ShadowCraft: Rogue spreadsheet for 4.0 - Page 20

Antiarc · 03/29/11, 5:30 PM

Aha, I know what's going on. The socket bonus gets divided among all the sockets in the piece, but that naively assumes that -none- of the sockets match. What I need to do is only divide the bonus among non-matching sockets to get a more accurate number for the socket bonus.

sinnaa · 03/29/11, 6:01 PM

Originally Posted by Antiarc

Aha, I know what's going on. The socket bonus gets divided among all the sockets in the piece, but that naively assumes that -none- of the sockets match. What I need to do is only divide the bonus among non-matching sockets to get a more accurate number for the socket bonus.

I was running into the same thing regarding socketing yesterday. From what I could gather, it just seems to update the gemming according to an individual's stat values. For my rogue, the value of spell hit had been decreased from 1.4 to 1.34 and after using this value instead, the numbers did work out where delicate was better than glinting. Perhaps there should be a little bit more info on the individual stat values and how they are used in the calculations. This could clarify things for many of the users. Thoughts?

Antiarc · 03/29/11, 6:40 PM

Originally Posted by sinnaa

I was running into the same thing regarding socketing yesterday. From what I could gather, it just seems to update the gemming according to an individual's stat values. For my rogue, the value of spell hit had been decreased from 1.4 to 1.34 and after using this value instead, the numbers did work out where delicate was better than glinting. Perhaps there should be a little bit more info on the individual stat values and how they are used in the calculations. This could clarify things for many of the users. Thoughts?

It should be using the EP value of the gem, plus the EP value of the socket bonus if applicable. Basically, it checks:

a) What is the best gemming setup (ignoring the socket bonus)?
b) What is the best gemming setup (taking the socket bonus)?

It then compares the two of those, and *should* select the one with the higher EP value. Obviously something is breaking it.

On a general note, I've implemented the 4.0.6 [Unheeded Warning] and pre-potting.

nextormento · 03/29/11, 8:26 PM

Originally Posted by Antiarc

On a general note, I've implemented the 4.0.6 [Unheeded Warning] and pre-potting.

Un-checking prepot is giving me a higher dps drop than unchecking the normal combat pot, which is unintuitive given that the prepot you seem to implement has a 2sec lower duration than that of the combat pot.

Antiarc · 03/29/11, 8:47 PM

You're right - I'll take a closer look at how the engine adds the combat pot. I just averaged the agility gain out over the course of the fight.

Limegreen · 03/29/11, 9:16 PM

Not sure if this has been stated yet, but when I optimize gems, my cogwheels come out random. I've seen crit, resilience and I think even spirit.

impatigo · 03/29/11, 9:39 PM

Originally Posted by Limegreen

Not sure if this has been stated yet, but when I optimize gems, my cogwheels come out random. I've seen crit, resilience and I think even spirit.

I had the same issue. BUT, looking at it if I optimize gems and optimize reforge it shows a DPS increase. Is it right? I don't know. It changed my hit cog to haste and reforged other gear in to hit.

sinnaa · 03/30/11, 12:07 AM

Originally Posted by Antiarc

It should be using the EP value of the gem, plus the EP value of the socket bonus if applicable. Basically, it checks:

a) What is the best gemming setup (ignoring the socket bonus)?
b) What is the best gemming setup (taking the socket bonus)?

It then compares the two of those, and *should* select the one with the higher EP value. Obviously something is breaking it.

I am well aware of what it should be doing, and I was trying to explain why the "error" is occurring, as i already see you addressed the one with multiple sockets available. I am not sure if what I am about to say plays into that at all, but i doubt it does. Let me just give you an example.

Say you have the Wind Dancer's Spaulders, which has a blue socket. You have the option of a Delicate Inferno ruby or a glinting demonseye. At an EP value on agil of 2.7, the delicate is worth 108 EP, based on the values found in the combat guide. Now, the glinting gem by itself is 20agil*2.7EP+ 20hit*1.4EP (assuming you already have special hit cap) = which has a total value of 84. Additionally, you get the 10 agil socket bonus so you have to add another 27, which results in 109... indicating the glinting is superior.

