Vontre's Magegraf, advanced theorycrafting tool for mages. - Page 4

LockApologist · 05/30/08, 1:31 PM

UD Lock output. Full Debug output.

Shadow Damage:	 1227

My armory shows 1247.

Mark of the Wild	 = 	spirit: 29 => 47.9
Divine Spirit	 = 	spirit: 67.9 => 117.9
Blackened Basilisk	 = 	spirit: 47.9 => 67.9
Blessing of Kings	 = 	spirit: 117.9 => 129.69

My armory shows 163 Spirit unbuffed. The script assumes 29 base. UD lock should have 27 Spi at lvl 1, so I'm not sure where 29 came from (143 @70). Additionally, Demonic embraces -5% spirit and Fel Intellect's +3% Int do not seem to be listed in the output. Are they accounted for?

Shadow and Flame	 = 	DoT bonus damage coefficient: 1 => 1.2

SnF doesn't affect DoTs.

Chaotic Skyfire Diamond	 = 	base critical multiplier: 0.5 => 0.55

CSD modeled as 10% increased crit, rather than 3%

How are you determining ISB uptime?

How is the Hyjal ring proc modeled?

The mana usage output seems to assume I do not have any int buffs, and is using my base mana pool. The graf assumes AI, MotW, etc, but does not apply the int increases to mana pool.

The amount returned by Life tap seems low. LT gives 580 + .8 x +shadow. I know I raid with ~1650 + shadow, so should LT for ~1900. Sim assumes ~1600.

http://magegraf.com/index.php?hash=8...55c7e819fefa57

LockApologist · 05/30/08, 3:39 PM

Also, can you add an option for Brilliant Wizard oil over Superior?

Thanks.

Vontre · 05/30/08, 3:43 PM

Glove enchant for spell damage had the wrong import id. This has been fixed.

Spirit calculations for warlocks have been neglected due to warlocks not using spirit. I have updated these fields for WLK preparation.

Nowhere does it say that SnF doesn't affect dots, it only affects shadow bolt and incinerate which don't have dot components. So this is actually correct.

CSD has been fixed to 3%

Fel Intellect and Demonic Embrace have been added.

The mana usage output is correct.

Life Tap is being calculated correctly, it just doesn't show the modified mana total until another bolt has been cast.

LockApologist · 05/30/08, 4:40 PM

Originally Posted by Vontre

Glove enchant for spell damage had the wrong import id. This has been fixed.

Spirit calculations for warlocks have been neglected due to warlocks not using spirit. I have updated these fields for WLK preparation.

Nowhere does it say that SnF doesn't affect dots, it only affects shadow bolt and incinerate which don't have dot components. So this is actually correct.

CSD has been fixed to 3%

Fel Intellect and Demonic Embrace have been added.

The mana usage output is correct.

Life Tap is being calculated correctly, it just doesn't show the modified mana total until another bolt has been cast.

1.) Thanks
2.) Warlocks don't use spirit for regen, but it makes a difference of 10-15 +dmg when using imp. Divine Spirit.
3.) True. As long as the dot coeff doesn't get applied to Immo or something, it shouldn't matter.
4.) Thanks
5.) 2.34: 9822.2816627907 This is the fire line in the Mana Consumption output. The sim is showing 10115 for starting mp (293 effective SB mana cost w/ 290 mp5). Now, I know i raid with just over 11k buffed mp, so this is not correct. I have 9790 unbuffed mp, and according to the sim, I gain 112 INT from buffs. That should result in 1680 more mana, not 325.
6.) And, you're right about Life tap. I incorrectly read the mana as mana before LT -> Mana after LT, when it's actually Mana before LT -> Mana after LT + SB cost. SB cost explains the 'missing' mana.

manly · 06/03/08, 8:11 PM

Found a small bug. Loading my gear works, although the 12 crit from CSD is not added to my crit rating.

embrauer · 06/05/08, 7:20 PM

Great little tool. I had a similar thing coded up for personal use, but I was far too lazy to take it to the level that you have it now. While I understand your primary concerns are working out bugs, I have a couple suggestions for cosmetic/utility improvements. It would be nice to have an 'overall' stat summary, as I'm sure most of us are concerned with final buffed numbers (damage, crit, hit, etc) rather than trying to mentally add the numbers to the unbuffed raw numbers that are currently displayed. And possibly included in the aggregate stats summary an indication of the % values of hit, crit, and haste. I don't always remember what 233 crit rating translates to in percentage terms. While I'm sure (and indeed you've mentioned it) Fel Armour and AI are implicitly assumed, it would be nice to see some indication of this for 'completeness' sake'. Finally, it's been mentioned but it would be nice to see spell strike added to the gloves enchant list.

