View previous topic :: View next topic |
Author |
Message |
MaximusDM
Joined: 13 Feb 2006 Posts: 876 Location: Massapequa, New York
|
Posted: Thu Nov 16, 2006 5:48 am Post subject: |
|
|
Whatever number they are spitting out at you on the extra stats screen has to be an estimate and can't be used for exact calculations. Plus there is already another page with a list of estimated cut offs. _________________
|
|
Back to top |
|
|
MJM249
Joined: 11 Nov 2006 Posts: 1237 Location: Minneapolis, MN
|
Posted: Thu Nov 16, 2006 5:52 am Post subject: |
|
|
Well, but from what I've seen, those estimated cutoffs still adhere to the old formula of [score/base + 2] = star rating. I have a 7.0 rating on Yes We Can at 189,000 in my scores, but the cutoff page still estimates it at 213,245, or 5x the base. |
|
Back to top |
|
|
JCirri
Joined: 04 Feb 2006 Posts: 4576
|
Posted: Thu Nov 16, 2006 10:48 am Post subject: |
|
|
Read my edit in the original post for updated information. _________________
|
|
Back to top |
|
|
Barfo
Joined: 10 Oct 2006 Posts: 2596
|
Posted: Thu Nov 16, 2006 11:27 pm Post subject: |
|
|
The base scores for M-Trogdor and H-Yes We Can are off on the "All Cutoffs" page. New formula implementation seems to be fine (interelationship between the numebrs is right), but the 4* and 5* cutoff is wrong compared to whats proven. For ex, H-YWC proven 4* cutoff is 91390, (thus a 45695 base), but the displayed cutoffs are 42,649 base and 85298 4* cutoff with higherscores seemingly calculated off those. M-Trogdor is similarly off too. Decimal score calculation on score display pages also seems to be using this data and coming to improper value as a result.
H-Trogdor, and H-BnH seem to be fine. |
|
Back to top |
|
|
JCirri
Joined: 04 Feb 2006 Posts: 4576
|
Posted: Fri Nov 17, 2006 4:36 am Post subject: |
|
|
You're right, I forgot to update one branch of my code. I've corrected the errors that were generated (by deriving the base_score as 5-star_cutoff/3, rather than 2.8 as it should've been).
I've updated the code, but I won't know for sure it works until the next 5-star cutoff is proven. _________________
|
|
Back to top |
|
|
Barfo
Joined: 10 Oct 2006 Posts: 2596
|
Posted: Fri Nov 17, 2006 3:53 pm Post subject: |
|
|
IMO you should fix base score as 4*/2, because (A) its much easier to get the 4* proof and i expect ot see many more 4* than 5* exact proofs pretty soon, and (B) assuming it truly is exactly double whatever numebr hte game has as the base score, 4* is always goign to be an even number (4 numbers all even suggests this but hte sample size is still too small as yet top know for sure) and thus 4*/2 will always be an integer, while 5*/2.8 can have a decimal and thus rounding errors might creep in if you use hte second formulation. |
|
Back to top |
|
|
JCirri
Joined: 04 Feb 2006 Posts: 4576
|
Posted: Fri Nov 17, 2006 4:37 pm Post subject: |
|
|
As soon as the first cutoff (either 4-star or 5-star) is proven for a song, the base_score is derived from that cutoff. If the 4-star cutoff is the first one proven, then the base_score is 4-star_cutoff/2. However if the 5-star cutoff is first to be proven, then the base_score is derived from that.
I think I've figured out the rounding issues, but it's possible there may be rounding errors that become known after the 4-star is proven (after the 5-star has already been proven). If that's the case I can simply update the base_score manually by a point to correct it (and possibly update my calculation accordingly). _________________
|
|
Back to top |
|
|
mattdaddy
Joined: 29 Mar 2006 Posts: 626 Location: Jonesboro, GA
|
Posted: Sat Nov 25, 2006 6:46 am Post subject: |
|
|
i see we have 8-star ratings now (i've got one on cherry pie, on hard) |
|
Back to top |
|
|
ES942
Joined: 02 Mar 2006 Posts: 3445 Location: Snalbans
|
Posted: Sat Nov 25, 2006 5:10 pm Post subject: |
|
|
i've got a couple on expert, if the 5.2 multiplier is correct. we just have to start proving some cutoffs. i need a cheat for not failing though. (maybe then i can pass heart shaped box ) |
|
Back to top |
|
|
wuLFe
Joined: 18 Aug 2006 Posts: 2679 Location: Richmond, IN
|
Posted: Mon Nov 27, 2006 4:34 pm Post subject: |
|
|
ES942 wrote: | i've got a couple on expert, if the 5.2 multiplier is correct. we just have to start proving some cutoffs. i need a cheat for not failing though. (maybe then i can pass heart shaped box ) |
Lol - I KNEW IT!!!! |
|
Back to top |
|
|
darthfisto09
Joined: 04 Sep 2006 Posts: 3
|
Posted: Fri Dec 01, 2006 5:09 am Post subject: |
|
|
quick question for anybody who has the answer
I see some songs on the GH2 list now have 6,7,8 stars.3 to be exact. anybody know when they rest of the songs will be like this? is it just a slow process to do them all? |
|
Back to top |
|
|
JCirri
Joined: 04 Feb 2006 Posts: 4576
|
Posted: Fri Dec 01, 2006 5:26 am Post subject: |
|
|
The added star ratings become activated as soon as the cutoffs are known. This means for every song that has either the 4-star cutoff or 5-star cutoff proven on the Song Stats page. So if you guys want to see more calculated ratings, help us get the cutoffs proven! Barfo has definitely been doing his share of work proving most of the GH2 songs on Hard (thanks again Barfo). _________________
|
|
Back to top |
|
|
Barfo
Joined: 10 Oct 2006 Posts: 2596
|
Posted: Fri Dec 01, 2006 5:46 pm Post subject: |
|
|
Quote: | anybody know when they rest of the songs will be like this? is it just a slow process to do them all? |
Right now i did a scattering of songs on H in order to target the easiest ones (basically based upon song length/low score), and I've started running down the set list (of hard) from the top (i'm at 'Woman' right now). After i finish about 90% of the list that way (might be skipping the really long songs, and ill definitely be skipping tough songs like CMH, PBFO, Six, thunderhorse), ill probably move on and hit whatever non-difficult Expert songs are left, but in order to do it at all efficaciously i need to be able to pretty much 97%+ the song so that i can easily establish a reliable baseline playthrough, so there will be a lot of expert songs that are outside my ability (at least using my current method). Right now it takes me about 10-25 plays per song per 4* cutoff (1-3 hours), mainly depending upon how rich the money section of the song is, and whether Riff's initial estimate of the cutoff is very accurate or not (if im flying blind i waste a lot of plays just trying to narrow it down). Also im extremely busy right now so im having hardly any free time to devote to GH2, so for the next few weeks ill be lucky to get 3 songs a week or so. By xmas it should open up and i hope to be mostly done with hard and moving on to expert by the start of Jan (if theres any X cutoffs left to do by then). |
|
Back to top |
|
|
jimmy-p
Joined: 25 Mar 2007 Posts: 15
|
Posted: Thu May 03, 2007 6:21 am Post subject: 8 stars |
|
|
can you not get a eight star on every song, i dont have any |
|
Back to top |
|
|
Parna
Joined: 15 Feb 2007 Posts: 169 Location: Sweden
|
Posted: Thu May 03, 2007 9:18 am Post subject: |
|
|
you can only get 8-Star on a few songs on each difficulty |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Copyright © 2006-2024 ScoreHero, LLC
|
Powered by phpBB
|