K404 Posted August 6, 2012 Share Posted August 6, 2012 Looks good so far Quote Link to comment Share on other sites More sharing options...
Massman Posted August 6, 2012 Share Posted August 6, 2012 clear punishments for the most common & worst OC crimes. (e.g. break rule X get 6 month ban. Break rule y, get a warning. Break it a second time get a one year ban) I think there should be a lot more detail than there is at the moment. This is what the work-in-progress got so far. Anything that should be added? A. Types of Punishment - blocked submission: submission blocked for points and ranking - warning: user informed that something is wrong - blocked account: user cannot submit results for X period - banned account: user no longer contributing to the team B. Offenses - stealing result - accidental shared submission - photoshopping result - using illegal tweak - benchmark hack - hardware/score sharing - verification incorrect - submitted in wrong category - selling/buying results - attach inappropriate screenshots/pictures C. Punishment System - Nothing - Warning - 1st degree user block (= two week suspension from submitting results) - 2nd degree user block (= four weeks suspension from submitting results) - 3rd degree user block (= two months suspension from submitting results) - 1st degree user ban (= three months suspension from loggin in + all results blocked) - 2nd degree user ban (= six months suspension from loggin in + all results blocked) - 3rd degree ban (= one year suspension from loggin in + all results blocked) - Eternal ban! D. System - 1 violation = 1st degree, 2 violations = 2nd degree, 3 violation = 3rd degree - # violation is defined by violation event - one event can include multiple results Nothing (moderator staff fix submission): - submitted in wrong category - incorrect verifcation - entered incorrect information in screenshot Warning - repeatedly making same submission errors - attach inappropriate screenshot/pictures User Block - hw/score sharing - using illegal tweak User Ban - score stealing - editing result info in benchmark verification screenshots - benchmark hack - selling/buying benchmark results Quote Link to comment Share on other sites More sharing options...
Christian Ney Posted August 6, 2012 Share Posted August 6, 2012 Massman for president Quote Link to comment Share on other sites More sharing options...
K404 Posted August 6, 2012 Share Posted August 6, 2012 Took me a minute to think through it all, but I think I like it Thankyou! Thinking (not criticizing) Is a two week block from submitting results a punishment? For example..... if that happened to me (which it won't, but for sake of discussion).....would I even notice? Would it scare people off making bad submissions if the team points were docked for a length of time if a sub submission was made by someone who'd be on the team longer than X months? (X months to stop people changing team to sabotage them) Quote Link to comment Share on other sites More sharing options...
Massman Posted August 6, 2012 Share Posted August 6, 2012 Hm, good idea. What if we make it like: - Nothing- Warning - 1st degree user block (= two week no submit + temp no points for results) - 2nd degree user block (= four weeks no submit + temp no points for results) - 3rd degree user block (= two months no submit + temp no points for results) - 1st degree user ban (= three months no login + perm no points for results) - 2nd degree user ban (= six months no login + perm no points for results) - 3rd degree ban (= one year no login + perm no points for results) - Eternal ban! Temp no points = user points and team contribution are set to "0", but reset to previous value when user block lifts. Perm no points = "result blocked" Quote Link to comment Share on other sites More sharing options...
K404 Posted August 6, 2012 Share Posted August 6, 2012 Sounds good What results would the "perm no points" apply to? Obviously, the result(s) that get a user blocked or banned would be deleted. Everything they ever submitted? (forcing them to start again) Can I mention two related things? Redcobra is still invisible in leaderboards, although his ban has been lifted (HWB needs a tweak) Remember the convo about what results need an FM link? I submitted a 2x top 20, no WR points (actually no points at all!) but HWB wouldn't let me submit until I added an FM link. The server has over-ridden the rule that only scores with WR points need an FM link Quote Link to comment Share on other sites More sharing options...
Massman Posted August 6, 2012 Share Posted August 6, 2012 "no points" always applies to the results itself. Results get no points -> user = sum of 'no points' scores -> user/team have no points. For user block, after the block the results get points again. For user ban, the results don't get points again. Another interesting idea is to block certain users from certain hardware categories after a ban ("you can submit in those categories anymore"). Technically doable, I think ... but is it a good approach? Quote Link to comment Share on other sites More sharing options...
