Well I posted this on the rogue forums in the EU, but though some people on EJ might want to take a look too.
I spent about 4 hours on this along with a Priest friend of mine (Macblood), and afaik it is the most accurate scaling analysis on the web at the moment.
What we did here was to take the most widely used player produced dps models for several classes, equipped each class in shadowpriest wishlist gear for current content, and then scaled all the stats on the gear up by 10% 20% 30% and 40%. (The average stat increase on tier 5 gear over tier 4 gear was 23% although it varied slightly for each class)
Trinket and melee weapon dps values were also scaled in the same way... 100 dps weapons became 110, 120 etc.
I appreciate that mages and moonkins might prefer slightly different gear to the warlocks and shadowpriests, but as hit caps are reached etc this becomes far less of an issue. As a result of this, actual starting figures may vary, but the scaling is more important especially in the later stages once caps have been reached. In other words, the later percentage values are what you should be looking at.
We then ran simulations on fights lasting 300 seconds with all the best buffs and Bloodlust etc.
These results were tbh totally unexpected. High scaling towards the beginning confirms (if any confirmation were necessary) the necessity to hit your cap values as fast as you can. After that, the classes settle towards a 7% scaling, all scaling together at the same speed.
I appreciate that this is only a rough estimate but believe this 7% scaling value post hit caps is likely what will be reached regardless of some gear tweaking. Of course it is possible to cheat and put in low hit gear etc knowing you will reach the caps anyway, but if you do that for the other classes too, I think you will come out with the same % values anyway. Even so, if mages/boomkins want to re-run the simulations with perfect gearing, I would be interested to see if they get different values.
Sukoden - "I refuse to play druid. In the end you get too many and you end up with a frickin zoo."
Thanks to Blackwing for noticing my Scaling Sheet had not been adapted correctly for combat, and therefore returned lower dps values than it should have.
I would think as a 7/51/13 build progressed in haste, hit and crit (ss glyph), it would allow a much tighter 4/5/5 or 3/5/5 with less rupture downtime. This is probably not modeled well as you are most likely sticking to a strict xS5R cycle.
I would think as a 7/51/13 build progressed in haste, hit and crit (ss glyph), it would allow a much tighter 4/5/5 or 3/5/5 with less rupture downtime. This is probably not modeled well as you are most likely sticking to a strict xS5R cycle.
The same can be said for mutilate, and I'd imagine other classes might see changes in their rotations as gear improves.
Mutilate would seem to be closer to the ideal already, with ruthlessness not scaling, you only get more 4x cycles as your crit rate increases. Since mutilate already has a fairly high chance for one hand to crit, it would seem an increase in crit rate would not translate directly into more combo points as fast as combat would with potency and SS Glyph. Even if CP/s does increase at the same rate as combat, you may at some point be forced into more eviscerate finishers because deadly is not stacking fast enough.
We may find at some point it becomes better to ignore ruthlessness for the increased frequency of 5CP finishers and more deadly stacks when we get to them.
Mutilate would seem to be closer to the ideal already, with ruthlessness not scaling, you only get more 4x cycles as your crit rate increases. Since mutilate already has a fairly high chance for one hand to crit, it would seem an increase in crit rate would not translate directly into more combo points as fast as combat would with potency and SS Glyph. Even if CP/s does increase at the same rate as combat, you may at some point be forced into more eviscerate finishers because deadly is not stacking fast enough.
We may find at some point it becomes better to ignore ruthlessness for the increased frequency of 5CP finishers and more deadly stacks when we get to them.
Keep in mind focused attacks procs giving us more energy.
When master poisoner was bugged, I was easily able to fit in 2 or 3 more envenoms than I would usually.
I would think as a 7/51/13 build progressed in haste, hit and crit (ss glyph), it would allow a much tighter 4/5/5 or 3/5/5 with less rupture downtime. This is probably not modeled well as you are most likely sticking to a strict xS5R cycle.
