The Ethics of Overclocking

[Gautam]

^^This post is full of more win than the rest of XS combined.

[poparamiro]

I think you're missing a vital point, Mike. What you're describing is why maybe 1% of the entire overclocking community is into overclocking and benchmarking, the other 99% is just there to have fun with hardware.

I agree, look at Kingpin, Sham, Hipro5 or the above mentioned Team Japan (and of course a lot others). They have great time overclocking, moding and pushing the limits. Not everything is about marketing crap, wr's with "all costs" or points.

Do overclocking for "the feeling" not for winning money and the most important thing, respect the work of other overclockers.

[cowie]

no dude the jokes on you. I'm laughing my ass off at that last post I made because it is the epitome of competitive overclocking. we need to stop taking ourselves so seriously and rethink what were doing here.

we're taking computer crap and trying to make it go faster than someone else's computer crap then we're pouring frickin liquid nitrogen on it to go even faster and if that wasnt enough, were hacking the software to get more points. and if that wasnt the icing on the cake we have threads complaining about the dudes getting more points cause they did the hacks and the dudes that dont know that hacks are pissed off cause their scores were lower and didnt get as many points as the other guy.
oh darn. cmon man you're in jersey of all places. Take it from a fellow newyorker that drives. the guy in front of me on the FDR with Jersey plates cuts me off on the houston exit and now hes in front of me. so I cut him off at ave D and I'm pouring Ln2 before he does because he still has to go through rush hour traffic crosstown and go through the holland tunnel. Meanwhile my 4th ram slot closest to the cpu is frozen stiff and I need to break out the hairdryer. the jersey dude is probably at a standstill in his car at hollland tunnel by now but Im at a standstill with my motherboard. whos gonna get the points first? whos gonna get more points? the jersey guy that cut me off or me for getting home 1st so I can overclock before he does? Maybe hes using tweaks in New Jersey of all places and the guy in NY he cut off is just happy to get home safely and have a full dewar and new batteries in the thermal meter. who gives a crap?!!! they both drive like jerks.

oh i see what you mean now you should have took the g.w....overclocking is like fishing
i think dynamite is ok bait

[Dumo]

Just overclock it and if you use any of those magic tweaks/mods then just share with others....simple as that. If you can do it then other peep can. The only deal breaker is your cpu.

Buy, bin and bench or get samples, bin and bench!!!

[Eeky NoX]

I think the "system" involved is so out of control right now that it is really unfixable at this stage. It would take an entire reinvention of the system to even begin to rein this in. I know guys on both sides of this, the newbies with no idea, and members of these 'inner circles', and there is no end in sight to this stuff.

Ya see that Mike ? Exactly what I stated in my last post

No real pessimism there but just tryin to think in a realistic obvious way. Before dancing we all gat to learn how to do so, or re-learn.
I don't care of high end hw nor points. What's fun in my way of OCing is just to catch up those with expensive systems with mid range componements maxed out ! Everytime a great achievement for me to see a score that matchs with another at lower frequs or better bang for a buck stuff. (I don't use tweaks so often cus I'm lazy ^^ rather work on real efficiency)

Optimising ---> Overclocking !

[Movieman]

more precise rules is not blasphemous and
this is needed because we need to accept the fact that everyone views things differently and righteousness is a different ideal in different people's minds.

I agree that hard and fast rules would be the best way to go.
Get together a group of guys that bench that particular benchmark( so they know all the potential tweaks or cheats) and let them come up with a good set of rules for that particular bench.
One group for 3D, one group for SP,etc..
I also think we have to keep pushing at people the honor part.
Drive it into their heads if necessary so that they understand what it means and it's value not just here but in all of life.

[massman]

To mike:

Regarding the high stakes.

My point is that 99% of the overclockers out there couldn't care less about the high stakes in live overclocking competitions. In fact, 99% of the overclockers don't have anything to do with extreme overclocking and just have a bit of fun with hardware. I don't see the necessity to make overclocking for them overly complicated by adding extra sets and subsets of rules to make sure they do everything in the right way.

Regarding consensus about ethics.

To be bold: yes, most people have a very good idea of what's 'okay' to do and what's not. You can do an experiment yourself and ask people how they feel about past (for me) unethical incidents and how they respond. In fact, the only responses I get are 'maybe the rules were not broken, but it was certainly a violation of the ethical code'.

