Overclock.net banner

Do you think the Intel Overclocking Competition is a good idea?

  • YES - I will be participating

    Votes: 15 71.4%
  • YES - But I will not be participating

    Votes: 5 23.8%
  • NO - I dont like the idea (please specify why in this thread)

    Votes: 1 4.8%

Official OCN Intel Overclocking Competition - Discussion Thread (We need your thoughts!)

5K views 76 replies 15 participants last post by  GENXLR 
#1 ·
Good morning/Evening all!

I have been in discussion with OCNs Events Manager and have decided to run an Intel Overclocking Competition!

What will the competition involve?

My Idea is to run an overclocking competition on a month by month basis for 6 months and entails (cpu requirements are just for example):

Month 1 - 2core/2thread Intel CPUs

Month 2 - 2core/4thread Intel CPUs

Month 3 - 4core/4thread Intel CPUs

Month 4 - 4core/8thread Intel CPUs

Month 5 - 6core/12thread Intel CPUs

Month 6 - 8core/16thread Intel CPUs

The competition will consist of users being asked to submit their top score in the chosen benchmarking software. Each user will have the option to submit two entries in their competing month. One entry to get into the competition, and then a second to try and beat their first score :)

Only users with the required CPU will be eligible to enter in the specified month. So for example if you have a 4690K, you could only enter in the Month 3 if the above schedule was final.

Each monthly competiton will be started on the first day of the month and will end on the day before the last of the month, giving me the fnial day to tally up the results and announce the winners. I will create an overclocking template, which will show what you need your competition submission post to look like to be eligible to enter.

I will create a spreadsheet of all the entries into the club, so that users can actively see who is the top of the chart throughout the competition.

Once each month has passed, the winners will be announced and a new thread will be created for the next month. I plan to start the competition from the 1st March 2015.

Prizes are currently being discussed :)

Rules Discussion (Your input is needed here!)

I have created a poll to gauge interest in the overclocking competition, please chose your answer carefully.

I have several ideas which I am pondering, and would like you to say which you think is the fairest option.

1. Should the competition be decided between 1st,2nd, 3rd place for ALL entries? Should the competition be split up by cooling and have two spreadsheets, one for LN2/Extreme overclocking and another for AIR/WATER overclocking entries?

2. Should different CPU architecture be handicapped, so as to keep the results fair? For example, in the month 3, the following CPUs can enter: 2500k, 3570k, 4670k, 4690k. Should there be a handicap against the 4690k so that the other CPUs have a points increase on their final score, with the 2500k recieving the highest handicap as it is the oldest architecture in the group? Or should the competition just be treated as all competition CPUs are equal in each month?

These are my pondering questions. Please post in this thread with your thoughts on the competition, and also your thoughts on the two questions. I look forward to hearing from all Intel CPU owners and hopefully this competition is a success for everyone involved. I will monitor this thread for the next two weeks before I decided to go ahead with the competition or not.

Thank you for your input!
 
See less See more
#2 ·
Just chiming in, should be limited to higher end air cooling / low end AIO cooling max temps 85c not really fair that someone else can win because they can spend more money.
Make sure that limiting factors are accounted for, example Sandy's high clocks vs other chips.
 
#3 ·
Quote:
Originally Posted by Xoriam View Post

Just chiming in, should be limited to higher end air cooling / low end AIO cooling max temps 85c not really fair that someone else can win because they can spend more money.
Make sure that limiting factors are accounted for, example Sandy's high clocks vs other chips.
Thanks for your input. So would you be in favour of a handicap system to accomodate for sandy vs ivy vs haswell cpus? Bearing in mind that a high clocked sandy will perform on par with an ivy bridge chip clocked a couple hundred Mhz lower.
 
#4 ·
Quote:
Originally Posted by $ilent View Post

Thanks for your input. So would you be in favour of a handicap system to accomodate for sandy vs ivy vs haswell cpus? Bearing in mind that a high clocked sandy will perform on par with an ivy bridge chip clocked a couple hundred Mhz lower.
Well it depends what your scoring method will be.
If it's just voltage speed and temps, definatly otherwise sandy will win.
there needs to be some sort of balancing method, otherwise people with more money and newer chips are just going to win by default.

And as mentioned previously, there should definatly be a sepration between ln2/extreme watercooling, and high end air/low end AIO water
 
#5 ·
I wanted to use software that is a quick and simple test of performance. Something along the lines of cine bench. So there won't be any prize for lowest temp or voyage etc just score based on cpu performance.
 
#6 ·
A few suggestions from my side.

1) I don't mind two separate categories for ambient & sub zero, but if people oppose that, then we can go the route of low clock challenge like they do @ the Bot.
4.5Ghz or 5Ghz.?

2) Instead of splitting in by number of cores & months, how about splitting the competition by Architecture & then a bit more sub divide for # of cores.
This would work better than doing 4 core competition then limiting Sandy to 5G, Ivy to 4.8 & Haswell to 4.5G.

6 months is too long, we'll have Broadwell-K by then.
Also about time we get some actual Overclocking done on OCN. Last competition was the 3D Fanboy competition last March.
 
#8 ·
Quote:
Originally Posted by Xoriam View Post

Yeah if you're using cinebench which I do not mind, some chips will definatly need a handicap, 4790k will just win by default obviously.
Actually if we do CineBench, XTU & SuperPi then we only need to split by Architecture.

Because for CineBench & XTU you just divide by the # of cores & get a per core score. They do it at HWBot right now.
And Pi is single threaded, so that doesn't matter.
 
