I'm not sure which way to use tbh, I tried it Ask Mr.Robot site and Reforge Lite but I'm not sure if it's the best way to do it.
However before I was using WoW Reforge but since atm we don't have any mastery/haste cap, it's useless, atleast for me.
Wowreforge doesn't handle Hit/Expertise, it is not really usable for now, or only manually.
I use reforgelite, just put in your statweights and let it do it's thing.
The Arcane only corrupts the weak..
Yes, you determine your own statweights and then input them into reforgelite, you know you can change the values in Reforgelite yes? Btw not to be mean but I can't understand half of what you're trying to say, perhaps it would be better to ask this question on a forum where they speak your native language.
The Arcane only corrupts the weak..
I use SimCraft for stat weights and then Notepad+Calculator to decide how to reforge everything because I'm a nerd like that.
SimulationCraft for stat weights and then ReforgeLite for reforging. However, I also have spreadsheets where I put in all items relevant to my main, currently all Intelligence items of ilvl450 and upwards, and calculate the optimal total stat value from each item. That way I know before I enter any dungeon or raid, what is an upgrade to me and what is not.
Yeah. I have too much free time.
I wanna be a human being, not a human doing. I couldn't keep that pace up if I tried.
David, I'm one of the people behind Ask Mr. Robot. If you need any help with your character, give me the realm and name and I can help you out. Any specific questions you have regarding the reforging results or stat weights?
IMO askmrrobot is a perfect tool for the vast majority of what players want to do, as far as using it as a computational tool.
Even if you disagree with the weights it uses to optimize, you can always lock certain gems and enchants, and put in your own custom weights you calculated with simcraft.
If you're concerned about forcing it to make hit caps exactly (which is the major criticism I see about it), all you have to do is give hit an artificially high stat weight up to cap, but not so high that it interferes with your gemming / enchanting strategy (but even then all you would have to do is lock specific gems / enchants in place anyway).
If you know what you are doing, it's a perfectly useable tool. If you don't know what you're doing, then you're better off just trusting what it says blindly anyway.
1. Option to reforge to rating. You can make this the advanced option, or whatever, but there are various haste thresholds that are better expressed by rating than by percentage.
2. The equivalent of 'at least', when it comes to rating. Neither your hard cap or soft cap describe this. Yes, it is a DPS loss mostly, but basically it is frustrating when you can specify a 'Hit Cap' build but it doesn't actually cap on hit.
3. Additional builds - at the moment, you have the Default and the Hit Capped build (which, like I said, doesn't actually respect the hit cap, but whatever). But when it comes down to it, no matter what you say about hooking AMR with SimC, it just seems to be a static selection of stat weights that is usually wrong. It's not your fault - it's just a fault of the methodology - if relative stat weighting changes, especially with regards to Int/SP vs. Secondaries, depending on item level, then having a static set of weights will result in you being wrong all over the curve.
Of course, you don't need to model the entire curve - Pre-Raid, Tier 14 LFR, Tier 14, Tier 14 Heroic will give you the proper spread.
4. Power Torrent over Windsong, Elemental Force
I'm confused in some aspects about reforge and gems. For example, 80 intellect+160 mastery isn't best then 320 mastery? Why Mr.Robot advice me to put a 320 mastery on my belt?
Which are the best gems in correspondence with sockets? For me it's:
Blue socket- 80 int+160 hit
Red/ Yellow- 80 int+160 mastery
But why some guides or even Mr.Robot says that 320 mastery it's better then 80 int+160 mastery? Why 80 int+160 mastery instead of pure int(160)?
I definitely don't go for hit cap(15%) I rather go all over mastery and lost 1-2% on hit rating.
So who wins? Intellect or Mastery? I'm sure it's intellect on general.
Look at the stats weights on AMR :
Intellect = 4.53
Mastery = 2.82
160 Int = 725
80 Int + 160 Mast = 362 + 451 = 813
320 Mast = 902
Stat weights vary with gear as they are so interrelated to one another. Sites like AMR and the preloaded information in addons like Reforgelite are often based on BiS gear lists that you don't have, and are probably not accurate for you. If the makers of those sites/addons could put a pop-up, or leave the entry fields blank with an option to "insert BiS stat values" with a warning that they may not be accurate for your character, that would quite possibly be the biggest help those things could offer people, rather than giving out information based on values that there is a high chance of being wrong for them, and in turn offering out bad advice that is detrimental to them.
Last edited by Jessicka; 2012-10-10 at 10:21 AM.
Wow, excellent discussions going on here. I'll try to answer every question:
1. We don't allow you to put in a haste rating cap for a reason: For haste you want to reach a specific breakpoint... let's say 20% in a raid. Well, you get a 5% Haste raid buff, along with possible other haste boosts. This raid buff is multiplicative, for example... To get to 20%, you'd need 14.3% Haste on your gear (1.20 divided by 1.05). It takes 425 Haste Rating to get 1% Haste. So you'd need 6,078 Haste Rating on your gear (425 * 14.3). Since it's the final % that you're trying to reach, that's what we want users to put in.
