Oneiromancer
Level: Legendary Smitemaster

Rank Points: 2936
Registered: 03-29-2003
IP: Logged

Hooray for Schik once more! There is a new, temporary highscores page up at the address http://www.drod.net/forum/highscores2.php (the same as the normal one, but with 2 after "highscores"). This page uses almost the same rules that I mentioned in this thread, with a few small changes. The biggest change is that the first person to get the best score in a room gets 1st place. If you tie that move count, you get 2nd place, along with everyone else who ties that score. Likewise, everyone that ties for the next best move count gets 3rd place. Keeping the 1st place rank separate makes the "bonus" for getting the first 1st place rank an inherent part of the system, and also makes getting "1st place!" in-game a much more special event, instead of wondering if you tied someone else or if you really are the best.

So if you compare the current high scores page and the new one, you'll see that not much has changed in the top 10...a few people switched places, but they were pretty close in score anyway. Below the top 10, there's a pretty good mixup as all the people who were just under the wire got some extra points and jumped ahead. The highscores2 page also give everyone in 8th place or higher 1 point, basically just for completing the room. There might be a bonus for completing a hold in the future too.

So long as you navigate to all the other pages from the highscores2 page, they will all be calculated in the new way. You should also be able to go to any page, and replace the "highscores" with "highscores2" in the URL, and get the new version (or vice-versa).

Over time, the highscores2 pages will get more and more inaccurate, as new highscores being uploaded won't be calculated correctly, and will be set to zero. So we should evaluate this quickly and see if it's something that we want to replace the old system.

Psst! Go here to mod Schik up!

Game on,

____________________________
"He who is certain he knows the ending of things when he is only beginning them is either extremely wise or extremely foolish; no matter which is true, he is certainly an unhappy man, for he has put a knife in the heart of wonder." -- Tad Williams

---

[Last edited by Oneiromancer at 07-27-2005 07:33 PM]

Watcher
Level: Smitemaster

Rank Points: 902
Registered: 02-04-2003
IP: Logged

OMG I dropped from #50 to #54! I totally hate this change!

Actually, I like it. It seems like a better way to calculate the scores. There's one thing I'd change, though: I'd set the ranking of a high score to the number of people who have done better, plus one. So if one person gets a move count of 125 in a certain room, and there's one person on 123 and three on 124, he'd get fifth place. My reasoning is mainly that if there's four people that have done better than you, it shouldn't matter whether their scores are identical.

Although I do like the exception that only one person can hold first place.

____________________________
Today the refrigerator, tomorrow the world!

Oneiromancer
Level: Legendary Smitemaster

Rank Points: 2936
Registered: 03-29-2003
IP: Logged

Watcher wrote:
There's one thing I'd change, though: I'd set the ranking of a high score to the number of people who have done better, plus one. So if one person gets a move count of 125 in a certain room, and there's one person on 123 and three on 124, he'd get fifth place. My reasoning is mainly that if there's four people that have done better than you, it shouldn't matter whether their scores are identical.

Yeah, Schik and I discussed that option. He suggested it, actually, but I don't really like it. First of all...I think it means that for the people who are not tied, their rankings don't change at all. Second, it doesn't help out the people who start out late as much. If 150 people are tied with the same score (in 1st and 2nd places then), and you get 1 move less by accidentally bumping into a wall, you can either get 3rd place with the current method, or 151st place with your method. Sure, you could redo the room to fix your mistake, but not everyone wants to do that over and over again, so this is a more immediate and visual reward.

I'm also thinking about the psychology of getting your ranking announced after beating a room. With both my method and yours, if you get 1st place you know you're the best (and the first to come up with that solution), and if you get 2nd place you know you're either a few moves short or tied for best. Now say you got 8th place...with the current new method that means there are 6 or 7 move count totals which are better than you, depending on whether the 1st and 2nd place people are tied or not. With your suggested method, it's not so clear...are there 7 individual different scores above you, or only 1? Maybe it's not such a big deal, but I'd rather get 25th place and think "wow, I really need to improve" instead of "bah, I missed only 1 move and I lost that many points?"

But I'm open to more persuasion, if lots of people agree with Watcher's method.

Game on,

____________________________
"He who is certain he knows the ending of things when he is only beginning them is either extremely wise or extremely foolish; no matter which is true, he is certainly an unhappy man, for he has put a knife in the heart of wonder." -- Tad Williams

---

[Last edited by Oneiromancer at 07-27-2005 08:22 PM]

larrymurk
Level: Smitemaster

Rank Points: 1911
Registered: 12-09-2004
IP: Logged

I prefer Watcher's method.

