Attack Rating Calculations

[paaco]

Nice Trev, compiling newest Rev now to check it out. Props to you and Andmetal for all the work and testing that went into this.

[KLS]

There's already a cap on worn atk in the code if I recall correctly so that's somewhat redundant if so.

I think it's fine staying just in showstats, the weapon skill isn't intended to be a part of the mitigation anyway.

[trevius]

Weapon skill makes up almost half of the Attack Rating Total. If attack rating is supposed to effect mitigation, then I don't see why weapon skill wouldn't be used for it. We don't need to emulate live, but I think if we wanted to, we would have to use the new attack formula. Then just dividing by 2 would get it very close to what it currently is. I am going to test it on my server a bit and see if it seems about the same or not. I don't mind leaving out of any combat calculations, but I am not really sure of any reason not to use it. Many MMO players care about the stats of their characters and can be pretty particular about them. I just think it might be nice if the server was using what their client says it is

As it is, you could switch between a 1hb that you have 300 skill in, and a 1hs that you have 1 skill in and you will hit for the same average hit on both (even though you will miss alot more with the 1hs with 1 skill).

I don't see any caps being used for attack currently. At least none that I could find or see in any of my testing. Using the newly adjusted #showstats command, it is easy to see what the servers sees your attack as, what the actual total attack is, what your worn attack is and what the worn attack cap is if you are being capped.

I am not trying to push the formula on anyone. Just trying to figure out why the current formula is better.

[KLS]

Hm maybe the ATK cap never made it in or *gasp* is broken =(

ATK is broken into 2 parts:

Avoidance ATK which bypasses a defender's avoidance AC. (We don't do this correctly at all). It is a factor of attackers weapon and offense skill versus the defender's defense skill. + whatever avoid and hit mods from gear and spells.

Mitigation attack which bypasses a defender's mitigation AC. (We do this mostly correctly tho I have a few choice tweaks I still am going to make.) This is a factor of strength and atk from mods, the strength part isn't completely proven or disproven because it affects the max hit you can do it's hard to tell without sony saying yay or nay.

AC is also broken into 2 parts. Neither of which we deal with properly tho we can calculate it almost completely we treat the lump as mitigation ac which is wrong.

Avoidance AC and mitigation AC which are basically the counterparts to the above.

ATK and AC aren't handled entirely correctly but I'd rather not replace one broken system with one equally broken... esp when the current broken system is pretty balanced until we can fix it.

[trevius]

I understand what you mean about replacing broken systems with other broken systems. If it isn't broken, why fix it? I don't think of the combat system as being broken, but I do think it could use some tweaks. And I am not trying to cause a headache here, so sorry if I am :P

As far as I understand it, attack rating is only supposed to effect how hard you hit for. Basically it just increases your average hit damage. I have never heard that attack actually effects your chance to hit. I think chance to hit is all based on skill levels + bonuses (and a few other things), but that is separate from attack. Anyone know this for sure either way?

I know as well as any server admin that making changes to these core systems can mean huge overhauls on content. Believe me when I say that I don't want to have to do that anymore than the next guy lol. Heck, I still change the avoidance bonus formula back to what it used to be every time I update the code on my server. But, I do think that by dividing the new code by 2 (or so), it would put things almost exactly back to what they are now, so little to no changes would be needed. And since it only effects player DPS a little, it wouldn't mean any real content changes.

Ideally I think it would be good to work towards a more refined combat system that is as close to accurate as possible. The only way that can happen is 1 step at a time. But, I definitely don't think any huge changes to combat would be taken well if things were changing more than once a year or so. So, I propose that we would try to completely fix the entire system and then add it all in 1 big change. If entire DBs need to be adjusted, at least it would only be 1 time then. I remember when the new AC system was added and that was a major nightmare, but the emu is better off for it now. I think a combat system overhaul would be useful too.

Basically, I think that it is fine to keep the new formula out of the combat system for now. If we can get enough stuff corrected and tuned with new and better systems, then we can put those in all at once when they are finalized. So, the new formula can stay on the back-burner till then (if it ever happens). I think it is good to know that it is out there and about as close to perfect as we will be able to get it. But, if any changes go in for attack rating (like caps), I think we might as well replace the whole system for it. Because, if we are going to make a big change like that, we might as well make it final and accurate.

I definitely don't plan to attempt to overhaul the combat system myself. Since most of it is just plain math formulas, I am able to figure out some of it on my own, but most of it doesn't have enough hard data to make accurate changes. Figuring out the formula for attack wasn't too bad, because I had clear numbers to work with client side. That isn't the case for most of the other systems. Most of it would be guess work, and I think most of the guess work is already pretty good.

Here is an idea; How about making a new rule (or a few rules) for enabling new combat system changes? So, we could let people decide if they want to use the current way, or try the new stuff. I think most servers would just want to use the old way until the new stuff was finalized, but newer servers might want to use the new ways, since they have less adjustments to make to current content. I know too many rules can bog down the code, so the number of rules would probably need to be limited. It is just an idea anyway. Just thinking out loud :P