Rawr.TankDK - Latest release: v4.2.0 (06/26/2011)

[Shazear]

Hey, I figured it might be beneficial to have a couple Rawr DK threads here on pwnwear.
While the main Rawr site is what I check most frequently, and so can provide the fastest response, Rawr comes up in conversation from time to time and here is a way to post those questions/ideas without having to leave to go start new threads on the main Rawr site. Also the EJ threads seem to have died down quite a bite since the change in policy.

Useful links:
Rawr Home Page
Rawr TankDK Wiki
Rawr Defect/Issue Tracker Feel free to write up issues here. I guarantee that I will look at them.

Old Blog Post/interview about TankDK (no longer accurately describes the state of the code, though)

I'm not going to go post all the dry changelist stuff again here. I do that in the EJ thread and the Wiki.

So feel free to post your TankDK related thoughts/comments/questions and let the conversation begin!

[Gravity]

I wonder Shaz, is Rawr architected in such a way that you can adapt it to the Cataclysm model (when the model is final, of course) easily or is it very hard to change the mechanics inside Rawr?

Also I sticked this post and added 'rawr' to the forum descriptor on index page.

[Shazear]

Changing talents around and such should be pretty straight forward. Executing mastery should also be easy enough. Things are pretty modular. The base app calls into the model to perform the calculations, and there are some supporting aspects within the base application to help the models get the information they need. So I believe the transition will be less costly in the basics than trying to root out the weird bugs that always make things a bit... odd.

[Insolence]

And Reforging etc?

[Gravity]

reforging... good point, that would be tricky to work out gear optimisation for eh.

[Shazear]

Yeah, we're discussing how we want to deal with that. Right now the balance is debating between more like enchants or more like gems. I already have my hands full dealing with just the DK stuff, so I try not mucking about too much in the base projects. About half the time I screw up one thing for fixing another just because I don't have the best awareness for how every model works.

[Cryptwidget]

Hey Shazear Still loving the work you do with this module..keep it up! Was wondering if I could get your feedback on something...

I'm trying to run an experiment in Rawr to create a profile for Heroic-only runs. The goal is to maximize my DPS while maintaining the minimum gear needed to reach the Def cap and 8% hit cap so that I can try and make those heroic runs just a smidge faster. I'm trying to do this by selecting the DPS gear I have available from my dedicated DPS set (blue-diamonding that gear so the Optimizer doesn't change out the gems, and restricting the enchant to whatever is on it currently), and setting the Options panel to level 82 mobs. For the Optimizer, Chance to be Crit <=0 and % Change to Hit >= 100. I'm also setting it optimize for Threat instead of Overall.

I'm wondering what else I can do to tweak it more to max my DPS while running the minimum needed mitigate heroic-level stuff. Should I adjust the Threat weight up and the Mitigation weight down? What else could I do? Any feedback is appreciated

[Shazear]

oh... good question, and I'm working on the new combat rotation stuff to help with that, unfortunately right now weighting by Threat may not quite give you what you want DPS wise since IT is such a large (7x) threat dump. The module may be favoring that instead raw DPS. I'll see about exposing the DPS numbers via the experimental switch and maybe get more of the new combat model integrated and use-able. I'm working on a post to get community feedback on what people would like to see in getting reasonable DPS/TPS numbers out of tank/dps.

In the current build, what could you do?
Be sure to set the number of mobs to 5 on the options page. Since most of your slow fights are due to trash pulls, that should help as well. And hopefully provide better numbers if you adjust the rotation to include DND and/or HB. Otherwise it sounds like you've got the optimizer settings cookin'. This is one of those situations where having separate discrete DPS and Tank DK models doesn't play to the strength of the class to be fuzzy on some of those lines.

[Cryptwidget]

oh... good question, and I'm working on the new combat rotation stuff to help with that, unfortunately right now weighting by Threat may not quite give you what you want DPS wise since IT is such a large (7x) threat dump. The module may be favoring that instead raw DPS. I'll see about exposing the DPS numbers via the experimental switch and maybe get more of the new combat model integrated and use-able. I'm working on a post to get community feedback on what people would like to see in getting reasonable DPS/TPS numbers out of tank/dps.