KevG
Level: Smiter

Rank Points: 333
Registered: 08-16-2004
IP: Logged

I like it. I really hated getting 50+ scores in JTRH even though I got the room perfect. I think my average ranking under this system is much more indicative of my actual skill than it was under the old system. Of course I'm in 15th place average ranking wise under both systems, but I still like the fact that the new average is much lower.

I would have preferred that everyone who got the high score get a #1 ranking, but it's not a particulary important point. Watcher's suggestion, however, is one that I would like to see implemented. As a general rule of thumb, once you earn a score your ranking for that score shouldn't improve unless you improve the score. With the currently implemented new system, everytime someone improves an existing score to make a tie, everyone else moves up a notch. To me this doesn't seem right.

Oneiromancer
Level: Legendary Smitemaster

Rank Points: 2936
Registered: 03-29-2003
IP: Logged

KevG wrote:
As a general rule of thumb, once you earn a score your ranking for that score shouldn't improve unless you improve the score. With the currently implemented new system, everytime someone improves an existing score to make a tie, everyone else moves up a notch. To me this doesn't seem right.

That's a good point, and I understand why people like this idea, but unless we change the rest of the scoring system, it might not make that much of a difference. Say there's this room where 10 people have the best score, 15 people have the next best, and everyone else scored worse.

In the original scoring system, the first 8 people to get the best score get points in order of who did it first, and no one else gets any points.

In the current new scoring system, the first person to get the best score gets the most points, the other 9 people with that score get almost as many points, the 15 people with the next best score get a bit fewer points, and so on, so that a lot more people get a reward for solving the room.

With Watcher's method, the first person to get the best score gets the most points, the other 9 people with that score get almost as many points, and no one else gets any points. It's almost the same as the original scoring system, except the people that get points all have the same amount instead of having a variety.

If most rooms only had 1 or 2 people tied for a higher place, then Watcher's method would perfectly reasonable to me. But with the amount of people who play the game, I feel that there will be many rooms where it's pretty much all or nothing...which is almost exactly what the complaint is about the current system, except that eventually you'll be able to redeem yourself by getting a perfect score. With my method, I think there would be less demo copying, because people would be happy to just have a score, have some points, and not feel that they have to copy demos in order to get on the scoreboard at all (which many people don't like as it is).

I await the rebuttal.

Game on,

____________________________
"He who is certain he knows the ending of things when he is only beginning them is either extremely wise or extremely foolish; no matter which is true, he is certainly an unhappy man, for he has put a knife in the heart of wonder." -- Tad Williams

agaricus5
Level: Smitemaster
Rank Points: 1838
Registered: 02-04-2003
IP: Logged

KevG wrote:
As a general rule of thumb, once you earn a score your ranking for that score shouldn't improve unless you improve the score. With the currently implemented new system, everytime someone improves an existing score to make a tie, everyone else moves up a notch. To me this doesn't seem right.

I don't want to seem stupid, but I'm confused. With the current new system, how would improving your score increase other people's?

For example, suppose 20 people get #1 (19 are then tied at #2), 10 hold #3 and 5 hold #4. I play the room, and get a move total that puts me in 3rd position (tied) so I get 9 points, but no-one else is affected. Now, I improve the score to tie with everyone in 2nd position, so I get 3 points more, but still, no-one else gets affected.

I then am inspired to get a new highscore, and beat the 1st player by 1 move. I gain 4 points, and only the person previously with the #1 score loses a rank. Now, someone else in 2nd position enraged by the fact that I hold a lower score, decided to copy my demo. He then moves into 2nd position, and everyone else loses a rank.

I don't see how improving a score could therefore increase the rankings of others.

Edit: Oh, I see - If a rank has only one person in it, then an increase will shift everyone up. Sorry.

Actually, I'd say that's not too much of a problem, since you can never move above 3rd like this, and it will require everyone in a position above you to have improved their score. It's possible if you're 7th, 8th or 9th, I guess, but if you're 4th or 5th, I'd think that pretty unlikely. And, even so, to make any noticeable impression on a person's position on the leaderboard, a lot of people are going to have to get busy, since 9 points compared to 1000, for example, is tiny.

