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 ... 5 6 [7]

Author Topic: A Noob's Insight on: Ships!  (Read 23230 times)

Vanshilar

  • Admiral
  • *****
  • Posts: 585
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #90 on: December 22, 2021, 06:40:49 AM »

I should say, raw damage statistics are nowhere near a complete picture of effectiveness. They are not useless, but do not give the whole picture. I'm merely trying to point out the ways that those numbers can fail to reflect underlying effectiveness, or ways they can overrepresent certain aspects of ships that aren't necessarily that valuable.

Damage numbers aren't necessarily the only thing you need to know, but they're probably one of if not the most important measures of a ship's effectiveness. After all, the purpose of a fight is to reduce every enemy ship's hull to 0, and the way to do that is by doing damage. They're basically a primary metric of a ship's effectiveness.

I would argue hyperion has a ton of value purely in it's tendency to end up behind the enemy and not die while doing that, causing them to turn around and expose their vulnerable engines to the rest of the fleet.

Going behind battle lines is actually a terrible thing for the AI to do, since you're opening up new battle lines and taking damage in a situation where (by definition) the enemy wouldn't be shooting anyway, and you're diluting your own forces to do so. Now it's great for the Hyperion to help draw off some enemy forces to help your fleet concentrate its firepower on fewer remaining ships, but that's something that can be done by other frigates, and 15 DP for each Hyperion is a lot to dedicate toward that (since you're looking at the ratio between the DP of the ship that you're using to draw enemy ships away, and the DP of those enemy ships). Every DP that goes toward splitting up the enemy fleet means less DP toward actually killing the enemy fleet. So this points to having maybe 1-3 such ships, and keeping their total DP low, otherwise your fleet starts losing too much firepower for the splitting up to be worthwhile. The fact that the Hyperion is such an expensive ship at 15 DP means that it has to distract a lot more enemy ships than other frigates doing the same task to be worthwhile. It doesn't mean the Hyperion isn't useful, but it means it's very expensive for its task, and not something worth building your fleet around. Basically support, like having a few cargo ships, although in this case Hyperion(s) would be support in combat rather than campaign.

Additionally, this means it's very expensive to dedicate skill points toward frigate-boosting skills such as Wolfpack Tactics on a high-end fleet, when there are many other skills which help boost the whole fleet or your personal flagship instead. Furthermore, having Hyperions rely on SO means you're going to be burning through CR and thus a lot of supplies for this.

r.e. dmg/sec
It doesn't matter at all unless CR/PPT is a major factor costing you supplies. A strategy that kills everything in 20 seconds is not worse than a strategy that kills everything in 10 seconds.

That doesn't make sense. There's a reason why DPS is a very important metric in pretty much all combat games, Starsector included. All other things being equal, a strategy (or fleet setup, loadout, etc.) that kills the enemy fleet faster is better than one that kills the enemy fleet slower. The faster strategy means that it can churn through more enemy ships before CR/PPT becomes an issue -- or, alternatively, means that you could've committed fewer ships (saving supplies) and still have had CR/PPT not being an issue. It means you have more wiggle room, time-wise. Being able to kill enemy ships quickly is a good thing.

The context of this is that you were claiming that a ship which kills enemy ships quickly may end up doing less overall damage than a ship which is trading blows back and forth, and I was pointing out that this is incorrect because it takes a lot of time for ships to back off, spend some time with shields up doing nothing, vent, re-enter combat, etc., during which time they're doing 0 damage. So if you look at the overall damage per unit time, a ship which is churning through enemy ships will end up with more overall damage dealt. Not quite sure why you're then claiming that damage/second is unimportant based on that.

Your fury is very heavy on efficient/anti-shield damage (double sabot, minipulser, ir pulse lasers), so it is not surprising at all that it is more effective against remnants that heavily incentivize shield damage with their super efficient shields.

Your claim was that the Hyperion could bypass shields due to TP, and therefore may do less overall damage because of it. Again, this conflates "per unit ship" versus "per unit time", but even aside from that, I was pointing out that this isn't what the numbers ended up showing. If the Hyperion were successfully bypassing shields, you would see it do an outsized portion of the overall hull damage. But I was pointing out that the Furies did a bigger proportion of the total hull damage than the Hyperions (even after accounting for 20 DP vs 15 DP). In other words, contrary to your claim, the Hyperions were actually worse at hitting hull, even though the Furies only had a single Cryoblaster for anti-armor/hull, while the Hyperions had that plus a Heavy Blaster.

