[3.14] Flame Golem 101 - Complete Guide To Golemancy, ✔️All Bosses Down, ✔️Necro, ✔️Elementalist
what spectres can i use?
solar guards ? i have a helm whith minion damage and burning damage | |
" Hey :) "Attacks with this Weapon Penetrate 30% Elemental Resistances" mean that only attack done with the one wielding the sword will penetrate 30% of resistances. This is not global debuff like EE and will only affect Animate Guardians damage. Its really going to waste on AG. " " Thanks for support :) " " " Hello :) I'm switching between my standard lvl 100 char with default links and gear and league one with Golems and Specs. I really like both and they play pretty differently. Standard gear gets much higher EHP with using Vertex and Rainbowstrides. The automated links are also great for fast clear 1-button without worrying about anything. Golem + Spec has lot more damage and defense with max block. Total clear speed however is much slower as minions have no charges and I have to cast Storm Burst to raise Offering up before more dangerous packs as CWDTs are at bare minimum. With +1 level Bones of Ulr and some extra curse effectiveness, this version also gets max level Temp Chains. Standard build is better for fast clearing overall and has more passive playstyle. Playing it is more mobile as its based on kiting and Phase Run keeping aggro down. Golem + Spec has massively more single target damage + defense but at the cost of slower clear speed and bit lower EHP. Playing it is more stationary as you keep casting Storm Burst and keep max block up. The AG curse should work if you have taken extra curse node for yourself and equipped curse on hit gloves for AG. It can be kinda hard to see but theres no reason why it would not work. SotL is as good as rest of your gear is. If you are not using lots of high ES items, the 100% increased ES wont give huge amounts of ES. However if you have lots of base ES to multiply, you end with as much ES than basic rare ES armor and + levels for Golems. With Specters the damage from Golems is bit less relevant so if you find some huge ES 6l you can try it. I really like SotL as its potentially cheap and has best balance between DPS and defense. I'd say that if you are good at kiting, like lazy play-style with fast clear-speed the default set is better. Id you focus heavily of Abyss rifts and high tier bosses, hate dying and want to end boss fights asap at cost of some clear speed the test layout (Golems + Specs) is better. " Hi :) The Hungry Loop is pretty much a trap item. You lose tons of ES and HP while using it and usually the ones interested in it are also using Bones or Ulr and some armor / evasion helmet for Specters, so they have already low EHP. I think it may be better to drop some links totally than lower defenses at cost of more stuff if you are trying to level at all. However I have not personally tested it yet and you might be able to use it with your items setup. I just recommend being cautious about how much you can lower defenses before it affects mapping ;) " Hello and thanks for support ^^ Id say that the second one is much better as the minion damage gem + added minion damage add huge amount of DPS. it has however pretty low ES, so i would max Temp Chains with +1 Bones of Ulr, Curse Effectiveness and run max block chance without Victario's Charity and Necro Aegis. Usually taking Spec with Golems means lower EHP, so you need more active defenses. You can check my league char for links and maybe adapt some of the ideas ;) " Hi :) I'm using Tukohama vanguards. They have best single target DPS and make boss fights as short as possible. Golems clear trash while TVs melt rare abyss mobs and bosses. Solars are bit too close to Flame Golems to really add anything more than total DPS. SGs also have less aggro range than FGs so, FGs have most likely killed trash mobs at the times Spec AI sees them. [3.12] Flame Golem 101 - Complete Guide To Golemancy, All Bosses Down, Necro, Elementalist https://tinyurl.com/y8k9fa2f Dernière édition par mika2salo, le 7 janv. 2018 11:53:33
| |
I have a question but first wanted to say what a great guide and what great support of the guide. Thanks for all your hard work as well as your partners in crime.
