Rawr (v2.3.23 released on 9/12) - Page 13

Treerobber · 03/08/08, 5:19 PM

Deleted: Wrong Forum

Hrank · 03/08/08, 6:11 PM

No more problem after loading it tonight. I guess I should have restarted the computer. Anyway works now.

twitch0115 · 03/08/08, 6:27 PM

Originally Posted by Caffeine

Did anyone manage to run Rawr under Mono on Mac OSX?

When I try to run it I get this exception:

Unhandled Exception: System.NullReferenceException: Object reference not set to an instance of an object
  at Rawr.ComparisonGraph.get_PrerenderedGraph () [0x00000] 
  at Rawr.ComparisonGraph.ComparisonGraph_Resize (System.Object sender, System.EventArgs e) [0x00000] 
  at System.Windows.Forms.Control.OnResize (System.EventArgs e) [0x00000] 
  at System.Windows.Forms.UserControl.OnResize (System.EventArgs e) [0x00000] 
  at System.Windows.Forms.Control.OnSizeChanged (System.EventArgs e) [0x00000] 
  at System.Windows.Forms.Control.UpdateBounds (Int32 x, Int32 y, Int32 width, Int32 height, Int32 clientWidth, Int32 clientHeight) [0x00000] 
  at System.Windows.Forms.Control.UpdateBounds (Int32 x, Int32 y, Int32 width, Int32 height) [0x00000] 
  at System.Windows.Forms.Control.SetBoundsCore (Int32 x, Int32 y, Int32 width, Int32 height, BoundsSpecified specified) [0x00000] 
  at System.Windows.Forms.Control.SetBounds (Int32 x, Int32 y, Int32 width, Int32 height, BoundsSpecified specified) [0x00000] 
  at System.Windows.Forms.Control.ScaleControl (SizeF factor, BoundsSpecified specified) [0x00000] 
  at System.Windows.Forms.ScrollableControl.ScaleControl (SizeF factor, BoundsSpecified specified) [0x00000] 
  at System.Windows.Forms.Control.Scale (SizeF factor) [0x00000] 
  at System.Windows.Forms.ContainerControl.PerformAutoScale () [0x00000] 
  at System.Windows.Forms.ContainerControl.PerformDelayedAutoScale () [0x00000] 
  at (wrapper remoting-invoke-with-check) System.Windows.Forms.ContainerControl:PerformDelayedAutoScale ()
  at System.Windows.Forms.Control.ResumeLayout (Boolean performLayout) [0x00000] 
  at Rawr.ComparisonGraph.InitializeComponent () [0x00000] 
  at Rawr.ComparisonGraph..ctor () [0x00000] 
  at (wrapper remoting-invoke-with-check) Rawr.ComparisonGraph:.ctor ()
  at Rawr.ItemComparison.InitializeComponent () [0x00000] 
  at Rawr.ItemComparison..ctor () [0x00000] 
  at (wrapper remoting-invoke-with-check) Rawr.ItemComparison:.ctor ()
  at Rawr.FormMain.InitializeComponent () [0x00000] 
  at Rawr.FormMain..ctor () [0x00000] 
  at (wrapper remoting-invoke-with-check) Rawr.FormMain:.ctor ()
  at Rawr.Program.Main () [0x00000]

I installed the latest version from Main Page - Mono and MonoDevelop runs fine, but afaik it uses GTK# and not WindowsForms.

So has anybody been able to run it on OSX. That was a little hard for me to understand. I am pretty bad with computers unfortunatly.

Astrylian · 03/08/08, 8:08 PM

Yes, I have run it on OSX... Beta 10, to be specific, but nothing has changed since then that I know of that would prevent it.

I didn't run it using Mono on OSX though, I ran it using Mono for Wine on a Wine port on OSX. "CrossOver" is the name of the app that is a port of Wine to OSX, so that you can run windows programs on OSX. It's a commercial program, but there is a free trial available. Once you've got that installed, install Mono for Windows in CrossOver, then run Rawr on CrossOver.

Saraya · 03/08/08, 9:19 PM

Question regarding the item rankings for the moonkin model:

Are you double dipping the benefit one gains from mana/mana regen stats for the "overall" contribution? For example in a mana limited situation(time to OOM < fight time), a belt of blasting with the +8 spirit and +8 int gems is ranked better than a belt of blasting with 2 +12 damage gems. This is pretty reasonable as mana regen outweighs most other stats if you're going OOM before the end of the fight. However, not only is the "Efficiency" portion of the item higher, the "Damage" portion is too. This part makes no sense to me, as the +12 damage gemmed belt should always have a higher damage rating even if it loses out miserably on efficiency.

Coming at it from another direction in case my explanation above was confusing:
The numbers I'm getting is 65red5blue for the 8 spirit/8 int belt, and 64red for the 2x12 damage belt. I do expect the red+blue overall rating for the spirit and int belt to be higher, but I do not expect the damage portion to be as well.

Roaminggnome · 03/09/08, 8:05 AM

The warlock module does not seem to be properly modeling Immolate. It lists the min/max hit and crit as 0, and the average hit as 13.4118. This of course leads the simulator to believe that not casting Immolate nets you a very large DPS increase, and quite probably is giving too much weight to +shadow dmg only items, at the expense of generic +dmg.

Kyuki · 03/09/08, 10:32 AM

Big /bow to Astrylian and all the others that are putting work into this tool. I'm looking forward to see more class additions to this.

I have a little suggestion for this thread in particular:
Add a change log to the bottom of your last post of the OP, and add all the diffrent rawrs that are out currently, and try to update it. That way it's easier for people to follow updates that comes along and see what has changed.
That ofc, if you feel that you have the time for it, but I'm sure it would help. I guess you need to request that the other developers let you know when they do another update etc for this to work without a hastle though.

