• Zerack

Download Link:
Current Version: DK_Optimize_2.5.3.zip (~619 kb).

DK Optimize is hosted on Google Code. The main link to the project is dkoptimize - Project Hosting on Google Code.

Installation Instructions:

• Uninstall any existing versions of DK Optimize
• Extract the archive and run SETUP.EXE
• Navigate to Start -> Programs -> DK Optimize
• Run DK Optimize

Basic Information:
The Death Knight Gear Optimizer is a program written expressely for the purpose of finding the absolute optimal setup of gear, gems, and enchants for any Death Knight of any spec. It functions on a system of weighted stat values in which each stat is assigned a value relative to others. Generally, Attack Power is given a reference value of 1.00 or 100 so that calculations and relative values are easy to understand. This is a commonly used system both for Death Knights and other classes seeking to maximize their performance, but it falls apart when attempting to build entire sets of gear as the most valuable stats (often hit or expertise) have a limit past which they provide no benefit. Loot Rank and Wowhead are great tools, but you can't just cherry pick the best item for each slot. This situation is where the program comes in, combing through selected gear in order to make absolutely sure it picks the best item for each slot in conjunction with an entire set.

Features:

• Calculates optimal gearing, gemming, and enchanting for any combination of items available in game.
• Pre-loaded items include all iLvl 200+ items and all gems.
• Support for professions shows how each one benefits your character
• Support for racial traits like Axe Specialization and Draenei Auras keeps the results relevant and accurate.
• Fully editable items, gems, and weights permits total customization and experimentation with theoretical or newly discovered gear.
• Supports weights for Tier 7, 8, and 9 set bonuses as calculated by the EJ community.
• Considers talents like Virulence, Rage of Rivendare, Veteran of the Third War, Nerves of Cold Steel, and Tundra Stalker to make sure expertise and hit caps are correct.
• Permits both Two-Handed and Dual-Wield gear and weapon selection.
• Ability to save multiple profiles of gear so that you can easily swap between sets for optimization.
• Ability to enable / disable specific enchants.
• Full multi-threading support.

Upcoming Features:
I'm constantly trying to improve this program, so here's an idea of what I've got in mind. If you've got a reasonable suggestion please feel free to mention it. No guarantees on anything, however.

• I'm not completely happy with how data is stored between versions, so I'll be tweaking that in the future.
• Jewelcrafting is naive in this new version, but the code exists to implement it intelligently. I'll be working on that.
• Tier 10 matching and weights are not yet implemented. This is coming when more is known about WoW 3.3.0 gear and patch notes.

Known Issues:

• Forcing both Spell Hit and Expertise caps can cause the program to seem to 'lock up'. The program is not frozen, but the calculation may be taking too long to be practical. Hitcap forcing does, however, function correctly when it completes in a reasonable time.

Version Notes:

