Massman

Seems like the best option. I'll add it to the planning. //edit: then again ... what about mainboard/memory/videocard?

In theory, yes. Practically, it's quite complicated, though, as it's not so easy to define the detection barrier in a simple definition. Not all CPUs require LN2 for 5GHz and what about videocards. We could add a simple pop-up whenever 'stock cooling' is selected, but then we'd be annoying those who do actually submit with stock cooling .

Most of those are user errors . It's nearly impossible to (automatically) identify what cooling was used based on a screenshot or even a system picture.

Time to tighten up security on competitions it seems . If I were you guys, I would wait for the limitations to come up before starting to assemble rigs.

You need a system picture too ... not just a picture of your dog :-P

Correct!

Yah, tried them on - R4E 1004 bios - R4E 1101 bios - Xpower2 1.2b1 bios This particular chip is rebooting all the way upto 57x, stops rebooting at 5843 or so. But no way in hell it'll load OS over 5250MHz. Other chips are behaving similarly (booting all over the place).

As I mentioned to some people, I'm re-trying the method on the 57x chips and it seems that C2 is behaving very different from C1. A specific chip is rebooting perfectly to 57x multiplier, but won't reach over 5250 in OS. The XPower-II is giving more 'accurate' results from my experience, but I need to cross-check more chips to be certain. To be updated! (PS: no chips have been send from Tones, so only crushed hopes! )

- MSI X79A-GD65 8D --- V1.5b2: download - MSI X79MA-SD40 --- V1.3: download - MSI X79MA-GD45 --- V1.3: download - MSI X79A-GD65 --- V1.1: download - MSI X79A-GD45 8D --- V10.2: download - MSI X79A-GD45 --- V1.4: download - MSI Big Bang XPower II --- V1.1: download - Foxconn Quantumian1 --- P02: download

You can buy 2 cards for 1/2way? Why keep talking about 4way? We are just repeating our own arguments over and over again. Everyone understands everyone else's point of view, but everyone still thinks their POV is "more correct". There is no point in repeating ourselves over and over again, I think. Change just leads to circular movement A -> B -> C -> A -> B without ever getting 100% of the people to like the idea. From what I can see, most people are just tired of "change" regardless of what the suggested change is.

How does the competition actually work? 1) email a cpuz result 2) 12 highest % oc travel to London 3) overclocking on SBE 4) whoever has the highest clocking chip wins an Ivy Bridge system ? Must be frustrating if you have a SBE that walls at 5G then :-/

Poland Owns! Nice coverage for the CC'11 winners in The Overclocker magazine! http://cde.cerosmedia.com/overclockers-TOC-theoverclocker/1D4f312793215cd012.cde

I see what's going on, it's a small bug. The engine ranks your submissions by #pts in order to find the 20 most rewarding (hw pts) ones to calculate your points in the user leagues. You have a bunch of 0.1pts scores (http://hwbot.org/user/mario1/#Points), which are ranked by #pts and then by date (I think). The problem is that for the enthusiast league, only the submissions with pictures count for points. The engine, however, doesn't check if there are 0.1pts scores with pictures. It's easily fixable, I think. Added for fixing: http://bugs.hwbot.org/browse/HWBOT-708

There are two problems with the sentence "... change to better and turn more equal the chances for all": 1) "More equal" is relative. Perhaps you think "focus on 1xGPU LN2" makes the game more equal, perhaps others think "spread focus for all categories" is more equal. What is "more equal" is a matter of opinion. 2) "change to better" is also relative. You are of the opinion that "more equal" is 100% change for the better, whereas others might think that it's better to have a system that is a bit "less equal", but more competitive. For example, some people are of the opinion that a flexible system where you can get points with 2xGPU gives them the option to run 2xGPU on LN2 and move up the ranks that way.

I understand the point of view and I think the complaint is valid in a way. But, as others already said, it's all a matter of opinion: some will say multi-gpu benching should be less rewarded, others will say they should be rewarded more. It's just not possible to figure out a system where everyone is 100% happy. Anyway, the essence of a competition is that the organiser makes up a set of rules and the participants tries to figure out how to play within that set of rules. The participant should adapt to the game, the game shouldn't adapt to the participants. Fwiw, a similar case could be made with the CPU-limited benchmarks. Ie, with a very good CPU, you can get a lot of easy points in not only all cpu benchmarks but also 3DMark01, 03, 05, 06 and Aquamark. Well, it's not plug and play, but with a small amount of effort you can get nice ranks ... you know what I mean.

Liking the ideas here .

Yeah, that's how SB and SBE usually works. Optimal temp range may vary from 2c2t to 6c12t, though. Viss: did you test C2 with older BIOS versions? I'm getting the feeling the one I'm using right now isn't cooperating very well with C2.

M.Beier too. He trained. He's now all the way upto four beers! Four beers! FOUR!

Sorry, there's a small bug. It's scheduled to be fixed asap.

KK, updated list. Looks like a lot of folks will be attending CeBIT

FYI, for the 3DMark Vantage ranking - 1xGPU => 33.6pts needs rank #36 - 2xGPU => 33.7pts needs rank #19 - 3xGPU => 33.6pts need rank #3 - 4xGPU => 33.7pts need rank #3 - 4xGPU => 34.9pts need rank #9 (with 15 3DWR points) So, even in the multi-gpu categories you need decent ranks to match 'average' score in 1xGPU. With 7970 the rankings are a bit skewed now, though ... but that's always the case.

Hm, I'm inclined to agree with Mean Machine here. In the tradition of UK, I assume lots of people will complain about extending the competition?

Either it's a type and supposed to say "Ram, bro" or someone wanted to befriend the chip and is calling it Rambo in a similar sense like Shambo, Vinbo, Tinbo and so on.

I'm now retesting the 57x chips at home and they are reacting quite differently. Same platform as I used in Tones office, so something else is going on.