NOW, the reason the program says that delicate is better is because its not using the ep values from the combat guide. I believe its using EP values based on specific gearing of a character (correct me if i am wrong here). So what is really happening in the simulation is this.

Glinting Demonseye = 20agil*(2.7ep)+20hit*(1.34EP) = 80.8. Again, taking into the 10 agil socket bonus, you get an additional 27 EP but now the total value of the gem, obtaining the socket bonus, is 107.8, which is less that 108, which is why it selects the delicate when people expect it to pick the glinting.

I believe these EP value imbalances are what is "breaking" it, and the possibility of it splitting bonuses across multiple sockets. I could be wrong, but the math differences, while slight, could explain why with one blue socket items with 10 agil bonuses people dont get what they expect.

ANYWAY, i was simply suggesting putting a source of the weights or mouseover popup information for users who are not as familiar with the way the values are used and change.

logist · 03/30/11, 9:12 AM

Originally Posted by 95Cedric

Just tried the reforge all button on the Shadowcraft tool and I am getting a 'Error contacting reforge service' at the upper right of the screen. This is in IE9.

Same as above. Previously was just black screen then the circle in the middle nothing happend, now I get same error upper right of the screen.
Something i forgot to enable?

Picko · 03/30/11, 11:12 AM

Awesome job with this dude, really loving it. The only problem I have is that Engineering seems to be giving me only 87.7 DPS while alchemy is giving me 246.3 DPS which is kinda odd considering Assassination Guide for Cata thread just the Synapse Springs from Engineering gives somewhere around 80 Agility and Alchemy also gives 80 Agility so it should be the same amount of DPS gain. I am not really sure if I am doing something wrong or it's some kind of a bug or even the other thread is wrong but I would like to figure it out. Thanks.

dinesh · 03/30/11, 2:16 PM

Originally Posted by Limegreen

Not sure if this has been stated yet, but when I optimize gems, my cogwheels come out random. I've seen crit, resilience and I think even spirit.

I am also still having a problem with the "Optimize Gems" suggesting I use a Shifting Demonseye, which should never be suggested for DPS and is obviously wrong. Pfintzer @ Dalaran-US - ShadowCraft

Originally Posted by sinnaa

NOW, the reason the program says that delicate is better is because its not using the ep values from the combat guide. I believe its using EP values based on specific gearing of a character (correct me if i am wrong here).

I'm not sure I'm really following your argument. The EP values in the combat guide are just guidelines, for use if you don't have anything better. The EP values in shadowcraft are specific to you and your toon, and are the ones which should be used to determine the best gemming for your specific toon.

sinnaa · 03/30/11, 8:22 PM

Originally Posted by dinesh

I'm not sure I'm really following your argument. The EP values in the combat guide are just guidelines, for use if you don't have anything better. The EP values in shadowcraft are specific to you and your toon, and are the ones which should be used to determine the best gemming for your specific toon.

....and how are you not getting what i am saying? i pretty much just said that same thing. notice how i used quotation marks around wrong and stuff like that??? I did that people people think the gemming is wrong but its probably not.

sp00n · 03/31/11, 8:03 AM

Is http://shadowcraft.mmo-mumble.com using a different (more recent) version of the underlying calculation engine than http://shadowcraft-test.appspot.com?
I'm trying to get the DPS breakdown for the various abilities, which isn't supported by Antiarc's version (yet?) but is by Caffeine's. However I'm seeing a 200+ DPS difference with the same gear and settings between both UIs.

(And just as I was checking the link generated by ShadowCraft, I noticed an additional jump of +220 DPS after I refreshed the page which was still being open in my browser from monday, bumping the disrecpany to a total of 449.96.)