____________________________
Resident Medic/Mycologist

---

[Last edited by agaricus5 at 07-27-2005 09:34 PM]

Maurog
Level: Smitemaster

Rank Points: 1501
Registered: 09-16-2004
IP: Logged

I don't like the new system, mostly because of the fact there is uncontrolled flow of points. In the old system, there were 55 points per room. In the new system, each room can bring thousands of points into the system. Let's analyze how it solved the entrance room problem: In the old system, the first eight people got (undeserved?) points for a trivial solution. This is kinda bad, but I don't think it matters that much. In the new system, everyone and their uncle get a second place, and it is as good as a second place in a 500+ moves room. Now, I'm all for giving out points and encouraging the new people and so on and so forth, but this is ridiculous. I don't find a 4-turn solution entrance room with 100 people at second place very amusing... do you?

If I had to invent a system, I'd probably go with the following - Each room comes with an initial solution from the architect as the 1st place demo. The first place gets an amount of points equal to the number of turns in his solution. Second place gets half as much, third place gets 1/3 and so on. Am I making any sense?

____________________________
Slay the living! Raise the dead!
Paint the sky in crimson red!

agaricus5
Level: Smitemaster
Rank Points: 1838
Registered: 02-04-2003
IP: Logged

Maurog wrote:
If I had to invent a system, I'd probably go with the following - Each room comes with an initial solution from the architect as the 1st place demo. The first place gets an amount of points equal to the number of turns in his solution. Second place gets half as much, third place gets 1/3 and so on. Am I making any sense?

Interesting, but if you're first in a room with 700 moves, and suddenly find an even better solution of 550, you'd lose 150 points, just for improving your score, which goes against the whole idea of the system in the first place. Also, the 1/x factor will immediately bias all scores towards the person in first place, which is perhaps a bit unfair if you're not as fast (there are quite a few rooms with reasonably long demos but which have very simple principles, so they are at present completely optimised by at least 10 people), and there's also the problem of people coming late only getting very few points, compared to the large size of the demo. In my opinion, your system is perhaps slightly worse than the current new system, since although it tackles point devaluation (everyone can get them) it introduces other problems that make it less friendly to newer players. Anyway, to copy every single demo in first place will require a lot of effort (many demos are complicated and >500 moves, and require you to complete the room first), and it is probable that anyone who actually takes the time to bother with this will eventually appreciate the techniques involved anyway.

____________________________
Resident Medic/Mycologist

Alneyan
Level: Smitemaster
Rank Points: 622
Registered: 07-06-2004
IP: Logged

I do like the new system for rank, but I'm not so sure about score: my average rank went up from 85 to 55 because of all those #2 ranks I got (instead of being somewhere between #10 and #20 on most of those rooms), but my score went up from less than 20 to 450, which is probably too much given my performance (pretty much average). StuartK seems to have gone down quite a lot in score, so perhaps the reward for #1 isn't high enough under the new system?

Alternatively, some rooms, either predesignated or rooms having a solution in "less than X moves", could bring fewer points, since doing well there is much easier; after all, being efficient in a tar mother room should bring in more points than stabbing a roach in the back on level 1. That way, the easiest rooms will not hurt new players like they do under the old system, but they will bring less of a reward than those nastier rooms where angels fear to tread; it would also not be dependent on how many players beat the room, unlike the current system.

pouzzler
Level: Delver
Rank Points: 33
Registered: 05-24-2005
IP: Logged

I seem to have gained two places and two thousand points with this new system; it makes me a little perplexed, but hey! Everything's good, isn't it?

leroy00
Level: Master Delver

Rank Points: 155
Registered: 09-30-2003
IP: Logged

Maurog wrote:
I don't like the new system, mostly because of the fact there is uncontrolled flow of points. In the old system, there were 55 points per room. In the new system, each room can bring thousands of points into the system. Let's analyze how it solved the entrance room problem: In the old system, the first eight people got (undeserved?) points for a trivial solution. This is kinda bad, but I don't think it matters that much. In the new system, everyone and their uncle get a second place, and it is as good as a second place in a 500+ moves room. Now, I'm all for giving out points and encouraging the new people and so on and so forth, but this is ridiculous. I don't find a 4-turn solution entrance room with 100 people at second place very amusing... do you?

