Most Valuable Submission of Week 32, 2015: 166.3 points by Dancop (Germany)

Last week we received about 3851 benchmark results from 891 registered overclockers around the world. The majority of the submissions is coming from Rookie overclockers representing 57% of the active community. They were responsible for 38% of the submissions. We had a peek at the most valuable submissions in a breakdown per league.

As you can see in the table below, the most valuable submission of the week was the Global First Place of Dancop in HWBOT Prime using a Core i7 6700K. The score is part of the ASUS/G.SKILL Skylake Launch Day Domination we covered in an article last week. With 166.3 points Dancop yields more than 100 points higher than Japan based Ikki . In the Rookie area we see Ziken scoring 31 points in XTU with the new Core i7 6700K.

The overclocking results submitted during Week 32 generated in total 590 World Record Points, 10023.6 Global Points (almost double from last week!!), and 6102.5 Hardware Points. The distribution per League is as follows: 35% for Elite, 28% for Extreme, 11% for Enthusiast, 6% for Novice, and 20% for Rookie. The representation of the active community is as follows: 3% Elite, 12% Extreme, 14% Enthusiast, 14% Novice, and 57% Rookie.



Most Valuable Submissions - Week 32, 2015

League Points Benchmark Overclocker
Elite 166.3 points HWBOT Prime 4xCPU Core i7 6700K Dancop
Extreme 60.6 points 3DMark Fire Strike 3xGPU GeForce GTX 980 Ti Ikki
Enthusiast 39.7 points 3DMark Fire Strike 1xGPU GeForce GTX Titan X Ximi
Novice 30.5 points Heaven Xtreme 1xGPU GeForce GTX 980 Ti IOWA
Rookie 31 points XTU 4xCPU Core i7 6700K Ziken


10

United Kingdom ObscureParadox says:

rookie XTU link takes you to a totally different score

South Africa DrWeez says:

we covered in an article last week. link brings you to this same page

Belgium Massman says:

Fixed all the links!

United States steponz says:

Wow are points totally jacked up... 4c hwbot prime is worth more than single FSE 166.3 to 126.6 XTU 6x is 231... seriously and we took legacy wr points away... All gpu should be higher than single cpu...think its about time to stop doing points.

Belgium Massman says:

steponz said: All gpu should be higher than single cpu


Hm ...

I suppose an algorithm based on participation is a problem when a simple benchmark is a lot more used than "complex" 3D benchmarks.

An arbitrary difficulty multiplier is a possibility.

United States steponz says:

Massman said: Hm ...

I suppose an algorithm based on participation is a problem when a simple benchmark is a lot more used than "complex" 3D benchmarks.

An arbitrary difficulty multiplier is a possibility.


its not a bad idea...

Belgium Massman says:

It might make things more complicated. What would do you think - devise a scaling algorithm or just gradations. IE: [list] [*]Grade C, easiest: 1.00x multiplier (for superpi, xtu, ...) [*]Grade B, more difficult: 1.25x multiplier (for single gpu benches) [*]Grade A, most difficult: 1.50X multiplier (for multi cpu/gpu benches) [/list] No idea what the effect on the rankings would be.

United States steponz says:

That would be interesting. . Some benchmarks should probably be higher like 32m.. maybe we should do it by number of pots.. like 32m always 2 pots.. single gpu always 2 pots..

Germany der8auer says:

[QUOTE=Massman;402124]It might make things more complicated. What would do you think - devise a scaling algorithm or just gradations. IE: [list] [*]Grade C, easiest: 1.00x multiplier (for superpi, xtu, ...) [*]Grade B, more difficult: 1.25x multiplier (for single gpu benches) [*]Grade A, most difficult: 1.50X multiplier (for multi cpu/gpu benches) [/list] No idea what the effect on the rankings would be.[/QUOTE] Maybe it's easier to adjust one benchmark instead of the whole point system :D Sounds more like looking for a solution instead of solving the problem. Adding grades would also make the whole point system even more complicated.

K404 says:

Wasn't this part of the reason for having WR points in the first place? It awarded 3+4-way GPU setups, assuming the benchmark was coded in a way that it scaled..... There's also a risk that multi-GPU setups (keeping in mind that 8-way is heading towards us) mean that it's gonna be more money-orientated..... 4 dual-die GPUs and/or Titans....... Running multi-card on stock has every chance of being worth more than single-GPU on LN2. (that already happens now, but will be interest be skewed?)

Please log in or register to comment.

Leave a Reply: (BBCODE allowed: [B], [QUOTE], [I], [URL], [IMG],...)