Why do you take a 10STR+10Crit-Gem and Nightmare Tear?
Nightmare Tear covers all 3 Gems you need to activate the Meta.

Zerack your link seems to be dead.

Anyway thank you again for the great work.

Link works fine.
The close button doesn't work and I'm missing a start menu entry.

Close Button: Purposefully disabled for this release due to some weird event triggering that I couldn't figure out. Just use File -> Exit

Start Menu: Purposefully omitted since I myself hate extra Start Menu entries. I'll add it back tomorrow.

404 Error: Uploading to a different site now. Should be fixed momentarily.

I hope that helps. If you've got other feedback that isn't in those points I mentioned, please feel free to let me know.

Edit: Try downloading from Here

Are all ToC and Ulduar 10/25 items included?

The socket bonus in the legs is +6 strength; it's entirely possible that 10crit is worth more than 4 strength (Not entirely sure due to the lack of stat weights as of yet, but this was sometimes the case in 3.1). Further, the gem will only cover red and yellow, and so the Nightmare's Tear is required to fulfill the blue requirement of the meta gem.

The current loot list includes everything that the previous version included by default from Ulduar in addition to everything from mmo-champion's 3.2 loot list pages as of today, unless I missed something. If you see something missing that could be added I will certainly do so.

well i downloaded and set it up quickly to see the results and im very sure it shouldnt look something like this

It is likely that you did not select gems sufficient to satisfy either CSD or RED meta requirements. It's a check I'm eager to put in, but I wanted to get this release out as quickly as possible since 3.2 dropped a full week ago. Go into the gems screen and add either a purple / green / blue gem or that and a yellow / orange / green gem. That should fix it. If it doesn't, it's a problem I haven't yet encountered.

Gemming takes really long. I've selected five gems and it doesn't go beyond calculating the time. It works instantly with one gem selected.
The windows with fixed sizes are annoying, they are either too small or too big.
Nothing gets saved.
I'm missing [Caress of Insanity].

Besides that it looks good .

Regarding UI problems: I know it's far from perfect. I'll be fiddling with it for a long time to make it nicer. I'm beginning to think I posted this too early.

Regarding speed issues: I 100% assure you that this is faster than the spreadsheet. The disconnect is coming from you running the pre-3.2 version spreadsheet to this program which accounts for all of the new issues of 3.2 (specifically, varied meta gems resulting from JC changes). There's more to check per-set now then there was then. I have a 3.2 capable version of the spreadsheet and running these two programs side by side is like night and day. Coliseum gear also has more sockets than Ulduar gear, which is a huge hit to performance. I noticed this quite prominently when I was entering gear this afternoon.

I'm sad to hear that you prefer the spreadsheet, and I hope that I can bring you around with UI improvements here. Out of curiosity, separate from the current annoying qualities of the interface, what else do you miss from the spreadsheet? If it's something feasible it's likely I could include it in the future.

Is it possibe to have a checkbox for enchants too? I have 4 specs that show that ap on boots is the best, but 1 where it wants me to have hit on boots. So that I don't have to switch enchant every time I respec, is it possible to have a checkbox for enchants too?

Great job on the application so far

With an orange and a purple gem the calculation takes 10s. When I add a red gem it jumpes to 3h.
Choosing more than 2-3 items per slot makes the optimizer take far too long to realisticly finish within time.

To explain this a little: The gemming currently implemented is fairly 'strict' in the sense that it makes several assumptions about what is optimal. Assuming first that hit < str, red gems are possibly optimal in every slot. However, something like a purple gem is never going to be better than a red gem in a yellow slot, so purple gems aren't even tested in yellow slots or red slots. What I'm saying is that selecting multiple red gems is going to greatly increase time. Selecting orange gems will increase less so, and purple gems will be the least. However, the biggest factor is the number of gems - I realize it's a pain in the ass because it takes a long time, but I really don't know how to make it much faster.

Choosing more than 2-3 items per slot will simply take a while. What do you mean by 'realistically'? I'd consider anything under a few days very obtainable, since the sets should remain valid for many times that period. The pause functionality was added specifically so that you could prevent this from nuking your CPU if / when you needed to do something on the machine running it. I'll look at complexity more tomorrow, but it's another one of those things that's difficult to optimize more without a revolution in how I check sets.

