|
On May 05 2012 08:26 virgol wrote:
So, to summarize this thread:
Grubby feels like there is not enough A-move potential in Void Rays right now
No, there is not enough potential in Void Rays to be part of an army composition in big fights. As in, the unit gets obsolete too quickly. Just like Phoenices get obsolete, and Carriers are useless by any measuring stick. It means Protoss air has nary a valid position in any army.
The theorising I do is lighthearted and curious. As you could see, I was hypothesising and doing poor one-off experiments. I don't have time to do more... I just wanted to provoke some thought and was interested by the response 
Thank you for your part of the response. xd
|
On May 05 2012 08:33 straycat wrote:Show nested quote +On May 05 2012 06:56 mskaa wrote:On May 05 2012 05:20 BronzeKnee wrote:
Why match up 126 vs 120? The upgrade costs 100-100 which isn't even the cost of a single Void Ray, so why do the unupgraded Void Rays get 6 extra on the side?
... Ehm the cost really isnt the point here... What do you think wins: 100 3/3 marines vs 200 0/0 marines.. But omg 200 marines is 5000 minerals - the 3/3 upgrade cost more.. Why they get so many??? If an upgrade was only worth its actual cost in ONE unit, there would be no reason to get it. Either way you have no idea what your talking about... Im actually shocked you are entertaining us with such a false theory. He means that it does not make sense to test 120 vs 126, since the 6 extra unupgraded is better to have than the upgrades. And so he is confused as to why Grubby had 126. And, one explanation he sees is that Grubby reasoned that the money that did not go to upgrades should go to extra void rays. But, as BK wrote, the upgrade cost does not even warrant one extra void ray, and so the rationale for the 6 extra is still a mystery. Show nested quote +On May 05 2012 06:56 mskaa wrote:
Either way you have no idea what your talking about... Im actually shocked you are entertaining us with such a false theory. Oops. Offensive gg but it turned out you were behind and lost.
The rationale for the 6 extra Void Rays is that with any other unit in the game, one single attack upgrade over such a huge sample size (with just about exactly a 5% different in unit count), would totally tilt the favor into the upgraded unit. What I meant to illustrate, though I did so poorly, is that the upgrade doesn't help VR enough, therefore there is insufficient late game potential to VR, unlike some other units.
If you don't believe me, try attackmoving 126 0:0 marines into 120 1:0 marines.
Or 126 0:0 BC's into 120 1:0 BC's.
Or 126 0:0 Roaches into 120 1:0 Roaches.
|
While testing this i've noticed some other odd behavior by the Voidray charge.
You can see it when you have the Voidray attack a bunch of SCV's. The 2nd charge will start when the 3rd SCV is attacked (1st, 2nd destroyed). This is strange because if it is true that the charge buildup (the buildup to level 2, not the 2nd charge level itself) is reset when the Voidray switches targets, the 2nd charge should start when the 2nd SCV is attacked or it should never charge up to the 2nd stage at all.
As I said in my previous post, the charge buildup does reset when attacking Probes (40hp), but apparently it doesn't when attacking a 45HP unit. Perhaps there is a tiny fraction of charge that bleeds over to the next target and 45HP is just enough to make it work? I dont know, its weird.
(Sorry, bit offtopic  )
|
On May 05 2012 08:34 Grubby wrote:Show nested quote +On May 05 2012 08:26 virgol wrote:
So, to summarize this thread:
Grubby feels like there is not enough A-move potential in Void Rays right now
No, there is not enough potential in Void Rays to be part of an army composition in big fights. As in, the unit gets obsolete too quickly. Just like Phoenices get obsolete, and Carriers are useless by any measuring stick. It means Protoss air has nary a valid position in any army. The theorising I do is lighthearted and curious. As you could see, I was hypothesising and doing poor one-off experiments. I don't have time to do more... I just wanted to provoke some thought and was interested by the response Thank you for your part of the response. xd
I tend to disagree a little. In PvZ, a really popular build going around from White-Ra, col.rsvp, and others is something like a voidray-carrier-HT-mothership-a few gateway units(normally archons for toilets or stalkers if they're going pure infestor corrupter). While using vortex to split the army up the mix of units make it very, very deadly and void rays seem to have a very good little spot in that army comp.
