hwbot "classic" rev8

[richba5tard]

What fundamental change to gl/wr do you suggest?

Less benchmark which give points is ok by me, but a separate discussion.

[Rauf]

Something easy. Like % of top score. Easy to understand and should be easy on the servers as only GFP/WRs require a full recalc.

[Noxinite]

Fml, people.

I just went through all the thread to start thinking about some of the issues, and I've managed to write a list of >30 different issues that people have mentioned. Some of them are related and some might be duplicated because I was getting confused. But this makes it hard to focus on any one thing.

Will continue thinking through things and try to come up with some ideas/proposals at the weekend.

[The_Silver]

a concern with a seasonal ranking is that people will delete and re put the score each season ... or keep backup

 

Edited November 21, 2018 by The_Silver

[websmile]

All this discussed here was meant to be adressed at rev7, you can also scale point distribution there by changing multiplier for benchmarks if you see the balance is not good, the season ranking problem was solved by comp points going down after 12 months and being replaced by new ones because the re-submission Silver mentions was adressed there and said to be non-solvable problem if hwresults are seasonal base and so on. Fact is that the database point distribution doesn´t work technically, that rankings don´t work and that the system, to give points in relation to top spot scared away lots of people on older hardware 3ds who dont bench new gen cpus on ln2. Again we sit here, after it took 18 months to implement rev7 from start to first release, and discuss about a new revision without changing the mistakes we have and the toxic stuff that kills oc. I hope to read some new ideas next days just to get some hope, but you will not hear nay proposals by me anymore, it is simply not amusing to read same stuff I read 5 years ago and that plans are made which include the same mistakes made as on former revs because again the wheel will be re-invented... I was always optimistic about hwbot, but things change

[ShrimpBrime]

2 hours ago, richba5tard said:

What fundamental change to gl/wr do you suggest?

 

straight forward answer.

No fundamental change needed. 

Rev 7 is fine. leave it alone and repair the odds and ends. 

Advertise overclocking to young crowds. I'm (we) not getting any younger. Need a larger crowd. It's harder to please the few than the many. 

Who's idea was it for this big revision change any ways? and WHY? (don't wanna hear about the way HWBot works but want to hear about the people running it and WHY the feeling or awareness of some feeling to make such a large drastic change... In other words I don't want the hardware answer)

"Sips coffee and leans back"

[ShrimpBrime]

OHH and also forgot!!!

W9 (the current fastest team on planet earth) is holding a holiday event. OC your processor and hit max % OC!!!

No points, no numbers, no hardware or cooling restrictions. no points, no prizes, nothing but good old fashioned fun. 

Register at Warp9-Systems and have a wonderful Holiday season!  http://warp9-systems.proboards.com/thread/778/w9-holiday-oc-competition?page=1&scrollTo=24121

My apologies for the off topic post. Thought it would be nifty to mention that here....

[mickulty]

Rev.6 was taking forever to calculate

Rev.7 is a buggy mess

If Rev.8 will be easier to maintain, I'm all for it.

I do wonder though, is it really a good idea in terms of code base and server load to bring in a seasonal ranking at the same time as trying to simplify things?  I'm neutral on the concept but maybe one thing at a time.  Mostly I just want hwbot to work and be reliable, we can worry about the meta later...

[unityofsaints]

6 hours ago, richba5tard said:

I obviously can't get rev 7 fully working due, do I?  Despite hundreds of automated tests, a test environment with a full production database copy, we are not able to reproduce the issues with rev7. Rankings just get fucked up on production and nobody seems to know why or how. If I can't reproduce it, and i've got 685.000 lines of 14 year old code to maintain, well guess what, I can't get it bug free. Even worse, all my spare time is spent on finding issues with the points and dozens of different rankings, I can spent time on improving other stuff. One major reason for rev7 was to free up calculation power, but in the end I had to write a tool which checks for broken rankings and repairs them, causing high load...

I need it easier to maintain, otherwise HWBOT will never improve quality wise.