#9 ·
Quote:
Originally Posted by rt123 View Post

Actually if we do CineBench, XTU & SuperPi then we only need to split by Architecture.

Because for CineBench & XTU you just divide by the # of cores & get a per core score. They do it at HWBot right now.
And Pi is single threaded, so that doesn't matter.
Yeah thats what I'm refering too if we use cinebench there needs to be a handicap on the new architectures because their single core scores will way to higher to compete with.
Personally myself and others as well have been holding off on the new setups simply because of the price of DDR4 and them ramping up CPU prices because they are the only ones available.
Not really fair in this competetion
 
#13 ·
In addition to the "by-core count" breakdown, probably relevant to have sub-zero and ambient/water cooled categories. An open class comp will be dominated by LN2 in any core count. R15 abnd R11 have an issue with the high precision timer in windows 8, 8.1 and 10 (server 2012 versions too) where the result can be greatly skewed by the user. XTU is fine, I don't know of any other cpu benchmark (aside from FM physics scores) that are not susceptible to this tweak/hack/cheat. Otherwise participation is limited to W7 back to XP.
blinksmiley.gif


Discussion thread at HWBOT

http://forum.hwbot.org/showthread.php?t=123642
 
#14 ·
Quote:
Originally Posted by Jpmboy View Post

In addition to the "by-core count" breakdown, probably relevant to have sub-zero and ambient/water cooled categories. An open class comp will be dominated by LN2 in any core count. R15 abnd R11 have an issue with the high precision timer in windows 8, 8.1 and 10 (server 2012 versions too) where the result can be greatly skewed by the user. XTU is fine, I don't know of any other cpu benchmark (aside from FM physics scores) that are not susceptible to this tweak/hack/cheat. Otherwise participation is limited to W7 back to XP.
blinksmiley.gif


Discussion thread at HWBOT

http://forum.hwbot.org/showthread.php?t=123642
yeah thats a good points cinebench scores can be completely hacked.
Something else that would work would be like firestrike physics score with HT disabled, in architecture categories.
because you need to validate, can't really mess with that.
 
#15 ·
Quote:
Originally Posted by Xoriam View Post

yeah thats a good points cinebench scores can be completely hacked.
Something else that would work would be like firestrike physics score with HT disabled, in architecture categories.
because you need to validate, can't really mess with that.
Unfortunately, FM only validates with a complete run (all tests) - selecting the physics test only in advanced mode will not verify that the RTC/HPET has not been tampered with (timer hack).
 
#16 ·
Quote:
Originally Posted by Jpmboy View Post

Unfortunately, FM only validates with a complete run (all tests) - selecting the physics test only in advanced mode will not verify that the RTC/HPET has not been tampered with (timer hack).
What exactly would be the issue would that, sorry if I'm missing it.
It would only take slightly longer to run the test.
 
#17 ·
Intrested. I'm running twin Quad cores in an Old D5400XS on Venomous X's I'd love to compete with, and my X5650, and X5670 rig. Also have my suprise coming up which may be unfiar to bring into this competition but I feel it's fair. Yeah limit to Air cooling and low end AIO cooling as people with money could just whip out phase change and LN2 and just shoot for the skies like i did at HWBOT with Skulltrail on phasechange for my 6.5ghz suicide runs.
 
#18 ·
Im definately interested in something like this, doing some other contests right now on another site. My .02 would be its kinda a bummer that someone with the money for only one chip can only compete once in the whole thing, maybe always have something going for like the i5's and i7's. Also, i think the cooling limit is pretty simple, no sub ambient cooling. While im all for it, it belongs in a class of its own, as no matter how good my oc is on water, a poor ln2 oc will still be better haha.
 
#19 ·
Quote:
Originally Posted by Xoriam View Post

What exactly would be the issue would that, sorry if I'm missing it.
It would only take slightly longer to run the test.
nothing is wrong with that,
 
#21 ·
Quote:
Originally Posted by PontiacGTX View Post

Which benchmarking software?
Then an i7 can enter two/three times?disabling HT and cores?
I was looking at cinebench, xtu possibly superpi?

I'm own to suggestions from people as to which software we use.

Also you can only enter with your cpu running as it is sold, i.e 4 cores, 8 threads cannot enter a 4 core, 4 thread competition.

So possibly looking at doing a sandy month, ivy month, haswell month etc. I would need to split the prizes up. I will speak to the events manager regarding what prizes we can look at. To be fair though I think the competition is not based solely on winning prizes, it should be about having fun too!

Thanks for your input guys thus far.
 
#22 ·
Quote:
Originally Posted by $ilent View Post

I was looking at cinebench, xtu possibly superpi?

I'm own to suggestions from people as to which software we use.

Also you can only enter with your cpu running as it is sold, i.e 4 cores, 8 threads cannot enter a 4 core, 4 thread competition.

So possibly looking at doing a sandy month, ivy month, haswell month etc. I would need to split the prizes up. I will speak to the events manager regarding what prizes we can look at. To be fair though I think the competition is not based solely on winning prizes, it should be about having fun too!

Thanks for your input guys thus far.
yes those are good options

what do you think about
3DPM-MT / 3DPM-ST
7Zip for i7s
WPrime
FryBench
 
#23 ·
I would say wprime or superpi would be a good test as well. They are quick but you can put a lot of time into getting a tweak to work correctly and they can prove difficult but are great ways to test OC's. In terms of cpu's and handicaps and such I'm thinking a bit on this and will respond on it later.
biggrin.gif
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top