Hyperhorn Posted August 6, 2012 Share Posted August 6, 2012 - selling/buying benchmark resultsLoophole-finders could trade benchmarks results without money/real value involved ("You get my Super-Pi-screen in exchange for your Vantage-result."). Quote Link to comment Share on other sites More sharing options...
Massman Posted August 6, 2012 Share Posted August 6, 2012 Good point. Change that to "submitting benchmark results you did not obtain yourself, either in private or in group session" ? Quote Link to comment Share on other sites More sharing options...
knopflerbruce Posted August 6, 2012 Share Posted August 6, 2012 I like that last idea. That punishment is more permanent:) Quote Link to comment Share on other sites More sharing options...
Hyperhorn Posted August 6, 2012 Share Posted August 6, 2012 I'm fine with that as it includes also donated results. Quote Link to comment Share on other sites More sharing options...
K404 Posted August 6, 2012 Share Posted August 6, 2012 I like that last idea. That punishment is more permanent:) I like it too For repeated offences, could expand to series or architecture. LOL....would it be possible to be banned from benching "Intel"? Quote Link to comment Share on other sites More sharing options...
I.M.O.G. Posted August 6, 2012 Author Share Posted August 6, 2012 Looks really good. I went through superpi, pifast, ucbench, and pcm05 specifically and they look good. Quote Link to comment Share on other sites More sharing options...
GENiEBEN Posted August 6, 2012 Share Posted August 6, 2012 Temp no points = user points and team contribution are set to "0", but reset to previous value when user block lifts This sounds like a possible future bug Other than that, can't wait to see these imposed (fairly certain I fail couple of them). Quote Link to comment Share on other sites More sharing options...
BeepBeep2 Posted August 6, 2012 Share Posted August 6, 2012 (edited) @Massman Is it possible to require motherboard CPU-Z too for 3D submissions? On some IGP, GPU-Z does not read correctly. For example, someone had recently submitted several scores for GMA 4500M that absolutely blew mine away: http://hwbot.org/benchmark/3dmark03/rankings?hardwareTypeId=videocard_1950#start=0#interval=20 But I'm afraid they are really GMA 4500MHD (GM45). There is no way to tell in GPU-Z, because 4500M (GL40) is read in GPU-Z as GM45 as well. See CPU-Z and GPU-Z here:http://d1ebmxcfh8bf9c.cloudfront.net/u21589/image_id_818868.png I'm not sure how often it happens, and I know it would be complicated/contrary to current rules but I'd like to know what you think on the matter. Edited August 6, 2012 by BeepBeep2 Quote Link to comment Share on other sites More sharing options...
Massman Posted August 7, 2012 Share Posted August 7, 2012 This sounds like a possible future bug Yes it does :D @MassmanIs it possible to require motherboard CPU-Z too for 3D submissions? On some IGP, GPU-Z does not read correctly. For example, someone had recently submitted several scores for GMA 4500M that absolutely blew mine away: http://hwbot.org/benchmark/3dmark03/rankings?hardwareTypeId=videocard_1950#start=0#interval=20 But I'm afraid they are really GMA 4500MHD (GM45). There is no way to tell in GPU-Z, because 4500M (GL40) is read in GPU-Z as GM45 as well. See CPU-Z and GPU-Z here:http://d1ebmxcfh8bf9c.cloudfront.net/u21589/image_id_818868.png We've included CPU-Z Mainboard tab rule for IGP-based results now. Would that suffice too? If you know the mainboard, you also know the IGP used right? I'm not sure how often it happens, and I know it would be complicated/contrary to current rules but I'd like to know what you think on the matter. Quote Link to comment Share on other sites More sharing options...
oanvoanc Posted August 7, 2012 Share Posted August 7, 2012 Rendered image differs too much from the original due to other software tweaking How can we clearly express "image differs too much" by definition? e.g. The image with LOD +8 differs more from the original than with LOD +2. So that would mean, that the same tweak will be allowed and disallowed at the same time (just by a different setting). Quote Link to comment Share on other sites More sharing options...