This isn't exclusive to Fury vs Hyperion, as if it's just because Furies were overpowered. Testing AI-controlled Odyssey vs Apogee vs Hyperion shows the same trend -- the Hyperions will underperform in terms of overall damage, and underperform in terms of proportion of the hull damage dealt relative to its DP, compared with the other ships. The Hyperions only did 10% of the overall damage and 11% of the hull damage, even though it's 19% of the DP. Screenshot attached.

The testing setup is as follows:
1. Take the desired fleet, in this case, 3 Odysseys (Cryoblaster/HIL on large energies, Hurricane on large synergy, Sabots on medium missiles, Xyphos/Longbow), 3 SO Apogees (Cryoblaster on large energy, Hurricane on large missile), and 3 SO Hyperions (Cryoblaster/Heavy Blaster/Ion Pulser), and attack a test fleet, in this case a 2-Ordos, 7-Radiant fleet.
2. Start off by grabbing objectives to get the full fleet deployed.
3. Gather the fleet together on the player side of the map.
4. Once the fleet is gathered together, set the fleet on Full Assault mode. (Note this cancels any previous player inputs.)
5. From here on out, no more player commands are given for the rest of combat. The player fleet is under complete AI control and is free to roam on their own, without any player guidance, until the battle completes. The player is merely a spectator once Full Assault mode is switched on.

It was done in 0.95a, but I don't think the results will change appreciably in 0.95.1a.

The main issue with Hyperions (and frigates in general, this is not unique to Hyperions) is that it tries to stay out of combat too often. (All officers were set to aggressive for all of my ships.) Even when it does decide to teleport in to fight, it'll take some shots briefly, then teleport out again, even though it's at low flux. It seems like by and large the Hyperions will say "welp looks like the enemy fleet is too dangerous, I'll just stay away until I see an opening" and wait for some other ship to create an opening (i.e. attack a ship and weaken it first) before coming in. (In RPG games, I think this is known as "kill-stealing", heh.) The biggest times where the Hyperions would commit to battle were 1) at the beginning, where there are a lot of isolated frigates to hunt down, and 2) at the end, when the whole fleet is concentrating their fire on the last few remaining Radiants. During most of the fight, the Hyperions would only go in sporadically, and jump out at the first sign of danger, letting the rest of the fleet do the bulk of the damage.

This is the opposite of what the Hyperion should do, especially since with SO it can teleport out to safety at will, so what it should do is jump in, shoot and absorb damage until it's at say 90% flux, then teleport out and let its flux dissipate, then rinse and repeat. But realistically the AI doesn't really "know" that it's commanding a Hyperion, and fighting against an enemy Hyperion doing this would be really annoying. So a lot of that potential is wasted.

Generally this seems to be the behavior of frigates, where they'll wait until there's an opening, or until there's a critical mass of them, before going in to fight -- they'll stay away until they feel there's enough to dogpile on a target, or until some other ship makes the first move. This is great for self-preservation for most frigates since they're generally pretty squishy, but it doesn't work out well for the Hyperion, who has the stats of a cruiser and should bully enemy ships around like it. But again that would make the Hyperion very, very annoying to fight against.

I think if you could somehow add up all the damage that was either fired at the hyperion and missed, or not fired at all because the ship was distracted by the hyperion, it would be a very large number, perhaps larger than the furies total damage tanked.

I highly, highly doubt it. The Hyperions simply aren't near the front lines often enough for that to happen.

r.e. the 2-Ordos test fleets as benchmark
I don't think a computer benchmark test is a fair comparison. At the end of the day, every computational task boils down to a bunch of basic operations/computations that need to be executed, and processors are measured by the number of computations per second they can do. On the other hand, different fights in starsector reward different types of damage, and different strategies unequally, so being able to deal damage quickly/efficiently in one fight doesn't necessarily translate perfectly to another.

Sure, at some level it's just some sort of basic operations/computations. Nowadays though processors aren't benchmarked directly by computations per second, but by how long it takes to complete representative tasks (or alternately, number of frames per second, which boils down to the same thing).

Granted, in Starsector, every single fight is unique. If you take any enemy fleet, and say add a frigate to it, I guarantee you the "optimal" fleet (whether you're measuring by minimum time needed to kill the fleet, or minimum DP needed, or whatever other metric), will be slightly different. Maybe the extra frigate means your fleet would be better served by something faster, or maybe the extra frigate has some missiles in which case it would've better had one of your ships had a bit more PD, or whatever. There is no single fleet that is going to be optimal (i.e. better than every other possible fleet) across many enemy fleets, much less across the spectrum of possible enemy fleets in Starsector, from the initial pirate D-modded frigate to Star Fortresses and Remnant fleets. So saying results from one test setup may not "necessarily translate perfectly to another" is a bit disingenuous.