My question is: looking at your current build and tree I don't see how you avoid getting stunned every 2 seconds. Is it you just don't care if you get stunned as it doesn't really matter as spectres and golems kill everything and your energy shield is high enough to absorb any damage you take while stunned or does your block chance keep you from getting stunned very often? As an aside, I tried slower projectiles in place of GMP and hated it but using both... that might be something I try. Keep up the great work! |
|
Alright so i hit it with some more chaos spam and hit this still low es sadly but i think i will stick with it now to try to get an enchant on it sucks every time u enchant it lose the spectres so i will wear the other for now I did see you doing that storm burst i tried it before i had the rumi's maybe i will try it again but for now till i get my enchant i will run like this
| |
How's mapping with the build and if u go the elementalist path, Liege of the Primordial, Paragon of Calamity, and what for the last 2 ascendancies O.O
Been looking into this build as the build I'll end the league with and just wanted some ideas xd | |
"Although your testing fits for a charge-based system, I don't believe it necessarily eliminates simple active/inactive states or some other system as a possible explanation. "I believe we're quite sure on this point. If not, it can easily be re-confirmed by timing the cooldown on summoning. "I'm skeptical about an 18-second wait making a difference, as you may see in the subsequent test results. If you feel it's worth doing, let me know and I can re-do the tests later. "I appreciate you laying out extensive lines of testing. However, I'm not sure all of that will be necessary to reach a practical conclusion. As you requested, I re-did the tests using Minion Speed (LVL 21), Faster Casting, and Spell Echo. I kept the test at 96 seconds so we have a baseline for comparison with the previous test (apologies for not making it an even 100s as the first test was just a random cut). I got the following results: Flame-Golem: 71 Magma Balls (previous test) Flame Golem-Minion Speed: 92 Magma Balls Flame Golem-Minion Speed-Faster Casting: 118 Magma Balls Flame Golem-Minion Speed-Faster Casting-Spell Echo: 186 Magma Balls (videos linked, i didn't count too carefully, apologies if the tally is off by more than ±1 or 2) I believe it's safe to assume that cast speed is the determining factor here and that it cannot be 3+1 charge-based system. 186 balls over 96s is over 2 shots/sec. For 3+1 charges, you'd need 1100% cooldown reduction to achieve that rate of fire (74x mostly perfect Harmony jewels). "I can't post one with similar defenses, because Elementalist will build around and slot different defensive tech in place of Bone Offering as I already mentioned. I did post one earlier though with ~150k Magma Ball DPS compared to ~100k on your NEC tree. It also had 500 more EHP. That was not even the extreme-offense version (which can hit 250k+). "Can you show us a Shaper or Uber Atziri video when you're ready? Or of somebody else? TV specs can carry even without golems, but I haven't been able to find a 3.1 example yet (since the TV totem nerf) and I'm curious how big the gap really is. "The last two don't matter too much. Grab the two cast speed nodes for that extra tiny bit of safety if you're a perfectionist. ⚡ Lightning Golems:
/1902593 ⛄ Ice Golems: /1914382 |
|
" It doesn't eliminate another system, but I do believe it eliminates a simple active/inactive state, because then why would it be variable, and gain more volleys after waiting the full 18 second cooldown? " This would confirm that it's working on the player, but that doesn't mean GGG didn't mess something up and also apply it to the golems themselves (It's been known to happen) " The 18 second wait is relevant for zero cooldown reduction testing. It's a test for the Charge system (or at least something that acts like it), where you wait for the expected recharge time of 3 charges. At zero cooldown reduction this is 18 seconds according to PoB. It's only a useful test when you notice the golems settle into a single ball every x seconds pattern after they fire their initial volley. Your tests at very high cooldown reduction would not benefit from this because they essentially never stop firing. " Your number count for spell echo, was it actually 372 magma balls total, but 186 casts? I shouldn't have suggested spell echo for testing, because the double cast might take longer than a single regular cast, so it makes the testing weird. With a charge system, spell echo also only uses one charge. I also don't believe that only testing at these extremely high cooldowns can lead us to a proper conclusion. We also need tests at zero cooldown, and median cooldown. This way we can find where cooldown is outstripped by cast speed. If cast speed at no cooldown gets significantly more Magma Balls in the same time frame, then we absolutely know something is up. Your tests at high cooldown could mean that your golems are gaining much more cooldown than we previously thought, or it could mean as you suggest, that the cast speed is somehow circumventing whatever it is the cooldown does, whether it be charges or some other method. Dernière édition par Shiverwarp, le 8 janv. 2018 01:38:20
|
|
"My apologies as I hadn't looked carefully enough at your 2nd gif. Interesting. Hmm... Could it be that for the first volley, the active frame timer doesn't begin until the first shot is active? And that the initial cast time is carried over and counted against active frame time in subsequent volleys leading to fewer shots fired? Actually no, nevermind. That could explain 2 shots dropping to only 1, but not 3 to 1. Hmm... might be worth asking on the Mechanics thread? Although hairier stuff like this tends to get ignored a lot :/ (edit: went ahead and posted the question) "I'm not really comfortable bringing bugs into the discussion just yet, otherwise it'll be difficult to come to any sort of conclusion. "You're right. In my rush to just get it done, I overlooked this. So then it was only 98 shots, which is less than the 3-link setup. Hmm... I could see being able to fit more individual shots than pairs into a certain frame of time as a rough explanation. But then I don't know how to reconcile that with your gif. At any rate, the non-Spell Echo test at 118 shots over 96s would still require around 640% cooldown reduction, a number that isn't possible even with a full 120 skill points and assuming GGG mixed up the stats on Harmony. "I honestly don't have the appetite for that much testing today. I'll have another look at it next weekend or so if you haven't done the testing yourself by then. Thanks for the error-checking. Until next time... :) P.S. What MTX are you using for your golems? I didn't see anything in the shop? ⚡ Lightning Golems: /1902593 ⛄ Ice Golems: /1914382 Dernière édition par tomatopotato, le 8 janv. 2018 02:31:58
|
|
I just did few tests as well, as a data point for you guys, but then I got into it.
