March 12, 20187 yr Author I know that and they are right 500/2 ~250 as shown to road to pro competitions tab. The thing is that if you sum up road to pro comp. +official comp.+ top 15 bench = not the right world ranking points as seen in first screenshots.
March 12, 20187 yr Well, well, well... As Einstein used to say with a bitter laugh, even math can some times be subjective and prone to different explanations. Logically 1+1=2, but who can argue that it can be 1+1 = 3-1. Is the glass half full or half empty ? The algorithm, upon which the whole point rankings calculations is based, is WRONG. And I would be most happy to prove to Mr. Frederick Collardyn that statement. It started on a wrong basis, has been revised so many times and it will never be correct. It is the math basis that is wrong. And if one or more "scientists" that designed this, are unable to figure out the problems, then there is a significant issue here. I DO NOT CARE. But the very few that might still be interested to keep this site alive, should be.
March 12, 20187 yr Rev 7 was published way too soon. This site is just a mess now. Imagine trying to understand how the points and rankings works for a newcomer... Just kill off the points system. Make it a pure database. Competitive overclocking is dead anyway.
March 13, 20187 yr Author Thanks Frederick,much better now.Please also add Galax competition points and we are done
March 13, 20187 yr Author 25 minutes ago, richba5tard said: Can you point me to the specific competition which does not give competition points? I did not find it on the schedule. Sure, http://oc-esports.io/#!/round/galax_goc17_finals http://oc-esports.io/#!/round/galax_goc17_qualifier
March 13, 20187 yr Administrators Phil, diue to the new settings, would you mind disabling points for your subs that are listed under your personal account? You know that with a team account, you cannot have a persoanl single account with hardware ranking This happened due to the recalculation and new settings I think
March 14, 20187 yr 22 hours ago, richba5tard said: Yeah some rankings where completely bugged, other rankins/leagues where not but the text did not match the algorithm. I've fixed all test and pushed the test version to production a few minutes ago. Mmmm...........still fucked up. http://hwbot.org/submission/3390025_mr.scott_gpupi_for_cpu___1b_phenom_ii_x4_960t_be_12min_54sec_7ms Global points can't possibly be correct.
March 14, 20187 yr Author 2 hours ago, websmile said: Phil, diue to the new settings, would you mind disabling points for your subs that are listed under your personal account? You know that with a team account, you cannot have a persoanl single account with hardware ranking This happened due to the recalculation and new settings I think Sure Michael,but do I have to do it one by one or there is an other way?
March 14, 20187 yr 21 minutes ago, Mr.Scott said: Mmmm...........still fucked up. http://hwbot.org/submission/3390025_mr.scott_gpupi_for_cpu___1b_phenom_ii_x4_960t_be_12min_54sec_7ms Global points can't possibly be correct. Indeed. 2nd place gets less points than 3rd and 4th getting 1/3 of 11th:http://prntscr.com/iqv49q@richba5tardcan you also enable calculation for competition subs? We posted 2 results on comp that are better than our previous general scores, and havent received points for them. 1st: http://hwbot.org/submission/3807692_ogs_gpupi_for_cpu___1b_core_i7_8700k_2min_28sec_50ms 2nd: http://hwbot.org/submission/3807688_ogs_cinebench___r15_core_i7_8700k_2329_cb Thanks
March 14, 20187 yr On 3/13/2018 at 8:43 AM, Fasttrack said: Well, well, well... As Einstein used to say with a bitter laugh, even math can some times be subjective and prone to different explanations. Logically 1+1=2, but who can argue that it can be 1+1 = 3-1. Is the glass half full or half empty ? The algorithm, upon which the whole point rankings calculations is based, is WRONG. And I would be most happy to prove to Mr. Frederick Collardyn that statement. It started on a wrong basis, has been revised so many times and it will never be correct. It is the math basis that is wrong. And if one or more "scientists" that designed this, are unable to figure out the problems, then there is a significant issue here. I DO NOT CARE. But the very few that might still be interested to keep this site alive, should be. ^^^^^ this, if it was right then the all the points calculations would be right with it, and clearly it isnt, not having a go at anyone here, theres been losts of changes, sure ..... but if you havent a clear and concise formula thats working correctly from the start, then the result will be what is happening now, the points calculations to this point dont know where theyre going to !!!
March 14, 20187 yr Administrators 11 hours ago, phil said: Sure Michael,but do I have to do it one by one or there is an other way? I think if you go to account settings you can chose that you do not want to participate in ranking with this account. I guess this should do the job
March 14, 20187 yr 1 hour ago, richba5tard said: Did you hit recalc on dancops #1 position? no - pressed it on our submissions that were listed under the comp. Maybe that fixed the issue
March 14, 20187 yr Members 15 minutes ago, richba5tard said: FYI points will be recalculated only when #1 spot changes in a ranking, rank position is always recalculated Any idea as to why the weird global points stuff is happening when it didn't on the uat?
March 14, 20187 yr 1 hour ago, richba5tard said: An automated test with a limited db does not cover all the scenario's which occur on a 2.2 million db with 1500 people making submissions. WoW........
March 14, 20187 yr 24 minutes ago, richba5tard said: I'm testing a build on UAT which has much improved speed for recalculating the rankings, as the problem with all rankings which are not ok now is not that the algorithm is wrong, but there are just too many recalcs going on from people pushing buttons. So you're saying if we all go and hit recalc on our profile that'll fix it right? Edited March 14, 20187 yr by yosarianilives
March 15, 20187 yr Members 3 hours ago, richba5tard said: An automated test with a limited db does not cover all the scenario's which occur on a 2.2 million db with 1500 people making submissions. Fair enough, possibly to disable people recalculating for a bit to help the database sort itself out?
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.