gems socket

Topics: Rawr.Base.Optimizer
Aug 18, 2009 at 7:35 AM

Hi,

First, sorry for my english ;)

There is something I don't understand. When I run the Optimize function, Rawr put a bad gems color into a wrong socket, for exemple a orange into a blue socket. I tried to run this function with and without "Override Regem" check, there is the same. In my gemming template I have only one custom template with gems into the good color socket.

I tried this with my caracter (Moonkin) and with a friend's caracter (hunt), same problem.

So, I think I miss something, but I don't know what. Could you explain it to me, please?

Thanks a lot.

Raoul

 

Coordinator
Aug 18, 2009 at 8:08 AM

That sounds fine, what's wrong with that? It'll use an orange in a blue, if that produces more DPS.

Aug 18, 2009 at 8:19 AM
Edited Aug 18, 2009 at 8:28 AM

Edited (wrong comment ;) )

But is there a way to force Rawr to use good gem color with good color?

Coordinator
Aug 18, 2009 at 8:53 AM

Yes, for comparsion charts only, you could enable only gemming templates that match socket bonuses. However, really question yourself on why you'd want to do that; that's a DPS loss.

Aug 18, 2009 at 5:38 PM

I've got the same problem than raoul2k9 but with the metagem in the helmet I lost the bonus of the meta because I need a red gem and rawr do not put red gem

 

Editor
Aug 18, 2009 at 7:55 PM

There is a simple way to make sure that Rawr puts enough of each color gem to meet whatever requirement a given meta gem might have - on the Stats tab, underneath your Character's name and server info, there is a checkbox called "Enforce Gem requirements.  Check it.

Rawr intentionally does not match all sockets to pick up socket bonuses.  Think of it this way (ignoring meta requirements): What if a given blue socket granted you +1 Attack Power(AP)?  Would you intentionally go out of your way to pick up that socket bonus by putting in a +10AP/+15Stam gem, if you could otherwise just socket in a +20AP gem?  From a purely DPS standpoint, that doesn't make sense.

But what if that socket bonus were +2AP?  Would it be worth it then?  Still, the answer is "No, it's not worth it".  The socket bonus would have to be worth a whopping +10AP to just be on par with ignoring the socket bonus.

The great thing?  Rawr calculates exactly where these breakpoints for matching or not matching socket bonuses lie.  That's the whole point of Rawr - determining the best TOTAL character setup, given the variables input as accurate.  So, Rawr calculates if matching the socket bonus gives more DPS than ignoring it for the best gem regardless of color, and makes its decisions accordingly.

There are of course a number of other variables I intentionally left out, but this is really the core concept.

 

Now, on the idea of having an option to "force" all sockets to be matched with a gem that would activate its relevant socket bonus, I can see some value in this, for certain situations.  Doing a Freya+3 attempt or potentially a Faction Champions encounter, the extra Stam (in this example) might be nice, despite the lack of purely DPS modeling .  A better case might be in a tanking model, where matching the socket bonuses generally lends more overall stats in a positive (to the end goal) way.  Even given these situations as arguments for the existence of a "match gem sockets with correct component color gems", though, I can't in my right mind advocate the existence of such an option within Rawr.  It already calculates the.  If you wish to consciously ignore its recommendations, that's perfectly alright - I do it all the time, for very measured reasons, of course.  But telling Rawr to blindly match every socket bonus is a BAD thing in the OVERWHELMING number of cases, and is not in line with the purposes of Rawr as a modeling and optimizing utility.

Coordinator
Aug 19, 2009 at 7:00 AM

And again, you can just only enable socket-matching gemming templates to get this effect if you reaalllllly want to.

Aug 19, 2009 at 7:28 AM

I'm OK (I've retest it and it's working) but sometimes Rawr do strange things ...