15 balls in 20 seconds, with just flame golem 21, not using a +1 chest, no other links. 26 balls in 20 seconds, only adding spell echo. I have 8 perfect harmonies and 3 perfect eminence. 15 x 8 = 120 CDR, 3 x 20 = 60 Faster I have a cooldown of 2.73 seconds, and a cast speed of 2.96/sec (with the spell echo in) 6 casts should take 2.02 seconds (3 charges) The math for 20 seconds, with it starting on cooldown at the start of the first cast, would get me 43 balls. Too high. The math for 20 seconds, with it starting on cooldown at the end of the first cast instead, would add .338 seconds to each cycle, would get me 39 balls in 20 seconds. The math for 20 seconds, with it starting on cooldown after 6 casts, 2.02 seconds, would get me 25 balls. The math for 20 seconds, with it starting on cooldown after 5 casts, 1.69 seconds, would get me 26 balls. So I did a longer test. 120 seconds, 164 casts counted. starting on cooldown after 6 casts, 2.02 seconds + 2.73 seconds per cycle, 152 balls starting on cooldown after 5 casts, 1.69 seconds + 2.73 seconds per cycle, 163 balls. This is the pretty accurate. Final test, added faster casting lvl 1 (20%) got 164 ba;;s in 120 seconds again. So cast speed isn't increasing my BPM (balls per minute) while I'm CDR capped. That being said, at no point did my golems stop shooting for 2.73 seconds in duration. They do however, seem to bunch up, then spread out again. I assume that's the difference between the 2.73 CDR and the 2.02 seconds to cast 6 balls causing it to be uneven. Also, as others have pointed out, they don't shotgun in bunches of 6, it didn't shoot 6 times, then wait 2.73 seconds, then shoot 6 more times. It shot 8 times with no noticeable delay, then paused, then within less than a second, shot another pair. It then continues to shoot regularly, with slight delays between waves. If they were recharging individually at 1/3rd of the 6 seconds, I should be able to shotgun into 10-12 before it starts pausing. If all the charges came back at 2.73 seconds, I wouldn't see the 7th and 8th shots without a pause first. So it's definitely recharged once before the 2.02 second mark, but still hasn't recharged a second time before the 2.70 second mark (two more casts duration), but then fires sub 1 second dual balls after that. I don't have a theory to explain what is happening with the charges, but it seems like 60 / (5 casts duration + CDR time) * 6 works out to an accurate number of balls per minute at least as long as your casts cast duration is less than your cdr. | |
" Looking at that thread, I don't think I'll bother. It seems to be for more well explored things than this. As far as I've seen, this thread is the only place people have really delved into the mechanics. " It's less about confirming a bug, and more about confirming what precise effects the Harmony jewels have on golems. We know more harmonies = more magma balls, but the interactions may be more complicated than that. " Is it somehow possible that the CDR is multiplicative? The fact that faster cast rate would circumvent the cooldown seems a bit bizarre to me. The logical way for it to work is for cooldown to set the upper limit of magma balls, and then more cast rate will let you actually reach that upper limit faster (or at all) " Yes sadly I already deleted my character or I would be testing it myself! Don't really have the time level up another to test, so thanks very much for the work you've already done. Really these tests are about trying to find the optimal cooldown vs cast rate setup, so that we can optimize DPS. If cooldown is not as important as we previously thought, we might be able to get quite a bit more damage. Dernière édition par Shiverwarp, le 8 janv. 2018 02:32:34
|
|