Max Calculator Thread

[Thai_tong]

Not accounting for them is about as inaccurate as not accounting for hit/miss data (just using 5-10h test for xp/h), so it's probably best to ignore that too for now :P.

Well that little fact sounds contrived. People are likely to have done many quests by the time they are interested in maxing and want to use this tool.

You should focus on what needs to be researched in order to get a working calculator that is off by 20 hours because it didn't count time to do quests. When it is working you can work on improving accuracy.

[Quyneax]

People are likely to be 70+ all stats too, for that matter. But that's not very interesting. Once people get to higher stats, with items already unlocked because of personal preference, it's going to be fairly easy to figure out what they should do. The hard part is taking into account quests and early levels. If you don't bother with that, it's really not very interesting to make a max calc.

 

And yes, the accuracy thing is only an estimate, but considering that not accounting for quests is a difference of 10ish hours of agility from WGS alone, I don't think it's far off.

 

In the end the calc has two purposes:

1) For people to use - reasonable methods (no 300+ apm etc.), most quests and early stats done, approximate solution.

2) For figuring out the real most efficient way to max ever - takes everything into account.

 

The first is less fun.

[Hedgehog]

I think the best way to solve this is to create the most efficient path to max, then calculate how much of the path has been completed.

 

Unless someone has an idea for a better algorithm, we can start working on creating the ideal path.

[Miss Lioness]

Well I think I already pinpointed which quests should definitely help in the beginning.

 

Also Hedge, you are forgetting 1 thing: WGS reward may cut-off 10 hours in agility. But if all the quests and WGS itself takes longer then those 10 hours + w/e cut-off in time remaining with the exp rewards, then it makes one slower to get maxed right? Then it is no longer efficient?

[Quyneax]

Depends, most people would probably want at least the full storyline quests complete (TWW including all optional rewards, that is most quests). Adding a toggle for 'quests', 'tasks' etc. is most convenient, not sure how that is coding-wise, I think comp is just as interesting as max to make a calc for as well, worth doing anyway.

 

@Hedge: You have to calculate the most efficient way to max from account creation (sadly, historical account creation can't be taken into account, so old accounts will always be less efficient) and from the current state, and then just figure out the percentage you've done. Because if your most efficient path involves slaying for 80-99 slayer and summoning, and someone has 99 slayer/90 summoning, you don't have 100% of slayer done, and you don't have 30% done either (or w/e % 90/99 is), you have more done. And if you have, say, 75 hunter, charm sprites might suddenly become efficient for a couple of levels, rather than slayer + jadinkos.

[Miss Lioness]

For now I would say: Screw hit/miss chances. At low level till about 60 this won't be any of concern. Hit/Miss chances starts to really pop in when you need to be there for hours on end.

[Thai_tong]

How would you decide how many times someone does daily activities?

[Miss Lioness]

How would you decide how many times someone does daily activities?

 

That is up to this person and not really something we can pinpoint in the calculator.

[Thai_tong]

How would you decide how many times someone does daily activities?

 

That is up to this person and not really something we can pinpoint in the calculator.

You have to include it somehow. An individual wont know the answer to that question either and you don't want someone to get stuck at the user interface so I suggest only getting them to put in their average hours/day which is something they do know. From their hours per day you can estimate the number of days they'll be playing for.

[Miss Lioness]

How would you decide how many times someone does daily activities?

 

That is up to this person and not really something we can pinpoint in the calculator.

You have to include it somehow. An individual wont know the answer to that question either and you don't want someone to get stuck at the user interface so I suggest only getting them to put in their average hours/day which is something they do know. From their hours per day you can estimate the number of days they'll be playing for.

 

That is something optional that only should be considered to put into the calculator once the global outline has been finished.

[Thai_tong]

How would you decide how many times someone does daily activities?

 

That is up to this person and not really something we can pinpoint in the calculator.

You have to include it somehow. An individual wont know the answer to that question either and you don't want someone to get stuck at the user interface so I suggest only getting them to put in their average hours/day which is something they do know. From their hours per day you can estimate the number of days they'll be playing for.

 

That is something optional that only should be considered to put into the calculator once the global outline has been finished.

Its part of the global plan :P

[Quyneax]

I do agree an hours/day average should be included, and maybe also a days/week option, as it probably matters quite a bit, with warbands, farming, daily challenges etc..

[The Floating Pen]

Will income/hour be factored in as well?

[Thai_tong]

Oh god here we go again... finding income when just out of the tutorial and up until you've maxed.

[Miss Lioness]

Personally this gets derailed too fast into all the optional things.