The cycle used for the combat rogue is 5s/5r (not using tott on cooldown), using this cycle it is possible to replace the slice and dice glyph by a blade flurry one (the bugged AR glyph should provide slightly more dps) and to drop 1 point from improved slice and dice to put the fifth point in sword spec. The higher dps I obtained for combat was using Calamity's grasp MH, Hailstorm OH with this 7/51/13 version and a 5s/5r cycle. Here is the Roguecraft spreadsheet configuration.
The cycle used for the combat rogue is 5s/5r (not using tott on cooldown), using this cycle it is possible to replace the slice and dice glyph by a blade flurry one (the bugged AR glyph should provide slightly more dps) and to drop 1 point from improved slice and dice to put the fifth point in sword spec. The higher dps I obtained for combat was using Calamity's grasp MH, Hailstorm OH with this 7/51/13 version and a 5s/5r cycle. Here is the Roguecraft spreadsheet configuration.
You get higher DPS by going 15/51/5, 2/2 Imp. SnD, 4/5 Sword Spec, changing BF glyph to SnD glyph and changing cycle to 3s5r5e. Also not sure where the gear comes to your uploaded spreadsheet but you might want to gem chest and change FoFF to Grim Toll.
Small things, probably wont change anything but posted still.
I'm unsure if my conclusions out of this post are correct, but I find rogue scaling less come Endgame...
Maybe I interpreted the data incorrectly though...
Due to the Fact, that certain stats (i.e. hit) get totally worthless after reaching the cap for all ested classes except rogues, the increased scaling of rogues early on gets somewhat diminished for them. This makes them scale less than almost any other class early on, but more or less on par in midgame.
What is important however, is the fact, that near the end (30% statincrease) we scale least of all the shown classes, as we still have not reached our Hit-Cap, but all other classes already have done so by a rather big margin.
If you now take a look at the overall situation (rogues no longer number 1 DPS ot begin with, bad scaling early on, similar scaling later - though we already lag behind) I really hope for some improvement in the form of glyphs or something...
Could you try running a test across different classes to see how much each class scales with raid buffs? I believe this is a contributing factor to poor rogue performance in 10-man and small group content: we are more dependent on raid buffs than others.
Could you try running a test across different classes to see how much each class scales with raid buffs? I believe this is a contributing factor to poor rogue performance in 10-man and small group content: we are more dependent on raid buffs than others.
I suppose standard raid buffs are already included in the simulation. Just look at the DPS - there is no way to get to 5k+ DPS for a rogue in a 5-man.
Something to check though:
Did you exclude raid buffs and base stats from scaling, or did you just increase overall stats and ratings (including buffs) by a fixed percentage? Not that it would make a huge difference, but it would be nice to know.
I'm unsure if my conclusions out of this post are correct, but I find rogue scaling less come Endgame...
Maybe I interpreted the data incorrectly though...
Due to the Fact, that certain stats (i.e. hit) get totally worthless after reaching the cap for all ested classes except rogues, the increased scaling of rogues early on gets somewhat diminished for them. This makes them scale less than almost any other class early on, but more or less on par in midgame.
What is important however, is the fact, that near the end (30% statincrease) we scale least of all the shown classes, as we still have not reached our Hit-Cap, but all other classes already have done so by a rather big margin.
If you now take a look at the overall situation (rogues no longer number 1 DPS ot begin with, bad scaling early on, similar scaling later - though we already lag behind) I really hope for some improvement in the form of glyphs or something...
This is something I worry about, and have since before release. My guild is currently in 10 man Nax, and I feel as a rogue I bring nothing to the table. At the moment, the only thing I have going for me is good gear (relative to the rest of the guild), and thus high-end dps. But if one were to min/max among classes with equal levels of gear, I really see no place for a rogue in a 10 man. There are lots of melee dps options, and many other options bring vital utility (battle rez, buffs, off-tank ability, etc.). We have tricks of the trade...like tanks have aggro issues.