Easy solution - get more eyes on the code. One man shows are never healthy, no matter how good that one man is

[ozzie]

1 question,

why is rev 8 even  needed, ??? coz you couldnt fix rev 6-7 properly for how long ? we went to 7 because of the same  , now lets go to 8, that'll fix it wont it lol, so we'll try with another rev and smash every ones team,  points , efforts, enthusiasm etc etc with it ???

and be bloodywell honest about it, fred/richyrich

 

[WhiteWulfe]

On 11/20/2018 at 4:04 PM, Mr.Scott said:

Revision reads like full on pay to play.

Pros and deep pockets need only apply.

Grinders are dead. Teams mean nothing.

Only thing left worth anything is Hardware Masters, and even that will have diminished value.

Pretty much my thoughts on it.  I've skimmed through a lot of things, but wow...  Having rankings mostly be focused on what's been done the past year?  What happens to growth?  What happens to those who actually work on improving, and most importantly, those who tend to focus on older gear because they can't afford to risk (let alone BUY!) the higher end newer gear?

 

Dropping 211 ranks in my country (20th to 231st, ouch) and going from the top 8% (118th out of 1505 is 7.84%) to the bottom third in my league (149th out of 219) just because I've been focused on real life instead of benching could very well coax people who had been active previously to just not return.  Sure, some might argue that trying to go after someone's score from ten years ago is a difficult thing, but I say eh, work harder on it - it's what I've done, and I've repeatedly used older scores from more knowledgeable overclockers as a target to push towards...  And eventually beat (yes, I'm still elated I beat ONE of Christian Ney's scores - it took over two years to surpass, of course I'm bloody well happy about it!!!  Still working on beating one of ftw's scores...)

 

And, like others have mentioned, the new method puts those who tend to rely on hardware points for their standings (due to personal preferences, financial viability of older platforms, or both) are automatically at a disadvantage when someone can get upwards of three times the points through Global/Benchmark points....  AND really does reinforce the whole pay to play aspect of it all, as the only real way to even remotely compete points-wise (or have a chance in going up the rankings) would be to go all-in on liquid nitrogen (alongside all of that expensive brand new gear!), which for a lot of people, is just way too expensive.  No, I'm not calling LN2 pay to win, I'm referring to the fact that the newest hardware is bloody expensive, but if you happen to have the bank to get it, you can rather easily jump up the rankings for quite some time with access to such.

 

I'm all for having the workload on the backend be easier, but at the same time, there's a reason why the current system wound up being the chosen compromise in the first place - those who focus on older hardware (and therefore having HUGE walls to climb just to even get remotely decent scores, especially with previously rather popular gear) were able to at least see some sort of reasonable reward for their work (and not just Hardware Masters rankings), those who focused on the competitive side of things were also able to make a good firm push into the higher rankings (I made it as high as 363rd worldwide last year thanks to competition points!  Or was it 327th.  Either way, still something I never would have even been able to remotely contemplate before rev7, and it helped give me additional motivation to push harder to see just what could be done), and those who had the higher up placements in benches were also rewarded.

 

Is any system perfect?  Nope, but I will say this - while going up in the worldwide ranks in nice and all, I'd rather not be going up a few hundred rankings globally just because others weren't able to bench as often (current info shows me going from 1442 to 1201 worldwide JUST BECAUSE OTHERS WERE INACTIVE - that same info that would penalize me 211 spots in country rankings because rl's been a massive pain this past year).  Tying competition points to a timer is one thing (otherwise those who've been competing for years would have a HUGE advantage - this was a point brought up back when rev7 was being worked on!), but tying the entire league and overall rankings to such?  Blatantly stupid, and a giant middle finger to those older members who may have wound up having real life take priority, and benching somewhat of a back seat for a while.  Hell, it's two giant middle fingers to the community in general...

[ozzie]

18 hours ago, avalanche said:

Is that how Rev 8 will look 2019?

Wow. I know every revision members efforts/ points have taken a hit & every time this tiny Warp9 Team has kept on trying.

 Those stats right there suck ass. I'll pull the plug OC finsihed 

thats ok jr, let em get over to our team as a team ( if they got the "balls" too ) and see how they go, well smack their arses there like we have done in here for years !!!!!!!!