I doubt anyone here is looking for "what is a fleet setup that is optimal for this one particular fight" nor "what is a fleet setup that is optimal for every possible fight" which doesn't exist. Rather, people are looking for "what is a fleet setup that is generally good across a wide range of fights" in Starsector. This thread is about evaluating the different ships, after all.

So you can certainly say "well, a 2-Ordos fleet isn't really representative of the fights that you'll encounter in Starsector". Sure. You can come up with your own set of 10, 20, 30, or however many different kind of fights that you think would be representative of fights in Starsector, and test out different fleets against those benchmarks to find what you feel is the "optimal" fleet in some sense, or the "optimal" ships to use. However, I don't have that kind of time. I prefer to test against one test fight, where I know that if my fleet does well against that test fight, it will do well against the vast majority of content in Starsector.

That's the purpose of using a 2-Ordos fleet as a test benchmark. If a fleet can do well against it, the fleet will almost certainly do well against pretty much everything else in Starsector, except for possibly some very specific fights. (I haven't gotten a chance to try this fleet against hypershunts yet nor the scripted phase fleet, for the simple reason that they were already over before I made this fleet.) It will do well against Star Fortresses. It will do well against regular faction fleets. It will do well against single Ordos fleets, the purpose of doing a 2-Ordos fleet (and sometimes a 3-Ordos fleet) is to stress test the fleet against difficult odds.

This is not the case if almost any other fleet were used as a test benchmark. The fight needs to be appropriately difficult, otherwise it loses its discriminating power between good and bad fleets (ships, loadouts, etc.). If you optimize your fleet against a faction fleet, there's no guarantee that it will do well against another faction's fleet, and it will almost likely fare horribly against Ordos fleets. It may or may not work well against stations.

So you're free go about making your recommendations on what ships are good or bad without accounting for how they would fare against Remnants. But that's going to lead people toward fleet setups which may do well initially, then get slaughtered when they venture away from the Core or against harder faction fleets, pretty much the definition of "newbie trap". I prefer to make recommendations based on how the ships fare against difficult fights, knowing that if they do well in those fights, they will almost certainly do well against easier fights.

This is basically the longer explanation for when I said earlier that a 2-Ordos is used as a test benchmark because I know a fleet that can defeat it can "punch down" against easier fleets, but if I use an easier fleet as a benchmark, it's hard to know if my fleet and the ships therein can "punch up" against more difficult fights. That's why the 2-Ordos fleet is relevant as a test benchmark.

IMO, any strategy that can beat all of the challenges in the game is pretty much indistinguishable from the others. Even differences in supply cost or money are pretty much irrelevant by the end of the game.

I'm not quite sure where you're going with this statement. My whole purpose for using a 2-Ordos fleet as a benchmark is that if my fleet can beat it, my fleet can beat almost every other fleet in the game (except for a couple of very specific fights) -- yet you say it's not a good benchmark, and then say this.

And if strategies that can beat the challenges are "pretty much indistinguishable from the others", well, why bother with comparisons? All comparisons need to be comparing against something, i.e. there is some sort of performance metric or figure of merit that's used to evaluate the value of different alternatives. Thus far you've said that supplies don't matter (i.e. it doesn't matter that the Hyperion, due to SO, uses up a lot of CR and thus needs a lot of supplies to recover afterward), DPS doesn't matter, how quickly a strategy kills doesn't matter, combat statistics don't matter, or at it least it mattered before when you yourself justified the Hyperion based on the combat statistics mod, but now that I can put up better stats using other ships, now maybe the damage stats from the mod aren't really that representative of effectiveness, or something. Then what's the purpose of a comparison when thus far you've said that pretty much everything used as a basis for comparison doesn't matter?

[attachment deleted by admin]
Logged

intrinsic_parity

  • Admiral
  • *****
  • Posts: 3071
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #91 on: December 22, 2021, 08:33:57 AM »

The last reply in this thread is more than two months old, and a significant number of the ships and weapons used in the original examples got rebalanced in the last patch.... Why do you still want to argue about things that aren't even in the game in the state that generated the initial discussion????

