knopflerbruce
-
Posts
4290 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Blogs
Posts posted by knopflerbruce
-
-
here we go, now we gonna have to pass a piss test to be able to submit!
Just wait till you have to pass the alcometer blow test to be allowed to enable ln2 mode:D
-
It's already here
The only downside I know of is that I get corrupted PI xml every now and then. Seemingly for no reason, running all sorts of benchmarks fully stable. Sounds strange that a sall file would be messed up due to instability, then.
-
Why all this talk about "voicing an opinion", There's nobody here trying to make some opinions disappear. However, it's also a place for discussion - so if you post something I disagree with, I use arguments to show that your arguments are incorrect.
-
Yep, same thing appløies to HWBot prime, Multicore Linpack and Multicore Pi. You can use whatever tab you like in CPUZ, and place it everywhere, including over the score field. For HWBot Prime the "save" box covers the score. It's still included in the file, so not sure how important it really is...
-
What hwbot did with people in Pro League?
Also you have lots of cups and points, but not the most of that from obscure hardware that nobody benched.
Maybe if the rules works this way... you will keep as the first...
But many others, never worked hard like you, and got lots of cups and points easy...
Also, you are talking ever about "top" ranking...
What about the guys that don"t have lots of hardware... and squeezed to the max their hardware to climb... be surpassed by guys with the most of obscure hardware... even benching without overclock with no one mhz more.
Check your profile. You still get both points and cups. I'm still 2nd in HW masters, and I've got a fair share of globals, too. The globals just don't count towards a ranking anymore. You've still got them, though.
Top rankings are different, but in the end the problem is the same - the quality varies alot in the low competition rankings, which must be taken into consideration. There is also a very limited number of such rankings, and the points drop quite fast. Not everyone can be number one... we've already got some fights with servers overclocked 50%+ in wprime. If wprime and cinebench, and other benchmarks that scale properly with many threads (24 and above), get WR points, that will help a bit if the globals for such rankings get a huge drop.
-
If there is no category, there are no results submitted to it, so there will not be enough results to be deserving of a category....so the circle goes around and around.
I think I suggested a few years ago that a category should have a certain number of submissions before it is given silverware or points, I still would not mind that happening. (I have a number of 2-point #1 scores, I am not having ideas that do not affect me as well) But... that is an extension of the changes that have just been made, I think the changes are fine for now. Maybe one day, they will need tweaked again, but in my opinion, HWB are on the right path
Thankfully, I don't think such a requirement will ever exist - for at least three reasons: we need a reason to buy and bench unbenched hardware. A few points and a set of cups... that's enough. A 2 point reward pr benchmark is already rather close to the "why bother spending time and money onm this"-limit. 2 points won't do much good alone, either, you need a couple of thousands before you start climbing the HW masters rankings
Also, points are already rewarded - you cannot remove points for hardware that people already paid for and spent time benching, for an already really tiny reward already. I'm pretty sure alot of the HW masters would quit instantly simply because it would take disrespectfulness to a whole new level A possible third reason: why would new users bench... say their laptops when they get no reward? We'd lose page hits and members = not good. -
It is not about hardware masters, xoc league or enthusiast.
It is thought that points, cups and medals, should not be distributed without reason ... and points, cups and medals should be conquered by merit.... and not because the overclocker bought something that no one more have.
So, to me... is a huge flaw to work this way.
"No reason" is subjective. As explained earlier, only the very top scores should be rewarded if you want to avoid points given to "no effort" subs. One can always argue that one shouldn't get points because one has better cooling (or better hardware) than someone else.
You're probably the only one making a point of cups... I never bragged about my 3000 ones. It's a curiosity that I have so many, nothing more than that. If I'm first in a ranking, I should get the gold cup - it's the definition of a top spot that you're unbeaten, regardless of competition. Others can compete if they wish, but maybe they're terrified by my results. I don't know. It's all speculation. There's also no ranking that's based on silverware, so you don't get much recognition for these things - only people that give you thumbs up for for your dedication.
-
Ticket ID: 1734
Priority: Low
ARES2 seems to be an overclocked version of HD7990, listed in HD7970x2. Why not merge the two?
The 7970x2 marking in price listings was there because 7990 was not officially released at the time. -
Nobody told is a fault to bench obscure hardware,...
The fault is to give 14 points and 7 golden cups for it.... or even 60 points and Gold Global cup for competing alone...
Further... still worst if the hardware was benched in stock frequencies.
Huge flaw in your reasoning: what are the 2 points good for? NOTHING, unless you're fighting in the hardware master's league, and in that league dedication means alot. It's a league for the insane, who can stand spending hundreds of hours running benchmarks in rankings, and getting scores worth 1-2 points pr half-hour benchmark. In XOL you won't get ANYWHERE with those 20x2=40 points that count towards the total:) Top 500, perhaps?