Well, your post addresses two entirely diffrent things which are not necessarily related. The question you've addressed in the second part of your post only relates to a small number of rooms (not even all entrance rooms fit the description), so maybe you should rethink whether this argument is a valid reason for criticising the entire system. This question has been addressed in another thread, where I suggested that there be a criterium for "trivial rooms" that would disallow high scores for that room, i.e. "no monsters/trapdoors/orbs/bombs/etc." If there's no puzzle, why give a high score? This seems to me to be the most elegant way to address this (rather minor) issue. There could be a flag for the hold author to set which would do this, but I feel sure that Schik would find a way to extract this information and automatically discard the room from the high scores list as the hold is made available on CaravelNet.

In reply to the first three sentences of your post: You fail to state why you find the increase in amount of points bad. It is not as if only others get more points; in all probability, you do also. The fact is that (not taking demo copying into account, which also skews the scores in the current system) the scores are now a more realistic reflection of performance now, since not so much emphasis is laid on the question of order of conquering a room, which in general is a question of who has the most time to play DROD and who goes through the holds most quickly. The high scores page in its current form isn't very interesting for new-comers, because, for many rooms, even a perfect solution puts them in a place below #8. (Currently, JtRH L25 has 115 explorers.) Admittedly, only a portion of all rooms has been ideally solved, but new players are currently excluded from this portion of rooms completely. True: If you are really worried about others gaining points relative to your standing, you may have to spend more time optimising your solutions in the future, but the new system changes nothing about the fact that one is an efficient/inefficient player. True: The new system may theoretically be more susceptible to demo copying than the current one, but I don't really believe that there are inefficient players out there who are so vain that they will copy demos for complicated rooms, as Agaricus pointed out. It seems to be consensus in the DROD community that watching demos is mainly good for learning new (general) techniques and less useful for appearing to be a better smiter than one really is.

-leroy

____________________________
You can hear happiness staggering on down the street -- footless, dressed in red.
-Jimi Hendrix, "The Wind Cries Mary"

Mattcrampy
Level: Smitemaster

Rank Points: 2388
Registered: 05-29-2003
IP: Logged

I believe the concern is that a room that is easy, and thus lots of people have the optimal solution, is worth the same amount as a hard room, in which only a few people have the optimal solution.

On the other hand, this merely requires a shift in perspective. Points under this system are exponential - a point at 5500 is worth more than a point at 2000.

____________________________
What do you call an elephant at the North Pole?

leroy00
Level: Master Delver

Rank Points: 155
Registered: 09-30-2003
IP: Logged

Mattcrampy wrote:
I believe the concern is that a room that is easy, and thus lots of people have the optimal solution, is worth the same amount as a hard room, in which only a few people have the optimal solution.

So in an extreme case, everyone gets the same optimal score, everyone has the same number of points (except the first conquerer), the field is levelled, and it is (very nearly) as if the room never existed. What's the problem? It is inherent to the game that the wheat will only be separated from the chaff in tricky rooms, not trivial ones. I really doubt anyone here expected anything different.

-leroy

____________________________
You can hear happiness staggering on down the street -- footless, dressed in red.
-Jimi Hendrix, "The Wind Cries Mary"

KevG
Level: Smiter

Rank Points: 333
Registered: 08-16-2004
IP: Logged

Oneiromancer wrote:
But with the amount of people who play the game, I feel that there will be many rooms where it's pretty much all or nothing...

Actually, that not true, there's currently a 1 point bonus for completing a room. Nobody walks away empty handed with either the current new system or with Watcher's proposed adjustment.

which is almost exactly what the complaint is about the current system, except that eventually you'll be able to redeem yourself by getting a perfect score.

But, that's a critical difference. The current high score system is broken because it's impossible for newcomers to lower many of the ridiculously high scores they get saddled with in JtRH. As long as it's possible to get points, there's no reason to make it easier; newcomers should be encouraged to improve their skills.

Points should reflect how well you did compared to others. In a room where numerous people have gotten the highest score, it seems perfectly reasonable to expect people to get the high score if they want more points than just the completion bonus.

Maurog wrote:
I don't like the new system, mostly because of the fact there is uncontrolled flow of points. In the old system, there were 55 points per room.

Which is why the old system is broken. With a constant influx of new people, there needs to more points brought into the system. A person's score is supposed to have at least some relationship to their ability. With newcomers being locked out of numerous rooms, they have no way of earning a score equivalent to their actual skill and progress.

In the new system, each room can bring thousands of points into the system.

So? Points don't have any intrinsic value. They only are important in so far as they determine how you rank compared to others. A fair system will introduce as many points as necessary to determine people's relative skills.

I don't find a 4-turn solution entrance room with 100 people at second place very amusing... do you?