I think in some cases void-rays have their spot in big fights, mostly against zerg.
|
On May 05 2012 08:57 Elp wrote:While testing this i've noticed some other odd behavior by the Voidray charge. You can see it when you have the Voidray attack a bunch of SCV's. The 2nd charge will start when the 3rd SCV is attacked (1st, 2nd destroyed). This is strange because if it is true that the charge buildup (the buildup to level 2, not the 2nd charge level itself) is reset when the Voidray switches targets, the 2nd charge should start when the 2nd SCV is attacked or it should never charge up to the 2nd stage at all. As I said in my previous post, the charge buildup does reset when attacking Probes (40hp), but apparently it doesn't when attacking a 45HP unit. Perhaps there is a tiny fraction of charge that bleeds over to the next target and 45HP is just enough to make it work? I dont know, its weird. (Sorry, bit offtopic )
This is super interesting.
|
On May 05 2012 08:57 Elp wrote:While testing this i've noticed some other odd behavior by the Voidray charge. You can see it when you have the Voidray attack a bunch of SCV's. The 2nd charge will start when the 3rd SCV is attacked (1st, 2nd destroyed). This is strange because if it is true that the charge buildup (the buildup to level 2, not the 2nd charge level itself) is reset when the Voidray switches targets, the 2nd charge should start when the 2nd SCV is attacked or it should never charge up to the 2nd stage at all. As I said in my previous post, the charge buildup does reset when attacking Probes (40hp), but apparently it doesn't when attacking a 45HP unit. Perhaps there is a tiny fraction of charge that bleeds over to the next target and 45HP is just enough to make it work? I dont know, its weird. (Sorry, bit offtopic )
I tested this a little as well. If the voidray attacks probes, it will not charge at all. The voidray needs 7 hits to kill a probe (6 damage per hit, 40 HP. 7 hits = 42 damage).
The SCV has 45 HP, so the voidray requires 8 hits. This additional hit brings the voidray to charge level 2. He will stay charged for ~5 seconds. If you haven't attacked within that time, the charge falters.
If you kill an additional SCV with another 8 hits, it will be fully charged. The voidray requires 8 hits in sequence to reach additional charges. With a speed of 0,6 per hit, that requires 5 seconds.
The charge of probes not carrying over perhaps has something to do with 'fazing'. Where charged voidrays could quickly change targets and deal tons of damage. I think you could also charge by quickly clicking between targets, however I'm not entirely sure about that. This was patched and fazing was removed from the game.
|
On May 05 2012 08:33 straycat wrote:Show nested quote +On May 05 2012 06:56 mskaa wrote:On May 05 2012 05:20 BronzeKnee wrote:
Why match up 126 vs 120? The upgrade costs 100-100 which isn't even the cost of a single Void Ray, so why do the unupgraded Void Rays get 6 extra on the side?
... Ehm the cost really isnt the point here... What do you think wins: 100 3/3 marines vs 200 0/0 marines.. But omg 200 marines is 5000 minerals - the 3/3 upgrade cost more.. Why they get so many??? If an upgrade was only worth its actual cost in ONE unit, there would be no reason to get it. Either way you have no idea what your talking about... Im actually shocked you are entertaining us with such a false theory. He means that it does not make sense to test 120 vs 126, since the 6 extra unupgraded is better to have than the upgrades. And so he is confused as to why Grubby had 126. And, one explanation he sees is that Grubby reasoned that the money that did not go to upgrades should go to extra void rays. But, as BK wrote, the upgrade cost does not even warrant one extra void ray, and so the rationale for the 6 extra is still a mystery.