FWIW, I still don't agree with a bunch of stuff you said, but there's literally no point in arguing about which strategy from the last patch is better. LMFAO
Logged

Vanshilar

  • Admiral
  • *****
  • Posts: 585
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #92 on: December 22, 2021, 08:53:59 AM »

The last reply in this thread is more than two months old, and a significant number of the ships and weapons used in the original examples got rebalanced in the last patch.... Why do you still want to argue about things that aren't even in the game in the state that generated the initial discussion????

Eh most of the points brought up were about the general methodology of evaluating between different ships and have little to do with the particular game version. Use of a 2-Ordos test fleet for example, or whether or not damage per second or the combat statistics mod are useful metrics. Unless you feel like a new game patch suddenly means those points no longer apply for some reason.

Thus far I haven't seen anything in the patch that indicate any drastic changes, although I guess it depends on if the Hyperion TP cooldown changes the AI significantly. I won't be able to re-test the fleet until January due to RL but I don't foresee having to make big changes either. (Biggest unknown is any AI changes that I'd have to account for in the ship loadouts.)
Logged

intrinsic_parity

  • Admiral
  • *****
  • Posts: 3071
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #93 on: December 22, 2021, 09:19:09 AM »

Well... I still would argue that no metric represents performance on its own, and there are tons of way to get value that don't reflect in the particular damage metrics you've chosen. I still would argue that Hyperion is uniquely able to safely TP behind enemies to distract them without putting itself in danger (like other ships would be on such extreme flanks), and I also would still claim that there's no point in arguing about which of two strategies are better, if both strategies can achieve all the same things and the only differences are incremental changes in time or supplies....

Also, you fury strat should be at the least 33% worse, since the Fury's DP got nerfed by 33%.... not to mention, that the cryoblaster and xyphos also got nerfed. I imagine it still works, but I don't think it's anywhere close to as 'optimal' now as you were saying it was then.

Hyperion got railed by AI changes. The AI is very over-cautious with it now, and I'm also pretty sure the changes to wolfpack (%ppt vs flat PPT) hurt SO builds quite a bit. As a player ship, it still seems decent but the PPT is still a bigger issue now IMO, and other stuff has gotten quite a bit better while it's fallen off.

So yes, I think a lot of stuff has changed that makes the original argument about whether furies or hyperions were better almost completely moot.
Logged

Vanshilar

  • Admiral
  • *****
  • Posts: 585
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #94 on: December 23, 2021, 09:28:35 PM »

Well... I still would argue that no metric represents performance on its own, and there are tons of way to get value that don't reflect in the particular damage metrics you've chosen. I still would argue that Hyperion is uniquely able to safely TP behind enemies to distract them without putting itself in danger (like other ships would be on such extreme flanks), and I also would still claim that there's no point in arguing about which of two strategies are better, if both strategies can achieve all the same things and the only differences are incremental changes in time or supplies....

You keep repeating the same assertions without supporting them nor addressing the points in any way. I've already addressed all of these earlier.

Also, you fury strat should be at the least 33% worse, since the Fury's DP got nerfed by 33%.... not to mention, that the cryoblaster and xyphos also got nerfed. I imagine it still works, but I don't think it's anywhere close to as 'optimal' now as you were saying it was then.

Fury was already set to 20 DP instead of 15 DP for the purposes of comparison, which you yourself mentioned earlier. All ships (Odyssey, Fury, Apogee, Hyperion) used 1 Cryoblaster so the nerf affects all of them fairly equally. (It may affect the Fury somewhat more since it has no other means of anti-armor/hull, whereas all the others had at least 1 other anti-armor/hull weapon.) Xyphos nerf just means losing 600-800 flux capacity for the ships that use it, out of ~25k for the Odyssey and ~13k for the Fury, hardly a major change.

Hyperion got railed by AI changes. The AI is very over-cautious with it now, and I'm also pretty sure the changes to wolfpack (%ppt vs flat PPT) hurt SO builds quite a bit. As a player ship, it still seems decent but the PPT is still a bigger issue now IMO, and other stuff has gotten quite a bit better while it's fallen off.

If anything I find the Hyperion more willing to stick around longer in 0.95.1a than before, which is an improvement. Previously as I stated above, the main issue with the Hyperion is that it'll teleport away unnecessarily at low flux, and it'll stay away from the fight hanging out at 0 flux unnecessarily as well. Now it'll actually let its flux ride up a bit before teleporting out which is a good thing.