[ozzie]

9 minutes ago, WhiteWulfe said:

Pretty much my thoughts on it.  I've skimmed through a lot of things, but wow...  Having rankings mostly be focused on what's been done the past year?  What happens to growth?  What happens to those who actually work on improving, and most importantly, those who tend to focus on older gear because they can't afford to risk (let alone BUY!) the higher end newer gear?

 

Dropping 211 ranks in my country (20th to 231st, ouch) and going from the top 8% (118th out of 1505 is 7.84%) to the bottom third in my league (149th out of 219) just because I've been focused on real life instead of benching could very well coax people who had been active previously to just not return.  Sure, some might argue that trying to go after someone's score from ten years ago is a difficult thing, but I say eh, work harder on it - it's what I've done, and I've repeatedly used older scores from more knowledgeable overclockers as a target to push towards...  And eventually beat (yes, I'm still elated I beat ONE of Christian Ney's scores - it took over two years to surpass, of course I'm bloody well happy about it!!!  Still working on beating one of ftw's scores...)

 

And, like others have mentioned, the new method puts those who tend to rely on hardware points for their standings (due to personal preferences, financial viability of older platforms, or both) are automatically at a disadvantage when someone can get upwards of three times the points through Global/Benchmark points....  AND really does reinforce the whole pay to play aspect of it all, as the only real way to even remotely compete points-wise (or have a chance in going up the rankings) would be to go all-in on liquid nitrogen (alongside all of that expensive brand new gear!), which for a lot of people, is just way too expensive.  No, I'm not calling LN2 pay to win, I'm referring to the fact that the newest hardware is bloody expensive, but if you happen to have the bank to get it, you can rather easily jump up the rankings for quite some time with access to such.

 

I'm all for having the workload on the backend be easier, but at the same time, there's a reason why the current system wound up being the chosen compromise in the first place - those who focus on older hardware (and therefore having HUGE walls to climb just to even get remotely decent scores, especially with previously rather popular gear) were able to at least see some sort of reasonable reward for their work (and not just Hardware Masters rankings), those who focused on the competitive side of things were also able to make a good firm push into the higher rankings (I made it as high as 363rd worldwide last year thanks to competition points!  Or was it 327th.  Either way, still something I never would have even been able to remotely contemplate before rev7, and it helped give me additional motivation to push harder to see just what could be done), and those who had the higher up placements in benches were also rewarded.

 

Is any system perfect?  Nope, but I will say this - while going up in the worldwide ranks in nice and all, I'd rather not be going up a few hundred rankings globally just because others weren't able to bench as often (current info shows me going from 1442 to 1201 worldwide JUST BECAUSE OTHERS WERE INACTIVE - that same info that would penalize me 211 spots in country rankings because rl's been a massive pain this past year).  Tying competition points to a timer is one thing (otherwise those who've been competing for years would have a HUGE advantage - this was a point brought up back when rev7 was being worked on!), but tying the entire league and overall rankings to such?  Blatantly stupid, and a giant middle finger to those older members who may have wound up having real life take priority, and benching somewhat of a back seat for a while.  Hell, it's two giant middle fingers to the community in general...

nicely written WW, you nailed it to a T ?  ?

[Guest]

On 10/29/2018 at 8:08 AM, Leeghoofd said:

I'm fully against any approach per year for members/teams/..., this makes about half of the members/ teams loose massive points, which makes motivation to continue to bench non existing... 

Only thing the bot will reflect who is still actively benching (probably the latest and greatest hardware) not about what they have ever achieved. Makes climbing the ladder even easier than it is now...

Especially for the hardware masters, well just skip the category and loose it all..

 

 

^ this pretty much. pay2win@suckyOC.com

50% cliff all those try hard enthusiasts scores dive off. Per year scheme uAT has my most experienced members standings tumble in the Team ranks.

Big team, lots of members, modern hardware, decent extreme line up ... race to the finish line each year.  Yahooo

[Guest]

The thing I liked about joining the warp 9 team.

Just a bunch of every day guys around the world with a common interest. Overclocking all sorts of hardware.

It was the old hardware that interested me, I'd see them chatting about the socket A days & the mods they'd do ... my interest was always tweaked

from their massive experience. This is HWbot's main asset. Not a bunch of rich kiddies with big gpu's. Not loose focus on the wow factor dudes.