The problem of using MIPMAP and DPP is not that they give more points. The problem is that they are widely accepted as not okay before it was used as enhancement in the competition. The problem of ethics lies in the fact that someone used a trick that anywhere would be regarded as invalid. I know some consider this as thinking outside the box and clever use of the rules, but I also know many others see it as abuse of the rules and damaging the integrity and respect towards overclocking.

By the way. I don't have anything against optimising sets of rules if the aim is to make it more clear for people. I just have a problem with re-creating a rulebook so people can take the time to find the loopholes and abuse those loopholes to win. To make this more practical: yes, I think 'mipmap disallowed' should be in the rules, but I expect every overclocker to tell that to the judges before the competition and not use it to win and deal with the rules afterwards. That would should of a total lack of respect to your fellow overclockers.

And yes, it's allowed to have no respect for your fellow overclockers. But don't expect people to give you any either.

The example of how things could go differently has already been given: at some event, one person had a bios that allowed more Vcore. It was passed around so that everyone could benefit from it. It was even said out loud: 'who needs this bios'. And not by the organisation, but by the overclockers.

Regarding leagues run by competitions. So me.

As you already kind of said yourself ('if were to'), this sketch of a situation is hypothetical. I agree that there's great potential for ethical mishaps, but let us not judge a situation before it has occured. Sure, you can accuse me of changing the league in a way that only benefits myself, but then again ... did I? Has my personal participation lead to unethical behavior? If yes, please tell me. Then I can resign from my position of HWBOT employee ;-)

Also, you need to look at the alternative. This alternative is a league or competition that is run by people who are not actively involved in the overclocking community. Run by people who have to rely solely on external information and experience to troubleshoot the competition. Would that be 'more ethical'? Or would this platform just give the same great potential of ethical mishaps? Why would someone who's not involved in the overclocking community not change the points algorithm in a way that would benefit specific users and teams?

In general about software tweaking.

The key element of overclocking is, obviously, tuning hardware systems to get more performance. I see where people want to innovate by finding new tweaks and software enhancements, but I honestly don't see why this should be made a part of competitive overclocking. I mentioned this in another thread before: when it becomes okay to optimise all software endlessly, where does it end? I mean, if there's no problem with mipmap, then why would there be a problem with wireframe? Why not just remove parts of the DX10 library so the VGA can render the scene even faster? I cannot see myself in an overclocking competition having to rely on a software engineer to hack the benchmark and remove as much as possible just to get a higher score. In fact, this defeats the whole purpose of overclocking and the hobby should then be called 'software hacking'.

There are small software enhancements that seperate the good from the great (eg: largesystemcache, running AM3 on 2 cores, etc). None of those involve direct benchmark optimisations and all of those are readily available for anyone to access.

[Oliver]

In general about software tweaking.

The key element of overclocking is, obviously, tuning hardware systems to get more performance. I see where people want to innovate by finding new tweaks and software enhancements, but I honestly don't see why this should be made a part of competitive overclocking. I mentioned this in another thread before: when it becomes okay to optimise all software endlessly, where does it end? I mean, if there's no problem with mipmap, then why would there be a problem with wireframe? Why not just remove parts of the DX10 library so the VGA can render the scene even faster? I cannot see myself in an overclocking competition having to rely on a software engineer to hack the benchmark and remove as much as possible just to get a higher score. In fact, this defeats the whole purpose of overclocking and the hobby should then be called 'software hacking'.

There are small software enhancements that seperate the good from the great (eg: largesystemcache, running AM3 on 2 cores, etc). None of those involve direct benchmark optimisations and all of those are readily available for anyone to access.

Wise words

[chew*]

@ trans am.........

not everything is as it seems.......just becasue your faster does not mean you are faster......

a simple example......2 results i did myself......one is faster.......but the other is faster

i guess it's all software right because no one sits there and tests various configurations for hours on end.

[uncle fester]

How has all this drama come up over a couple of people cheating ????
Cant we just put all the crap behind us and enjoy what we do ???
i wasnt around for OPB days but did that start this sort of spiraling cluster ???
Its simple people who cheat eventually get caught.
If you figure out something you think might not be legit just message massman or other people, they will tell you if its cool or not.