Uhm, I think thats the problem.. In such huge numbers an upgrade should be worth many units.. I agree that to prove that +1 attack means less overall damage, it would have to be a fight in even numbers.
However, this shows that a +1 upgrade for voids is mostly useless which ofc it shouldnt be (It probably changes against different unit comps, and with how many voids decide to attack the same target and so on. Basically it makes it very hard to know what will happen in a fight with a lot of voids, it punishes focusfire (if uncharged), and theres basically no way to micro it correctly unless you're a computer). Like Grubby already said earlier, try moving 126 with 0/0 of any unit against 120 of the same with +1 attack, and the upgraded should win every time. The bigger the numbers the more important the upgrades should be, however thats not the case with voids which is what I was getting at.
|
Huh. Never really thought of that happening, but it's true.
|
On May 05 2012 09:04 HansK wrote:Show nested quote +On May 05 2012 08:34 Grubby wrote:On May 05 2012 08:26 virgol wrote:
So, to summarize this thread:
Grubby feels like there is not enough A-move potential in Void Rays right now
No, there is not enough potential in Void Rays to be part of an army composition in big fights. As in, the unit gets obsolete too quickly. Just like Phoenices get obsolete, and Carriers are useless by any measuring stick. It means Protoss air has nary a valid position in any army. The theorising I do is lighthearted and curious. As you could see, I was hypothesising and doing poor one-off experiments. I don't have time to do more... I just wanted to provoke some thought and was interested by the response Thank you for your part of the response. xd I tend to disagree a little. In PvZ, a really popular build going around from White-Ra, col.rsvp, and others is something like a voidray-carrier-HT-mothership-a few gateway units(normally archons for toilets or stalkers if they're going pure infestor corrupter). While using vortex to split the army up the mix of units make it very, very deadly and void rays seem to have a very good little spot in that army comp. I think in some cases void-rays have their spot in big fights, mostly against zerg.
yup, and they normally charge against Corruptors, if you have enough.
|
Interesting observation, but I don't see this as a design flaw. It's just an emergent property of their unusual attack dynamics. It may have some interesting consequences, e.g., players will need to spread VRs better or manually choose different targets to get the most out of them.
|
He's absolutely correct. When the base damage of the VR was buffed, from 5 to 6 back in 1.1.2 it was a huge nerf. VRs no longer charged on a lot of things they used to, making them worthless in the early game. BUT, if the opponent gets an armor upgrade, VRs will charge like they used to, at least against unarmored units, mainly zerglings. Even against armored units, they charge better.
Protoss is a weird race. One that relies on splash damage a lot, because so many units are designed to prevent focus fire from being effective. (Zealots, sentry, templar, VR, phoenix, in most cases should not be focus fired. Stalkers, Collossus, and Carrier are the only units that you should always focus fire) Once VRs are charged, you can focus them again, but before that, the VRs are a liability, and them doing less damage until charged, so that they'll charge faster and more reliably, is exactly what you want.
|
this is stupid.
let's make the estimation, with such close numbers, that in average the first wave of attacks will kill half of the numbers of void rays you have if left uncontrolled (which means, randomly, 126 void rays will kill 63 void rays before they change their target).
i do have the mathematical formula to actually calculate, in a probabilistic approach, how many void rays you can expect to kill , but i am quite convinced that my intuitive conjecture is the right one (if i had a way to actually check the real values...oh god.)
so having this conjecture in mind, and assuming that grubby's conjecture is false, and just saying 1-0 has no effect over 0-0, let's run the numbers
1st wave : 120 vs 126, the 120 kill 60, the 126 kill 63
2nd wave : 57 vs 66, the 57 kill 29, the 66 kill 33 (let's be generous with our upgraded void rays)
3rd wave : 24 vs 37, the 24 kill 12, the 37 kill 18
4th wave : 6 vs 25, the 6 kill 3, the 25 kill..."12"
5th wave : 0 vs 22
now i know that the probabilities should be iterated at every wave and everything, but as a general way to convince yourself that 6 void rays makes a HUGE difference, i think this was pretty spot on.
