually identical (.45 to .44). This means that a Front-shielded ship, has no advantage over an Omni-shielded ship in terms of shield cost. If Front-shielded ships had shield upkeep roughly half of their Omni-shielded counterparts, there would be a distinction but no such distinction exists in-game. In terms of coverage arcs, Front-shielded ships do have considerably wider arcs on average: 215 vs 150. It's not quite double but it is significantly more. In terms of efficiency, which is not 1:1 with shield-type, Front shields do tend to be slightly more efficient on average than Omni shields (.88 vs. .94) but I believe that to be more coincidence than intentional.

***Whether a ship has Front or Omni shields doesn't seem to have a lot consistency to me.  The Astral has a 360 degree Front Shield at .6 efficiency while the Paragon has a virtually identical shield, save that it is Omni. The Enforcer and Mora have Omni shields while the Dominator, Legion and Onslaught do not. All high-maneuverability High Tech ships have Omni shields but the Wolf does not. Front vs. Omni within doctrine types seems kind of all over the place, too.

Winners and Losers/Suggestions

Winner. The Omen has the best overall shields in the game. At .3 upkeep:dissipation ratio, .6 efficiency, 360-degree Omni shields (which can be converted to Front to halve even the paltry upkeep), the tiny little Omen is surprisingly durable with over 4,000 effective shield HP. (Of course, the Monitor is a special case and really has the best shields in the game!) The Medusa is also right there with the Omen, it just doesn't have as much shield coverage.

Winner. High Tech. They have the best shield efficiencies in the game along with the best shield upkeep:dissipation ratios. However, their values aren't vastly ahead of Midline or Low Tech to the point where its imbalanced. This is why Energy Weapons, despite their short range and generally inferior stats, have to be balanced against the High Tech doctrine of good shields.

Loser. The Tarsus. The poor Tarsus arguably has worse shields than the Conquest. Not only does it have the second-worst upkeep:dissipation ratio, it also has the second-worst shield efficiency. The one thing it might have better than the Conquest is a better shield arc (120 vs 90) but at least the Conquest has an average/good upkeep ratio (.4 vs. .664). Suggestion: Make the Tarsus better defensively than the Buffalo since it costs more. Shield efficiency of 1 (from 1.2) and shield upkeep of 50 (.4 ratio).

Loser. Civilian ships, fuel tankers specifically. They have poor efficiency and poor shield upkeep ratios. The Atlas, in particular, has the smallest shield arc in the game at 45, though the rest of its shield stats are average. No suggestion here: this is all very justifiable as civilian ships tend to be very vulnerable.

Loser. The battlecruisers. The Odyssey and Conquest suffer from Inefficient Shield Syndrome because of their overall speed. As mentioned, I think their efficiencies could be tuned a little more in-line with the rest of their fleet doctrines for consistency's sake, though I don't believe where they are at is bad, per se.

Loser. Inherently Front-Shielded ships. Front shields are supposed to be more efficient but those that are built-in with them are not. I didn't test shield opening speed (perhaps they are indeed faster to open) but they definitely don't get a discount vs. Omni shields as it pertains to upkeep. Having a Front shield simply keeps you from getting the very nice Front Shield Conversion. Suggestion: Don't break the game but discount the upkeep of all natural Front Shielded ships. It could just be a built-in Hullmod that reduces upkeep by 25% or something rather than re-tooling all the values. I don't know if I'm seriously suggesting this but the game isn't internally consistent on this topic.

Winner. Stabilized Shields. I have a new appreciation for how much Stabilized Shields can affect my ship's flux profile. It won't come before maxing vents but when the average ship loses 40-50% of its dissipation to simply maintaining the shield, cutting it in half for 3/6/9/15 doesn't seem like a bad idea. The Paragon, for example, saves 375 flux/sec with a 15 OP hullmod. Or to put it another way, adding 60 Vents doesn't even cover the cost of raising the shield!

So...if you read this far, thank you! I'm sure .95 will skew everything but as for what we have, that's what I've noticed. Feedback and discussion welcome.

-Made edits due to incorrect info
« Last Edit: December 20, 2020, 05:53:29 AM by FooF »
Logged

TaLaR

  • Admiral
  • *****
  • Posts: 2794
    • View Profile
Re: Shield Stat Outliers
« Reply #1 on: December 19, 2020, 08:47:25 PM »

While for some ships Stabilized Shields may be an optional shield efficiency upgrade, for Paragon/etc it's more like a near-mandatory OP tax.
Logged

Wyvern

  • Admiral
  • *****
  • Posts: 3803
    • View Profile
Re: Shield Stat Outliers
« Reply #2 on: December 19, 2020, 10:16:14 PM »

Personally, I don't find 'shield upkeep as percentage of base dissipation' to be a particularly useful stat, for all that it's stored that way in the data files. A combat-ready ship generally has a large number of extra flux vents, and some hulls will almost always see stabilized shields installed, too. I'm not entirely sure what the actual useful stat here is, though.