Where do we want this to take us ????

[trans am]

@ trans am.........

not everything is as it seems.......just becasue your faster does not mean you are faster......

a simple example......2 results i did myself......one is faster.......but the other is faster

i guess it's all software right because no one sits there and tests various configurations for hours on end.

huh? how is this even comparable to my point? yeah you're ram is number of the beast in the 1st one. of course its going to be better.. cmon man gimme some credit.
I need to quit posting sarcasm. people dont get me.

[chew*]

huh? how is this even comparable to my point? yeah you're ram is number of the beast in the 1st one. of course its going to be better.. cmon man gimme some credit.
I need to quit posting sarcasm. people dont get me.

Maybe you should look again and think before you post because last time i checked a higher time is not a faster time......

You complained about having higher clocked results.....but not as good results.......I just showed you one case how this can happen with "hardware" not software.

[trans am]

yes but the cas made up for the drop in cpu frequency. Is am3 more dependent on ram timings than several hundred mhz in clocks speed?
should I not use win xp for am3? gimme some tweak settings man. i need points! stop being so secretive and help me out.

[chew*]

yes but the cas made up for the drop in cpu frequency. Is am3 more dependent on ram timings than several hundred mhz in clocks speed?
should I not use win xp for am3? gimme some tweak settings man. i need points! stop being so secretive and help me out.

The drop in frequency or cas is not gonna make up 18 secs bud. The answer is clearly obvious.....the 4:16 divider has issues, the 4:20 does not.

Guess who found this out........me, guess how, my own testing, just like everything else i have found. Thats what makes an ocer. Knowing your hardware extremely well.

Not sure about other guys but when you see the post make sure you know there are countless hours/days put into each result. We don't pull them out of our asses.

Software is the easy thing to blame......he has a tweak i don't........sometimes it's best to look at your hardware and tuning before using the easy excuse

[trans am]

oh yes. of course duh... 420 is always the key.
thats why you got good points going on.

[chew*]

oh yes. of course duh... 420 is always the key.
thats why you got good points going on.

Yah i put that much effort into a whopping 1 point

[mike]

To Jan-Pieter:
First off let me just say that I have the highest respect of you and the HWbot. I enjoy very much following the community rankings and such. Richbastard ( owner of Hwbot) and you have done a fantastic with your database and have been a big part of the evolution of the overclocking community.
It is clear that you have given a lot of thought into the various issues I am trying to discuss so let me respond to your post that was specifically directed towards me. Before that let me stress the point that my intent with this thread was not to discuss my particular thoughts on specific ethics issues or to "call out" specific people or entities, but when asked, as long as I have something to say I will try to respond.

To mike:

Regarding the high stakes.

My point is that 99% of the overclockers out there couldn't care less about the high stakes in live overclocking competitions. In fact, 99% of the overclockers don't have anything to do with extreme overclocking and just have a bit of fun with hardware. I don't see the necessity to make overclocking for them overly complicated by adding extra sets and subsets of rules to make sure they do everything in the right way.

I will try to re-word what I had said in my previous post. I understand that the needs for HWbot rules in particular are different for the hobbyist than for group of people that look at overclocking from the competitive and professional end of things. My sense of this is that good rules should not hinder the hobbyist but only protect them and enable the professional overclocker to maximize their enjoyment of the sport on competing on Hwbot.

However, I think its may be the case that we have also different motives in this. Your main motives are to protect the hwbot from becoming inaccessible for the hobbyist. I can clearly see that it is a very difficult tight rope walk to keep both groups happy. If you feel that for the HWbot it is not feasible to create more defined rules to suit both, maybe it is best to only focus on the core audience of the hwbot members and not attempt to instill your ethics on the rest of the overclocking community.

My motives in this discussion are to create a platform where overclocking can possibly make the transition and evolution into a professional sport. From my point of view, for that to happen, there has to be a clear framework of rules, and a game that is designed for competition. Will this be possible? I can't answer this at this point. But for sure we should try. And the key principle for me is that the framework that has to be developed has to come from the community and not from any particular special interest group with motives which might not concur with the community as a whole.

Regarding consensus about ethics.

