Massman

Wrong. It's about how much competition there is. If we'd reward mainstream hardware, we'd be seeing huge points for the 8500GT, 8600GT, 9500GT. 8800GTX was high-end at it's time, as well as the E8500 (which is more competitive than E8600).

That graduation did not represent the effort one had to put into benching. Why should we be handing out points for free?

That's one of the issues that crossed our minds months ago. Given that we want the overclockers league to represent skill more than the size of one's wallet, we opted to reward the one who has to compete against more people more.

So: - You mainly bench close to stock speeds - You only use air cooling - You almost never compete in highly competitive rankings BUT - You want loads of points

Okay, let's give 0 points then.

Interesting case you bring up. Given an equal amount of financial input, given an equal ranking in their respective category, give a different degree of competitivity in the category, which of the following results should be rewarded more: A) Effort to: insulate, voltmod, find appropriate OS/driver/tweak combination, test for first time, re-test for ranking, get LN2 ... B) Insert 4 graphics cards, bench Note that the competition in category A is much heavier than in category B. So, you can drop in ranking much faster. Which of the following should be rewarded more?

The downside of allowing everyone to score in all categories is that you make it mandatory to compete in all 4 categories in order to be somewhat competitive: - Financial input is extremely large (5870 + 2x 5970 needed) - Financial input is rewarded over skill/effort (those who can affort to bench, get points) - The 3D aspect of hwboints is not balanced with the 2D effort (24 3D benchmarks, 7 2D) These are the three main problems we faced when splitting rankings further. There were a couple other minor ones, but I can't remember them all at the moment. The 5 categories is most definitly a viable solution, but I'm not that eager to apply this kind of major changes at this point. Rev3 has been up for less than 48h, which means that the real paradigm shift has yet to happen. Currently, everyone's still applying the Rev2 benching tactics to the Rev3 game, which inevitably leads to "wtf is happening"-situations. Before we can uncover the real flaws of Rev3, and apply fixes where needed, we need people to actually test the system. Both purely for boints and purely for rankings.

*zucht* Why should we allow a bugged result?

It's indeed not your problem. It's ours. We solved it.

Hardware is under NDA, therefor not applicable to rankings.

FYI: I've declared myself world champion of "Wheel-on-wheel-racing". I hope I get a million dollars now.

I fail to see how this revision is any worse than rev2 for those who are less fortunate in terms of financial input: - Rev2: 2x dualGPU needed - Rev3: 1x singe GPU with loads of effort OR 1x dual gpu at stock At the moment, it's not thát impossible to get into the top40 1xGPU or top20 2xGPU. It takes a bit of effort ... yes. That's the whole point of rev3

Ok, I hear your problem. Here's what I'm thinking about - All points you gather are added to you team's total. If you are 2nd and Tsan is 5th, it's better than you nowhere and Tsan 4th. - The underlying cause is the multi-gpu being not so popular, not the inherent lower reward for this category - You affect the popularity as well - The "poor" bencher has the possibility to play in the single GPU category, whereas this used to be: you need dual gpu. As far as I can see, it's still possible to get in the single gpu top 20 by using stock cooling (various examples available). This means that the we have the same effect as Rev2, but without the necessity to have to buy two dual-GPU graphics cards in order to get somewhere.

Yes, especially when it's you. From the creator to the hwbot staff: He requests your score to be blocked.

Okay, another bug.

Did you overclock only one core? This invalid-thing often happens when one fore is clocked a lot higher than the other cores.

We don't ask anyone to go slower, we challenge everyone to race in a highly competitive competition. Why should one be rewarded for racing alone?

Fixed

The creator of MaxxMem has confirmed that the score is a bug caused by the application: the hardware was incorrectly recognised which resulted in an impossible latency result. Will be blocked.

Because there are other reasons why you bench besides HWBOINTS?

I suppose you will be rebenching the card anyway?

Oh. Yes, yes it is

No, I have to "approve" the posts made using the frontpage comment system. When approved, they show up as normal + are added to post count of user .

I prefer the CPU-Z information - is a more solid base for adding specs.

I'm using FF