Edit: Also this is happening in the effect timing output

Shadow/Destro
2.36: Icon of the Silver Crescent
2.36: Destruction Potion
2.36: Destruction Potion
4.2: Icon of the Silver Crescent
4.2: Destruction Potion
4.2: Destruction Potion
4.2: Bloodlust
6.03: Icon of the Silver Crescent
6.03: Destruction Potion
6.03: Destruction Potion
6.03: Bloodlust

I assume this is just an issue with the output and not actually doubling the effect of the destro pots.

manly · 06/05/08, 7:57 PM

The doubling of destruction potion is intentional. Its a side effect from the backbones of the code. Basically, destruction potions are coded as 2 buff effects; one of which is the 2% more crit, the other one being the 120dmg. It is, in other words, mostly a display issue.

Aaliyäh · 06/05/08, 10:23 PM

I had a little problem with the armory import because of the umlauts in my name.

Edit: Sorry, I just noticed that the umlauts issue was already addressed.

A solution is to visit a UTF table, for example Unicode/UTF-8-character table and see what is the hex representation of the special character, and then use the % escape character.
For example, ü = c3 bc. So replace it with %c3%bc.
Then, when you enter your Server name, instead of "stürme" for example, type "st%c3%bcrme"

Gort · 06/08/08, 6:41 AM

Small bug: [Reckless Noble Topaz] isn't in the DB, it seems, and sockets with that gem are treated as empty.

--

I can see this sucking far too much of my time. Is there a way to adjust talents used? I've been dorking around with a fairly suboptimal 18/0/43 "Man, I wish I had a Shadow Priest" build for the odd Karazhan run, and I'd be a bit interested to see what it looks like as compared to a 10/0/51, etc.

Arnath · 06/09/08, 11:56 PM

Most simulators suffer from a pretty high level of stat granularity. What I mean is that, for example, haste rating only shows up as a bonus in dps at certain breakpoints and thus doesn't show up as a useful stat in comparison. Can I ask what Magegraf does to avoid this?

frijolefrito · 06/10/08, 12:09 AM

Can someone link me how destro pots equal more damage then mana pots as far as dps.

Docjowles · 06/10/08, 10:57 AM

I am going to go out on a limb and say it's fairly obvious. If the encounter is structured such that you won't go OOM, the added mana from a mana pot does nothing for your DPS, whereas the added damage and crit of the destro pot does. If you have a well-played shadow priest in your group, judgement of wisdom and/or a shaman dropping mana totems, that is not an unlikely situation.

To be fair, it has been concluded that swapping from destro to mana pots is the first tradeoff you should make if you have mana troubles (http://elitistjerks.com/658230-post3191.html). But if your mana pool can handle going all-out, destro pots surely do boost your damage.

LockApologist · 06/10/08, 1:16 PM

Originally Posted by Docjowles

I am going to go out on a limb and say it's fairly obvious. If the encounter is structured such that you won't go OOM, the added mana from a mana pot does nothing for your DPS, whereas the added damage and crit of the destro pot does. If you have a well-played shadow priest in your group, judgement of wisdom and/or a shaman dropping mana totems, that is not an unlikely situation.

To be fair, it has been concluded that swapping from destro to mana pots is the first tradeoff you should make if you have mana troubles (http://elitistjerks.com/658230-post3191.html). But if your mana pool can handle going all-out, destro pots surely do boost your damage.