2. Getting exactly to hit caps is usually a DPS loss, actually. Now, to be fair, it's so minimal that either way you do it, you won't notice in a raid. But, let me explain. First, I'm sure we can all agree that any points OVER the hit cap is wasted. Next, we can also agree that there is a finite set of ways to get exactly to the hit cap, based on the rules of reforging, etc. Now, let's say your best secondary stat is Mastery, and your worst secondary stat is Crit. Now, let's say to get exactly to the hit cap, you'd have to reforge all of your Mastery into Hit/Exp. You'd be exactly hit capped, but at the expense of your best secondary stat. Now, let's say Mr. robot can get you to 0.05% of the hit cap, but reforges all of your Crit into Hit/Exp. That would actually do higher DPS because you preserved your Mastery. I have some more info on this here.
Rustjive AND Jessicka:
3. Stat weights. We generate our default stat weights in the following manner: we build a program that interfaces with SimC and runs a ton of simulations with various gear sets. The program continues to run until it converges on a stable set of weights. That way, these weights can work across most gear levels. Sometimes, at very high gear levels, different stat weights actually change the optimizer results. In those cases, we add an extra stat weight preset. Then, when someone loads their character we check to see if they meet the conditions to require the different stat weights. If they do, we automatically select it for them. Fancy, huh?!
The other thing about having a stable set of stat weights is that your BiS gear list will be right. If you put in stat weights for only your current set of gear, it will change the BiS list that Mr. Robot suggests. You want the BiS list to be what you are going for, if that makes sense.
4. I'll look into the enchants. Here's more information of the stats you get from enchant procs (and trinket procs... it's on the same page). We model out the proc rate and uptime with a simulator, in order to determine the average stat increase you'd get.
It looks like zum answered your question. I have one thing to add: the socket bonus. If you're looking at a red socket and the bonus is worth it, Mr. Robot will put in the orange gem. If the socket bonus isn't worth it, or it's a yellow socket, Mr. Robot would put in the Mastery gem.
I understand that it is a DPS loss in most situations, because often reforging to an 'at least' cap will bring you above the actual rating needed. However, what AMR does is it restricts the user on the basis of 'oh, we know better' - but nothing is further from the truth. There are legitimate reasons to want to cap on Hit - maybe people just hate losing their rhythm due to misses. But the other reason to implement 'at least' instead of just the concept of Hard and Soft caps is because Haste thresholds work in the complete opposite manner as Hit caps. Especially for Affliction, in between thresholds Mastery is simply a better stat than Haste, and the value of Haste continues to diminish up until the point of the threshold. The marginal value of that 1 single point of rating that brings you above a threshold is huge. There's no way to model this with hard and soft caps short of trial and error.2. Getting exactly to hit caps is usually a DPS loss, actually. Now, to be fair, it's so minimal that either way you do it, you won't notice in a raid. But, let me explain. First, I'm sure we can all agree that any points OVER the hit cap is wasted. Next, we can also agree that there is a finite set of ways to get exactly to the hit cap, based on the rules of reforging, etc. Now, let's say your best secondary stat is Mastery, and your worst secondary stat is Crit. Now, let's say to get exactly to the hit cap, you'd have to reforge all of your Mastery into Hit/Exp. You'd be exactly hit capped, but at the expense of your best secondary stat. Now, let's say Mr. robot can get you to 0.05% of the hit cap, but reforges all of your Crit into Hit/Exp. That would actually do higher DPS because you preserved your Mastery. I have some more info on this here.
I know how this works. The issue is that right now you only have 1 build, 1 set of stat weights (ignoring the hit cap build). This set of stat weights is nowhere close to what stat weights for current gear levels are like - therefore at this moment in time AMR is just simply wrong.3. -snip-
The reason PT > WS is because it doesn't proc on DoTs, and it procs on other things that are free (Main Gauche, for example). The PPM mechanic works properly, but the underlying assumptions are wrong.4. I'll look into the enchants. Here's more information of the stats you get from enchant procs (and trinket procs... it's on the same page). We model out the proc rate and uptime with a simulator, in order to determine the average stat increase you'd get.
Last edited by Rustjive; 2012-10-11 at 03:00 PM.
1. Tanks: not hitting hit cap is unacceptable if you depend on "active mitigation" effects happening 100% reliably.
2. Relying on Crowd Control / Utility effect hitting: highly situational, but still possibly very relevant. As of early Cata this is not relevant to interrupts anymore, but certain effects such as hunter feigning and such, if relevant to how an encounter is executed, expose this issue.