• Version 2.5.3
  • Updated MH / OH speed to work correctly with the default value of weights from Kahorie's Simulator.
• Version 2.5.2
  • Fixed an error that would sometimes cause Main Hand, One Hand, and Off Hand designations to be ignored.
  • Fixed an error that would sometimes cause uniqueness in One-Handed weapons to be ignored.
• Version 2.5.1
  • Fixed a stat error for Gloves of Smoldering Touch
  • Fixed an error that would cause boot enchants to calculate incorrectly in some situations.
  • Added Arena PvP weapons for iLvls 245 and 258.
• Version 2.5.0
  • Re-release version written in Visual Basic.NET
  • Fixed stat errors for several items.
  • Added all non-proc based items available from the Onyxia encounter.
  • Added the ability to enable and disable boot, wrist, and glove enchants.
  • Fixed an error that would cause the calculations to crash when forcing a hit or expertise cap was turned on.
  • Added features resulting in a speed boost in overall execution.
  • Added the ability to save multiple configurations of gear.
  • Added a screen to configure various constants such as hit caps and rating statistics.
  • Added additional methods of output, including vBulletin, Plain Text, and XML.
  • Updated weight sets for Unholy and Frost to be current with WoW 3.2.2.
  • Updated weight sets for Unholy to be current with preliminary WoW 3.3.0 data.
  • Fixed problems with the installer that would cause it to not install on many systems.
  • Added file associations so that *.dko files can be opened from the file explorer.
  • Updated Jewelcrafting to be naive in its implementation. Revisions at a later time.
  • Added multithreading support to both candidate set selection and gemming. Large speed gains.
  • Fixed errors resulting from multi-region use.
  • Added rudimentary error logging to report causes of program crashes.
• Version 2.1.2
  • Updated Expertise to no longer calculate a truncated weight per recent theory changes.
  • Fixed a stat error for Shoulderplates of Trembling Rage.
  • Added Band of Draconic Guile.
  • Fixed a stat error for Blackhorn Bludgeon.
  • Fixed a stat error for Maexxna's Femur.
  • Added a weight for weapon speed in 0.1s increments.
  • Unholy weights updated but out of date again at time of this release. Blood and Frost unchanged.
  • Significant improvements to gearing and gemming algorithms should result in increased speed for all calculations.
• Version 2.1.1
  • Fixed several items which had incorrect stats.
  • Added a feature to write calculation results to a text file.
  • Region compatibility should really be fixed this time.
  • Updated trinkets to separate proc statistics from base item values.
  • Added consideration for the ArP, Hit, and Expertise caps from trinket procs.
• Version 2.1.0
  • Fixed a stat error for Comet's Trail.
  • Added further improvements to the gear iteration system.
  • Fixed a bug where long calculations would sometimes crash the program.
  • Added Nitro Boosts for Engineers.
  • Updated weights for Dual Wield Builds.
  • Fixed a bug where new items in the gear and gems screens would display 'a' instead of a checkmark character when selected.
  • Fixed a bug that would result in lost data when items were sorted by iLvl.
  • Added all Trial of the Crusader, Trial of the Champion, Ulduar, and Naxxramas leather and plate items.
  • Enabled the feature to force hit and expertise caps.
  • Fixed a bug where Blood Elf race was being incorrectly handled.
• Version 2.0.3
  • Updated information storage to work correctly in regions using ',' as a decimal point.
  • Fixed a stat error for Starfall Girdle.
  • Added Titanium Razorplate.
  • Added Titanium Spikeguards.
  • Updated weights for Unholy Two-Handed specs.
• Version 2.0.2
  • Fixed a bug that caused the uniqueness property for One-Handed weapons to be ignored.
  • Fixed a stat error for Stylish Power Cape.
  • Added Pyrite Infuser.
  • Added Caress of Insanity.
  • Improved the speed at which gearsets are iterated before the gemming process.
  • Fixed a bug that caused the calculation screen to malfunction if used twice during the same session.
• Version 2.0.1
  • Improved gemming process for faster gemming.
  • Improved the usability of the Gems, Weights, and Gear screens.
  • Fixed a bug that would make the results screen overly large on some systems.
  • Fixed a bug that forced me to disable the close buttons on many windows.
  • Contents of the main screen are saved from session to session.
  • Streamlined the process for saving data when the program is closing.
• Version 2.0.0
  • Initial Release

 

Afabar

Sorry to come with bad news, but the correction do not seems to work on my french worksation. However, I have replaced with text editor the "." by "," and it works.

Great work.

 

• Zerack

Ah, I understand. What you're saying is that with a text file that contains ',' as the decimal point, it works fine. However, the problem is that I distributed a file with values already populated containing '.' as the decimal point. That's a silly oversight on my part. I most likely won't be able to get a fix out for that until Monday, but as you mentioned the solution is a quick Find / Replace in your text editor of choice.

Let me know if I interpreted that incorrectly. Thanks.

 

dr_AllCOM3

I've selected the minimum number of gems and it still takes ages. It won't go beyond "Calculating...". It only works in a foreseeable timespan when I select very few items. I just don't need an optimizer programm, when I more or less know which items to pick. You see the dilemma .

How do you do the gemming process? I'd just check for gemming by color vs best gem spam. Then take away best gems (aka red gems) for some Exp/Hit. That's how I gem manually.
Or at least give us an option to put the best (red) gem in every slot, that would help me a lot.