To be bold: yes, most people have a very good idea of what's 'okay' to do and what's not. You can do an experiment yourself and ask people how they feel about past (for me) unethical incidents and how they respond. In fact, the only responses I get are 'maybe the rules were not broken, but it was certainly a violation of the ethical code'.

Well that indeed bold question. I am not sure if I understand you correctly, but I think you were asking me to have other people judge you. I don't want to make this a discussion about you, but since you asked, I do think there are a few incidents from your own past that I would consider at least questionable.

But before getting into that, just to demonstrate that this is not personal I will talk about some of my own incidents which I would consider questionable, at least in a world with better rules.
For example in a past contest I intentionally changed my brand of a videocard to that of another brand. It wasn't against the rules in that contest, but it had clearly an impact of the outcome in that contest ( since the other brands' card was much better...) but no rules were broken. Why did I do that? I did that because the one manufacturer that I do not want to name was trying to sabotage the contest, and I did not want to support the kind of sabotage, because in the end, the overclocking community would end up being the loser.
Other times, I have intentially hidden my harddrive subsystem, to protect myself from being copied. Clearly many people would consider me unethical for that behavior but many others wouldn't. From my point of view, I wanted to protect myself from copycats...I have been aware of hardware and result sharing, even helped other benchers to achieve better results.
The list goes on, and is actually pretty long if I was to judge myself, but at the end I was always fine with my own decisions and in a world where I am the only who can regulate myself, that has always been enough.

Now to get back to the questions of your behavior. I have to consider you as two seperate entities. One is the competitive overclocker Massman, the other is the representative of HWbot.org
As the competitive bencher Massman I would never ever doubt any of your results, neither could I imagine in a million years that you would cheat. I have the highest respect of your own ethical mindset, even if sometimes I might not agree with you 100%

However as the representative of the HWbot.org, I can think of at least a few situtations where things are not as clear cut. For example when you were sabotaging an online contest by publicly advising others on how to break the rules of the contest. Not sure if you broke a law there, but to me this is the sort of thing that I would consider crossing the line.
Another example would be in a very recent thread where I almost felt like you were extorting a manufacturer publicly to pay the hwbot.org in order for their community to submit scores on the hwbot.org.

But to come back to the main point - it all comes back to my first post. What is ethical and what isn't? Who is to judge who and what ethics do we want to abide to. To me the answer is clear cut. There has to be a sanctioning body to rid of all these questions. Which also brings me back to my motive. I want to see a sanctioning body that is not for profit, and has the interests of the overclockers as their primary goal. As you are well aware , I had intitated the concept of a non profit sanctioning body a couple years ago.　

The problem of using MIPMAP and DPP is not that they give more points.... The problem of ethics lies in the fact that someone used a trick that anywhere would be regarded as invalid... .. abuse of the rules and damaging the integrity and respect towards overclocking....

...To make this more practical: yes, I think 'mipmap disallowed' should be in the rules, but I expect every overclocker to tell that to the judges before the competition and not use it to win and deal with the rules afterwards...

What is the difference between MIPMAP and LOD again?

To me a big problem for me also lies in the fact that the current form of competitive overclocking, that we are unable to police and enforce the rules that are developed. This is why I am also asking for the community to re-think the approach of competitive overclocking.

For the hobbiest member on the hwbot this might be a non issue, but for the competitve member it is a big problem.

How could you act currently if "hypothecially speaking" known cheaters are holding some of the top spot on the bot? What could we do about it. Nothing without the ability to properly enforce a clear set of rules.

Regarding leagues run by competitions. So me.

Incorrect. I was trying to make it as broad as possible, there are many more leagues than just the HWbot.org. There is also the FM, we had the F1 league, and there are other forms of leagues that might not be as tightly run as the HWbot.org. Hopefully even XS will soon have its own form of a league for some new approaches in overclocking that we want to explore. So the same issues apply to other entities outside of the bot

As you already kind of said yourself ('if were to'), this sketch of a situation is hypothetical. I agree that there's great potential for ethical mishaps, but let us not judge a situation before it has occured. Sure, you can accuse me of changing the league in a way that only benefits myself, but then again ... did I? Has my personal participation lead to unethical behavior? If yes, please tell me. Then I can resign from my position of HWBOT employee ;-)

But here is the thing, and it gets to the heart of what I am saying. You can say "I will resign if someone can say I have been unethical" but those are words with little force because saying that something is unethical is absolutely not the same as it being unethical. Its an opinion and one which, after all, you personally are likely to disagree with. So who gets to decide? What you are saying is basically the same as saying "I will resign if I, the sole judge and jury for my behavior, find myself guilty."