Either way *a million thumbs up*

Kyuki · 03/09/08, 10:34 AM

Or well.. You could just Highlight the wikilink, and caps it out or something. Just noticed it.

*edit

Dont want to spam the thread so I'm just gonna edit this into this post:
It seems that the Moonkin model is abit bugged, or is getting influenced by the mage model. Dont get any type of cycles, and the mage T6 setbonuses are around and basically the mage gear is beeing showed as upgrades (the tier sets).

I tried rebooting the program a couple of times, and used the other models with no problems, it's just the moonkin model.

Am I doing something wrong or is it a flaw in the model?

Benita · 03/09/08, 12:40 PM

Originally Posted by Saraya

The numbers I'm getting is 65red5blue for the 8 spirit/8 int belt, and 64red for the 2x12 damage belt. I do expect the red+blue overall rating for the spirit and int belt to be higher, but I do not expect the damage portion to be as well.

You can set the fight lenght to 1 min instead of 4 min trivializing any mana regen. That should lead to the results you look for on the item page.

It should be the default for that im guessing as most dont seek an answer to any mana question and only check for max dps gear. If you want to check when you will be running oom, you need to configure it anyway with the buffs, pots, innervate, spriest and so on.

Saraya · 03/09/08, 1:28 PM

I'm not looking for any particular results, rather, I'm trying to get how this is working straightened out. I might be interpreting the data wrong, or maybe there's some miscalculations going on. Either way it's a good thing to get resolved(for me, at least).

Mana regen is always an issue for me, so being able to compare some item socketed with different gems side by side is very valuable information. I've played around with the utility a little more, and if I just sort by the red bar and only compare the red bar, the results are much more reasonable. I guess I'd just like some clarification on how the red and blue bars should be interpreted after all.

And yeah, seeing all those nice mage tiered gear as upgrades is kind of depressing since they are always above the equivalent moonkin one.

----
And I forgot to say this in the previous post, but thanks very much for the effort put into expanding rawr to including other classes and specs(ok well, I only care about the moonkin one after all). This is very cool.

Laupen · 03/09/08, 2:51 PM

New version looks ace! Been using the Bear model since it appeared on Druid Wiki and it makes all my choices for me now

Switched my main to my Elemental Shammy recently and was overjoyed at the prospect of Rawr.Elemental. Any further news on development?

Console · 03/09/08, 5:45 PM

Originally Posted by Laupen

Switched my main to my Elemental Shammy recently and was overjoyed at the prospect of Rawr.Elemental. Any further news on development?

I don't have a beta available yet, but I am still working on it. In the words of Blizzard "soon..."

thedopefishlives · 03/10/08, 8:38 AM

Originally Posted by Saraya

Question regarding the item rankings for the moonkin model:

Are you double dipping the benefit one gains from mana/mana regen stats for the "overall" contribution? For example in a mana limited situation(time to OOM < fight time), a belt of blasting with the +8 spirit and +8 int gems is ranked better than a belt of blasting with 2 +12 damage gems. This is pretty reasonable as mana regen outweighs most other stats if you're going OOM before the end of the fight. However, not only is the "Efficiency" portion of the item higher, the "Damage" portion is too. This part makes no sense to me, as the +12 damage gemmed belt should always have a higher damage rating even if it loses out miserably on efficiency.

Coming at it from another direction in case my explanation above was confusing:
The numbers I'm getting is 65red5blue for the 8 spirit/8 int belt, and 64red for the 2x12 damage belt. I do expect the red+blue overall rating for the spirit and int belt to be higher, but I do not expect the damage portion to be as well.

I'm the maintainer of the Rawr.Moonkin module. I can say that no, I'm not "double dipping" the regen stats for the overall contribution. It is possible that I may be overrating int slightly (because int does, after all, add spell crit); I'll do some playing around with your gear today and see if I can get an answer for you.

[e] To clarify how the red and blue bars work: The red bar is raw DPS. The blue bar is time to OOM, in seconds. The formulas are roughly based on the Moonkin DPS spreadsheet by Efejel. The reason you're seeing an upgrade in DPS from adding more regen is because the DPS calculation includes the fact that you're in a mana-limited situation. If you are able to increase your time to OOM, the model takes that into account and increases the amount of damage you are able to do accordingly. That's probably why you're seeing regen give you a DPS upgrade over raw +damage.

Hrank · 03/10/08, 12:28 PM

Voidheart Leggings (warlock only) appear in the list of upgrades for Moonkin.
Moonglade pants rate higher than the Tempest Leggings. I don't know enough about Moonkins to know if that is correct but it struck me as odd they would.

Edit: in addition could it be possible to filter out cloth? Cloth is often a better option than leather, but I'd still prefer to have leather's additional armor since I need to solo too. A checkbox would make me a happy cow.

thedopefishlives · 03/10/08, 12:38 PM

I'm afraid the mechanics of Rawr are such that I can't eliminate cloth at run time. Astrylian has stated that in a future version of Rawr, class-specific items will be filtered. Also, I can remove cloth entirely if it should be discounted as an option for top-end moonkin; I just don't know enough about top-end moonkin to be able to make that call. As for Moonglade pants vs. Tempest leggings, there are a lot of variables in play in the moonkin model. I'd have to check the spreadsheet with the specific gear/buff setup to see if my math is right, but so far, it's proven reliable.

[e] I just tested out your character in the spreadsheet and in Rawr. It doesn't show the Moonglade Pants as superior to the Tempest Leggings. In fact, the difference between the two is almost exactly what the moonkin spreadsheet shows. Can you tell me if you did anything different in the options panel?