I do tend to agree on the Odyssey's shield being rather poor for what it is; it'd be okay if it had an actual maneuverability system rather than the current lunge-forwards system, but it doesn't, so it's not.

The Sunder actually makes a pretty good shield tank, despite the low efficiency - with its flux stats designed to accommodate heavy energy weaponry, it works out pretty well.
Logged
Wyvern is 100% correct about the math.

TaLaR

  • Admiral
  • *****
  • Posts: 2794
    • View Profile
Re: Shield Stat Outliers
« Reply #3 on: December 19, 2020, 11:12:03 PM »

I do tend to agree on the Odyssey's shield being rather poor for what it is; it'd be okay if it had an actual maneuverability system rather than the current lunge-forwards system, but it doesn't, so it's not.

Player-piloted Odyssey can work around forward-only system well enough to count as one of best possible player ships. First, it's a broadside ship, so you don't even need to aim nose at enemy. 2nd, you need to plan maneuvers way ahead of time. For example, approach into combat range and immediately start turning for future retreat, all while firing broadside. If enemy retreats somewhat in process, backpedal in their direction.
Or use system in diagonal direction to dodge incoming shots and drop shield + hold fire in process to lower flux levels somewhat.

Of course AI doesn't have a shred of flexibility like that and uses system only for straight approaches.
Logged

SCC

  • Admiral
  • *****
  • Posts: 4143
    • View Profile
Re: Shield Stat Outliers
« Reply #4 on: December 20, 2020, 12:47:37 AM »

I don't think that Conquest performs badly enough to warrant a nerf buff, though admittedly, I rarely ever let AI pilot it. It is hard to build, but that's hardly something you can help with by changing shield stats.
Decreasing upkeep for all front shields would indirectly benefit midline and low-tech more than high-tech, I'm not sure if it's needed. Lasher, Dominator, Onslaught, Legion vs Brawler, Wayfarer, Hammerhead, Sunder, Drover, Falcon, Eagle, Gryphon vs Wolf, Hyperion, Apogee.

Edit: yes, FooF, I meant to write "buff".

SafariJohn

  • Admiral
  • *****
  • Posts: 3023
    • View Profile
Re: Shield Stat Outliers
« Reply #5 on: December 20, 2020, 06:01:05 AM »

Frankly, I never look at shield upkeep except to see if Stabilized Shields is worth it.
Logged

FooF

  • Admiral
  • *****
  • Posts: 1389
    • View Profile
Re: Shield Stat Outliers
« Reply #6 on: December 20, 2020, 06:06:01 AM »

Personally, I don't find 'shield upkeep as percentage of base dissipation' to be a particularly useful stat, for all that it's stored that way in the data files. A combat-ready ship generally has a large number of extra flux vents, and some hulls will almost always see stabilized shields installed, too. I'm not entirely sure what the actual useful stat here is, though.

I do tend to agree on the Odyssey's shield being rather poor for what it is; it'd be okay if it had an actual maneuverability system rather than the current lunge-forwards system, but it doesn't, so it's not.

The Sunder actually makes a pretty good shield tank, despite the low efficiency - with its flux stats designed to accommodate heavy energy weaponry, it works out pretty well.

Didn't I feel foolish when I realized shield upkeep was a stat in the ship files...I manually input all the shield upkeep and base dissipation numbers by hand from the Starsector Wiki... :idiot: You're right, it's right there in the files so the stat isn't all that useful.  I thought the upkeep numbers were independent somehow and not some fraction of dissipation. So, my working assumption was wrong and I thought this was new info. Well, it was new(s) to me, I guess! (I also learned the Wiki's data is wrong and cleaned up the OP) The meta-analysis might still be useful though.

@SCC

Did you mean to say "buff" not "nerf?" I was suggesting a buff in shield efficiency. Agreed, I don't think the Conquest is performing badly. My argument is almost purely from this hyper-focused vacuum of looking at shield efficiency numbers and doing meta-analysis on them. That's why I don't know if I'm seriously suggesting Front Shield ships getting a shield upkeep buff across the board.

Also, to clarify my language in the OP, I should have been using the word "stable" not "more efficient" when describing Front Shields because "efficient" has its own meaning when talking about shields. Front Shields should have less upkeep, not be more efficient (i.e. take less damage when hit). 
Logged

DatonKallandor

  • Admiral
  • *****
  • Posts: 718
    • View Profile
Re: Shield Stat Outliers
« Reply #7 on: December 20, 2020, 11:12:35 AM »

The Tarsus really doesn't need a defensive buff because it has burn drive, the best defensive system a freighter can possibly have.
Logged