|
Stats:
Health of 126 0-0 voids: 31500
dps of 126 0-0 voids: 2100 (exact)
Health of 120 1-0 voids: 30000
dps of 120 1-0 voids: 2200 (exact)
death rate of 126 voids: 14.31 seconds (just straight 1-0 voids killing the 126)
death rate of 120 voids: 14.285 seconds (just the straight 126 voids killing the 120)
I could implement the algorithm to show that with the death rate and such the 126 excel over the 120 1-0 do to sheer numbers. As such this is a bad test for grubby's theorem. While his idea is good in basis and may be true in actuallity with the test he used it is clearly not true. Which is sad because i am a huge grubby fan.
I recomment all you theory crafters out there actually crunch some numbers before just watching a replay, As it is clear with number crunching this out come can be expected.
|
On May 05 2012 10:18 Toxi78 wrote:
this is stupid.
let's make the estimation, with such close numbers, that in average the first wave of attacks will kill half of the numbers of void rays you have if left uncontrolled (which means, randomly, 126 void rays will kill 63 void rays before they change their target).
i do have the mathematical formula to actually calculate, in a probabilistic approach, how many void rays you can expect to kill , but i am quite convinced that my intuitive conjecture is the right one (if i had a way to actually check the real values...oh god.)
so having this conjecture in mind, and assuming that grubby's conjecture is false, and just saying 1-0 has no effect over 0-0, let's run the numbers
1st wave : 120 vs 126, the 120 kill 60, the 126 kill 63
2nd wave : 57 vs 66, the 57 kill 29, the 66 kill 33 (let's be generous with our upgraded void rays)
3rd wave : 24 vs 37, the 24 kill 12, the 37 kill 18
4th wave : 6 vs 25, the 6 kill 3, the 25 kill..."12"
5th wave : 0 vs 22
now i know that the probabilities should be iterated at every wave and everything, but as a general way to convince yourself that 6 void rays makes a HUGE difference, i think this was pretty spot on.
^^^ My post just helps demonstrate the actuallity of your statment
|
I think that with such huge numbers, the unupgraded voidrays still do too much damage to actually charge. So then it's a simple matter of 7 damage versus 6 damage. In smaller numbers there might be a bigger difference. I conducted a small test myself where 10 voidrays with +1 focus fired on 10 voids without upgrades. The unupgraded voidrays were attack moved. The +1 voidrays could not charge up until the very end, whereas several unupgraded voids did charge. The unupgraded voidrays won with 3 left.
|
On May 05 2012 06:56 mskaa wrote:Show nested quote +On May 05 2012 05:20 BronzeKnee wrote:
Why match up 126 vs 120? The upgrade costs 100-100 which isn't even the cost of a single Void Ray, so why do the unupgraded Void Rays get 6 extra on the side?
Furthermore, Void Rays can fully charge on each other whether or not they are 0:0 or have +1 attack, thus the premise of the theory isn't actually tested by the test. The premise is that when you upgrade a Void Ray it doesn't get a chance to charge up because of how fast it kills things, which means this should be tested against a lower HP unit to see if the Void Ray doesn't get a chance to charge up, thereby doing less damage in a large battle. Hydralisks might be a good choice here, but I'm not sure if Voids fully charge on them or not.
To prove this theory, Grubby needs to find a unit which the Void Ray kills right after fully charging, then test it with upgrades to see if it allows the Void to kill the unit before it fully charges, thus meaning it takes longer for the Void to fully charge, and thus down lower damage over time. Either way the Void Ray should charge sooner after changing targets, thus the benefit would be very situational and short lived.
