Best dps 5.2

02/06/2013 12:24 PMPosted by Dinan

The ones that are played by players who know their class, the raid mechanics, and can adapt to situations on the fly.

I have nothing to add here.

No, I kid. But that was an excellent point.

Something that Ghostcrawler has brought up in the past is that top players will tend to hone in on a specific spec, and wring the most out of it. Parses are posted showing the results, and others follow suit. Soon the parses get dominated by players doing very well with a particular class and spec. Now, we know from internal testing that other options might be competitive, or even better, but because of the... uh, I guess I'll call it the 'echo effect', those specs never really get a chance to shine because everyone has already found something that works, and works well.

Of course, that's not always the case. Sometimes those other specs might not be getting spotlight time because they're much harder to get optimal performance out of, or other factors. Still, it's a real phenomenon.

The other thing to keep in mind is that we (as a community) have developed a strange relationship with damage numbers. Once upon a time, there were pretty broad differences in dps between the highest and lowest dps classes. Bringing the wrong class could have a real impact on your ability to win a fight. Over the years I think we've gone a very long way toward fixing that, and the gap between the different classes and specs has narrowed considerably. Instead of 50 and even 100% differences in output, now we're usually talking about 5 and 10% differences. At these ranges, skill, timing, and knowledge play a more central role than ever before in determining how much practical damage a class is capable of, and even the classes that aren't topping the charts are contributing essential and meaningful amounts of damage.

Don't get me wrong, I have fun vying for the top spot too. I'm just saying that even if your class isn't the #1 top damage for a particular encounter, you're still in a position to help make an important contribution to the fight and ensure that your raid group is killing bosses.

Not trying to quote out of context, but a 10% difference will never sit well with the bulk of the community. I'd even go so far as to say a 5% difference is probably too much. Yeah it's come a long way, but it simply doesn't work in the community's perspective on damage output.

I know, and I get it, but I also think that's kind of unfortunate. Due to vagaries in individual player capabilities, differences in class ability mechanics, latency, encounter design, the moon phase, etc., getting dps numbers within a couple % points of each other is extremely difficult. In some cases, it's not even desirable. What I mean by that is, it's cool that some classes perform better in specific encounters when their unique traits become valuable. As I said earlier, being top dps is fun, if it's you. It can be fun OR frustrating to shoot for that target if it isn't you. Regardless, the point of good dps is defeating encounters, which is much more a product of the raid's efforts as a whole, and their collective skill and knowledge.

Would we like dps (and tanking, and healing) output to be even more balanced, just on principle? It would be awesome. Is there a point of diminishing returns where constantly tweaking those things doesn't significantly impact the success or failure of players and raid groups in general? I think that's also the case. I'm not saying we can't do better, but we do have to prioritize our time to yield the best overall experience we can provide. We've made progress over time, and we'll continue to try to do so. In the meantime, I hope that most players feel like they can choose the class they want to play, and not feel like that choice is holding them back from being a productive member of their group.

This notion that a thin layer of elite players is leading the masses down the wrong path is cute. It's way more likely that your lab results, for whatever reason, don't translate perfectly in a live environment.

Oh, we know for a fact that lab results don't always translate into a live environment. I didn't use the word 'wrong' because it would be... well, wrong. Our theory crafting community is smart, and their analysis is sound. The specs and classes they advocate have been proven to be strong and effective, and are the top performers in many cases.

What I'm suggesting is that often other specs can still be viable, competitive, and better than they're presented to be by parses, but that's not always reflected because they didn't make the initial theory crafting cut.