It's always been about doing the best with what you have. The warp9 team does that.

________

Won't look that way Rev 8. Magnificent Team efforts are demolished 2/3 of total null & void  #1 team crunched under the shifting goal posts

 

 

[ozzie]

its not about W9.        its about any team, just coz were no1 it dont mean other teams havent tried too, but theres 1 thing the other teams havent got like our small miniscule team thats the best in the world, theyve all tried to beat us, theyve beat us in the team cup yes, thats what team cup is for to beat other teams, other teams have much more LN2 , DICE  benchers than we do, even when we were classic platforms we were never under the no5 team in the world in my time since i was there. always competiive in every way

not to worry, the worst thing for anyone is to fall on the sword they initiate in the first place, the pain to come is excruciating for them ?

 

[Fasttrack]

My latest 24 hours regarding HWBOT.

1) Over 1000 e-mail messages, FB messages, phone calls and SMS messages.

Common parameter in all - Frustration, anxiety, elevated stress levels.

2) Common questions - sell and get out ? --- Wait and see ? ---- Sell old HW and an arm and a leg and buy modern expensive HW ?

Turbulence, confusion.

Agreed, most people do not have the necessary self-discipline and life experience to see a hurricane hitting their head, while an hour ago all was calm.

But, for a playground ?

This is not family, business or whatever else, where a sudden event can bring life upside down.

This is supposed to be a hobby.

A hobby can be done at home alone, or thru broader channels ( Team Work ).

UTMOST CONSIDERATION MUST BE EXHIBITED BEFORE REACHING FINAL DECISIONS.

Shame that after so many revisions, a stable platform can't be found.

My best wishes for a viable solution. Otherwise I predict a real avalanche wiping whatever is left

 

[Guest]

Change & upheaval agaaaaaaaaain

https://www.google.com.au/search?rlz=1C1ASUM_enAU820AU820&ei=kmr2W_WmHZO8rQGP_qr4CQ&q=change+%26+upheaval+meaning&oq=Change+%26+upheaval+mean&gs_l=psy-ab.1.0.33i21j33i160l2.3622.5619..7214...0.0..0.326.1337.2-4j1......0....1..gws-wiz.......0i71j0i22i30.M7eiskjAmk0

Well there's a ripple effect each year happening revision time = Unknown the result

[richba5tard]

The TPP is a drain on resources and maintenance, but if the concept is loved I'd consider maintaining it. The big drawback(imho) is that it is quite static but so be it.

[Fasttrack]

1 minute ago, ozzie said:

answer me you bastard, that is your name isnt it, without the rich in it, ???

No, no and no Shane.

Please delete your post. That is way beyond any acceptable level.

 

[ozzie]

well which one is it ??? richbastard or frederick

either 1 i dont care, answer me

[richba5tard]

You have been warned ozzie. I'm letting you cool down until Sunday.

[GeorgeStorm]

Removing globals completely from career makes them even more elite focused, as they normally require the biggest expenditure, and if they only count for a year then are worthless fewer are going to be willing to compete, meaning even more bias towards those who might get stuff for free/some kind of support.

 

If globals are considered too big, then surely just adjust the max available, so it's less of an issue? If hwpoints had a limit of 50 (and now various depending on bench), why not the same for globals, make the max 100, or even 50, a global 1st place could still get double a hw 1st place which is worth it no?

[Guest]

Some good ideas there George Storm  I don't know what you HWbot guys will decide on. I hope the weighing of enthusiasts scores are better.

uAT makes some elite benchers like 4500 points ... but the enthusiasts are way down bulk of the 500 & lower.

_________

Really far from a level playing field. If we all overclocked same hardware, same cooling ... the results would be very close  

[Guest]

56 minutes ago, richba5tard said:

The TPP is a drain on resources and maintenance, but if the concept is loved I'd consider maintaining it. The big drawback(imho) is that it is quite static but so be it.

It's the little man all the time ... plugging away @ his scores on his meager budget. The backbone of HWbot members is guys who cannot afford best hardware.

They still submit scores with their best parts all the time ... score that are below the horizon where LN2 soars.

Keep the team spirit alive RB ... I'll fight for my team