Cataclysm Rogue Changes - Page 38

kindath · 07/23/10, 10:39 AM

I'm sure this is unnecessary, but I wrote up the formulas in excel while bored at work and figured maybe they'll be useful to someone.

speed =baseSpeed/(1+(haste/100))
ticks =FLOOR((baseDuration+baseSpeed/2)/speed,1)
duration =ticks*speed

Results aren't consistent with what I've seen in this thread, actually. Hrmmmmmmmmm.
EDIT: Herp derp, I should have been Ceiling to dura-speed/2 instead of flooring to dura+speed/2. Wasn't sure exactly how the game would run that.

speed =baseSpeed/(1+(haste/100))
ticks =CEILING((baseDuration-baseSpeed/2)/speed,1)
duration =ticks*speed

Speed = time between ticks
Ticks = number of ticks
Duration = obvious
Haste in whole numbers

Actually, it's probably more likely that the limiter on ticks is =CEILING((baseDuration-speed/2)/speed,1), half a hasted tick, not half the base tick. Otherwise you get weird durations at high amounts of haste that don't stay close to the original duration. I'm getting inconsistent results, though, compared to Pyriana's rupture results...
When using half the hasted tick length, I'm getting a break point of 615 haste rating to go from 9 to 10 ticks, which is consistent, but a break point of 1025 to get from 10 to 11 ticks, which is not.
When using half the original tick length, I'm getting a break point of 1093 to get from 10 to 11 ticks, which is consistent, but a break point of 656 to get from 9 to 10 ticks, which is not.

Back to the drawing board.

I'm probably overthinking this, I'm sure blizzard did something lazy like ticks=ROUND(baseDuration/hastedTickSpeed,0)

Soulvex · 07/23/10, 11:55 AM

Not to detract from the current discussion, as I find it very interesting, but for those who do not check MMO frequently (and I haven't seen anyone mention this yet) there was a new build with some more minor rogue changes.

Combat

* Backstab now causes 175% weapon damage. Up from 150%.

Subtlety

* Shadowstep now causes 30% additional damage (up from 20%) but no longer reduces threat.
* Initiative is now a 2 points talent, down from 3. Now has a 50%/100% chance to proc. (Up from 33/66/100%)

Seems like they're "trying" to open up more possibilities to grab the fun stuff in Sub.

Amerilina · 07/23/10, 12:41 PM

I took Pyriana log and made a program to filter it to just rupture or Garrote. I figure it might be useful.

Garrote_Filter.txt

Rupture_Filter.txt

fourdots · 07/23/10, 1:33 PM

Apologizing in advance if this has been asked before, but I can't seem to find this anywhere, so here goes......

Looking at the new trees, I just noticed that "Vendetta" does not have the word "cooldown" in its description. Cause for celebration, or just a mistake on the part of the person who wrote the tooltip? It would obviously be awesome without the cd, but really, it seems way too good to be true.

Milou · 07/23/10, 1:41 PM

Both the talent tree and the spell description on mmo-champion show the 2 minute cooldown.

Vendetta - Spells - Sigrie

• Aldriana · 07/23/10, 1:53 PM

Originally Posted by StoicRoivaS

That's sort of what I was getting at. That last point is definitely going to be a "freebie" of sorts, but with how wide these gaps are, and how non-discrete our control over the stats are (i.e. gems only being changeable in huge chunks) it sort of feels like, in a practical sense, this is mostly just a matter of intellectual curiosity. The last cataclysm green gem I saw was +30 to a stat? So the blue gems are somewhere above that even. And like you said, my intuition is also that haste is fairly constant over a wide range, with just a bit of a "bonus" here and there. So really, unless you're neck and neck with a break point and you can swap a minimal amount of points around somehow, I just don't see this as something your every-day spreadsheet user will end up super concerned with. That's not to say that getting the theory crafting absolutely right isn't the way to go, because it is.