And either way Grubby has no idea what he is talking about. I am actually shocked you guys are entertaining such a theory... Ehm the cost really isnt the point here... What do you think wins: 100 3/3 marines vs 200 0/0 marines.. But omg 200 marines is 5000 minerals - the 3/3 upgrade cost more.. Why they get so many??? If an upgrade was only worth its actual cost in ONE unit, there would be no reason to get it. Furthermore, 120 voidrays will probably not attack 120 different voidrays on the other side, but will attack a lot of the same voids. That means the voidrays actually wont charge up coz they die too fast.. Ofc if you have 10k+ apm you can individually make everyone of your voids attack a different enemy voidray, but for the rest of us mortals we will have to make due with several voidrays attacking each other and thus not getting charged up. Either way you have no idea what your talking about... Im actually shocked you are entertaining us with such a false theory.
The whole idea was that Void Rays do less damage in big battles without the damage upgrade (read first line of OP). Not who wins in mass Void Ray battles. No where did I purpose any theory, I was simply stating the proper way to test the idea, and showing that the idea itself is very situational, and any bonus is very shorted live in most situations (when the +1 Voids do indeed charge, they'll easily surpass the damage output of the charged unupgraded Voids in a short amount of time).
Throwing massive amount of Void Rays at each other does nothing to test Grubby's idea. As I stated a single unupgraded Void Ray should be tested against a unit it kills right after fully charging, then test it again with +1 upgrade to see if the upgrade allows the Void Ray to kill the unit before it fully charges, thus meaning it takes longer for the Void to fully charge, and thus down lower damage over time. Either way the Void Ray should charge sooner after changing targets in big battles, thus the benefit would be very situational and short lived. So while in an extremely small set of cases Grubby is correct (and said circumstances would never be in big battles because the upgraded Void Rays would change targets and charge very quickly), if you were planning to get +1 attack for your Void Rays but are thinking of not getting it now because you think the Void Ray will actually do less damage, you are incorrect.
Now back again to mass Void Ray battle, even unupgraded Voids in tandem will easily focus down a +1 Void Ray before they have a chance to charge, thus both sides suffer from this, and both players would need the 10k+ apm you speak of in order to maximize damage. ACRcoldflame did the math a few posts above this one to simply show that the extra health and damage from the 6 Void Rays on one side overcomes the extra damage from the +1 on the other, and thus case closed. If there were any truth to Grubby's idea than 120 unupgraded Voids would beat 120 upgraded since they would charge faster, but that is not the case. The extra 6 Void Rays is like the extra bullet the magician has hiding in his teeth. It sure looks like he caught the bullet fired from his assistant's gun with his teeth (just like it looks like unupgraded Voids do more damage than upgraded Voids), but when you look at it closely, it is just smoke and mirrors.
|
On May 05 2012 11:45 BronzeKnee wrote:Show nested quote +On May 05 2012 06:56 mskaa wrote:On May 05 2012 05:20 BronzeKnee wrote:
Why match up 126 vs 120? The upgrade costs 100-100 which isn't even the cost of a single Void Ray, so why do the unupgraded Void Rays get 6 extra on the side?
Furthermore, Void Rays can fully charge on each other whether or not they are 0:0 or have +1 attack, thus the premise of the theory isn't actually tested by the test. The premise is that when you upgrade a Void Ray it doesn't get a chance to charge up because of how fast it kills things, which means this should be tested against a lower HP unit to see if the Void Ray doesn't get a chance to charge up, thereby doing less damage in a large battle. Hydralisks might be a good choice here, but I'm not sure if Voids fully charge on them or not.
To prove this theory, Grubby needs to find a unit which the Void Ray kills right after fully charging, then test it with upgrades to see if it allows the Void to kill the unit before it fully charges, thus meaning it takes longer for the Void to fully charge, and thus down lower damage over time. Either way the Void Ray should charge sooner after changing targets, thus the benefit would be very situational and short lived.