I'd rather the room not count at all towards the score; but giving everybody points neutralizes the effect of trivial rooms. It's definitely preferable to destroying newcomers average ranking by forcing them to get unlowerable scores of 100+.

Maurog
Level: Smitemaster

Rank Points: 1501
Registered: 09-16-2004
IP: Logged

You know what? I don't care about it anymore... Use whatever system, they are all wrong anyway. Make it so whenever you complete a room you get a random amount of points, I bet that will encourage new people. Hell, call them experience points and make people level when they get enough. Or even better, award people with items they can "wear" on their character. Or invent a deterministic system that awards points regardless of other people, only based on your achievements. Then you won't have to worry about old people spoiling it for new people, right? Why do we encourage people to make shortest solutions anyway? If we make it so the *longest* solution is the best, we won't have to worry about ever reaching the optimum. That will give everyone a fighting chance... there will always be room for improvement. Or maybe award points for monsters killed instead of moves made. So he stayed longer and the queens spawned, but he killed more roaches, isn't he a better delver then? Oh, I know, just give the last person visiting the room all the credit and nothing to the rest, *that* will encourage everyone to re-visit the rooms. Nothing like some friendly competition... fun fun fun.

I had a point, but it got lost in the way somehow. Ah, whatever.

____________________________
Slay the living! Raise the dead!
Paint the sky in crimson red!

pouzzler
Level: Delver
Rank Points: 33
Registered: 05-24-2005
IP: Logged

I feel I arrived at the perfect time to evaluate both systems fairly:

the old system IS hard for newcomers, and I had to work quite a lot to get where I am, even beginning JtRH only a month and a half late.
People beginning a year from now? High scores would hold NO incentive for them because they would be ranked 123th or something equally devoid of meaning.

Moreover, almost every highscoring system in the world works the way the new system works:
everyone who gets the same time (sports), grade (exams), lack of respect (politics), number of true answers (quizzes), ...., gets the same rank.
There must be some value to something independantly developped and rediscovered time and time over, musn't there?

To go on with the argument, I go from 35th to 33th with the new system.... what a difference! Ranks hold no more meaning! The establishment collapses!

Err.. sorry, I might have gone a little overboard in the last paragraph

Rabscuttle
Level: Smitemaster

Rank Points: 2460
Registered: 09-10-2004
IP: Logged

How about:

"First's the worst,
Second's the best,
Third's the one with the hairy chest."

rowrow
Level: Smiter
Rank Points: 432
Registered: 08-17-2004
IP: Logged

When I used to say that
I wondered what does it have to do with having a hairy chest. . C'mon, I was 3 years old.

____________________________
B'hakhgra Du S'tra Moth'ness Ti!

---

[Last edited by rowrow at 07-29-2005 04:17 PM]

agaricus5
Level: Smitemaster
Rank Points: 1838
Registered: 02-04-2003
IP: Logged

pouzzler wrote:
Moreover, almost every highscoring system in the world works the way the new system works:
everyone who gets the same time (sports), grade (exams), lack of respect (politics), number of true answers (quizzes), ...., gets the same rank.
There must be some value to something independantly developped and rediscovered time and time over, musn't there?

To go on with the argument, I go from 35th to 33th with the new system.... what a difference! Ranks hold no more meaning! The establishment collapses!

Well, since the #1 score is what distinguishes those who actually bothered to innovate, as opposed to just copying, maybe just a little more bias on this idea might help reduce the problem.

How about a slight modification:

Onto the new system, we add "innovation points". Basically, if you get a #1 score for any room at any time, you gain a fixed number of points. These points are permanent, and so even if you then lose your #1 position later, you still keep these points as reward for having come up with a better solution previously (although you can only get the points once, so if you beat the person in first place again, you won't get the bonus again). This way, people who bother to try for the #1 place will do better on average than just someone who copies all the winning demos on Caravel.net, even if they don't happen to be quite as good at other players at retaining their #1 places. Of course, there will be a problem with there being no record of this for anyone but the current holders of 1st place demos, but I guess the number of completely unoptimised rooms isn't that large, so in the long run, this problem may become less important.

Does anyone think this is a good idea?

____________________________
Resident Medic/Mycologist

Oneiromancer
Level: Legendary Smitemaster

Rank Points: 2936
Registered: 03-29-2003
IP: Logged

At first I was worried about your idea, Wesley, because I was know it's pretty much impossible to quantify "innovation." But then I read it, and it sounds like a decent idea. However, it's something that would just be extra--like getting 1 point for completing a room even if you don't get in the top 8 (which Schik had already coded up before the alternate scoring system page, and probably would have added eventually). So I don't think it makes much of a difference when deciding between the two alternates presented in this thread.

