However when I run my toon through simcraft, I am getting a dps loss by putting on Circle of Ossus. I've tried running different Iterations for different lengths and it always comes out as a loss.
What am I doing wrong with simcraft? I assume I'm having a configuration issue.
I have plenty of blue sockets. I only break socket if the bonus is <5.
I assume I am doing something incorrect with simcraft to give me this strange result.
I'm comparing me with this profile. The only difference should be the belts themselves as I've even gemmed them the same.
I was just planning out my next eof purchase and trying to get the biggest bang for eof. I took it as a given that Circle of Ossus is better than Cord of the Patronizing Practitioner. Looking at the stats I still believe that. I must be doing something wrong when I am running simcraft.
I have plenty of blue sockets. I only break socket if the bonus is <5.
I assume I am doing something incorrect with simcraft to give me this strange result.
I'm comparing me with this profile. The only difference should be the belts themselves as I've even gemmed them the same.
I was just planning out my next eof purchase and trying to get the biggest bang for eof. I took it as a given that Circle of Ossus is better than Cord of the Patronizing Practitioner. Looking at the stats I still believe that. I must be doing something wrong when I am running simcraft or there is a tipping point in stats (I would assume haste) where having too much would throw off other items.
No professions on the second profile. Adding professions=alchemy=450/tailoring=450 raises the DPS from 10486dps to 10556dps.
I have been running some Sims and I have noticed that a slight change in actions is giving me 40 more DPS for Warlock_T10_56_00_15 and I am not sure why.
Also I was doing some tests as to what sequence of actions would give me the most DPS with T10_4p Corruption rolling and without Improved Scorch. I tried conditionals such as actions+=/shadow_bolt,time<=1 and actions+=/shadow_bolt,isb_trigger=1 before Corruption but in the end the following gave me the most DPS out of all. I don't understand why the DPS isn't calculated without proper Corruption rolling in the OP since the gap is quite big.
got a quick questions. I was wondering about the 56_00_15 Specc. Why do we skill 2/2 improved Life Tap if we also skill Dark Pact? I thought with using Dark Pact, there is no need for Life Tap anymore especially it gets some help to our healers.
So i'm a little confused now. Would be very nice if you guys can help me witha quick awnser.
As a 00_56_15, I keep a close eye on my spell power value for demonic pact, especially when comparing gear/rotation options that can often result in very close dps values. I noticed a few things about the Raid-Buffed Spell Power output value that anyone tracking spell power should be aware of, and I did not find these already mentioned:
(1) I get two different values for RBSP that happen to differ by 10%. Based on the text logs I believe it depends on whether Demonic Pact takes effect (i.e., felguard crits) before or after the snapshot_stats action is executed. This can be misleading if the difference in spell power happens to coincide with changes in gear or rotation.
(2) The first life_tap action is after snapshot_stats and so the 20% spell power from spirit via Glyph of Life Tap is not included in the RBSP value. Of course this won't matter after spirit is phased out for dps.
(3) I use Hellfire to proc Muradins Spyglass before the fight starts, so technically that is another 180/198 spell power that is not reflected in the RBSP.
If I have not misinterpreted anything, the first one might be considered a bug for the 56/15 users, though I'm fine with it as long as I understand what the output means. The second can presumably be accounted for by modifying the actions list if desired, and I don't know yet if there is a way to force a trinket to be at max stacks at time=0.
Demonology action list improvements:
1: Remove curse of agony from the action list, soulfire > coa
2: Immolate above corruption
3: Move the life tap triggered by mana<14k all the way down to just above shadow bolt
4: add "soul_fire,decimation=1,molten_core=1" 1 line above corruption
5: make immolation only trigger when metamorphosis duration < 15s, or 4xT10 bonus is up.
All these changes are only minor dps increases, but they all do increase your DPS, My action list right now looks like this:
Again, these are all very small DPS increases, but theyre all improvements. For example, the DPET of immolation aura went up from 24840 to 27032 thanks to fishing for a 4xT10 proc
The mind is like a parachute. It works best when it's open.
The last two lines of that action list are redundant; essentially, you have "Life Tap if [stuff] otherwise Life Tap." Also, since "/life_tap,buff_refresh=1" is a higher priority action, the more restrictive "/life_tap,mana_percentage<=20,buff_refresh=1,moving=1" will never be executed at all.
It's better to cast CoA than nothing, so "/curse_of_agony,moving=1" before the final Life Tap action provides a tiny DPS increase on movement fights (33 DPS for you on Helter Skelter) without affecting Patchwerk DPS.
I'd be concerned that prioritizing "/soul_fire,decimation=1,molten_core=1" before "/immolation,if=buff.tier10_4pc_caster.up" might make you miss the 4T10 buff on that Immolation, though I admit it's not a very likely occurrence. Swapping those two actions gives a whopping 1 DPS increase on Patchwerk, clearly within the margin of error, so I'm probably just nitpicking.
In the discussions of how to properly set up the demonology tree for testing, I was wondering if there was a way to only show the amount of demonic pact buff you contribute. I am still relatively new and this may be common knowledge but i was trying to read through and understand the program and couldn't seem to find the information.
Basically I am the only Demonologist on our 10/25 man raids and I wanted to know what my dps would look like if only my demonic pact is calculated ( and I apologize if this is already done). This is to show the worth and merit of spirit over crit for a rDPS spec.