Edit: Now that I think about it in a bit more detail, it makes more sense that the point of haste that passes you to the next break point spikes up in EP, but every point after that slowly climbs back down. I mean, it's clear that the first point past each break point is going to be worth more, but it's equally clear that the value of haste won't exponentially rise like armor pen does now. So somewhere the EP has to come back down to the "average" after each spike. The value of that very first threshold point is worth the most, relatively, but really that 2nd point is worth a fair bit also, because in order to get that 2nd point, you had to make some gear change that (in 99.999% of cases) swapped more than one point of haste around. So going two points over the break point is worth almost as much EP and doubling the value of the first point over the break point, if that makes sense. If that's not clear, say for illustration the current weight of haste is 2.0 at some arbitrary level, and it jumps to 2.5 when you're 1 point shy of a break point. That one point is worth a fair bit more by comparison, but equally so, there will be exceptionally few (probably zero) cases where you can only get that one point. So say you can get 5 points as your smallest increment, it seems more likely, at least how my intuition parses it, that the value of those 5 points is a lot closer to (2.5 * 5) than it is to (2.5 + 2 + 2 + 2 + 2).

So really, at least how I'm imagining, the EP values of haste probably look a lot more like:

Forgive the 5-minute MSPaint job, but after thinking about it for a bit, this case seems far more likely than (2.5+2+2+2+2). And in case it's not clear, the reason I went with an arc-tangent looking curve between break points instead of just a simple linear decline is pretty much due to what I described above. We don't really have the option of picking our haste in units of 1. Granted, this isn't really something you can model into a spreadsheet, at least I don't think, but in theory, that feels right to me.

I disagree. It's going to *spike* up, not *step* up. That is, you're proposing that the value of incremental points of haste will go something like: 2.0, 1.9, 1.8, 2.5, 2.4, 2.3 around the breakpoint - i.e., will steadily decrease to it and then jump up and start descending again. I can't think of any reason why this would be the case - I think the value of incremental points of haste, assuming reasonable refresh mechanics, will be more like 2, 2, 2, 50, 2, 2, 2. That is: the value of individual points of haste before and after the breakpoint are about the same - but the act of crossing that breakpoint gives a significant bonus.

Pyriana · 07/23/10, 2:03 PM

also as mutilate with no other energy regen it was taking about 3.6-3.7 seconds with 1310 haste to regenerate enough energy to mutilate again (55 energy)

jonnnney · 07/23/10, 3:19 PM

Has anyone checked if haste is effecting deadly poison and recuperate? Also I am still not clear about how rupture works when you refresh it before the last tick. Is the last tick counted as part of the old rupture or the new one in terms of duration?

I understand that how haste is effecting rupture is more interesting and involved question, I just dont want the other questions lost in the murk if at all possible.

StoicRoivaS · 07/23/10, 3:22 PM

Originally Posted by Docrev

As described in this link, that's not how reforging works: Blizzard's Greg Street on new systems in Cataclysm - Gameplanet Feature

You don't lose stats in total. Rather than going from 100 hit to 40 haste and 0 hit, for example, you'd go from 100 hit to 40 haste and 60 hit.

I specifically recall it being stated at one point that reforging will be a loss in total stat points when you use it to swap itemization around. Furthermore, I specifically recall them saying they didn't want reforging to be done to every single piece of gear simply for min-max purposes, that their intent was to allow pieces that may have been sharded (due to hit or spirit typically) to be used by someone. Has this drastically cahnged? If you can just take 50 points from one stat and put them into 50 points of a better stat, then isn't the end result really something closer to "When you get an upgrade it now costs you X gold to gem + Y gold to enchant + Z gold to reforge?" I thought the whole point was to not reforge every piece?

The blue post I'm looking at now says "As a general rule of thumb, you’ll be able to convert one stat to 50% of another stat." Now I understand that this probably came before GCs detailed explanation, and it is certainly less detailed, but I read this as "You give up all of one stat to get half as many points in another stat". Maybe they just changed their mind at some point? I really don't think reforging every piece was the intended goal, but if it indeed does not lose you any itemization, and just allows you to swap it around, we will be reforging pretty much every piece, wont we?