I agree with what pouzzler said...I devised this alternate scoring system because I was thinking about an exam in a math class...it doesn't matter who turns in their paper first, everyone that gets the same score gets the same grade in the class. The little "bonus" we have for first place is like getting extra credit for coming up with a particularly elegant solution on the math exam. I'm not worried about having a surplus of points, because there's still a maximum that any one person can have...and it's not like they're going to be spent on anything, it's just a way of comparing yourself to everyone else. While the other method that Watcher presented is reasonable for one room at a time, when you look at the whole picture I think it makes it harder to compare yourself with everyone else, because it is still easy to do very well, and still not get any points at all.

Those who are at the very top of the rankings, because they've done the most holds, are likely to stay there for a long time, with either scoring system. But if you want to see how well you performed on one hold compared to everyone else who solved that hold, I'd rather have a gradual scale than an all-or-nothing scale.

Game on,

____________________________
"He who is certain he knows the ending of things when he is only beginning them is either extremely wise or extremely foolish; no matter which is true, he is certainly an unhappy man, for he has put a knife in the heart of wonder." -- Tad Williams

agaricus5
Level: Smitemaster
Rank Points: 1838
Registered: 02-04-2003
IP: Logged

Oneiromancer wrote:
At first I was worried about your idea, Wesley, because I was know it's pretty much impossible to quantify "innovation." But then I read it, and it sounds like a decent idea. However, it's something that would just be extra--like getting 1 point for completing a room even if you don't get in the top 8 (which Schik had already coded up before the alternate scoring system page, and probably would have added eventually). So I don't think it makes much of a difference when deciding between the two alternates presented in this thread.

Those who are at the very top of the rankings, because they've done the most holds, are likely to stay there for a long time, with either scoring system. But if you want to see how well you performed on one hold compared to everyone else who solved that hold, I'd rather have a gradual scale than an all-or-nothing scale.

But wouldn't this "innovation bonus" be a sort of intermediate grade?

It's a bonus that several people can get, so it's not like the 1st place bonus, which is by definition unique, but there is a limit to how many can get it, since you can only optimise a room so far. This way, it's not like the room completion bonus or the 2nd place bonus, which everyone can eventually get with enough effort. Also it rewards trying to find better solutions as opposed to just copying for 12 quick points (you end up with more then 12 even if you do slip to second some time later). The effort hierachy might then be something like this:

1st place score - 16 (Only one person can get this)
Innovation bonus - +3 (Anyone who gets a first place at any time can get this, but since rooms can only be optimised so far, probably only 5-15 people can get this per room)
2nd place score - 12 (Anyone who bothers to copy the 1st place demo or finds a variation on it can get this, and everyone has the potential to get this eventually for any room)
Completion bonus - +1 (Anyone who bothers to complete a room will get this, so everyone will have the potential to get this eventually for any room).

To show you how this may work in context, take for example this scenario for an imaginary 50 room hold:

Player A is a very experienced DROD player and currently holds 30 first place demos and 20 second place demos, although for 18 of these rooms, he was previously in first place but was beaten by someone else. He would then get 30*16 (1st place) + 20*12 (2nd place) + 48*3 (innovation) + 50*1 (completion) = 914 points.

Player B is enthusiastic, but not as good as player A, so although at present he only holds 5 first place demos and 45 second place demos, for 25 of the rooms in which he is second, he had been first at some time in the past, even if only for a little while. He would get 5*16 (first place) + 45*12 (2nd place) + 30*3 (innovation) + 50*1 (completion) = 760 points.

Player C is only interested in rising up the leaderboard as quickly as possible. To do this, he has completed every room, downloaded the winning demos from DROD.net, and copied them directly. Because of this, he managed to gain 50 second place demos in only 4 hours, which greatly improved his standings. So, he would get 50*12 (2nd place) + 50*1 (completion) = 650 points

Player D is a slow and careful player who wishes to optimise his scores in a room first before moving on to the next. Because of this, he has only played 20 rooms, but of those, he is first in 12, and second in the other 8. In those 8 second place rooms, he was previously first in 5 of them before being beaten by someone else. He would get 12*16 (1st place) + 8*12 (2nd place) + 17*5 (innovation) + 20*1 (completion) = 213 points

Player E is, however, not too bothered about his actual scores; he only wants to complete as many holds as possible. Because of this, he has only obtained 2 sixth place scores, 1 seventh place, and 3 eighth place scores. He would get 2*3 (6th place) + 1*2 (7th place) + 3*1 (8th place) + 50*1 (completion) = 61 points

Finally, player F is a beginner player, and is more concerned with actually completing all the rooms in the hold than the actual scores or ranks he gets for his demos. He only has completed 27 rooms up until now, and none of his demos are above 11th place. Because of this, he only gets 27*1 (completion) = 27 points.