I'm a bit confused when looking at some of the numbers in the Detailed Output section of the first post. When looking at "Spell Stats" different categories are broken down (ie: power, hit, crit etc..) followed by 3 different numbers, like this: power=4223|3425(3425). So if we were to replace the numbers with letters ie; power=x|y(z), what do each of those letters represent? X=max?, Y=base?, Z=numerical representation?
I'm a bit confused when looking at some of the numbers in the Detailed Output section of the first post. When looking at "Spell Stats" different categories are broken down (ie: power, hit, crit etc..) followed by 3 different numbers, like this: power=4223|3425(3425). So if we were to replace the numbers with letters ie; power=x|y(z), what do each of those letters represent? X=max?, Y=base?, Z=numerical representation?
Thanks for you help in advance.
stat=buffed|unbuffed(amount from gear)
So for, e.g., crit=41.61%|34.98%(1094) the simulated gearset has 1094 crit rating, the Warlock has 34.98% unbuffed crit chance - which would include that crit rating, as well as crit chance from intellect and talents - and 41.61% crit chance including raid buffs at the pull.
Note that if you just switch life_tap vs. immolate, you loose ~10dps. You need to enhance your mana buffer pre 35%, because there won't be any life_tap after you go execute.
The new simcraft syntax might be a little bit more complex, but note that you should really switch if you got a buff you have to react to, like bloodlust, molten_core,tier10_4pc.
Edit: Yeah, forgot about demonic pact, but I wouldn't call it a "lot", it's about 8 sp (398 spirit * 20% * 10%). Still, that would be more raid dps than the ~20 personal dps you gain.
Im thinking if we add a line somewhere slightly above actions+=/shadow_bolt, check if glyph of lifetap has less than... say 4 seconds left on it- then life tap; should increase DPS slightly. We'd increase glyph of lifetap uptime and reduce the number of times where we have to lifetap before we can refresh on of our dots.
Demonology is still going OOM sothe extra life taps shouldn't be wasted GCD's- just better spread GCD's.
[edit]
After some brief testing, it seems that the 14k mana trigger already produces the same behaviour. I think. Either way the DPS gain is < 5.
Last edited by Warlocomotif : 05/14/10 at 2:33 PM.
The mind is like a parachute. It works best when it's open.
I have a question about dps output vs machine performance, I haven't seen much discussion about this. I am running an older machine (read as damn near antiquated, single core p4 with 1.2gb ram and shotty video card). Based on spreadsheet analysis and simcraft I should be running an average of about 6500dps (no raid buffs). I know thats ideal, and would be happy if I could get my target dummy dps anywhere close to the 6k mark. I generally put up about 4.5k target dummy dps, I know some of it has to do with human error. I am also wondering how much of that can be attributed to machine error as well.
Using easy numbers for napkin math, say my machine "loses" (due to slower performance and lag time between the server) 10 seconds of fight time for every minute of in game fight time. At an average dps output of 5k, that amounts to a loss of 50k damage done for that 60 seconds period, or 833.33dps lost for that minute of fight due to machine performance.
Am I way off track in thinking this is an actual issue? Does anyone know how to calculate the "lost" time?
Thank you for any input/advise.
I have a question about dps output vs machine performance, I haven't seen much discussion about this. I am running an older machine (read as damn near antiquated, single core p4 with 1.2gb ram and shotty video card). Based on spreadsheet analysis and simcraft I should be running an average of about 6500dps (no raid buffs). I know thats ideal, and would be happy if I could get my target dummy dps anywhere close to the 6k mark. I generally put up about 4.5k target dummy dps, I know some of it has to do with human error. I am also wondering how much of that can be attributed to machine error as well.
Using easy numbers for napkin math, say my machine "loses" (due to slower performance and lag time between the server) 10 seconds of fight time for every minute of in game fight time. At an average dps output of 5k, that amounts to a loss of 50k damage done for that 60 seconds period, or 833.33dps lost for that minute of fight due to machine performance.
Am I way off track in thinking this is an actual issue? Does anyone know how to calculate the "lost" time?
Thank you for any input/advise.
Quoting the addon's page:
"This add-on monitors the time between finishing one cast and starting another, providing this feedback to the user so that they can reduce this lost dps time. It's aim is to help improve this ABC - always be casting. It currently rates the gap as follows, with a summary at the end of combat."
If you have a good latency to the sever ( 100ms < ) this addon should always show something between 0 and 50ms. Lesser is preferable. Zero is no dps loss.
But if you get some big values (150+) then you are experiencing some kind of problem, and if your ISP is good, that's probably hardware latency.
At the end of the fight it shows a summary, with a % of lost time. That's a good index in patchwerk-style encouters, where your dps lost should be something less than 1%
It also does some cast rotation tips which I always ignore lol
Changed a few pieces of my gear and now I am over hit cap, tried many items combo that brings me as much as possible the hit cap without loosing much DPS stats and all I am getting is strange results, even being over the hit cap, SimCraft tells me that the better gear combo is that which gives me more hit,.
I know the gloves is 13 Ilvl higher and gives more spellpower/haste etc, but I am not that confident with the results, it just go way way over hit cap and ignores much crit.
Checked the "Plot - Hit" and it shows the SimCraft "know" I am already over hit cap (M% = 0), the graphs show that if I get more hit the DPS won't change. I don't get why so overestimated haste/hit values.