Author Topic: realtalk 2.0  (Read 865 times)

Exile

  • Professional Jumper
  • ****
  • Posts: 471
  • fhhh
    • View Profile
Re: realtalk 2.0
« Reply #15 on: April 20, 2017, 02:56:32 PM »
bring back skillsrank

catfoosh

  • Professional Jumper
  • ****
  • Posts: 368
  • brock is a weeb
    • View Profile
Re: realtalk 2.0
« Reply #16 on: April 20, 2017, 03:02:08 PM »
I don't think attempts is a good idea because it includes random people who join that have no intention on beating the map.

gibus

  • Advanced Jumper
  • ***
  • Posts: 142
  • MEDIC!
    • View Profile
    • Steam
Re: realtalk 2.0
« Reply #17 on: April 20, 2017, 04:01:51 PM »
I don't think attempts is a good idea because it includes random people who join that have no intention on beating the map.
that would affect all maps evenly tho so it wouldnt matter
huha! stream highlight!

catfoosh

  • Professional Jumper
  • ****
  • Posts: 368
  • brock is a weeb
    • View Profile
Re: realtalk 2.0
« Reply #18 on: April 20, 2017, 04:24:21 PM »
I think its a random thing I don't really think its an even spread. I think beginner maps would get a lot more attempts do to lower skill players joining and not finishing

gibus

  • Advanced Jumper
  • ***
  • Posts: 142
  • MEDIC!
    • View Profile
    • Steam
Re: realtalk 2.0
« Reply #19 on: April 20, 2017, 04:30:50 PM »
beginner maps would also get a lot more completions for the same reason tho

honestly the only way to find out if it works is to test it
huha! stream highlight!

scotch

  • Moderator
  • Advanced Jumper
  • *****
  • Posts: 211
    • View Profile
Re: realtalk 2.0
« Reply #20 on: April 20, 2017, 05:43:49 PM »
I think thats a fairly ambiguous metric to track but keep the ideas coming

Ladyboog

  • Advanced Jumper
  • ***
  • Posts: 202
    • View Profile
    • ladyboog
Re: realtalk 2.0
« Reply #21 on: April 20, 2017, 05:50:46 PM »
Just make sure to gimme some credit in the new point system for being incredibly smart.

879m

  • Professional Jumper
  • ****
  • Posts: 272
  • Techa mengu, go!
    • View Profile
Re: realtalk 2.0
« Reply #22 on: April 20, 2017, 06:16:55 PM »
Keep completion points and tiers, but get rid of top ten. Instead, wr will have the same amount of points, and all places beneath it will have points based on the time difference between them and the wr. So if you are 2nd place by 10s you will get less points than someone who is 2nd place by 0.1s.
I'm not sure how the scaling would be done, perhaps last place is worth 0 extra points. Of course this doesn't work particularly well for maps with low completions. For those with high completions it'd be great though, now any improvement is awarded with points, even if you don't go up a ranking.

scotch

  • Moderator
  • Advanced Jumper
  • *****
  • Posts: 211
    • View Profile
Re: realtalk 2.0
« Reply #23 on: April 21, 2017, 01:08:20 AM »
That was the original plan to encourage improving your PR.

Mario

  • Newbie
  • *
  • Posts: 16
    • View Profile
Re: realtalk 2.0
« Reply #24 on: April 21, 2017, 04:30:37 AM »
For completion, the ultimate idea was what catfoosh said - the more people who have beaten the map, the less points you get for completion. I think that this is a really good idea, and my only concern with it was that shitty maps that nobody likes to play would start getting farmed more because they are worth more than better maps of the same difficulty. Not sure what the perfect solution is but I definitely like a dynamic completion based point system over strict tiers.

I think this could work for maps which are t4 or harder. From memory, I think maps of the same difficulty at these tiers generally have roughly the same number of completions after a few months. e.g littleman, soar and valley (which is much newer) have similar numbers of completions and difficulty (haven't checked, but I think they are roughly around 200?).

There are some exceptions like exilenpals which has like 30 completions even though its t5 whereas a lot of t6 maps have about 80.

For lower tiers, popular/old maps like beef have disproportionate amounts of completions relative to other maps of the same difficulty since they are played a lot more and a lot of these players don't stick around/play all the maps.

What about making more tiers / subtiers? About a year ago I remember someone made a spreadsheet giving each t5/t6 map a more accurate rating (e.g pharaoh would get something like a 5.7 to reflect it being harder than iT which could get a 5.1). Idk if anyone still has the link to that.

Keep completion points and tiers, but get rid of top ten. Instead, wr will have the same amount of points, and all places beneath it will have points based on the time difference between them and the wr. So if you are 2nd place by 10s you will get less points than someone who is 2nd place by 0.1s.

I'm not sure how well this would work in reality. Being 30 secs off WR on beef is not that great whereas being 30 secs off WR on pharaoh might get you 2nd or 3rd place. Even if you considered the difference as a % it wouldn't work either (e.g being twice as slow as WR on beef is awful but being twice as slow as WR on pharaoh is still pretty amazing.)
I prefer the % or exponential decrease systems suggested.

Some other games specify certain time bands (which are unique for every map) you can come in to get more points. However, that looks like a lot of effort and it might have to constantly be updated as new strats are found.