...plus, there's very little skill involved in most of the popular categories as long as you have a decent hardware and proper cooling. Boot'n'run as well, nothing has changed - apart from adding vaseline and some toilet paper for insulation - not exactly "overclocking skill" in a normal sense. Not to mention the huge amount (10+ in many cases) of points you get, for "nothing", compared to the small amount of 2 you get for running unbenched HW.
-
Thank you, sir! These chips are hard... never seen a chip eating LN2 like that before. With some luck, maybe I can push it another half multi... more than that seems like a no-go.
-
I think some of you underestimate the skill needed to be 1st out of 10-15. It's just as hard as being ~10th in a very popular hardware ranking, even though the reward is much less. FOr no-competition stuff it's different, but you don't need much competition before it gets alot harder to be number 1.
-
Must be great to add a retail CPU every once in a while
Thanks! -
I know
I've been refreshing some rankings from time to time to see the impact.However, is it possible to get some more details on the algorithm? You've covered the changes for the 'obscure' rankings, but you've also mentioned changes for "medium-popularity" rankings...? I cannot find the post, maybe it's my brain that's messing with me.
-
2 things:
1. Higher participants increase probability that one is competing against many non-noobs. Lower participants increase the probability that one is the only one playing who knows what he is doing. With the breadth of hardware and variety of rankings and rewards, I think this algo update is a step in the right direction... It isn't perfect, but it isn't too difficult to do, and its a more logical way to reward participants that scales both with high or low competition. With so much hardware and so many rankings, I can't think of a feasible solution that would be better at addressing actual skill.
Well, I can think of another solution, but it probably wouldn't be feasible... Instead of only awarding points based on number of participants, the algorithm could also account for the quality of participants you beat. For example, if you beat 9 other participants, that would be worth 10 points. If 2 of those participants were ranked highly within their league, you get additional bonuses because you beat other skilled Overclockers. Probably too complicated to implement, however it would address both the amount and quality of competition.
Ultimately though, I am the opposite of rasparthe on this issue. I don't believe awarding the same points for beating 2000 people or for running hardware no one else bothers to run is any good, especially when it rewards stock clock submissions that rank people higher than others legitimately trying to overclock.
2. How do inactive rankings lose points? Is there a decay element in the algorithm? When does point decay start? I have never seen this personally - I've only seen points decrease when my submissions are beaten.
PJ could probably give you the detalis.
http://hwbot.org/benchmark/wprime_-_32m/rankings?start=0&cores=1#start=0#interval=20
60 points for this one does not match the # of users competing. I bet it would be 100+ under normal circumstances, and who knows what will happen when the "obscure ranking fix" is done. Maybe it will drop further? Not sure how the fix works, if it counts activity or participants.
-
Ticket ID: 1729
Priority: Low
Stock:\r\n\r\nhttp://valid.canardpc.com/show_oc.php?id=2852637\r\n\r\nUnlocks to 5x and 6x.
-
-
Sad day for HWbot, too bad core misunderstanding of what drives HWBOT still alive and well. Can't wait until all points are disabled unless you are using latest CPU (4770k) and latest GPU (Titan). Very disappointing, I'm sure policies that treat certain segments of the site like second class members will continue to grow and expand the membership.
Sadly, points are mostly rewarded by the number of "noobs" you beat, rather than the actual skill required to beat your score. Not only that, inactive rankings get devalued simply because there are few recent subs, not because the submission skill level has decreased.
-
So tell me how to show NDA stuff on validation movie?
Seems like you have two main options - 1: break NDA, or 2: don't use NDA mods or whatever.
Option 3: edit the video so that the soldering point is greyed out.
No wonder no-one else got the mods if they're under NDA...
And I wonder if you take advantage of mods/tweaks others have shared? If yes, what's the big deal if you share a mod YOU figured out? -
I guess refclock/mem freq are also exceptions to the screenshot rule.
-
Don't forget that it's more than likely that we'll ban those who cheat from HWbot for a few months
-
Stupid request: can you try to run hyperpi 32m?
I'm having a bad time getting my A1 chips to run fully stable... -
What's the limit for a stock card? Maybe highest seen so far + 50MHz? Just throwing out ideas... Golden cards do exist.
-
I don't care for rumours and even less for the people spreading them. But I guess that's how things go in life.
But I am quite disappointed that people would think that I or HWBOT would invent this rule. Then again, it is nearly impossible to defend or disprove any rumours started on the internet. Especially if people prefer to listen to noise-makers.
Anyway, if anyone would ever ask me, I am in clear with my conscience on everything I do for overclocking and to support it.
I agree with this. Thinking that WE would want to ban these inventions that we have praised before is ridiculous. The reasoning behind that conclusion has to be borderline ret*rded. C'mon!
Disclaimer:
This is the opinion of me, as an individual, and does not necessarily represent the views of other crew members. I'm also sorry if anyone got butt-hurt by reading this post.
-
True story
HWBOT R5 Bugs/Features 2.0 - Report bugs or request new HWBOT features here
in HWBOT Rev.5
Posted
The last one could very well be the activity thing kicking in.