Lanterns still firing far slower than advertised [With lantern/tower/trap speed tests]

[Azzy]

Just tested with a grade 4 gem, max lantern skill, tooltip shows gem firing at exactly 1 shot per second. Left 1 enemy alive, froze it in lantern range, and timed it for a minute. 60 seconds should give 60 hits. Only 31 hits. Attack speed of -52%, which in a regular tower is 2.08 shots per second, do display is correct.

Upgraded to 30.22 shots per second, over a 60 second period - 845 hits, opposed to the 1,813 it should have done.

Upgraded to 60.6 shots per second, over a 60 second period - 851 hits, opposed to the 3,636 it should have done.

Upgraded to 102.23 shots per second, over a 60 second period - 840 hits, opposed to the 6,133 it should have done.

So, not only are they initially firing at half speed (still), they are also capped at 30 shots per second a someone else said n a different thread. Also, tested it out with 1x speed. When done at 3x speed, 60 seconds of real time (so 180 seconds game time), still only generated 846 hits.

[ruphus]

They have attack speed cap, which haven't displayed properly yet.
Also yes the fire rate does get 'slower' at faster speed since again theres a hard limit of how many shots they can do in one frame.

[Azzy]

Even if a max speed cap made sense (which it doesn't, why would it?), still leaves lower speeds being half as effective as they should be.

[pmnoxx]

It would be great how have an option to reduce speed of all towers, by X, and increase damage they do X times, where X you could control from settings. this would help with performance late game, where you have turrets firing a few thousands of times per second and bringing FPS down to 1 ;/

I mean both damage and mana leech.

[Marosh]

Nice test!

I didn't do a test that specific, I only used my usual setup in my end run. With the reduced max targets and new target setting, my bleed lantern couldn't really keep up with the monster stream while my trap was still fine (x3 game speed). And that didn't make sense.

We assume the 30 speed cap and a stream of mobs.

30 x "Target limit" = max hits per second
30 x 6 = 180 max hits per second. This is with 0 points in lantern.

A Tower is speedcapped at 120 attacks per second. I think, traps are also capped at 120. So in this category, lanterns win even with the target reduction, because the 6 target limit is the base. If a trap is capped at 240 (who knows, I don't), you would need 10 points in lantern. With a endgame talisman, you have already +25 in lantern,

Maybe the range + attackspeed is the problem. For a trap, all 120 hits are concentrated on it's one own tile. Because of the 30 attack speed of the lantern, maybe some fast mobs slipped through. I test it out today. Maybe my setup was just aweful.

[Marosh]

Originally posted by Azzy:

So, not only are they initially firing at half speed (still), they are also capped at 30 shots per second a someone else said n a different thread. Also, tested it out with 1x speed. When done at 3x speed, 60 seconds of real time (so 180 seconds game time), still only generated 846 hits.

I've realised, that confirmes the FPS-hit-bug.

[Gulch]

Originally posted by Azzy:

Just tested with a grade 4 gem, max lantern skill, tooltip shows gem firing at exactly 1 shot per second. Left 1 enemy alive, froze it in lantern range, and timed it for a minute. 60 seconds should give 60 hits. Only 31 hits. Attack speed of -52%, which in a regular tower is 2.08 shots per second, do display is correct.

Upgraded to 30.22 shots per second, over a 60 second period - 845 hits, opposed to the 1,813 it should have done.

Upgraded to 60.6 shots per second, over a 60 second period - 851 hits, opposed to the 3,636 it should have done.

Upgraded to 102.23 shots per second, over a 60 second period - 840 hits, opposed to the 6,133 it should have done.

So, not only are they initially firing at half speed (still), they are also capped at 30 shots per second a someone else said n a different thread. Also, tested it out with 1x speed. When done at 3x speed, 60 seconds of real time (so 180 seconds game time), still only generated 846 hits.

Thank you for testing it!

[pmnoxx]

Can you check whenever this affects traps as well?

[Marosh]

Originally posted by pmnoxx:

Can you check whenever this affects traps as well?

That the hits are tied to framerate? No. Not affected. Only lanterns.
That the tooltip shows way higher numbers? Yes. Traps are capped at 180shots per second. Also towers.

[Azzy]

So, testing with tower:

Speed: 30.2. Hits: 905. Expecting: 1,812
Speed: 60.4. Hits: 1549. Expecting: 3,624
Speed: 120.5. Hits: 3,088. Expecting: 7,230
Speed: 150.3. Hits: 3,595. Expecting: 9,019
Speed: 180.6. Hits: 4,194. Expecting: 10,812
Speed: 213.1. Hits: 4,116. Expecting: 12,786
Speed: 256.2. Hits: Accidentally killed test mob :<
Speed: 280.9. Hits: 4,116. Expecting: 16,854

Key notes with tower tests: They don't register hits until it actually hits the monster, so have to unsocket gem whilst paused and then let the remaining shots carry on o find out how many it actually shot. All tower tests done with mana leech gems, using beam with reduced range to autosocket for better timing.

Speed cap around the 180/second mark.