In the current build, what could you do?
Be sure to set the number of mobs to 5 on the options page. Since most of your slow fights are due to trash pulls, that should help as well. And hopefully provide better numbers if you adjust the rotation to include DND and/or HB. Otherwise it sounds like you've got the optimizer settings cookin'. This is one of those situations where having separate discrete DPS and Tank DK models doesn't play to the strength of the class to be fuzzy on some of those lines.

Cool, thanks for the reply and info A few more then...
- Should I be seeing the stat numbers change when I adjust the number of targets? Currently, I'm not.
- Would it behoove me to use the Experimental Code option in 2.3.21 right now, or stick with the standard model?
- Changing my rotation around manually to add in D&D/Pest and what not is not saving...was gonna put a ticket in on the Rawr tracker, but it looks like someone already did

[Shazear]

- Should I be seeing the stat numbers change when I adjust the number of targets? Currently, I'm not.

You will only see a change in TPS values when those number change if you have abilities in the rotation that support multiple targets (Heart Strike, Howling Blast, DnD, etc.)

- Would it behoove me to use the Experimental Code option in 2.3.21 right now, or stick with the standard model?

The experimental code is mainly for me to be able to change between "normal" mode and the new code paths w/o completely destroying the model. You can play with it, but take the results with a pretty hefty grain of salt.

- Changing my rotation around manually to add in D&D/Pest and what not is not saving...was gonna put a ticket in on the Rawr tracker, but it looks like someone already did

Yeah, I'll be looking at that over the weekend, see if I can get it fixed. It was working at one time, but I probably broke it trying to get some other thing working.

[Insolence]

Wondering if you can confirm a rumor from a while ago. Was said Blood Presence Threat Modifier was reduced from 2.0 to 1.5, can you confirm?

Tried to see for myself but no Threat Reduction for DPS + no Omen makes it hard.

[Shazear]

Are you asking about Blood Presence for Cataclysm or in Wrath?
I don't believe the presence modifiers have changed since I did that partial update for threat on the Tankspot forums.

As far as Cata data, I'm not in the beta, and haven't done any work specifically for it just yet.

[Insolence]

Meant Cata Blood Pres sorry yeah.

Well was a Datamined thing from early Alpha, was probably wrong anyway.

[Shazear]

Hey, So I keep writing, and then loosing/deleting this ask for feedback post, so I'm just going to do it.
Is there anything in that other DK model (DPS) that you'd like to see brought into TankDK? I'm working on writing up a Rawr.DK.Common model that will do alot of the similar work between the 2 models to help with consistency and reduce the amount of C&P code I'm generating. I've been finding that as I see something that I fixed in Tank crop up in DPS, and vice versa, but maybe there are key features that I don't even think of as being useful across the 2, or just a way one model works compared to the other? Let me know.

[Gravity]

I've always found the Options tab: threat, survival, mitigation weights confusing.
Are they ratios to one another? threat:survival:mitigation? How do I know what's normal, bad, or best practice? Maybe better represented with a triangle, each point being a stat, and you choose your setting by moving a point around within that triangle.

Also, the Stats: 'threat points' and 'modified threat' are like 13000 for me, using default settings on Rawr 3. The tooltip suggests that's 13K tps, which surely isn't possible in my gear. Perhaps update their tooltips?

Is there any value in the distinction between 'summary' and the 'overall stats' anyhow? If they're being multiplied by the Options settings, then surely I only care about the final figure and the 'summary 'data is just a figure used in the calculations? Maybe hide them, and display those raw summary stats on mouse-over?

I find the rotation system hard to input, because I don't think in the terms of # attacks per x time. In Cataclysm, it'll be even less so, I'll think in terms of priority. I'll want fevers up all the time, scarlet fever next priority, then death strike, rune strike, heart strike, minus y blood runes per 60s for tank cooldowns. It might be painful.

Maybe you and Kahorie could share a code-base for the priority system, put it into a library so that both simulators can benefit.

The new boss handler in Rawr is impressive; I hadn't seen it before. Haven't tested it, but the UI makes it look very promising for DPS simulations where you have a start/stop and multiple targets, etc, (eg. much closer to real gameplay). Perhaps have a scenario for tanks, which lets you specify how many tank cooldowns to use which inputs to the priority system I mentioned before.