Link to Antiarc's ShadowCraft:
Surprise @ Wrathbringer-EU - ShadowCraft

Import string for Caffeine:

{"gear":{"head_socket_meta":"68778","head":"65129","head_socket_0":"52211","head_enchant":"62369","head_reforge_from":"exp","head_reforge_to":"hit","neck":"59517","neck_reforge_from":"mastery","neck_reforge_to":"exp","shoulders":"65083","shoulders_socket_0":"52212","shoulders_enchant":"62346","shoulders_reforge_from":"crit","shoulders_reforge_to":"haste","back":"65035","back_enchant":"60663","back_reforge_from":"crit","back_reforge_to":"haste","chest":"65239","chest_socket_0":"52212","chest_socket_1":"52220","chest_enchant":"74250","chest_reforge_from":"0","chest_reforge_to":"0","wrists":"65050","wrists_enchant":"85008","wrists_reforge_from":"crit","wrists_reforge_to":"haste","hands":"65073","hands_socket_0":"52212","hands_enchant":"74198","hands_reforge_from":"mastery","hands_reforge_to":"haste","waist":"99999265371","waist_socket_0":"52212","waist_socket_1":"52212","waist_reforge_from":"crit","waist_reforge_to":"haste","legs":"65242","legs_socket_0":"52212","legs_socket_1":"52212","legs_enchant":"85067","legs_reforge_from":"crit","legs_reforge_to":"haste","feet":"65144","feet_socket_0":"52212","feet_enchant":"74252","feet_reforge_from":"mastery","feet_reforge_to":"exp","finger1":"65082","finger1_enchant":"0","finger1_reforge_from":"crit","finger1_reforge_to":"haste","finger2":"62362","finger2_enchant":"0","finger2_reforge_from":"mastery","finger2_reforge_to":"exp","trinket1":"58181","trinket1_reforge_from":"hit","trinket1_reforge_to":"haste","trinket2":"65026","trinket2_reforge_from":"0","trinket2_reforge_to":"0","mainhand":"65094","mainhand_enchant":"74246","mainhand_reforge_from":"crit","mainhand_reforge_to":"haste","offhand":"68601","offhand_enchant":"74246","offhand_reforge_from":"hit","offhand_reforge_to":"haste","ranged":"68608","ranged_reforge_from":"mastery","ranged_reforge_to":"hit"},"settings":{"race":"night_elf","spec":"combat","glyphs":["adrenaline_rush","sinister_strike","slice_and_dice"],"buffs":["crit_chance_buff","bleed_damage_debuff","armor_debuff","all_damage_buff","melee_haste_buff","physical_vulnerability_debuff","guild_feast","spell_damage_debuff","spell_crit_debuff","agi_flask","attack_power_buff","short_term_haste_buff","stat_multiplier_buff","str_and_agi_buff"],"cycle":{"use_rupture":true,"use_revealing_strike":"sometimes"},"gear_buffs":["potion_of_the_tolvir"]}}

Caffeine · 03/31/11, 8:21 AM

Originally Posted by sp00n

Is ShadowCraft - Rogue DPS Spreadsheet using a different (more recent) version of the underlying calculation engine than ShadowCraft-Online?
I'm trying to get the DPS breakdown for the various abilities, which isn't supported by Antiarc's version (yet?) but is by Caffeine's. However I'm seeing a 200+ DPS difference with the same gear and settings between both UIs.

The results should be the same. However until last week or so the FAQ on shadowcraft-test.appspot.com was wrongly stating that fight duration was set to 6 minutes, it's set to 5 minutes (with no way to change it). Antiarc's UI has a 6 minute default (and can be changed in the settings). Apart from some performance improvements in the engine version used by Antiarc, a changed Unheeded Warning and the option to prepot there should be no difference in DPS.

sp00n · 03/31/11, 8:40 AM

Originally Posted by Caffeine

The results should be the same. However until last week or so the FAQ on shadowcraft-test.appspot.com was wrongly stating that fight duration was set to 6 minutes, it's set to 5 minutes (with no way to change it). Antiarc's UI has a 6 minute default (and can be changed in the settings). Apart from some performance improvements in the engine version used by Antiarc, a changed Unheeded Warning and the option to prepot there should be no difference in DPS.

Hm. DPS is raising by another 200 if I reduce fight duration to 300 seconds.

// Edit
With every buff and debuff removed, the difference is still 228.14 DPS. Stats are exactly the same, so one of both is doing something different.