Killing Machine issues

90 Orc Death Knight
7780
Yikes, are you even reading what you're typing? I'm surprised you're so willingly agreeing with (and demonstrating) everything I'm saying =[

Tell you what, I'll give you this - you can have the last word. Go ahead, please post after this. At least I'll know I was able to help contribute some small amount of temporary contentment to you.

Hopefully someday things will get better for you.
Reply Quote
90 Draenei Death Knight
11090
12/20/2012 03:32 AMPosted by Lailala
As long as the second happens half of the time, you've broken even. Any better ratio than that and you're ahead.


That's not the correct way to look at it.

In my previous example, if you KM-FS (scenario 2), you gain 46.75k damage (that is, you have lost x/2: x - x/2 = x/2). If you overwrite KM (scenario 1) but can use it on OB afterward, then you gain 93.5k damage (that is, you have lost x: x + x - x = x). If you overwrite it and can't use KM-OB (scenario 3), then you have lost 3*x/2 (because you'll eventually KM-FS unless you let it expire), however, that's not a scenario we need to consider.

But, the comparison between scenario 1 and scenario 2 is ignoring the second window for scenario 2, yet considering it for scenario 1. That's not an accurate comparison because you're not looking at the same thing anymore.

If, in scenario 1, you are able to KM-OB with the second KM, then in scenario 2 with the next KM (which you left out), you will also be able to KM-OB (and with a T75 proc you would be more likely to anyway from a resource-availability standpoint). This brings scenario 2 to 93.5k + 46.75k = 140.25k. This is only a loss of x/2 (x + x - x - x/2 = x/2).

If you don't agree with that, then you were comparing scenarios with two different initial rune states, which wouldn't make sense at all.

Thus, scenario 2 has only lost x/2. Scenario 1 has lost x.

That is, you were better off using KM-FS anyway.
Reply Quote
90 Human Death Knight
12620
You're trying to apply an opportunity cost of something that wasn't going to happen.

If you let a KM that was going to be used on a frost strike instead be overwritten, then you've lost the value of KM on a frost strike, which is X/2.

Again, you're trying way too hard to make this more complicated than it is.
Reply Quote
90 Human Death Knight
6270
12/20/2012 07:32 PMPosted by Lailala
Again, you're trying way too hard to make this more complicated than it is.


Isnt that the whole point of the counter arguement against you? Your talking about a trivial amount of dps gain (or not) by getting all mathmatical and what not? Id go do your thing, and feel confident that your method is the best. If your right then cool your the best.. if your not right, you will never know/admit it anyways...who cares?
Reply Quote
90 Human Death Knight
12620
Again, you're trying way too hard to make this more complicated than it is.


Isnt that the whole point of the counter arguement against you? Your talking about a trivial amount of dps gain (or not) by getting all mathmatical and what not? Id go do your thing, and feel confident that your method is the best. If your right then cool your the best.. if your not right, you will never know/admit it anyways...who cares?


That is a valid argument against it, if you choose to take the simpler route. That does not, however, mean that it's not a gain to go ahead with the more advanced approach.

He's trying to over complicate the math, though, due to coming to his conclusion first and attempting to make the math fit, rather than looking at the math and then coming to a conclusion.
Edited by Lailala on 12/20/2012 11:01 PM PST
Reply Quote
90 Human Death Knight
6270
The thing that gets me, is that I feel your more concerned with being right, over the original issue... Why does it bother you, if no other death knights are using your "advanced approach."
Its not like anyone has said, "well I dont think it works, so take it out of the game!"
Edited by Steahl on 12/20/2012 11:33 PM PST
Reply Quote
90 Human Death Knight
12620
If other people choose to use it is up to them. My concern is that people have the correct information when making that decision.
Reply Quote
90 Draenei Death Knight
11090
12/20/2012 07:32 PMPosted by Lailala
You're trying to apply an opportunity cost of something that wasn't going to happen.


You can call it that if you want, but I'm simply giving KM a constant value. Ideally, you would want all KM's to be used on OB, which is the reason for picking that value. It is the ideal limit.

There is, however, the reality you have not dealt with yet, though: the fact that scenario 2 is ahead by x/2 in terms of damage dealt. That's the bottom line and directly conflicts with the claim that you are trying to sell to me.

12/20/2012 11:00 PMPosted by Lailala
He's trying to over complicate the math, though, due to coming to his conclusion first and attempting to make the math fit, rather than looking at the math and then coming to a conclusion.


"Make the math fit?" Does it honestly look like I am the one trying to shoe-horn something here? Is that your fallback argument when I use your own examples against you? I'm simply stating what the math says and going from there.

