Math vs. Feeling - New Wizards

Wizard
Either of those aps is above the break brotha.

Excellent point Tekk. The models are as accurate as the designer and the circumstances they are representing.
09/02/2013 08:20 PMPosted by Melkor
Excellent point Tekk. The models are as accurate as the designer and the circumstances they are representing.
It's a balancing act, and that's why I like your post about math vs. feeling. Healthy skepticism is good. We don't want to be bull-headed and dismiss good models and "math" out of hand, but likewise we should be very wary about fully embracing incomplete models. And even if we had a perfect model, the "feeling" aspect of each player would still strongly bias that player's perceptions of what's "good" and "bad." Some of us think lower breakpoints (<2.73aps) are junk, while some of us are perfectly fine with it. Same thing with mitigation or sustain, etc. And this is friendly disagreements between very competitive players that understand the core game mechanics. Thus, it's very complicated, and often we can't rely on math to quantify "a satisfying gaming experience." It's dreadfully inconvenient! :p
that's why i chose not to use the CM/WW and archon setup anymore. It's too common these days, as for most of you have been using the same build over and over again. I don't feel unique at all. And when it comes to efficiency in farming ,id admit it's not the best, but still comparable specially when the "fun factor" comes in.
09/02/2013 08:09 PMPosted by MrDuMa
maths = fun..?

No "?" about it. Math is fun. I used to run around and write the phrase "math is fun" on all the chalk/whiteboards when I was in high school. Ticked off my friends who didn't think math was fun. Nerd then, nerd now.

Evidence, however, is absolute.
hahahaha run around writing "math is fun" on the boards. What a rebel.

I vaguely remember a nerd kid getting bullied back in grade 5 (or something).. he finally snapped at the bullies and said something along the lines of, "You wanna settle this like real men?" The bullies thinking they were gonna physically beat him up agreed... he screams out, "Let's have a MATHS COMPETITION."

Their enthusiasm disappeared.
09/02/2013 08:48 PMPosted by ChangBooster
maths = fun..?

No "?" about it. Math is fun. I used to run around and write the phrase "math is fun" on all the chalk/whiteboards when I was in high school. Ticked off my friends who didn't think math was fun. Nerd then, nerd now.

Evidence, however, is absolute.
I disagree! "Perfect" evidence is absolute, and shoddy evidence is well... flawed and uncertain! :p Evidence is just evidence, and should always be critically evaluated based on its quality.
Loroese:
I agree with your post completely. The main thing that gets me is when people post claims that we can show to be untrue mathematically, then continue to refuse to even consider the math behind our statements.

At the end of the day, it's your play time and your character. You can do with it what you want, and play how you like. It's also fine to share builds that you find to be fun or interesting or even MP10 viable. Just don't go around saying your build is better than SNS or Archon without some hard proof. Also don't go saying that your gearing is the best way to do something when there's numerous posts by people who have put a lot of time and effort into proving otherwise, unless you have some evidence to back up your claim.

Basically, as long as people don't start claiming opinion as fact, I'm fine with just about anything said around here. Posing a theory and asking for opinions is perfectly fine also.

I agree with this, but I also find the flip side to be true. I see a LOT of people on this forum innappropriately using the "math," such as the simulators, as "proof" of their opinions. A lot of the time, the models can only mathematically predict certain portions of the mechanics, and we still have some very glaring gaps in our understanding of the game mechanics. Thus, I think it's also important not to get carried away with the math, and treating something as "fact" in cases where our knowledge is still incomplete. Models are still useful predictors, but they should be just that... useful predictors and not proof. I'm very leery of people using models, particularly Steve's simulator (http://d3cmww.com/), as proof that certain setups are "better." Models should never be taken as primary proof, rather they are good supporting evidence and guides to make novel predictions (again not "proof"). :)


I totally agree with that too. Part of the reason I didn't put more work into my simulator was because we just didn't understand everything at the time, so I didn't like that I had to use a bandaid fix for the SA procs. I was hoping to end up with something very accurate, but it ended up basically the same as the d3cmww one, which is just a close approximation. Still, most of the math involved in the simulator has been backed up with in game testing, which is why I dislike when people post claims counter to what the simulator predicts without any backup evidence.

In the end, the math is just a working theory that has been shown to be true and just hasn't been disproven, but can't really be proven to 100% certainty without inside information about how the coding in the game works. If I'm interpreting your post correctly, that's along what you were saying, namely that math shouldn't necessarily be taken as complete truth either, or at least shouldn't be spouted without full understanding of the meaning behind the math. Like people telling others to raise their APS to 2.73 breakpoint without saying anything about why they should.
09/02/2013 07:42 PMPosted by Melkor
Are we "losing you" to console?

I'll still be around I'm sure.

09/02/2013 05:43 PMPosted by ChangBooster
I'm gonna be a console concubine tomorrow. Self found 10k DPS wiz FTW!


Think I might have been wrong about the 10k DPS thing though.
http://us.battle.net/d3/en/forum/topic/9792970549
might be faked cb
09/02/2013 09:07 PMPosted by TekkZero
I disagree! "Perfect" evidence is absolute, and shoddy evidence is well... flawed and uncertain! :p Evidence is just evidence, and should always be critically evaluated based on its quality.


I stay away from shoddy evidence. You better tell everyone from here on out who posts a ghom test that they gotta redo it with you in their house standing over their shoulder making sure the sample isn't tainted.
I'm going to use the pronoun "you" in place of players who are new or utterly lost.