Mage mana mechanics are significantly different than warlocks (and we definitely hear about it every time 'who gets the spreist?' is asked

), so it's not exactly that cut and dry. While what you stated is probably correct for mages, locks have no other trade off to make than destro vs mana pot (You should always be using Dark runes, unless healthstone is required). Generally, locks will use their mana pool several times over for most encounters, so 'will I go OOM?' is not a valid metric.

For locks, the questions are:

a.) Do I have to life tap at all?

This is not usually an option, but some fights are short enough, that with BoW, JoW, a good spriest, etc you can go a little over 2 min before you'd need to tap. In that case, destro pots are clearly superior.

b.) Will I have free life tap time?

Fights with a lot of movement can give free tap time, reducing the value of mana pots. For example, Gurtogg is a good example of where mana pots are not a strong as destro (well, the fight mechanic helps with that as well). Illidan and Kalec are others. Lurker for a T5 example.

c.) Will I get a Bloodlust/Heroism?

Generally, if you can get through the duration of the buff w/o tapping, destro pots will be better than mana for that particular cooldown.

Docjowles · 06/10/08, 2:00 PM

My fault, I didn't catch that he was asking as a warlock.

Frosty2 · 06/10/08, 2:35 PM

I would like to start off by saying thank you to Vontre for doing such a good job on this tool. I am pretty sure that there is an error that has occured, because last time iused this i was at 1791dps (+1592 mana), but for some reason i'm getting 1724dps (-14298 mana) and none of my socket bonuses are working.

here's the link
http://www.magegraf.com/index.php?ha...926a87a79de342

cbags · 06/10/08, 3:48 PM

The Spell Hit to gloves is not listed, I know most people T5 and above don't have to worry about it....but I still suck. But I am working on not sucking so much, please ignore armory.

Beregon · 06/10/08, 4:44 PM

Originally Posted by cbags

The Spell Hit to gloves is not listed, I know most people T5 and above don't have to worry about it....but I still suck. But I am working on not sucking so much, please ignore armory.

When I model it in Rawr, even uncapped in Spellstrike/Spellfire I get slightly more from the +20 damage enchant than from the +15 hit enchant on the gloves. There may be examples of situations where that isn't the case, but in most cases given +hit vs +damage is a 1:1 tradeoff on gems the +hit to gloves is an inferior enchant: If you gem +hit and enchant +damage you can get more of both.

cbags · 06/10/08, 5:06 PM

My raid gear is 3/3 Spellfore and 2/2 Spellstrike, and I am at 164 hit...I was just commenting that it's not listed. For the most part I am at opr slightly over cap, but I have stuck with the thought that Hit to Cap, worry about the rest later. Unless you can't GET to cap.

The thing is that right now, in that gear, I am not capped, and I know that the gloves HAVE the +15 Hit on them and it's not being reported.

Vontre · 06/10/08, 8:25 PM

Originally Posted by Arnath

Most simulators suffer from a pretty high level of stat granularity. What I mean is that, for example, haste rating only shows up as a bonus in dps at certain breakpoints and thus doesn't show up as a useful stat in comparison. Can I ask what Magegraf does to avoid this?

The stat comparison test for intellect, spirit, and haste rating are all +100 value. This seems to work alright for the stat analysis mode, but the regular graph does indeed have some breakpoints, but they are rare. You are probably used to a simulator which uses a fixed duration and snipes off the last spell. Magegraf actually finishes the last spell cast and calculates your dps over the actual duration (301.2 seconds or whatever), so the only breakpoints occur due to scorch timing and timers. I could potentially mesh the scorch and fireball rotation into an average to alleviate this problem. The quick fix, though, is to just look at the spell detail value, or enable fire vulnerability so the model just skips over scorch.

Swindley · 06/11/08, 6:39 AM

I'm curious about the calculations/benefits on flame cap and destro pots.

http://magegraf.com/index.php?hash=1...366f5829373eb3

First of all, it seems to be missing some of my +dmg bonus total, and I can't really figure out where (+22dmg missing)

I did the calc with some common raid buffs, with the exact same stats, but switching both mana gems & pots for flame caps and destro pots, and the dps increase only went up +18.
Is the advantage of theese consumables really that low or is the calulations off?