I could see adding the ability to enable or disable specific enchants. The logic would be similar to how it works now, just with some flagging. I may add it in a future release.

I had 3-6 items per slot selected and I was at 4000h for the items alone. The average user will choose even more, since he doesn't have the knowledge to only select the seemingly best items per slot.
Does gemming interact with the items? I.e when gemming can get you to Exp cap, then the item with Exp is traded for a better one?

You can always assume the best gem as the base gem. Maybe check if gemming for color provides a better benefit. Then swap base gems for Hit/Exp to get to the cap if selected. It's not necessary to check every gem selected. There's always one best gem for each slot and one best overall.

I'm having problems getting the program to function on my office PC (says I need to update some system files, but then when I click ok it doesn't actually seem to update anything and I get the same message once I've re-booted). That means I can't just have it running during the day while I work.

If I have to do all the calcs at home that'll give me a lot less time to work on it and will slow down my updates considerably, so in the mean time if people want to PM me with their test results for certain specs/factions I'll get updating.

edit: Also if stuff needs changing this patch on gemming, enchanting etc I'd be grateful if people could let me know what needs doing; as I said I'm not playing WoW atm so I'm not up to speed on any of the 3.2 changes really. Thanks.

You can run it without installation. Just extract DK Optimize.CAB manually, and run DK Optimize.exe. If you see an error when you click on the View Results button after calculation, then create a folder named gemicons and move all ~gem.jpg files into it.

This is true. The installer exists primarily to put the correct ActiveX controls in the right place. If you extract manually and don't have one of the files needed, the error message is quite clear and a google search for the file needed should allow you to run the program with no problems.

http://elitistjerks.com/f72/t55505-o...6/#post1344621 now updated a bit more.

Another day, another version. I've posted a new version of my optimize utility after some interface work today that will hopefully make it a bit less painful to use. I'm still not happy with the results screen and will most likely end up scrapping it completely in favor of something else, but it should be readable for everyone now. More importantly I adjusted the requirements for gemming, so you should see an increase in performance when using Nightmare's Tear in your gemming sets. Let me know how much (if it all) this helps.

Download Link: Here (2206 KB)

Other Changes:

• Gemming methodology slightly improved.
• Gem, Weight, and Gear windows can now be resized.
• Gem, Weight, and Gear windows should now behave in a more user friendly manner when entering data.
• Results screen should display at a more manageable size for many users.
• All window close buttons are re-enabled throughout the application.
• The selections on the main screen are now saved along with gem, weight, and gear choices when quitting.
• The program will no longer attempt to find the best set for an invalid combination of gems. Meta requirements must be attainable.
• The method for saving gear, weights, and gems when exiting the program has been greatly streamlined.

More to come in the next few days, so let me know what else still bugs you. I've got a few ideas to increase the speed at which it parses sets (pre-gemming), so hopefully I can get those going this week. The easiest way to reinstall is to delete your existing DK Optimize directory completely and fresh install.

If you've modified your gear.txt, gems.txt, or weights.txt file, they are still valid for this release. Simply re-copy them back in along with a blank status.txt file after reinstalling and you should be good to go.

[Pyrite Infuser] is still not listed under Trinkets (minor version of Blood of the Old God). [Caress of Insanity] is also missing. Also, the 'Unique' drop down on DW weapons does not work. I made [Malice] unique and it still listed it as being MH and OH. I think it is also worth noting that many Leather based items are now coming up in high end gear due to the stat balance used by Blizz and stacking stam on plate. As an example [Soul-Devouring Cinch] is the best DW belt (and I believe comparable to [Starfall Girdle] for all other specs). Also, items like [Chestguard of the Recluse], [Garona's Guise], [Guise of the Midgard Serpent], [Proto-hide Leggings], [Concealment Shoulderpads] are better than all/most current Ulduar gear in terms of DW itemization (on a piece by piece comparison - excluding hit and tier piece bonuses in the current examples).

Here is a link to Wowhead weights: Armor - Items - World of Warcraft