From this, we can see that...

Fewer complete rooms (player F) < more complete rooms (player E) << fewer complete rooms, but high efficiency (player D) << more complete rooms, all at 2nd place (player C) < more complete rooms with holdings of a few 1st places (player B) << more complete rooms with holdings of many 1st places (player A).

Basically, the more effort you put in, the greater the reward. Innovation points are a sort of extra bonus, but they can only be obtained by the few (relatively) players who are willing to try to improve a demo rather than just copy it.

____________________________
Resident Medic/Mycologist

---

[Last edited by agaricus5 at 07-31-2005 08:04 PM]

SuperV
Level: Delver

Rank Points: 34
Registered: 11-30-2004
IP: Logged

Wow good!

There's my scores : http://www.drod.net/forum/highscores2.php?action=playerinfo&id=1170

____________________________
D.R.O.D. Progress :
KDD : Level 8 - 3 rooms remaining
JTRH : Level 8 - 3 rooms remaining

SuperV hint : BUY JTRH!

Maurog
Level: Smitemaster

Rank Points: 1501
Registered: 09-16-2004
IP: Logged

This innovation stuff is a good idea, but you must be sure it cannot be abused. For example, Player E (not from previous exaple, E for evil) visits a room with current best demo at 580. Studying the room carefully, E notices that if you get to this room from another exit, it is much simpler and can be completed in 380 moves. So, what E does is complete the room but at turn 279 wait 199 more turns and exit, getting a score of 579 moves and innovation bonus. Then repeats waiting 198 moves and so on and so forth. That gives E in total +600 points innovation bonus +16 first place bonus +1 completing the room bonus. Also, for innovation bonus you will need some sort of architect demo/score as mentioned previously. For people who are first to visit the room etc.

____________________________
Slay the living! Raise the dead!
Paint the sky in crimson red!

agaricus5
Level: Smitemaster
Rank Points: 1838
Registered: 02-04-2003
IP: Logged

Maurog wrote:
This innovation stuff is a good idea, but you must be sure it cannot be abused. For example, Player E (not from previous exaple, E for evil) visits a room with current best demo at 580. Studying the room carefully, E notices that if you get to this room from another exit, it is much simpler and can be completed in 380 moves. So, what E does is complete the room but at turn 279 wait 199 more turns and exit, getting a score of 579 moves and innovation bonus. Then repeats waiting 198 moves and so on and so forth. That gives E in total +600 points innovation bonus +16 first place bonus +1 completing the room bonus.

You should only be able to get the bonus once, so it can't be abused in that way. You could try to deny the chance of anyone else getting it by yourself optimising the room, but that amounts to protecting your first place score.

Also, for innovation bonus you will need some sort of architect demo/score as mentioned previously. For people who are first to visit the room etc.

If you're first to play a room, then you're the first to get the bonus, since you discovered the first "optimal" solution. Obviously, this will benefit fast players in the sense that the first players to complete a room will get the bonus, but unless the player is lucky (or planned) enough to get the optimal solution immediately, there's no reason why someone else can't improve on it and get the bonus too afterwards. It's not entirely fair, but perhaps encourages people to aim to get the lowest move counts. For transit rooms with very short demos that become optimised by the first person, maybe a variation could be added:

The innovation bonus will not be awarded to the first player to get first place in any room, unless someone else beats that demo (i.e. you can't get the points by making a demo and then beating it straight after). If this happens, both the previous holder of 1st place and the new holder get the bonus points. This way, no bonus is awarded for immediately optimised rooms (no improvement is possible), which may be fairer as the first person only gets a 4 point advantage, not 7, but people are still not discouraged to try to get first place scores, since it is still worth it in the long run - being first to get 1st place will get you 16 points, and even if someone else beats you later, you will still get 12+3 for having been first, compared to only 12 for just getting a second place score.