Aside from that, I am just using your words to structure an example of what I mean. In actuality, the very last thing I would like to see is for you the resign for any reason whatsoever. Your voice is extremely important in the overclocking community, you don't have any fear of voicing touchy subjects, and you are a leader in overclocking community precisly for the values which you represent.


Wow this was a long reply - sorry I just felt like it was important to be thourough. As a final point, I want to emphasize that I don't want this to become a personal back and forth. I really do bhelieve there are some larger issues to hash out here and I think there is a lot of value in hearing from as much of the community as possible on these matters.

[trans am]

Hey chew seriously thanks man. I dont come off as a jerk. I have a unique sense of humor. I never knew that about that ratio. and if u didnt tell me I wouldve never even bothered to check.

[chew*]

Hey chew seriously thanks man. I dont come off as a jerk. I have a unique sense of humor. I never knew that about that ratio. and if u didnt tell me I wouldve never even bothered to check.

Just trying to open your eyes a little

harware is quirky and can have huge impacts.....software gains are minimal at best if legit.

[Computurd]

its not the legit ones that people are concerned about.

[chew*]

its not the legit ones that people are concerned about.

Quite true but the above example without an explanation looks like a cheated result at least to some people.

[trans am]

Just trying to open your eyes a little

harware is quirky and can have huge impacts.....software gains are minimal at best if legit.

http://hwbot.org/community/submission/1032626_

Man I need to go back and redo this one at 4:20am at 4:20 ratio. Chew, do you know if this ram ratio bug includes all 1156 boards or is it just p55? or is it just that asus board in particular? I was on h55 ud2h.
I'm sorry I dont want to turn this thread into a guide to overclock clarksdales. I'm just really excited that chew turned me onto this one. This is without question the best thing ive gained from this thread.

[chew*]

Man I need to go back and redo this one at 4:20am at 4:20 ratio. Chew, do you know if this ram ratio bug includes all 1156 boards or is it just p55? or is it just that asus board in particular? I was on h55 ud2h.
I'm sorry I dont want to turn this thread into a guide to overclock clarksdales. I'm just really excited that chew turned me onto this one.

Not quite sure tbh my guess is it's p55. I think it applies to using 4:16 ( 2:8 ) only atm....whether board dependant or not....no clue atm.

Chatting with asus about it.

[trans am]

I'll check it out on the next refill.

actually screw it I'll check it out right now. I'm on the max III formula atm. l'll do that one first to see if its board specific.

[Computurd]

Chew is awesome, but TA you must resist derailing the thread...
must...resist...derailment....


Chew, you did make a very good point, and do a fine job at illustrating it.

------------------------------------------------------------------------------------------------------------------------------------------------

but there are other issues going on as well, not only the types of things weve been talking about here.
hardware sharing amongst teams is a humongous problem, and very very widespread on the hwbots. hell as a noob when i first started in on this hobby they spoke openly about it on a forum i will not mention.

also, has anyone went back and checked the HWBot database for any similar results on the top scores out there? I bet there are a lot more than meets the eye here.
how many #2 and #3 and #4 (etc.) spots are populated with the same sort of Andre-esque type file sharing?
wanna #5 spot on a bench? let me borrow your cherry chip for a week?

what about teams? why dont we start comparing the top results for particular benchmarks, amongst their very own team-members results? so one team has a few guys in the top spots on a benchmark, lets do some targeted study of these results, see how they compare?

You wanna really see the skeletons come out, lets start digging for them. i guarantee they are everywhere.

and the worst part of this is, i guarantee that the mods at the bot probably know a lot more about this than they are willing to reveal, or do something about.


It is overclockign though, and it has always been a bit of a 'shady' area of computing, the bad boys of the computer nerds! the nerds with pizzaz.

so these types of things are to be expected, but to allow it to go on at the levels that it is currently happening, is ridiculous.

If there were a professional organization keeping track at the bot, like a sanctioned body that they have in other sports, they would need a whole ing truckload of banhammers, i assure you of that.