Change to Wolfpack's PPT bonus just means a 20-second difference for the Hyperion, in a fight that lasts for roughly 5 minutes. You were the one saying the time doesn't matter anyway so this shouldn't be an issue.
Logged

intrinsic_parity

  • Admiral
  • *****
  • Posts: 3071
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #95 on: December 23, 2021, 11:05:18 PM »

I really can't comprehend how you can sit here and argue that nothing has changed and previous results are valid..... the fleets you are testing against are not even the same (new skills change AI core performance, radiant DP nerf etc.), the officers you were using are not the same (skill changes), there were relevant hullmod adjustments (hardened shield nerf), relevant weapon adjustments, relevant ship adjustments etc. Many of the changes do affect all of the ships tested, but that doesn't mean they affected them all to the same extent... When you change the parameters of a scenario, you have to re-test, you can't just claim that everything is the same with no evidence. Based on my experimentation on the patch so far, many things play very differently, and I have no reason to believe this would not also play differently.

Fury was already set to 20 DP instead of 15 DP for the purposes of comparison
No, DP was not changed for previous testing. I merely pointed out that the damage metrics for the ships that I was seeing (in my inexact testing) were very similar per DP if you arbitrarily changed the DP values when analyzing. That does not translate to actually adjusting the DP and testing. I did not adjust the DP values of furies, test them against multiple ordos and present results.

If anything I find the Hyperion more willing to stick around longer in 0.95.1a than before, which is an improvement. Previously as I stated above, the main issue with the Hyperion is that it'll teleport away unnecessarily at low flux, and it'll stay away from the fight hanging out at 0 flux unnecessarily as well. Now it'll actually let its flux ride up a bit before teleporting out which is a good thing.
I have directly observed (via damage metrics) that my AI hyperions with the same loadouts consistently do less damage this patch than they did previously (something like .5-.66 times the damage based on my very non-scientific observations) , so I have no idea what you think improved, but it has not improved for me, and in fact has worsened enough that I no longer use AI hyperions much.

Change to Wolfpack's PPT bonus just means a 20-second difference for the Hyperion, in a fight that lasts for roughly 5 minutes. You were the one saying the time doesn't matter anyway so this shouldn't be an issue.
As to the wolfpack PPT change, 20s is literally 20% of SO hyperions PPT (100s now with the change). Hyperions do not last 5 minutes, they need to do tons of damage very fast to get enough value to justify their use. They were able to do that last patch, but no longer do that in my experience. The fact that you think 20s is irrelevant indicates to me that you may not have actually experimented much with SO hyperions.

You keep repeating the same assertions without supporting them nor addressing the points in any way. I've already addressed all of these earlier.
All you did was make your own unfounded assertions about how going behind enemy lines is bad (with no actual evidence, just theory crafting), when I have direct personal experience that it works very well...  At the end of the day, I don't really care if you don't agree with me. I value my personal experience and testing over your theory crafting.

Both of us have just been spouting off massive amounts of theoretical arguments with zero actual evidence (videos, extensive sim testing etc.) on either side, so it's hilarious that you accuse me of not presenting any evidence as if you haven't been doing it the entire time.

It's obvious you're going to keep arguing every point I make forever so I'm going to stop responding after this.
Logged

JUDGE! slowpersun

  • Admiral
  • *****
  • Posts: 614
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #96 on: December 23, 2021, 11:55:55 PM »

Well, I totally missed this gem!  I'll only comment on OP, don't really have a reason to wade through 7 pages of agree/disagree.  It is hilarious and nicely cogent, especially for someone who has stated that English not exactly their strongest suit since second language I guess (or just fluent in Google Translate, but it works).  They don't have many posts, and haven't been on in nearly 2 months (hope they doing OK), but compared to most self-proclaimed noobs, just a breath of fresh air!
Logged
I wasn't always a Judge...

Vanshilar

  • Admiral
  • *****
  • Posts: 585
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #97 on: January 05, 2022, 10:31:32 PM »

I really can't comprehend how you can sit here and argue that nothing has changed and previous results are valid..... the fleets you are testing against are not even the same (new skills change AI core performance, radiant DP nerf etc.), the officers you were using are not the same (skill changes), there were relevant hullmod adjustments (hardened shield nerf), relevant weapon adjustments, relevant ship adjustments etc. Many of the changes do affect all of the ships tested, but that doesn't mean they affected them all to the same extent... When you change the parameters of a scenario, you have to re-test, you can't just claim that everything is the same with no evidence. Based on my experimentation on the patch so far, many things play very differently, and I have no reason to believe this would not also play differently.