Lets first try and create a calculator that can calculate the time it will take to max combining what one has left to do.

 

Take me for example: I got all quests completed, so no worries on that one! I've banked all but 1 skill (smith) to 99 using the fastest method. Except for Herblore, that I do with Prayer Renewals.

 

How long would it then take me to max, taking in account the leveling to included the rise in exp/h however slight!

 

EDIT: Excluding combat ones, as they are really iffy with recent changes!

[Quyneax]

Oh god here we go again... finding income when just out of the tutorial and up until you've maxed.

You can't tell people to do broad arrows to 99 fletching @ Livid when they don't have the money for runes/arrows. You need to take into account +/- gp/h for all methods you're going to use training. GP is a crucial variable just like xp. I think of a maxing calculator as the solution for the smallest possible sum(a₁:a_n) of a sum of vectors: a₁[] + a₂[] + a₃[] + ... + a_n[], where each vector has 26 (or more) components, being xp/time in all 25 skills, gp/time, maybe effigies/time or produce/time as well. You start at a state x, with so and so much xp in those skills, this amount of gp etc. etc. - these and those vectors are available at state x, now you have to find a combination of them to get to point y, using the least amount of vectors possible, taking into account that more vectors may become available as you move.

 

Idk, it makes an easy to way to think about it but I can't do the actual math :(.

[Miss Lioness]

But I wonder why we should take that all into account when we just haven't started on writing the basic calculator?

[Thai_tong]

Oh god here we go again... finding income when just out of the tutorial and up until you've maxed.

You can't tell people to do broad arrows to 99 fletching @ Livid when they don't have the money for runes/arrows. You need to take into account +/- gp/h for all methods you're going to use training. GP is a crucial variable just like xp. I think of a maxing calculator as the solution for the smallest possible sum(a₁:a_n) of a sum of vectors: a₁[] + a₂[] + a₃[] + ... + a_n[], where each vector has 26 (or more) components, being xp/time in all 25 skills, gp/time, maybe effigies/time or produce/time as well. You start at a state x, with so and so much xp in those skills, this amount of gp etc. etc. - these and those vectors are available at state x, now you have to find a combination of them to get to point y, using the least amount of vectors possible, taking into account that more vectors may become available as you move.

 

Idk, it makes an easy to way to think about it but I can't do the actual math :(.

Thats not what I meant. In an actual calculator I think you should let someone choose their method, you don't force them to fletch broads. If you wanted to you could calculate their cost for their method but I was not talking about the actual calculator. I was talking about the part you think is fun which is finding the fastest way to max from the start. Depending on how far you are progressed in the game your income will change which means it may be worth making oak larders if you are not progressed and mahogany tables if you are well progress. Now good luck to you in your quest to find the fastest time to max, you'll have to decide whether a fast expensive method is worth doing at a given level which is impossible really.

[Quyneax]

No it's not? You don't have to tell whether it's worth it, that is for the program to decide. All you have to do is show the possible methods.

[Thai_tong]

No it's not? You don't have to tell whether it's worth it, that is for the program to decide. All you have to do is show the possible methods.

The program can't decide itself because you don't know what someone's income will be at every point in their goal to max and your program doesn't know either.

[Hedgehog]

I think you could solve that problem by creating a hierarchical list of incomes and when the requirements are met for one, then that becomes the highest income.

[Quyneax]

Well it's true there will be a lot of possible incomes and methods, for example mining methods change every level due to increased success chance etc. so that'd be a pain to put in. It may be easier to let the program calculate some values, such as the congold xp rate from 80-99, and of course ignore methods you can exclude beforehand, if you can prove it's always worse (includes obvious things like ivy, but could be other things). It may also be worth collapsing chunks of levels into one, for example 1-15 woodcutting, because the method simply never changes - that way you get one method, with average xp/h, for 14 levels.

 

Anyway, I don't see why you would need anything but a list of methods to find the best route?

 

There's a point A and B, and you want to connect A and B with as few steps as possible. So taking a sort of map of which states lead to which other states, you can calculate the path from A to B with the fewest steps.

[Hedgehog]

The best way to do it (I think) is to find the best way to go from 34 to 2475 total and try to connect to that in the fewest number of hours. Let's start brainstorming what that path is.

[Thai_tong]

How about we start off with a simple skill, herblore. I think the calculator would have to check the productivity of each potion as new ones are unlocked so there would be a lot of steps to the calculation. At 96 the xp that would be got from the herbicide from slayer and possibly combat training should be added.

[Hedgehog]

Crafting is probably simpler because it doesn't affect income like overloads and extremes do, so I'd start with that.