Amerilina · 07/23/10, 3:33 PM

Originally Posted by jonnnney

Has anyone checked if haste is effecting deadly poison and recuperate? Also I am still not clear about how rupture works when you refresh it before the last tick. Is the last tick counted as part of the old rupture or the new one in terms of duration?

I understand that how haste is effecting rupture is more interesting and involved question, I just dont want the other questions lost in the murk if at all possible.

It does not look like Deadly scales with haste. This is from Pyriana's log. You can get the haste value he had from his test info on the previous page.

Start

7/22 22:06:16.997  SPELL_AURA_APPLIED,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF
7/22 22:06:20.016  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:06:23.035  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:06:25.987  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:06:29.023  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:06:29.023  SPELL_AURA_REMOVED,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF

Ended Ticks: 4


Start

7/22 22:07:52.902  SPELL_AURA_APPLIED,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF
7/22 22:07:53.700  SPELL_AURA_REFRESH,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF
7/22 22:07:55.918  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:07:58.871  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:08:01.922  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,246
7/22 22:08:04.941  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,380
7/22 22:08:04.942  SPELL_AURA_REMOVED,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF

Ended Ticks: 5


Start

7/22 22:11:35.625  SPELL_AURA_APPLIED,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF
7/22 22:11:38.676  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,361
7/22 22:11:41.978  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,361
7/22 22:11:44.614  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,234
7/22 22:11:47.649  SPELL_PERIODIC_DAMAGE,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,234
7/22 22:11:47.649  SPELL_AURA_REMOVED,0x010000000009060D,"Pyrianatest",0x511,0xF13079AA000027C2,"Heroic Training Dummy",0x10a28,2818,"Deadly Poison",0x8,DEBUFF

Ended Ticks: 4

The log post did not have any refreshes of rupture so checking that would have to wait.

Ends · 07/23/10, 3:46 PM

Originally Posted by StoicRoivaS

I would hardly call that a solution either though. The amount of stat points you reforge is not something you can change, at least that I recall, it's always 40% or whatever. Secondly, since reforging always loses stat points in total, reforging is not something we're actually going to *want* to do. It might be something that we do end up doing under certain circumstances, but as I understand it, it's never something we will enjoy doing. Say for instance you're in a guild that does the absolute hardest content, and there's only a single option for a certain slot at the highest item level, and that specific piece has a stat that is decidedly bad for you (for instance if you're crit capped and you want to swap crit into arp from the 284 heroic LK-25 axe). In that situation, it may be worth taking the total stat point hit to swap some itemization from a stat you don't like at all to the stat you like the most. But like I said, that's a fairly specific circumstance. In general, we don't want to reforge at all, and arbitrarily reforging pieces to hit some minimally-impacting haste break point almost certainly won't be optimal, at least as I understand it so far.

I misunderstood the way it worked, I didn't realize it was an all or nothing situation, I thought it was up to 40/50% i.e you could change it in 1 point increments which would be ideal for situations like this, and other caps like expertise and hit where the value of the stat takes a dive after a certain amount.

Correct me if I'm wrong, but unless there is a very good balance amongst the EP values of reforgeable stats you will always want to reforge a piece to give the maximum amount of your highest valued stat. So if reforging were available in wrath, any piece with crit on it would be changed to arp, regardless of what ilvl the piece is. Its the same concept as gemming. The only way they can dissuade people from doing this is, as you said, if they invoke a "reforging tax" which is currently not implemented.

Pyriana · 07/23/10, 4:01 PM

Originally Posted by jonnnney

Has anyone checked if haste is effecting deadly poison and recuperate? Also I am still not clear about how rupture works when you refresh it before the last tick. Is the last tick counted as part of the old rupture or the new one in terms of duration?