If you're having fun, go for it. If not, stop it and do something else.

There are guides here designed to improve your play experience, should you choose to follow them. Guides here aren't absolute, they're recommendations and suggestions, nothing more.

Read. Use the search function. Read. Read. Did I mention read? Read. Please. There are many "optimal" builds out there tested time and again for you to run should you choose to go the min/max, efficiency route. There are also many "for fun" builds out there tested time and again to be "interesting but not quite powerhouse level." There are numbers and gameplay videos everywhere at your disposal. Use them. Digest them.

There are reasons some builds are deemed more powerful than the others. Numbers. Experiments. Theories. While no one is perfect, many here strive for perfection. We then rely on numbers as much as possible, hoping to one day make all theories concrete proof. Then there are some of us who don't care about effectiveness. Acknowledge the numbers and keep them in the back of your mind should you choose to make tweaks in your build or gear. But do not discount the numbers unless you did some serious number crunching of your own (I personally go out and test numbers on a whim, but I'm no math wiz). We're all constantly learning and hungry for more knowledge (we're wizards after all).

Again, go out and have fun your way. Just remember there are always ways to make your build/gear set better, which can potentially lead to more fun for you. That's where the numbers and theories come in.

/latenightrambletimetosleep
________________________________________________
Diablo III MVP | Forever a Wizard
Archon Wizard Video Guide Series:
- http://us.battle.net/d3/en/forum/topic/9793089550
Twitter: @Jaetch
Loroese:
Still, most of the math involved in the simulator has been backed up with in game testing, which is why I dislike when people post claims counter to what the simulator predicts without any backup evidence.

In the end, the math is just a working theory that has been shown to be true and just hasn't been disproven, but can't really be proven to 100% certainty without inside information about how the coding in the game works. If I'm interpreting your post correctly, that's along what you were saying, namely that math shouldn't necessarily be taken as complete truth either, or at least shouldn't be spouted without full understanding of the meaning behind the math. .
It's that, but the simulator doesn't completely replicate all the game mechanics either. And my view is that someone should always use the simulator as supporting evidence for any claim made: both positive and negative claims. :) The thing is that a simulator can be right a lot of the time, but we should be wary of accepting any numbers as fact unless it's backed by someone doing an independent in-game test.

It's like taking Wikipedia as fact. It's a great starting point for people writing a research paper, but one should never reference Wikipedia in a paper. Likewise using most of the math and models we spout is a great starting point, but one should always double-check everything. For example, I'm wary of people overinterpreting say how crit chance or attack speed breakpoints impact the effective DPS multiplier of a SNS wizard. We know there's a link, but I'd be wary of using the simulator to tell us what that number is. I see a lot of people posting the results of the simulator, and then argue it as fact. I'm very uncomfortable when I see those numbers, since factors like latency or incomplete knowledge of game mechanics (like CM or SA procs) can strongly impact the result of the very model they are using. And it's awesome to use the model, but unless they have an accompanying reference of in-game tests, I'm anxious about the validity of their conclusions.
09/02/2013 09:48 PMPosted by ChangBooster
I disagree! "Perfect" evidence is absolute, and shoddy evidence is well... flawed and uncertain! :p Evidence is just evidence, and should always be critically evaluated based on its quality.


I stay away from shoddy evidence. You better tell everyone from here on out who posts a ghom test that they gotta redo it with you in their house standing over their shoulder making sure the sample isn't tainted.
Heh, I like Ghom tests for single-target testing, since they are pretty easy and straightforward.

I worry more about other kinds of comments, like extending single-target data on Ghom to general farming (or extrapolate to another person's experience). Or trying to use simulators+models to "prove" a point. Basically anything that tries to extrapolate data beyond its ability to predict! :lol You could say I'm just trying to be a healthy skeptic, in that I'm worried that very often we try to generalize more than can be supported by data. I'm of the opinion all evidence should be evaluated critically and considered only within the proper contextual boundaries it was collected. :)
09/02/2013 08:09 PMPosted by MrDuMa
Gotta hand it to Loroese and Comrade though.. I'd be one of those idiots who stack IAS not realising I have 2.729APS rather than 2.73APS.. at least they gave us clear goals.

It appears they didn't give them clearly enough. If you had 2.729 APS, you would reach the breakpoint.
09/02/2013 08:48 PMPosted by ChangBooster
maths = fun..?

No "?" about it. Math is fun. I used to run around and write the phrase "math is fun" on all the chalk/whiteboards when I was in high school. Ticked off my friends who didn't think math was fun. Nerd then, nerd now.

Evidence, however, is absolute.

I am a math major. I always thought math was absolute....until I learnt the squeezing principle in uni.

I love me some math too. Probably because my strict Chinese parents beat me until I liked it.
09/02/2013 10:52 PMPosted by MindsMirror
Gotta hand it to Loroese and Comrade though.. I'd be one of those idiots who stack IAS not realising I have 2.729APS rather than 2.73APS.. at least they gave us clear goals.

It appears they didn't give them clearly enough. If you had 2.729 APS, you would reach the breakpoint.

Umm.. you did read what I wrote, right?
09/02/2013 11:01 PMPosted by Ryoka
Probably because my strict Chinese parents beat me until I liked it.


probably most Asian parents do

Join the Conversation

Return to Forum