____________________________
Resident Medic/Mycologist

---

[Last edited by agaricus5 at 07-31-2005 03:08 PM]

StuartK
Level: Smitemaster

Rank Points: 564
Registered: 06-10-2003
IP: Logged

Many, though not all rooms with trivial solutions could be excluded from the highscores if a few filtering rules were added. e.g. rooms in which no monsters are killed (and arrows traversed/trapdoors dropped/tar stabbed?) and the solution is X moves (the distance from the entrance point to the nearest exit using brained pathfinding logic?)

Also, once a room has X number of solutions exactly the same, perhaps that room could be excluded from the highscores (though demos would still be available) except for the 1 point room completion bonuses? That way only rooms for which there is scope for genuine competition are scored...

KevG
Level: Smiter

Rank Points: 333
Registered: 08-16-2004
IP: Logged

agaricus5 wrote:
Obviously, this will benefit fast players in the sense that the first players to complete a room will get the bonus, but unless the player is lucky (or planned) enough to get the optimal solution immediately, there's no reason why someone else can't improve on it and get the bonus too afterwards. It's not entirely fair, but perhaps encourages people to aim to get the lowest move counts.

What's going to happen for most dungeons is that the first few players are going to really rake in the bonus points. It almost amounts to a bonus for playing a dungeon early. That's a good thing.

The first few players who play a dungeon are at a major disadvantage. They have no idea what target score to shoot for. This means they have no idea which rooms they've done the best in and which rooms they have room for improvement. A seemingly perfect #1 now might not even be in the top 10 in a few weeks. Giving them the better chance to earn some bonus points evens things out a bit.

I like it, but I do have a question about implementation. Is this only going to be applied to new dungeons? Is it going to be applied to new dungeons with the bonus applied to everyone with a current #1? Or, does the information exist to give the bonus to everyone who has held a #1? Personally, I would prefer #2 since I suspect it's the option that would benefit me most.

agaricus5
Level: Smitemaster
Rank Points: 1838
Registered: 02-04-2003
IP: Logged

KevG wrote:
I like it, but I do have a question about implementation. Is this only going to be applied to new dungeons? Is it going to be applied to new dungeons with the bonus applied to everyone with a current #1? Or, does the information exist to give the bonus to everyone who has held a #1? Personally, I would prefer #2 since I suspect it's the option that would benefit me most.

Schik would know, since he's in charge of the whole scoring system, but if it were me, I'd assume that all rooms are currently optimal, and use the variant rule:

The innovation bonus will not be awarded to the first player to get first place in any room, unless someone else beats that demo (i.e. you can't get the points by making a demo and then beating it straight after). If this happens, both the previous holder of 1st place and the new holder get the bonus points.

If a room actually is optimal, then the player in first position will have the 4 point advantage anyway, and if not, then he should have the credit for having been first when he gets beaten eventually. Of course, if Schik does have the data stored away, then using it to calculate bonuses would be the best solution, but if not, then I think this would be best.

____________________________
Resident Medic/Mycologist

---

[Last edited by agaricus5 at 07-31-2005 05:11 PM]

Tscott
Level: Smiter

Rank Points: 382
Registered: 02-10-2003
IP: Logged

agaricus5 wrote:
1st place bonus - +16 (Only one person can get this)

I'm not understanding this part of your proposal. So the first person to get 1st place gets 16 points? Wouldn't that just be the first person to complete the room? Or do you get 16 points until someone beats you and then those 16 points get taken away from you and added to the person who beat you?

____________________________
And I can recall our caravel: a little wicker beetle shell with four fine maste and lateen sails,
its bearings on Cair Paravel. O my love, O it was a funny little thing to be the ones to've seen.
-Joanna Newsom "Bridges and Balloons"

Oneiromancer
Level: Legendary Smitemaster

Rank Points: 2936
Registered: 03-29-2003
IP: Logged

agaricus5 wrote:
But wouldn't this "innovation bonus" be a sort of intermediate grade?

Actually, Wesley, I was mostly trying to say that I view your "innovation bonus" as something that could be added to any high score system, and thus deserves being discussed on its own merits, but doesn't necessarily have any bearing on the discussion between the proposals we have so far...as far as I can tell, at least. I suppose if it were much more complementary to one proposal over the other, and people really liked it, then it could be considered in support of that proposal.

Game on,

____________________________
"He who is certain he knows the ending of things when he is only beginning them is either extremely wise or extremely foolish; no matter which is true, he is certainly an unhappy man, for he has put a knife in the heart of wonder." -- Tad Williams