I walked into this scenario in Beta with the same mindset: that stalling had to be a gain. So far, the math hasn't supported it. So far, my sim hasn't supported it. So far, logs have not supported it.

I would be crazy to somehow think it still has merit. Yet, I haven't walked away yet, and that should tell you something. I enter arguments waiting to be proved incorrect. Hoping, actually, because if that happens, it means there is still more for me to learn. This thread has not fulfilled that.
Reply Quote
90 Human Death Knight
12620
12/21/2012 02:29 AMPosted by Euliat
You're trying to apply an opportunity cost of something that wasn't going to happen.


You can call it that if you want, but I'm simply giving KM a constant value. Ideally, you would want all KM's to be used on OB, which is the reason for picking that value. It is the ideal limit.

There is, however, the reality you have not dealt with yet, though: the fact that scenario 2 is ahead by x/2 in terms of damage dealt. That's the bottom line and directly conflicts with the claim that you are trying to sell to me.


I have dealt with this over and over and over and over and over and over and over and...

Yes, if you overwrite, you lost x/2.

But, if you didn't overwrite, you gained x/2.

If they both happen the same amount, you break even. Any better than that and you're ahead.

What the hell do I have to say to get you to grasp this?
Reply Quote
90 Human Death Knight
12620
And, as long as you absolutely refuse to put those facts together, and have never and are not willing to ever run your sims in accordance with that reality, yes, you're simply trying to fit the math to the result you've decided on.
Reply Quote
90 Draenei Death Knight
11090
12/21/2012 03:05 AMPosted by Lailala
What the hell do I have to say to get you to grasp this?


It's pretty simple as far as I can see.

If you overwrite KM while waiting, you deal 93.5k extra damage. If you don't overwrite KM because you used KM-FS, you deal 140.25k extra damage.

140.25k > 93.5k

Unless your story now is by waiting that you would get 2 KM-OB's from waiting like that, which I'd doubt.
Reply Quote
90 Blood Elf Death Knight
3015
I stopped reading page 3, so if this is already mentioned, my apologies.

Getting KM procs on the most number of Oblits without wasting resources is pretty straightforward... I'm a new nub that very rarely uses my frost dps spec, and I've already figured this out.

You get into a rhythm. Start off your gcd luls with an oblit and FS dump after. Time your oblits to happen just after your swings that proc KM. FSing immediately after in the "cant KM" window, and repeat.

The timing isn't always perfectly flush, so you'll need to adjust your cycle a bit as the windows of KM moves out of sync from the Oblit-FS dump.

The most essential key element is making sure you're not just chaining oblit oblit, any oblits immediately after another one has a really low chance to get KM, and zero chance if you're timing it with your swings.

You don't have to intentionally wait for anything, you just need to be hitting oblit when KM could be proccing and hitting FS when you know KM can't proc. No wasted resources, and simple.
Reply Quote
90 Human Death Knight
12620
12/21/2012 05:09 AMPosted by Euliat
What the hell do I have to say to get you to grasp this?


It's pretty simple as far as I can see.

If you overwrite KM while waiting, you deal 93.5k extra damage. If you don't overwrite KM because you used KM-FS, you deal 140.25k extra damage.

140.25k > 93.5k


Ok, you're doing this on purpose, you know you're wrong, I give up.
Reply Quote
90 Human Death Knight
12620
And, for those just jumping in, if you're wondering how it's become so clear that he's just trolling:

He literally left out the possibility that's the entire reason you would wait in the first place, which is that you don't overwrite and you get a KM oblit.
Reply Quote
90 Human Death Knight
12620
12/21/2012 05:09 AMPosted by Euliat
What the hell do I have to say to get you to grasp this?


It's pretty simple as far as I can see.

If you overwrite KM while waiting, you deal 93.5k extra damage. If you don't overwrite KM because you used KM-FS, you deal 140.25k extra damage.

140.25k > 93.5k

Unless your story now is by waiting that you would get 2 KM-OB's from waiting like that, which I'd doubt.
'

and now for math proofs using the numbers he decided on

Value of KM on oblit = X
X is 93.5k

No runes to oblit, KM procced

Waiting
No overwrite, oblit for full damage. Value of KM1 = 93.5k
Overwrite, value of KM1 = 0

Not waiting:
Frost strike uses KM. Value of KM1 = 46.75k

46.75k = 93.5k/2

therefore if waiting results in the first possibility at least half the time, KM1 averages a value higher than 46.75
Reply Quote
90 Human Death Knight
13075
Don't forget Rime.
Reply Quote
90 Human Death Knight
12620
I'll rime ur face.
Reply Quote
90 Draenei Death Knight
11090
12/21/2012 07:57 AMPosted by Lailala
He literally left out the possibility that's the entire reason you would wait in the first place, which is that you don't overwrite and you get a KM oblit.