And either way Grubby has no idea what he is talking about. I am actually shocked you guys are entertaining such a theory... Ehm the cost really isnt the point here... What do you think wins: 100 3/3 marines vs 200 0/0 marines.. But omg 200 marines is 5000 minerals - the 3/3 upgrade cost more.. Why they get so many??? If an upgrade was only worth its actual cost in ONE unit, there would be no reason to get it. Furthermore, 120 voidrays will probably not attack 120 different voidrays on the other side, but will attack a lot of the same voids. That means the voidrays actually wont charge up coz they die too fast.. Ofc if you have 10k+ apm you can individually make everyone of your voids attack a different enemy voidray, but for the rest of us mortals we will have to make due with several voidrays attacking each other and thus not getting charged up. Either way you have no idea what your talking about... Im actually shocked you are entertaining us with such a false theory. The whole idea was that Void Rays do less damage in big battles without the damage upgrade (read first line of OP). Not who wins in mass Void Ray battles. No where did I purpose any theory, I was simply stating the proper way to test the idea, and showing that the idea itself is very situational, and any bonus is very shorted live in most situations (when the +1 Voids do indeed charge, they'll easily surpass the damage output of the charged unupgraded Voids in a short amount of time). Throwing massive amount of Void Rays at each other does nothing to test Grubby's idea. As I stated a single unupgraded Void Ray should be tested against a unit it kills right after fully charging, then test it again with +1 upgrade to see if the upgrade allows the Void Ray to kill the unit before it fully charges, thus meaning it takes longer for the Void to fully charge, and thus down lower damage over time. Either way the Void Ray should charge sooner after changing targets in big battles, thus the benefit would be very situational and short lived. So while in an extremely small set of cases Grubby is correct (and said circumstances would never be in big battles because the upgraded Void Rays would change targets and charge very quickly), if you were planning to get +1 attack for your Void Rays but are thinking of not getting it now because you think the Void Ray will actually do less damage, you are incorrect. Now back again to mass Void Ray battle, even unupgraded Voids in tandem will easily focus down a +1 Void Ray before they have a chance to charge, thus both sides suffer from this, and both players would need the 10k+ apm you speak of in order to maximize damage. Coldflame did the math a few posts above this one to simply show that the extra health and damage from the 6 Void Rays on one side overcomes the extra damage from the +1 on the other, and thus case closed. If there were any truth to Grubby's idea than 120 unupgraded Voids would beat 120 upgraded since they would charge faster, but that is not the case. The extra 6 Void Rays is like the extra bullet the magician has hiding in his teeth. It sure looks like he caught the bullet fired from his assistant's gun with his teeth, but when you look at it closely, it is just smoke and mirrors.
I get your point, that this doesnt prove that +1 attack is actually worse than no upgrades..
It does however tell something about how bad the attack upgrade is, because of the charge mechanic.
126 to 120 is only 5% more, and it doesnt even win!.. Wouldnt you agree that an upgrade should be worth more than that? To compare, a marine gets 16% increase in damage with +1, and yes, i know that is one of the highest scaling units. But still even a stalker gets a flat 10% (or 7ish% against armored) and they certainly dont scale well..
Voids do theoretically scale as well as marines, with around 16% dmg increase, but through its charging ability it somehow gets knocked down enough to lose against 5% more voids without the upgrade..
Along with that, it actually discourages micro, since messing with your voids means they could switch target and lose their building charge, and focusfiring can actually hurt you more than help you since you might miss your charging. Basically you have an unreliable unit that may or may not kill everything, and its not even up to your micro to determine it, but to what the AI decides to; attack first / which upgrades they and you have / what else is attacking that unit / and so on.
tldr;
I get you'r point, grubby's little experiment doesnt prove what he was trying to prove. But voidrays are still messed up :D
|
On May 05 2012 10:28 ACRcoldflame wrote:
Stats:
Health of 126 0-0 voids: 31500
dps of 126 0-0 voids: 2100 (exact)
Health of 120 1-0 voids: 30000
dps of 120 1-0 voids: 2200 (exact)
death rate of 126 voids: 14.31 seconds (just straight 1-0 voids killing the 126)
death rate of 120 voids: 14.285 seconds (just the straight 126 voids killing the 120)
I could implement the algorithm to show that with the death rate and such the 126 excel over the 120 1-0 do to sheer numbers. As such this is a bad test for grubby's theorem. While his idea is good in basis and may be true in actuallity with the test he used it is clearly not true. Which is sad because i am a huge grubby fan.