No, when parameters change, you look at how the parameter changes may affect the metrics of interest. You don't need to re-test to know that Heavy Needlers are still better than Cryoblasters against shields even though both weapons were adjusted. You don't need to re-test the horsepower of a car just because it's painted a different color. As I mentioned, the changes in the latest update were basically nip and tucks to the ships involved. It means the conclusions will still be broadly correct, even if there may be a few percentage points of change here and there. (An example of a major change that needs re-testing would be phase ships due to the new phase mechanics, but that's not applicable here.)

Besides, most of your objections (Remnant fleets don't matter, damage results don't matter, damage per second doesn't matter, time to kill doesn't matter, supplies don't matter, etc.) have nothing to do with "you need new data due to the update" but are along the lines of "your method of testing and metrics of comparison between ships are not valid", an entirely different animal.

Regardless, testing the Odyssey/Apogee/Hyperion fleet again in 0.95.1a-RC6 still shows pretty similar results as before the update, as attached in the screenshot. I'm still seeing the Hyperion severely under-performing for its DP. It doesn't mean that a fleet shouldn't have a Hyperion or two, since a fleet needs some frigates for anti-frigate use and to help split up the enemy fleet, but claims that Hyperions provide cruiser-levels of damage are overblown.

That does not translate to actually adjusting the DP and testing. I did not adjust the DP values of furies, test them against multiple ordos and present results.

And the difference is, I did. When I said that Fury over-performed relative to the Hyperion, it was based on actually setting the Fury's DP to 20 in ship_data.csv for testing as well as comparing it at 20 DP to Hyperion at 15 DP for analysis. That's the basis for my statements.

If anything I find the Hyperion more willing to stick around longer in 0.95.1a than before, which is an improvement. Previously as I stated above, the main issue with the Hyperion is that it'll teleport away unnecessarily at low flux, and it'll stay away from the fight hanging out at 0 flux unnecessarily as well. Now it'll actually let its flux ride up a bit before teleporting out which is a good thing.
I have directly observed (via damage metrics) that my AI hyperions with the same loadouts consistently do less damage this patch than they did previously (something like .5-.66 times the damage based on my very non-scientific observations) , so I have no idea what you think improved, but it has not improved for me, and in fact has worsened enough that I no longer use AI hyperions much.

Nope I'm seeing Hyperions do roughly the same amount of damage; in fact sometimes somewhat more (since each run tends to vary). They generally did around 10-15% of the overall damage pre-update, and post-update they're still doing around 10-15% of the overall damage, though one run got up to 19%. (Since Odyssey is worth 45 DP, Apogee 18 DP, and Hyperion 15 DP, then on a per-DP basis the Odysseys should do 58%, the Apogees 23%, the Hyperions 19% of the overall damage.)

I haven't bothered to record videos of the fights post-update, since it's fairly annoying to set up to record for me, but an example of such a fight from pre-update (from 0.95a-RC15) can be found here: https://www.youtube.com/watch?v=eUj-ggxn5Jk

As stated previously, the setup is to capture objectives to deploy the whole fleet, and have the fleet gather up. At that point, "Full Assault" mode is activated, and from there on out, no more player commands are given; my ships act autonomously under pure AI control, until the 2-Ordos enemy fleet is destroyed. All ship officers are aggressive, so there is no difference in personality. So the fight progresses with aggressive officers under "Full Assault".

Even so, the Hyperions can plainly be seen staying out of the fight for the most part. They engage enemy ships only about 15% of the time, meaning for any given Hyperion, around 85% of the time it's staying away from the fight, neither attacking enemy ships nor attracting enemy fire. In fact there are long stretches of time where all 3 Hyperions are just staying on the back lines (such as around 4:15 to 5:00, 5:14 to 6:40, 7:13 to 7:45, 7:50 to 8:24, and many others if I bothered to continue to keep track). Most of the time they just go in for brief bursts, fire for a few seconds, and then TP away at low flux (such as at Hyperion 1 at 5:00, Hyperion 1 at 6:40, Hyperion 2 at 7:08, Hyperion 2 at 7:45, Hyperion 2 at 8:35, and many others if I bothered to continue to keep track). They don't really use much of their available flux capacity at all.