That would be because it was left out when the first time the scenarios were mentioned. It's suddenly a huge glaring error now apparently and wasn't then. Sure, trolling if you will. Let's blow off the 1 person who has a shot at validating this theory with simulations instead of cherry-picked logs.

Of course delaying FS to get a KM-OB is the desired outcome. The question was the opportunity cost, which was lost in poor communication between both parties.

Your theory hinges upon being able to delay usage long enough to get KM-OB's, even allowing for KM overwrites in the process. My sim didn't allow overwrites so I will need to reconfigure it when I get back.

However, you're constantly saying that as long as you can delay in that scenario more than 50% of the time successfully, you'll break even. You've mentioned that several times but you can't even give me a ballpark on that probability.

My hesitation is the probability that you can actually delay that long without wasting resources. You don't buy the stalling leading to crammed GCDs argument but I watched it happen in my sim.

There is a 64% chance that KM won't proc on that swing, which means you need to have the resource configuration that will allow you to stall that duration over 78% of the time without waste. This will give you that 50% overall. That's if you're only considering the scenario in which you either KM-FS, KM-OB, or overwrite, which is ignoring the equally important scenario considering 2 KM windows that I laid out previously. Delaying could cost you a KM-FS because you could never KM-OB before the overwrite. The odds are worse there.
Reply Quote
90 Orc Death Knight
8835
I haven't followed this thread for the last few pages, are we debating the issue like it's black or white? 'Wait vs don't wait' ?

I decided to put more attention into my rotation and while I never tell myself "Stop and wait for the next swing" unless my rotation is going dry, it's more of an issue of balancing when to oblit and when to fill between swings with FS.

To put it short, as long as I have a re-charging rune of all types, plus at least 1 fully depleted rune of any type, then I'm using FS during the swing, and only oblitting immediately after the swing. In other scenarios such as no fully-depleted runes (FS wastes my RE) and both runes of 1 type are up (no recharging), then I will oblit no matter what.

There's no "stop and wait" involved since every GCD is filled, but it more-or-less moves my oblits to land after each swing, without sacrificing a non-crit oblit and without causing a failed RE proc.

Ideally, I don't want to be in a situation where I burn my runes inside a swing, then have nothing but a large pool of RP left when the swing occurs.

My question: If my RE procs aren't being wasted, and runes are always re-charging, is there any drawback from moving my oblits to AFTER the swing, but still using FS between swings? This essentially guarantees all of my KM's will go to Oblit, unless my runes go dry to bad RE procs (which is unavoidable to everyone), and I never actually stop and wait for KM procs. In short, the goal is to be filling every GCD, but always have 1 pair of runes available after each swing, for KM Oblit.

Ex: Dd ff Uu , inside of a swing with no current KM proc.
In this situation most people would follow their priority and Oblit. In my case, I would FS.

If RE proc doesn't occur, I would keep FSing until the swing occurred (the swing will probably occur during my GCD, there's no downtime) and then decide if I want to Oblit. If KM does proc, then I just generated a free KM oblit with no loss, and other people following the normal priority rotation would have lost that KM oblit.

If RE does proc before the swing (Dd Ff Uu) then I will oblit no matter what, since FS will waste an RE proc, and my only other choice is to stop and do nothing. If my rotation is going dry on the next gcd (Not enough RP for FS), then I wait for the swing. Otherwise if I have RP, and it's early in the swing, I can attempt at least 1 or 2 more FS's to get an RE proc for an Oblit on the next swing (for the KM proc).

For me, it's a much more complex issue than "I'm going to wait for KM procs" since the priority literally changes every swing. My current runes and RP, as well as my position in the swing timer, make things more complicated for controlling KM procs, since the idea is to both minimize downtime and maximize KM oblits.

tl;dr Instead of talking about "waiting" for KM procs, is it more of an issue of moving your FS and Oblit priorities around to compensate for when KM procs occur?
Edited by Zionic on 12/22/2012 10:28 PM PST
Reply Quote
90 Draenei Death Knight
11090
12/22/2012 09:58 PMPosted by Zionic
tl;dr Instead of talking about "waiting" for KM procs, is it more of an issue of moving your FS and Oblit priorities around to compensate for when KM procs occur?


Yes, that's more or less what it's about.
Reply Quote

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)

Reported!

[Close]