DocsDK UI
Simulator

 

• Zerack

I appear to have introduced a bug in the latest build. I'm going to check it out now.

Edit: I've nabbed the culprit, and it's something I'm going to have to figure out a way around as our gear scales higher. It's not a bug, just unfortunate timing combined with ugly scaling in a complex problem. In previous tiers, the value of pre-cap hit was always below the value of strength point for point, meaning that unless you were purposefully forcing the hitcap, there was never any reason to put anything but a red gem in a red socket. This simplifies gemming significantly, since you can just pass through and drop +20 str (essentially) into every red socket.

However, I'd written the code to account for when hit exceeded strength at the pre-cap level and it has just now been hit by the newest weight sets distributed. When hit is > str on a point for point basis, it becomes possible that a yellow or orange gem is optimal in a red socket. However, I can't simply reverse the str > hit logic and use yellow as my baseline since hit is a cappable stat.

To answer your question as to the gemming methodology, it's something like that: Fill red slots with red gems. Yellow slots and blue slots are looser, so they are iterated with the rest of the gems including the prismatic Nightmare's Tear. This just falls apart when hit > str, and it's something that I need to find a resolution to.

In the short term, simply make your Hit To Melee Cap weight field less than your strength weight, and you should be able to run 5-6 (maybe 7, depending on color) gem calculations reasonably. As long as the sets come out more or less at the hit cap (which they should / will), the relative optimal-ness of the set is conserved. As a note, adding another red gem to the mix is the fastest way to increase complexity. Additionally, forcing the hitcap imitates some of this logic, since you want to sacrifice reds in red sockets in order to meet the hitcap specified.

It's late on a Friday so I won't work out a solution to this tonight, but thanks for pointing it out to me. Sorry for it to have cropped up right as I made a new thread about it, but it just happened to coincide with that timing. Let me know if you have any other problems.

 

• Asphyxialol

I'm not entirely sure why, but your newest release doesn't seem to be taking overall stats into consideration.

As an example it is stating that using both Grim Toll and Mjolnir Runestone is the optimal setup for 2h Blood, and that the 4p t8 is worth breaking to replace the shoulders and gloves with Shoulderpads of the Eternal and Bladebreaker Gauntlets. Both of these have been proven to be poor swaps previously. It is also stacking hit to very high ratings without concern for the melee and spell hit cap it would appear (I am getting swaps for 2h blood with 400+ hit).

 

dr_AllCOM3

Plus selecting Hit/Exp cap makes it much slower or it even crashes (with melee cap selected for me).
Without those it's nice and fast . Very fast.

I'm missing a way to save to precious results, besides making a screenshot.

Edit: Is the Arp cap taken into account?

DocsDK UI
Simulator

 

Sakuratei

I ran my first test today and the gearset it suggested showed 313 Hit rating, and spell hit cap 100%. However, the Frost thread (and my own understanding) puts spell cap at 11% from gear which is closer to 360 hit rating. Did I miss something along the way?

 

Qrio

Quick observation - in the 51/0/20 weights, ArP is valued at 5.67 by default which makes it almost double the value of 1 STR for the spec.

According to the BiS thread, post #459 by Afabar, which is the one I assume you based your stats on (as the rest of the slats mirror it exactly) it should be 2.67.

Also - I seem to be missing the Valorous Darkruned Shoulders in my items list - if I manually add that, do I have to do a special setting so it is counted as meeting the t8 bonuses? To clarify : if I just type in Valorous Darkruned Shoulders as the name, will it be recognized as a t8 piece?

And last question: with the automatic install, do I need to uninstall the previous version before I install the new one in the same folder? I've done it so far to avoid conflicts, but if it's not needed I can just skip this step

Edit: Added the last question

Remember everything, forget nothing!

 

Xaer

Scrap this, missed the manual install part.

 

• Zerack

Asphyxia - Can you upload me your gear / item / status text files that are giving that set as a result? I'll take a look at it.

dr_AllCOM3 - I'll try and put in a way to text export your results in the next release. Also I hadn't added the ArP cap, but will do so in the next version.