The video was recorded in November; I made multiple such videos in October and November as a more general commentary on AI issues in Starsector (front line ships running forward while back line ships staying away, omni shields facing wrong direction, firing on farther away ships that won't get killed while ignoring closer ships that are overloaded or close to dying, inability to move around hulks, plasma burning into hulks then flaming out and zooming into enemy fleet, venting next to enemy ships, etc.), but didn't have time to review them until December, when the update hit.

Chances are it'll be some time before I record more such videos, but suffice it to say, I still see the same AI issues in 0.95.1a-RC6. For Hyperions, though, even though they still stay away a lot, they're at least letting their flux go up to over halfway before TP'ing out a fair amount of time now. So they sometimes get a higher overall damage than pre-patch, even though overall it's still pretty similarly low most of the time. They don't put out cruiser-levels of damage either pre-update nor post-update.

Both of us have just been spouting off massive amounts of theoretical arguments with zero actual evidence (videos, extensive sim testing etc.) on either side, so it's hilarious that you accuse me of not presenting any evidence as if you haven't been doing it the entire time.

That's projection. You've yet to give a single shred of evidence to back up your claims this entire thread. For example, you made a blanket statement -- that Hyperions put out more than cruiser-levels of damage -- without posting a single screenshot, details about fleet composition (which cruisers were it being compared against? What loadouts did they use? How was the fleet structured?), enemy fleet(s) it was tested against, testing methodology, rationale behind testing methodology, performance metrics, quantitative results of those metrics, and so forth. I've provided all this and more for two very different fleets (player-piloted Doom with Furies and SO Hyperions, 3 Odysseys/Apogees/Hyperions under pure AI control), all of which you summarily rejected, without providing any similar documentation of your own, nor any contrary evidence whatsoever.

You make repeated references to Hyperions doing a lot of damage based on the combat statistics mod, without ever posting any screenshots of it, yet when presented with actual screenshots showing the opposite, it became "well, actually, it's not that useful to gauge effectiveness".

Whereas you could have chosen instead to post your own screenshots of your combat results, then it would have been a much more illuminating discussion. For example, maybe you chose different enemy fleets to fight. Perhaps the Hyperion does much better against faction fleets or pirate fleets, which aren't as "threatening" as Remnant fleets, and thus the AI is more willing to jump in and fight, resulting in more damage. That would be a useful nugget of information, that for easier battles it'll help you "win faster" but that you'll need something more "heavy" in damage-dealing for endgame fleets. (Sort of like how frag weapons are good against low-armor ships but need something else to get rid of armor first against higher-armor ships.) Or maybe the cruisers you were comparing it against were slower (since I use high-mobility ships -- Odyssey, Fury, SO Apogee) thus the Hyperions could jump in early and get more time to do damage by comparison (while the cruisers were busy catching up). That would be useful to know too.

Instead it's just been a blanket rejection of anything to the contrary, and the only evidence put forth is "I see it in combat results (which are never shown), I see it with my own eyes, it's my personal experience" which on an internet forum basically amounts to "because I say so".

It's obvious you're going to keep arguing every point I make forever so I'm going to stop responding after this.

Perhaps it's for the better, since after multiple posts I've yet to see any credible objection and it's pretty obvious that different posters have different standards of logical consistency and evidence behind what they post. Realistically at this point I've moved on to other fleet setups and other topics anyway, so there wasn't much need to further revisit this beyond verifying that it still held true post-update.

[attachment deleted by admin]
Logged

TontonBoo

  • Ensign
  • *
  • Posts: 41
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #98 on: January 06, 2022, 05:12:34 AM »

Not much value all around since it's a very personal taste and all that but :

Lashers being the worst is just a big no-no. For sentimental reasons probably. You can smash an Onslaught with half a dozen of these babies. That's 24 deployment. For a big bastard like the Onslaught. Individually weak, sure. But in a fleet and en masse they can transcend their on-paper stats. Like most ships do in fact.

That's why such a list - thumbs up it's really well done by the way - is at best very subjective and depends a lot on your way of doing things. And works only in a vacuum.
Logged

Serenitis

  • Admiral
  • *****
  • Posts: 1458
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #99 on: January 06, 2022, 05:48:54 AM »

I really appreciate the effort people put into these teir lists, even if they're literally meaningless for anyone other than thier creator.