I recomment all you theory crafters out there actually crunch some numbers before just watching a replay, As it is clear with number crunching this out come can be expected.
This is true if each VR was shooting at only one VR. But I think the argument here is that, if a group of 1-0 VR's are attacking a single target (like they usually do in real combat), the target tends to die more quickly, and therefore the VR's are less likely to get charged. Then VR's switch target and attack another target, and again they might not get charged, and so on. Whereas a group of 0-0 VR's take more time for killing a single target, but are more likely to get charged. Therefore, their dps will be increased from the start of the fight. The mere difference of 6 VR's is not the real issue here.
I think the issue here is that, VRs effectiveness (hypothetically) significantly decrease as the damage output of the protoss army increases. In the end, 2 VRs in a 50 supply army might be doing significantly more damage (percent wise) than 4 VRs in a 100 supply army (for example).
Can your algorithm take this into account?
|
doesnt anyone take into account that the smaller the number of the void rays, the better off they are? For example. in a 1v1 void ray fight, the upgraded one wins (more chance to charge up). As the number of void rays goes up, the chance of getting charge up decreases as units die quicker. What I'm trying to say is that 126 void rays is ridiculous, because the supply limit would allow this. Nor 100. 50 is probably a more accurate figure in a real situation (including 50 probes, that would be 200). Without probes, its 66. See how that works out.
|
On May 05 2012 08:57 Elp wrote:While testing this i've noticed some other odd behavior by the Voidray charge. You can see it when you have the Voidray attack a bunch of SCV's. The 2nd charge will start when the 3rd SCV is attacked (1st, 2nd destroyed). This is strange because if it is true that the charge buildup (the buildup to level 2, not the 2nd charge level itself) is reset when the Voidray switches targets, the 2nd charge should start when the 2nd SCV is attacked or it should never charge up to the 2nd stage at all. As I said in my previous post, the charge buildup does reset when attacking Probes (40hp), but apparently it doesn't when attacking a 45HP unit. Perhaps there is a tiny fraction of charge that bleeds over to the next target and 45HP is just enough to make it work? I dont know, its weird. (Sorry, bit offtopic )
the match behind is very simple.
lvl1 = standard attack
lvl2 = precharge (level where voidray doesn't loses his precharge when switching targets)
lvl3 = full charge (dmg booost)
it takes 6 attacks for a Voidray to reach lvl2
this means:
+0 Voidray:
6*6 = 36 dmg (vs. light) = thus lvl2 can be reached 1 voidray if it attacks a probe alone
+1 Voidray:
6*7 = 42dmg (vs. light) = thus lvl2 can't be reached since unit dies charge ticks are lost when switching to the next target.
(i.e. can't charge on probes anymore)
+2 Voidray:
6*8 = 48dmg
(i.e. can't charge on scvs/ unupgraded marines anymore)
+3 Voidray:
6*9 = 54 dmg
same applies for the armor bonus dmg
10 * 6 = 60dmg needs to be dealt to a single armored unit before it dies
+1
11*6 = 66
+2
12*6 = 72
+3
13*6 = 78
and this is the optimal preset where we assume only the Voidray attacks the unit.
If other units shoot these too and voidray can't reach lvl2 and never reach lvl3 because charge is lost every time the focussed unit dies.
Especially in lategame this gets even worse, because splash damage comes to please which reduces the overall hp of single units so Voidrays most likely can't deal their 6 attack ticks on a single unit.
The charge mechanic is just broken in general imho.
(I mean where seriously is this unit used? except from early game harass vs. Zerg is has become absolutely useless unit)
|
|
|
|
|
|
EPT
