Java-based optimal Rogue reforging calculator
 

Fast is never fast enough! Now basically instant.

Split off from Korner's thread to avoid polluting it with bug reports that don't apply to his tool.

This is a reforging calculator that calculates the correct reforgings to achieve optimal EP. It's (loosely) based on Korner's tool, but since it's written in Java rather than Perl and has several optimizations, it's quite a bit faster (20 times or more in most cases).

Reasons to use my version:

• It's even faster than Korner's C version!
• You like the convenience of a web-based tool.

Reasons to use Korner's version:

• It's been around longer and has fewer bugs
• It has more options like customizable EP coefficients and caps
• It works (manually) with random enchant items, which my tool doesn't handle at all

You can get Korner's original version from his thread:
http://elitistjerks.com/f78/t115053-...reforger_tool/

If you want to try my version, there are three versions: the applet version (which runs on a web page), the runnable JAR version (if you know how to run Java JARs) and the source code version (if you are a programmer and want to modify/examine my tool).

Most people should use the applet version.

The applet version of my program needs permission to access Armory and Wowhead. You'll see a big scary security warning when you load the page. If you don't trust me (and why should you?) you are welcome to download the source code and review it yourself. Alternatively you could run it in a VM or a machine that you don't do anything important on. I don't have to tell you how common keyloggers are. I can assure you that there is nothing malicious about my program, but of course someone trying to hack your account would say the same thing.

Typical solution times for my Java version are instant to a few seconds, depending on your hardware and how HotSpot is feeling.

If you don't have Java, download it from java.com

Applet version (recommended for most people):
Reforger

Runnable JAR version (faster than the Applet if you're familiar with Java):
http://powerofb.wsnetwork.net/reforger/reforger.jar

Source code:
http://powerofb.wsnetwork.net/reforger/src.zip

As of 1.4, the source code is released under the Apache License, Version 2.0.

Version history:
1.8:

• Disable multi-threading. There was a weird race condition that would sometimes give bogus results. The single-threaded version is negligibly slower and doesn't have the problem.

1.7:

• Fix another combat calculation bug.
• Fix a performance bug - now up to 10x faster than 1.6, 100x faster than 1.5 and 50x faster than Korner's C version!

1.6:

• Rewrote calculation code - up to 50x faster than 1.5 and even faster than Korner's C version!

1.5:

• Fix combat reforging formula.
• Print list of non-reforged items.

1.4:

• Handle non-English character profiles
• Correct EP values for Combat rogues
• Auto-convert simple character URLs to advanced character URLs
• Add license information to source files

1.3:

• Improved detection of random enchant items that will not result in false positives.

1.2:

• Print an error when encountering items with random enchants rather than give bad results
• Print an error message if you don't enter the advanced armory URL rather than give bad results
• Correctly calculate gem/enchant bonuses rather than using Armory's (bad) values. Stats should now match in-game values.

1.1:

• Fix an EP calculation bug
• Add better error messages
• Add version info

How I will fix that problem? solutions?

Surround in a try-catch block to handle the exception gracefully.

In my case, it was most likely due to my Engineering helm.

Hi, i'm intrigued by the two new reforge tools. I oped to use the Java-based tool, as it is somewhat faster. I left it to run over-night, and came back to check the results, which seemed at least a little odd:

. . .
Seeing as how the tool reforged everything to hit/mastery, it looks like it's possible that the tool is using Assassination calcs across the board. Anyone else running combat have a similar experience?

The EP values and caps are the same right now for Combat and Assassination in my program, so, yeah, it's wrong.

Thanks for the confirmation, should probably have noted i was using the Applet version, not the JAR version.

Recieved the same message. I also have an engineering helm equiped.

odd question
 

Is there a combat reforger option?

I've fixed the problem with false-positive random enchant detection on items that have no reforgeable stats.

Stats with recommended reforgings (EP = 6114,0):
Hit: 1333
Mastery: 1829
Haste: 837
Expertise: 449
Crit: 414

Anyone having trouble with cycle stability with less than 500 expertise or only me?

java.lang.RuntimeException: Random enchant item found. These items are not yet supported.
at net.wsnetwork.reforger.Reforger.runReforging(Reforger.java:153)
at net.wsnetwork.reforger.ReforgerApplet$ReforgerRunnable.run(ReforgerApplet.java:99)
at java.lang.Thread.run(Unknown Source)

Probably engineering tinkers(nitro, haste on gloves, parachute)

This appears to be the problem item:

["Mistral Circle" Stats=[Hit=0 Mastery=0 Haste=0 Expertise=0 Crit=0]]

Solving this problem is complex because neither Wowhead nor the Armory item pages report the stats on these items correctly. The stats will need to be pulled from your Armory page, which is more complex since the item might already be reforge and the HTML is harder to parse.

I'll see if I can get a fix done tomorrow.

Some steps to resolve the random items issue can be found here: http://elitistjerks.com/f78/t115053-...l/#post1849030

So if we hardcode all the random items and stats in the program, the item variant we need can be pulled from Armory by that suffixID, and the reforge option on the item can be determined from the same URL from "&re=144" part.

Perhaps blocked from Armory?

Full error below.