The main thing to understand is, in order to be a good space captain you need to know how to hold a good grudge.
That might mean not doing business with a specific faction, not doing this one common thing because "reasons", or just outright refusing to use certain ship types because they're "cursed" (or you just don't like they way they look).
So I might disagree with many things said in this list, but I still enjoyed reading it.
Logged

TericAthnos

  • Ensign
  • *
  • Posts: 4
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #100 on: September 25, 2022, 11:56:20 PM »


- Hyperion? In S-TIER?! Are you out of your mind?! That pile of garbage is B-tier at most. Yes, it has impressive, destroyer-grade flux stats and weapon mounts, but almost the logistics of a freaking capital ship, the most pathetic peak performance time in the game, and on top of it all, is made out of paper. Its ship system is pretty useless now for anything else than quickly closing the distance, and/or instantly putting itself in a dangerous spot where it cannot easily escape from and it dies. I don't see any practical use for it. The amount of nerfing this ship has received between the updates is absolutely comical.

Hyperion is absolutely A tier, possibly S when you stack all the relevant skills on top of it. If you put SO on it, it can use its teleport at will. With shield upgrades/skills and can take a surprising amount of damage and getting behind bigger ships to unload Heavy Blasters or Ion Pulsers is just too easy. Personally, I put a Heavy MG in the universal slot to bring down shields but you can do a lot with that mount. I like it without SO but getting out of situations is trickier because of the need for 0-flux boost to teleport. Other than that, it’s ridiculously good and easily worth the DP cost.

SO on a ship which already has the lowest PPT in the game and a tremendous CR per deployment cost? I'm surprised that thing doesn't start falling apart before it even gets to shoot at anything...

I beg your pardon? Do you actually used Hyperion? I know many players got repelled by x2 maintenance and ppt, i was one of them. But Hyperion really worth that much, it's a cruiser punch in a frigate body. With wolfpack and skilled officer it rips space and time on the battlefield. Clearly it's an SSS+ tier and 100% pick in frigate tactics, and it even good choice for admiral without red skills because how convenient piloting it is. PPT boosted like crazy also with said officers and wolfpack, and with reliable subsystems.
Scarab also sss+ tier just because 18 no cooldown sabots build exist.

Yes, the whole reason why I'm posting this is how many times I've been coming back to this frigate, and every single time the game kept reminding me how it's absolutely not worth it to touch it with a 10 meter stick.

Nope, It's not a cruiser punch in a frigate body lol. It's a destroyer punch with a survivability of a kite, a cost of (almost) a battlecruiser, cosplaying as a frigate, bundled with a gimmick system.

Being able to teleport behind enemy ship and blast it with ion pulsers sounds amazing, until you realize low tech ships aren't the only enemies, and omni shields exist. And if you REALLY want to do that, just go grab a Harbinger. That one doesn't even have to get behind the opponent to disable it.

There is nothing this frigate can do which another ship in the game cannot already do for a fraction of the cost (and can be deployed more than once). The fact that it needs officers and specific character skills to function just adds insult to the injury, and only proves my point.

Pretty much the only things Hyperion has going for are the teleporter (which gets more and more crippled with every update, and - again - right now is most useful for jumping right into a spot where Hyperion will die the quickest), and its flux stats, which indeed, are VERY impressive, but are still nowhere near a justification for its absolutely monstrous costs.

It's a definition of impracticality. It's kinda like a Hummer in real life - it's pretty, it's loud, expensive, burns a TON of fuel and unless you're going to use it to tow a freight locomotive, you really would rather drive any common city car for every day routine.

Hence, I'd place it at B rank at most.

If you actually have a basic braincells on how to pilot ships at starsector then you will realize that even in S tier hyperion is still an underrated ship

provided with support it could easily face to face against a capital ship (remember the sinking the bismar mission)
Logged

TericAthnos

  • Ensign
  • *
  • Posts: 4
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #101 on: September 26, 2022, 12:00:44 AM »

Hyperion was a under the classification of a super ship meaning it performs better or maybe a little bit overpowered (Im looking at you doom) if it was piloted by an experienced player
Logged

CaptSaltee

  • Ensign
  • *
  • Posts: 1
    • View Profile
Re: A Noob's Insight on: Ships!
« Reply #102 on: October 02, 2022, 06:00:07 AM »

Two things I want to notify readers about.

Hey,
1st post here. I just wanted to say that I actually signed up and created an account just to reply to you. Thanks for taking the time to put all this together.
I've had this game sitting on the backburner for a while but have always been on the fence about buying it. It really wasn't until now I've decided picking it up. I don't have vast amounts of time to play games, and when I do it's always nice to have a sort of jumping off point to help me along. Looking forward to your weapons ranking.
Logged
Pages: 1 ... 5 6 [7]