I understand that how haste is effecting rupture is more interesting and involved question, I just dont want the other questions lost in the murk if at all possible.

from my anecdotal experience, it does not effect recuperate, depending on how today goes I might be able to do more testing tonight, saturday and sunday, one thing I noticed is that (obviously) a 4 point rupture can not refresh a 5 point rupture, and I think procs are independent of refreshing rupture, however vendetta isn't, as soon as vendetta would drop I would have to let rupture drop to refresh it.

and without mods @ 80 with 1310 haste ~50% crit, dps testing the heroic target dummy I was able to keep rupture up fairly well and depending on rs procs get between 2 and 3 envenoms per rupture, up to 4 if cold blood was up, it also appeared that about 70% of my damage was nature, I'll see if I can find the log if you guys are interested in breaking it down, but I may have deleted it.

edit:

Originally Posted by Ends

Correct me if I'm wrong, but unless there is a very good balance amongst the EP values of reforgeable stats you will always want to reforge a piece to give the maximum amount of your highest valued stat. So if reforging were available in wrath, any piece with crit on it would be changed to arp, regardless of what ilvl the piece is. Its the same concept as gemming. The only way they can dissuade people from doing this is, as you said, if they invoke a "reforging tax" which is currently not implemented.

Yes, until you start hitting a point where another stat out values it such as when mutilate meets a balancing point between ap and haste, I believe blizzard is doing this to help separate the casuals and hardcore or tcers a bit more, since gear itself will be the same in general. I also feel that it will really help with the caps, if you do go over the expertise cap (and expertise is the best stat) you can take it off of some gear and gem for it, and if that still doesn't reach the cap to your satisfaction you can take one of your most suboptimal stat boosts and switch that to expertise instead of your "secondary" best stat, I don't think that we'll have that issue with any stat but haste though, hit can float around where ever it wants above the poison cap as long as we're not crit capped, unless they decide to hit starve us for this tier.

the other benefit of reforging (not sure how much it costs to do), is if it's cheap and you are working on a progression boss that is a caster that casts constantly, you can quite possibly drop a ton of expertise as iirc mobs don't dodge while casting. Now you can do this currently with gems but gems require a real world component which can get expensive to get and time consuming to get cut assuming you don't have the cut, whereas reforging is just an npc.

Rosvall · 07/23/10, 4:47 PM

Originally Posted by Aldriana

I disagree. It's going to *spike* up, not *step* up. That is, you're proposing that the value of incremental points of haste will go something like: 2.0, 1.9, 1.8, 2.5, 2.4, 2.3 around the breakpoint - i.e., will steadily decrease to it and then jump up and start descending again. I can't think of any reason why this would be the case - I think the value of incremental points of haste, assuming reasonable refresh mechanics, will be more like 2, 2, 2, 50, 2, 2, 2. That is: the value of individual points of haste before and after the breakpoint are about the same - but the act of crossing that breakpoint gives a significant bonus.

I do not really see where the large 'spike' would come from. The only I can think of is that you will spend 0.04 combo points less per second on using ruptures. Energy usage would be the same due to RS. You will even proc less poisons due less energy free ruptures.

So it's basically "Gain combo points - Lose poison procs" when you step over a break point. I fail to see that this would actually be a huge jump at all. Care to elaborate for me please?

This assumes 100% uptime while refreshing the clock cycle it runs out, obviously.

• Aldriana · 07/23/10, 5:00 PM

Whether you're above or below the breakpoint, each point of haste increases the damage per time of an active rupture and reduces the duration slightly. However, when you cross a breakpoint, you gain an extra ~2 seconds of rupture duration. Hence, the point that actually crosses that breakpoint gives you an effect akin to Glyph of Rupture - you gain significant benefits in terms of CP/Time needed to maintain it. So: when you cross a breakpoint, you gain a one-time benefit of about half what Glyph of Rupture gives; all other points have the usual tradeoff of DPS vs duration, and as such should be of similar value.

jonnnney · 07/23/10, 8:06 PM

Originally Posted by Pyriana

the other benefit of reforging (not sure how much it costs to do), is if it's cheap and you are working on a progression boss that is a caster that casts constantly, you can quite possibly drop a ton of expertise as iirc mobs don't dodge while casting. Now you can do this currently with gems but gems require a real world component which can get expensive to get and time consuming to get cut assuming you don't have the cut, whereas reforging is just an npc.

It costs the vendor sell price of the item