Testing with traps:

Speed: 3.2. Hits: 89. Expecting: 192
Speed: 15.3. Hits: 327. Expecting: 918
Speed: 30.1. Hits: 458. Expecting: 1,806
Speed: 60.6. Hits: 919. Expecting: 3,636
Speed: 90.5. Hits: 1980. Expecting: 5,430
Speed: 120.3. Hits: 2,730. Expecting: 7,218
Speed: 180.2. Hits: 2,733. Expecting: 10,812
Speed: 213.2. Hits: 2,772. Expecting: 12,792
Speed: 490.1. Hits: 2,775. Expecting: 29,406

Key notes with trap tests: no idea wtf is happening around the 15-60 shot per second region, those are WAY lower than expected. Seems to dip right down then come back up. The 3.2 speed was as standard. Tested traps multiple times because of this. All trap tests done with a fresh mana leech gem in the trap, using barrage (only increases special) to autosocket for better timing.

Definitely a speed cap around 120/second, as Blauer said.


Another confirmation I gleamed from all these: actual shots are also tied to your fps performance, mine was around 15fps, which explains why most of the tests are roughly half of the expected values. Towers were lowering fps to 13ish on the higher speeds, though doesn't completely account for the drop from expected.

Also confirms why mana farms drop off when reaching grade 30 or so, since the attack speed increases don't have any effect, and it's purely the mana leech part that is increasing the farm. Also why killgems drop off in effectiveness as well.

In conclusion: Lanterns are capped at 1 shot per frame/30 per second, traps at 4 per frame/120 per second, and tower at 6 per frame/180 per second.

So, due to special increases traps will always reign supreme for mana farms.

Applying debuffs, unless you are letting mobs roam the entire screen (not advised), lanterns will hit up to 180 monsters per second, if hitting 6 per shot (this increases with skill points), and gives the same strength debuff as traps and towers. Towers can shoot 180 per second, reach the whole map, but you shouldn't need to do so. Therefore, without any points in lanterns, they are on par with towers. However as you add points into lanterns (which you get passively from talisman fragments), they will outperform towers for applying debuffs.

Now, the main one, killgems: Do not use lanterns to kill. Unless it is early wave swarmlings, they are not good for killing. So, tower or trap. This all comes down to skill point availability. However, without going into it completely, if a gem deals 50% more damage in a trap than it does a tower, it is better in a trap. Of course that is a very basic evaluation that doesn't take into consideration armour reductions, the need for range etc. The ability to boost traps and not towers though, leads me to speculate that lategame, traps will be a lot better, despite having a lower firing speed cap.

[Marosh]

Thank you so much for that!

[sresk]

Even at WL 500 my kill gems in traps where better than in towers and at wl 1500 they are substantially better. The main problems is that every now and then a trap kill gem will leak a monster and at that point you need a second or third trap or a tower. also the traps can get resocketed by monster deaths which isnt a problem for towers.

So I'm finding my best results are a trap kill gem or 2 and one lower level tower (about 2-3 grades lower seems to work fine) with 8 amps for killing bosses and beacons. If anything leaks past the traps then swap the kill gem to the tower and throw a bolt on it until things are under control again.

[Marosh]

Originally posted by sresk:

Even at WL 500 my kill gems in traps where better than in towers and at wl 1500 they are substantially better. The main problems is that every now and then a trap kill gem will leak a monster and at that point you need a second or third trap or a tower. also the traps can get resocketed by monster deaths which isnt a problem for towers.

So I'm finding my best results are a trap kill gem or 2 and one lower level tower (about 2-3 grades lower seems to work fine) with 8 amps for killing bosses and beacons. If anything leaks past the traps then swap the kill gem to the tower and throw a bolt on it until things are under control again.

Try a 3 trap setup. Back to back. That works really well. They don't need to be the same grade. The gem closest to your orb has grade X. The middle gem grade X-1 and the furthest away gem X-2. That doesn't cost that much more and a nice bonus is, you see when you can't handle your current enrage gem. They walk over the first one and get killed by the second and so on. With the hit farm from the lantern analysis, you don't need to worry about the hit multiplier.

[Azzy]

Quick update: just tried traps/towers running at 3x speed.

A tower at speed 60ish performed as expected, hovering around 4,400 hits in the timed 60 seconds.

A trap at speed 60.89 just clocked in 5400 hits (same gem gets 950 hits at 1x speed)

3.2 speed, 235 hits (89 at 1x)
29.4 speed, 1,841 hits (479 at 1x)
60.9 speed, 5,466 hits (954 at 1x)
91.6 speed, 8,532 hits (1,902 at 1x)
121.1 speed, 8,478 hits (2,838 at 1x)
180.5 speed, 8,514 hits (2,838 at 1x)

So, when running at triple speed, once you hit the caps then they work out to roughly the correct hits for each speed mode, however it seems that you hit that cap a LOT earlier if running at 3x speed. It looks like a definite bug, and based on the unusually low hits seen from gems running at 1x speed, I'd say it's not quite calculating lower speed traps properly when running at 1x speed.