Off course, I'm assuming it wouldnt know how to perfectly time this with other cooldowns, (destro during icy veins+heroism under 20% etc), but it seems it wouldnt be worth switching for such a small dps increase at the cost of risking so much of your mana pool? (or maybe you'd need a healthstone use or evo gets broken)

My experience with flame caps and destro pots is rather limited tbh, because most of the time I need the mana instead, and short unimportant fights I don't find it worth to actually spend them on, costwise, but hoping for some input from others that use them more often.

Also, it seems to be missing Timbals Focusing Crystal, and the option to use elixirs instead of flasks.

Lysistrate · 06/13/08, 9:24 AM

This tool is really nice... Thanks a lot...

One minor bug, which does not need fixing right now: you can select a 2H staff and offhand at the same time and get stats from both... and, I played around a bit, had a MH OH combination equipped and imported the original char from armory.. the OH stayed equipped and was not cleared, so maybe you want to reset all slots before import.
Another thing, is of course the changed trinket stacking which has been mentioned already... I am really interested if serpent coil + skull of guldan beats hex + skull, because stacking (destro+skull+mana) is still possible with the mage trinket but not with hex.

Great tool this saved me a lot of time, I was already thinking in programming the exact thing as well... If I can assist you with coding, feel free to PM me.

edit:
Timing of coldsnap is not right as well. it should be used of course in the execute phase, because double icy veins within bloodlust and trinkets etc is of course better.

Naefr · 06/13/08, 12:28 PM

its not picking up some gems.

http://www.magegraf.com/index.php?ha...cd08e4f25d3b0a

Vivid Chrysoprase - WoWWiki - Your guide to the World of Warcraft - heroic drop
Forceful Seaspray Emerald - WoWWiki - Your guide to the World of Warcraft
Shining Fire Opal - WoWWiki - Your guide to the World of Warcraft - heroic drop

embrauer · 06/14/08, 1:53 PM

Originally Posted by Beregon

When I model it in Rawr, even uncapped in Spellstrike/Spellfire I get slightly more from the +20 damage enchant than from the +15 hit enchant on the gloves. There may be examples of situations where that isn't the case, but in most cases given +hit vs +damage is a 1:1 tradeoff on gems the +hit to gloves is an inferior enchant: If you gem +hit and enchant +damage you can get more of both.

Well, not entirely. If you have gloves with two sockets, putting in two [Great Lionseye] and enchanting +20 damage, is for all intents and purposes, the same as putting in two [Runed Crimson Spinel] and enchanting +15 spell hit. The former yields 20 hit and 20 damage, and the latter 24 damage and 15 hit. Whichever scenario you choose will be dependent on your level of hit and gem bonuses (throughout your gear). While I agree that having 20/20 is marginally better than 24/15 (assuming you need 5 hit to get to the cap) having the Spell Strike enchant is useful for modeling purposes.

Roywyn · 06/14/08, 4:10 PM

Originally Posted by embrauer

Well, not entirely. If you have gloves with two sockets, putting in two [Great Lionseye] and enchanting +20 damage, is for all intents and purposes, the same as putting in two [Runed Crimson Spinel] and enchanting +15 spell hit. The former yields 20 hit and 20 damage, and the latter 24 damage and 15 hit. Whichever scenario you choose will be dependent on your level of hit and gem bonuses (throughout your gear). While I agree that having 20/20 is marginally better than 24/15 (assuming you need 5 hit to get to the cap) having the Spell Strike enchant is useful for modeling purposes.

2 spinels + hit enchant = 24 dam, 15 hit.
1 lionseye + 1 pyrestone + damage enchant = 26 dam, 15 hit.

2 rubies + hit enchant = 18 dam, 15 hit.
2 dawnstones + damage enchant = 20 dam, 16 hit.

4 rubies + hit enchant = 36 dam, 15 hit.
4 topazes + damage enchant = 40 dam, 16 hit.

The spell hit enchant is lower in budget, which is why it will alway be inferior, unless you can't cap hit otherwise.

Vannik · 06/16/08, 2:38 PM

This may be nitpicking, but could you add values for the Y-axis on the plots?