b80f87b74934437aa8809cce41895434
I have also been fiddling with the fight length and the bis system and it’s telling me to stack haste as a 2h arms warrior on all cases. Am I maybe just missing something and I somehow just need to click a button somewhere that makes it work better?
I’m not seeing it suggest that you “stack” haste. I do see the iron counterweight becoming a higher-ranked enchant at certain fight lengths. The total value of those weapon enchants is pretty small, so when you end up in an area where 20 haste is the difference between an extra auto attack or not, it likes the haste. But it is only getting less than 2% haste.
I’ve gone back and forth on whether or not I should smooth that out or not (essentially allow a fractional number of auto attacks in the calculations), but I find it interesting to show people that the real difference between the different enchants/items can be affected a lot based on the fight length in reality.
it also suggests changing the meta gem, and it was really bad when I did the bis setup, recommended I drop to about 5% hit, and use just about every haste item/enchant available in phase 1. It might be a good idea to bounce ideas off the classic warrior discord, fight club, especially since the only widely available sim for warriors on there have their own problems (like having to manually input socket bonuses and not stacking cds quite right), so it could also be good advertising if people put up good parses using askmrrobot.
But yeah the haste might be scaled a little too high, the pally tank guildies I got to try this out on also may have memed similar stuff with it suggesting their bis trinket be dst.
When I look at BiS with the snapshot you gave me, it gets 0 haste. I’ll need you to make me a snapshot showing a build where it is trying to get a lot of haste for an arms warrior. Even setting it to phase 5 I’m only seeing 10% haste and no haste enchants or gems.
The prot paladin stuff just makes sense - they produce their threat mostly with consecration and SotR. Attack speed/haste is a very good stat for increasing threat via SotR. There is a setting to increase targets hit by consecration, which would move the focus away from haste and more towards spell power for threat. DST is an over-budget trinket that should show up in a lot of BiS lists as far as we can tell.
I’ve been over the warrior calculations so many times… I’m pretty confident in them at this point. Dropping hit rating is often a more efficient use of item budget, depending on the setup. You can use the customize section to force the hit cap if you don’t like playing with a higher miss chance. I know a lot of people just don’t like having that chance of missing.
2a185c53634f43ec890b81825135ebe2
I suppose the biggest parts of what has me confused at the sim is where it rates Thundering Skyfire Diamond and Abacus trinket over the vastly more commonly seen other options, the last hour or so I’ve been attempting to find a top end parse for arms anywhere that actually uses thundering meta gem (which if the haste meta was really 1% better than the crit dmg meta I’m sure it should have popped up somewhere on the first two pages of gruul or mag). I suppose I’m also confused as to why the sim is willing to calculate badge of the swarmguard or mark of the champion only when it’s equipped and seems to drop badge from the list when it views abacus as even slightly an upgrade when I have it calculate over and over again from 30s to 40s in 1s increases.
Another thing that has me very confused by the bis setup is why it’s recommending I wear the armorsmith only chestpiece and the swordsmith only sword at the same time. I’m told this is possible to do, but the chestpiece doesn’t work nearly as well if done.
Regarding the meta gem and the trinket: The differences between those items is too small to empirically measure. If you took those same players and swapped the meta gems and had them do the fight again, they’ll get results that are so close you could never tell the difference, statistically.
This is one of my biggest gripes with the gear design in WoW over the years, btw. The differences in items are so, so, so small that you cannot empirically feel a difference, much less prove a difference.
Is my model “right” where everyone else’s is “wrong” about those tiny differences? I don’t know. No one can know. I do know that I’ve gone through the math in my model and it picks all the gear you would expect 98% of the time. Why would it be wrong about these two items if it picks everything else “right”?
We don’t consider classic-era items unless you have them. Trying to keep the amount of calculations down so the site doesn’t choke. Warrior is the slowest spec for calculations already.
Right now I don’t think there is a way for us to know what blacksmithing specialization you have, so you might have to manually exclude a piece that you can’t effectively use from a different specialization.