Sakuratei - If you have full raid buffs, the hit cap is 289 hit rating. This is the value that is used as the hit cap in the program.

Qrio - Ah, I'll fix that weight in the next go-round. Just a typo in the file on my part. Regarding Tier sets - you can simply add Valorous Darkruned Spaulders - the string match for Tier 8 is 'Darkruned' so you shouldn't have any problems. You don't have to reinstall - just select 'No To all' when you are prompted to overwrite in the installer and you should be fine.

 

Xaer

A small notice, after running the manual install, I also had to manually add the gemicons folder to my install directory and copy all the images to there since it was giving errors with pathing.

If possible could u update the item list to display weapon speed ?

Also, when selecting dual wield there is no option to double up on specific one handers as there is no mainhand/offhand selection.

 

Sakuratei

Ah thanks, Zerack, I was being a noob and looking at melee hit rating when checking percentages. I didn't know hit rating affected melee and spell hit differently.

 

• Asphyxialol

I resolved the issue Zerack, I hadn't noticed at the time of my post that you had changed the weightings for each spec. However now I bring to you another concern (again, Armor Pen related). I don't believe the optimizer understands the soft armor penetration cap vs the hard cap and the actual effect trinkets have on it.

As an example, per the weights Doc posted in the Blood topic (which have been similar to my results), it says that I should use both Grim Toll and Mjolnir Runestone and then have about 650 passive armor penetration with the two trinkets, obviously this is not correct as Armor Penetration cannot exceed 100% and that puts me *massively* over the cap whenever even 1 trinket procs. Because of how the trinket values are entered into the sim as static values I don't believe it understands how the soft and hard armor cap works and how using a static value on armor penetration is not effective for optimization the way the program works.

There should be a way for it to calculate out the soft cap of 443 (+665 from Mjolnir is 1108, or 90.008%) and then the hard cap of 1108, and how that anything excess from trinket procs is simply wasted armor pen (ie, the weight of armor pen after 443 is greatly diminished in the event you are using a trinket). Is there a way we can solve that issue?

Obviously the value I am speaking of as 443 is in Blood, but that is the only spec where Armor Pen becomes worth stacking to extreme levels without regard for a balance in other stats, which is why I am encountering the issue as Blood and not as Frost / UH.

 

Twinbladez

It's mostly a personal nibble but buttons to mass select based on Availability (e.g. picking up everything in Ulduar) Would be nice to save time

 

norg

Couple of small things I noticed, Zerack:

Band of Callous Aggression -- the heroic and non-heroic versions have their AP the wrong way round
Ring of the Violent Temperament -- the non-heroic version is missing its crit stat (50)

 

• Zerack

I've posted a new version (2.1.1) above. Please make sure to uninstall your existing version before reinstalling. The changes are listed above, but in response to people in the thread, here's where we stand:

dr_AllCOM3 - The ability to export results as text has been added. There's quite a bit of info in there, so hopefully it's sufficient.

Norg - The items you mentioned above as being incorrect have been fixed.

Asphyxia - I've both updated trinkets to be proc-based (with uptime as an entered value) and added consideration for caps being reached from the procs for partial value. This should help Blood optimizations immensely.

Kahorie - I'm fairly certain I got the region problem sorted out, but give it a try and let me know. If all else fails, do the search / replace again, open region.txt and make sure it has a value of '1' on the first line.

Xaer - The manual install archive has the paths preserved. Make sure you are preserving relative paths when you extract the archive.

Twinbladez - Given how many variations there are on the 'source' field, adding something like that would really be an exercise in tedium. Selecting the ones you want really shouldn't be that bad. This annoyance may be alleviated by multiple 'sets' of configurations in a future release.

New Version (Installer): Here
New Version (Manual): Here

 

• Asphyxialol

Not to be a persistent bugger or anything, but with the default blood settings it is now making Bandit's Insignia out to be better than Mjolnir Runestone, which basic math clearly suggests otherwise.