Crew Turrican Posted August 7, 2012 Crew Share Posted August 7, 2012 How can we clearly express "image differs too much" by definition? e.g. The image with LOD +8 differs more from the original than with LOD +2. So that would mean, that the same tweak will be allowed and disallowed at the same time (just by a different setting). lod changes textures and not the "rendering" itself. what is meant, is when for example some stuff isn't rendered at all and missing. Quote Link to comment Share on other sites More sharing options...
Gunslinger Posted August 7, 2012 Share Posted August 7, 2012 Rendered image differs too much from the original due to other software tweaking Doesn't black screen benching violate the spirit of this rule? Quote Link to comment Share on other sites More sharing options...
BeepBeep2 Posted August 7, 2012 Share Posted August 7, 2012 Yes it does :D We've included CPU-Z Mainboard tab rule for IGP-based results now. Would that suffice too? If you know the mainboard, you also know the IGP used right? I'm not sure how often it happens, and I know it would be complicated/contrary to current rules but I'd like to know what you think on the matter. Sounds good, I think most are detected correctly by GPU-Z but in case it happens the user can just be asked for confirmation... I mean, it should be a no issue with recent APU/intel GMA HD but for that example there is no way to tell without looking at the mainboard. I think most ATI/NVIDIA are detected correctly though. (AMD HD4xxx IGP weren't for some time in the beginning) Quote Link to comment Share on other sites More sharing options...
Massman Posted September 4, 2012 Share Posted September 4, 2012 Btw, this is the work-in-progress for the general rules page (contains everything aside from the benchmark-specific rules): http://hwbot.org/news/7914_rules_update_2012_hwbot_general_rules_and_guidelines Currently this is just one big wordpress blog page, but I assume it would be more convenient to turn it in a hardcoded page with different sections under different tabs? Quote Link to comment Share on other sites More sharing options...
Mr.Scott Posted November 11, 2012 Share Posted November 11, 2012 UPDATE THE GOD DAMN RULES !! Do you think rules are not that important? How do you intend to enforce accurately what isn't yet written? The old rules don't cut it anymore, due to many 'un-written', yet enforced rules, and quite frankly, I'm F'ing sick of it. The picking and choosing whats enforced and on whom is getting ridiculous. Quote Link to comment Share on other sites More sharing options...
Vinster Posted November 15, 2012 Share Posted November 15, 2012 I had a member of my team ask me where the rules were this week... I still can't find them other than the links from this post... I told him what I know and as for the screenshot I told him to look at the Top10 for that bench and do what they do... but this is getting silly... if I click on the screenshot example for SuperPi I get a thumbnail view... I can't see what it has on there??? and I've tried IE.Chrome and FF... they are all doing the same thing... I think the rules (lack of them being written and easily found) is starting to hinder peoples ability and confidence to bench right now... if they don't have direction they will get flammed for making a mistake they were not aware they made... that's not fair to them. also on the Submit page... why can't we get to AS SSD and other benches... sorry that's another discussion. Vin Quote Link to comment Share on other sites More sharing options...
I.M.O.G. Posted November 15, 2012 Author Share Posted November 15, 2012 I don't know that they are that hard to find, just under the benchmarks tab in the header: http://hwbot.org/benchmarks/processor http://hwbot.org/benchmarks/videocard There are however a lot of items uncovered and awaiting clarification. Many of those are covered in the work-in-progress which is at least a sideways step in the right direction (because it is unfinished, and a timeline is uncertain on it becoming the reference for what is enforceable): http://hwbot.org/news/7914_rules_update_2012_hwbot_general_rules_and_guidelines Quote Link to comment Share on other sites More sharing options...
Vinster Posted November 15, 2012 Share Posted November 15, 2012 I don't know that they are that hard to find, just under the benchmarks tab in the header:http://hwbot.org/benchmarks/processor http://hwbot.org/benchmarks/videocard Thank you for that, I was looking under the submission section as that's where I remember it from the previous revisions Vin Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.