In a 25 man, sure you bring rogues since you have all buffs covered and may as well not shard loot.
We are melee dps (high chance of death and drain on healers as lots of raid damage ranged can avoid), we bring no buffs, have no AoE ability (fan of knives is fun but a gimmicky) but to counterbalance that...we have the same DPS as other damage specs. This isn't intended as a whine, as all content is certainly doable with rogues in the raid so we aren't useless, but from a pure theorycrafting best-raid-setup point of view, is there any place for rogues?
This is something I worry about, and have since before release. My guild is currently in 10 man Nax, and I feel as a rogue I bring nothing to the table. At the moment, the only thing I have going for me is good gear (relative to the rest of the guild), and thus high-end dps. But if one were to min/max among classes with equal levels of gear, I really see no place for a rogue in a 10 man. There are lots of melee dps options, and many other options bring vital utility (battle rez, buffs, off-tank ability, etc.). We have tricks of the trade...like tanks have aggro issues.
In a 25 man, sure you bring rogues since you have all buffs covered and may as well not shard loot.
We are melee dps (high chance of death and drain on healers as lots of raid damage ranged can avoid), we bring no buffs, have no AoE ability (fan of knives is fun but a gimmicky) but to counterbalance that...we have the same DPS as other damage specs. This isn't intended as a whine, as all content is certainly doable with rogues in the raid so we aren't useless, but from a pure theorycrafting best-raid-setup point of view, is there any place for rogues?
I'd say it depends largely on what encounter you are optimizing for.
If you don't have a ret pally or an arms warrior, then rogues can debuff the boss.
If you don't have a hunter, then tricks of the trade is useful for adding a threat buffer for the tank, or switching threat on the fly to a secondary tank, etc.
Also, even though rogues are in melee range, they have numerous cool down abilities to help them avoid damage.
I'd say it depends largely on what encounter you are optimizing for.
If you don't have a ret pally or an arms warrior, then rogues can debuff the boss.
If you don't have a hunter, then tricks of the trade is useful for adding a threat buffer for the tank, or switching threat on the fly to a secondary tank, etc.
Also, even though rogues are in melee range, they have numerous cool down abilities to help them avoid damage.
Although those are all true and honest, they are also situational and often unnecessary and/or better done by other classes. I saw somewhere on these forums the ratios for crit scaling per agil. point and it drops off more for rogues than other classes. That's why rogues start off easy to play and get caught in mid-early leveling then struggle to equal the damage of other classes later on.
You are looking at the glass as half full and that's always appreciated. Thanks for giving the tactics and tricks that can actually work. Realistically, there's some truth in saying that the best reason to bring a rogue is to pick locks though.
Until I got into Naxx 25 mans gear I was around 7th to 5th in dps. Now in Naxx 25 mans I'm 3rd to 1st. This is with equally geared players. I'm 15/51/5 swords.
Off the top of my head what a rogue bring to an instance: Kicks to stop casting, poisons to slow healing, more aggro (Tricks of the Trade) for tank, dismantle when it gets ugly, kidney shot to stop a whirlwind or help a healer whenever mob gets away from the tank, and I find that I don't really take much damage tbo. They can expect a melee to do better where casters will struggle and it balances out in the end. In 25 vault I was in aprox 15th or more away from most damage taken. I'm just not seeing the scaling problem either. Now I did see horrible dps before Naxx when compared to others but not now at all.
Would tend to agree with bradadad here. It's all very much annecdotal but with a complete epic set our once godly hunter (in similar gear) is now getting increasingly in reach. I have a few minor tweaks (switching slice glyph for adren rush till they patch it and pulling my old TBC head out of retirement re-gemmed/chanted) but in naxx10/naxx 25 (bout half and half) gear i'm usually around 3rd.
Well 300dps behind Warlocks and 100dps to the mages, not considering Movement puts us nowhere close to Ranged DPS
Movement works both ways.