Even using the Simulator's data it should be better to use Mjolnir Runestone. It states the proc is 665 Arm Pen and it is up 0.183, making it effectively a passive 121.695 * 2.67 = 324.92565 and then 102 crit rating * 1.84 = 187.68 which totals 512.60565 EAP, where Bandit's is calculated at 350, yet it still shows Bandit's at a superior EAP.

 

• Zerack

Can you upload the text dump of the exact set you ran to get that result? That's more helpful to me than anything else, really, since it allows me to duplicate it. It's entirely possible something is messed up - I didn't get a lot of time to test it.

 

• Asphyxialol

http://www.gamerenders.com/uploads/l...ia%20Blood.txt

There is the log - removing Bandit's Insignia gives me accurate results, however so long as it is activated under any of my gear sets it is shown to be the top dps piece available.

 

• Zerack

I've re-uploaded version 2.1.1 with a very small change fix (I didn't change the version number) that should fix the problem with trinkets. Basically, I had copy and pasted some blocks of code from my gear set evaluation w/ trinket procs into my final gemmed set evaluation and neglected to change a variable. Unfortunately that variable was still in scope in the gemming, so pulling from it didn't cause any errors.

You should see appropriate results for the trinkets now - at least I did in my limited tests. I didn't notice this when I was testing before since it doesn't show up if you don't run with more than 2 trinkets. Regardless it is now fixed and the OP and my previous post today both point to fixed versions.

Let me know if this fixes it for you.

 

Canadianloaf

I encountered a similar issue with ArP trinkets. I used Grim Toll and Pyrite Infuser for my test since they bring the same non-proc stat (+Hit). To simplify the test, I chose only a single option in each slot except the trinkets. With my gear list both trinkets push Hit to the Alliance hit cap so the difference in the non-proc value of the two is negligible. The gear is also well under the soft ArP cap. In this case, the proc value of Grim Toll (612*.2*2.67) should well exceed that of Pyrite Infuser (1234*.173). However, the optimizer selects Pyrite Infuser.

The text log is available at: http://files.me.com/canadianloaf/peuun6

Edit: Didn't refresh before posting.. will test whether the most recent patch fixes this.
Edit2: Awesome, Zerack. Confirmed that this test case is fixed.

 

norg

I'm in the process of double checking the Blood sets and I'm a little confused by the results I'm getting.

For the 'No restrictions' set, it is recommending a set with Mjolnir Runestone and 414 static ArPen, which obviously makes perfect sense when looking at the soft/hard caps.

However when I go to run the 'No heroic hard modes', (ie remove all items which are ilvl 258 or higher, and add in a couple more 245 ones to suit), it recommends a set with DC:G instead of Mjolnir and then starts adding in as much static ArPen as it can (526 with the items I made available, though presumably if I had made more available in the way of gems, for instance, it would have used more).

Can anyone think of a good reason why it would stay at softcap w/ Mjolnir but then ignore Mjolnir and shoot for the hard cap with slightly different items? It seems strange because I can't see any examples where I've removed an ilvl 258 item without having a direct ilvl 245 replacement available (Garrosh's Rage for Might of the Nerub, for example).

I just want to make sure it's working properly regarding the ArPen caps before I go and re-do all the BiS Blood sets for the twentieth time!

 

• Asphyxialol

Because the other stats on the items outweigh the 18.3% uptime on for an effective 122 Armor Pen, thus making Mjolnir excessive and rather than aiming for soft cap they push on for a passive armor pen of 526 (which is only ~100 more, and thus equivalent to Mjolnir, so when you think about it... it's pretty close, you just need enough itemization on the other items to push the equivalent of 102 crit and 20 armor pen, which realistically isn't hard).

 

norg

Actually, I can only assume I didn't add DC:G to the equation when I initially did that 'No restrictions' set, as when I double checked it just now, the optimizer eschewed Mjolnir in favour of DC:G, even when static ArPen was under the softcap.

The overall scores were 14688 when using Mjolnir and 14713 when using DC:G. All other items were the same.

From that I can only assume that the final verdict is that DC:G > Mjolnir for Blood, even if you are able to utilise the entirety of Mjolnir's proc. Does that sound right?