Something like Grobbulus I have a really tough time competing with the Rogues and DKs that I typically pass by a large margin.
Either it's a fight where you're losing DPS time cause you're off the boss, or it's a fight where I lose DPS time by having to reposition myself while you maintain full rotations.
I disagree with the spec you choose, i understand the thinking with 30% more rupture but the 30% Critical Strike damage and chance at extra combo points after using Rupture, Eviscerate and i think even SnD by this build out way that. Talent Calculator - World of Warcraft Build) using Fist/Dagger.
As to how Rogue dps is setting ATM, i feel it may be a little lower than projected ATM but then again I am not fully optimal gear and i am having some hang ups with Rupture and SnD. (to much time left on SnD after finishing a 5r/5e section or not having enough AP to refresh rupture and setting on five combo points and three seconds left on rupture)*suggestions to this, 4s/5r/5e rotation ATM. I use Calamity's Grasp/ Webbed Death ATM but have not been able to test it in Naxx 25 yet was using Kel'Thuzad's Reach. My gear is statistically better than everyone else in my guild and on most Rogue friendly I am 1-3. The thing that concerns me is that i am better geared than most people, and beside some slight situational hiccups with rupture or having to refresh SnD to continue a rotation. I am still running near optimal rotations on most fights but am not toping the charts on allot of fights.
I think that all our players are equally skilled and it concerns me to see Fury Warriors, Death Knights, Mages, Locks and even sometimes our Feral Druid(Tank/dps spec) beating my dps on more than a few fights.(Not mentioning Hunter because they are OP ATM) It concerns me because they have yet to come close to the amount of optimal gear that my character has. I will have a chance to test this tomorrow in our 25 man Naxx run with Calamity's Grasp and will compare those WWS to others and link them here. I just think that for the lack of things we bring to a raid, (not counting situational encounters), that our dps is met or beaten to easily by other classes that do the same thing(Melee) but bring more to the raid. (No were not ment to top charts of everyfight, dont think OP like that)
I will link a spot to our WWS tommorow in this spot and i would only like people to look at dps done on what are considered Rogue friendlish fights. Patchwork, Kel'Thuzad, Gluth, and Noth.(am i missing any?) Also im going to attempt to not have to move in and out on Patchwork because of hatefull strikes and dps the whole time without getting (Slime Debuff).
I will try to add in hunters as well when I get home from work, just as a comparison. Mainly, I'm wondering how bad this steady shot nerf really hurt us, if at all.
You get higher DPS by going 15/51/5, 2/2 Imp. SnD, 4/5 Sword Spec, changing BF glyph to SnD glyph and changing cycle to 3s5r5e. Also not sure where the gear comes to your uploaded spreadsheet but you might want to gem chest and change FoFF to Grim Toll.
Small things, probably wont change anything but posted still.
I've seen a lot of posts on EJ lately discussing these two options. Every configuration I try on the spreadsheet favors 7/51/13, 4s/5r, 5/5 swords with BF glyph using CG and Hailstorm. I've tried messing with the raid buffs, different weapon/poison combinations, different levels of gear, and I just can't get 15/51/5 3s5r5e to come out ahead. They are usually within 20-30 dps of each other. Since I assume that most people who say "this is the best I can find" have also tried both, I must be missing something about how others are configuring the spreadsheet. Any ideas on what it is I could be missing that causes me to keep coming up with 4s/5r no matter what I do?
For those of you getting 3s/5r/5e as your max, are you skipping any buffs your raid doesn't have but many other raids probably would? Are you using an unconventional offhand? Are you gemming for agility or AP (or are you using gems to cap expertise or hit even though that often leads to a dps decrease?)? Are you using anything that is profession-specific? If I could do more dps with 15/51/5, I'd really love to know how so I can help my raid more.
12/10/08, 1:36 PM
Similar Threads
