The official keeph8n

[Broken APEX49]

If 9900k is too advantageous

[yosarianilives]

6 minutes ago, GTI-R said:

If 9900k is too advantageous

It's per core scoring 

[Leeghoofd]

Still think that there is more involved than just having the best processor to win this....

[keeph8n]

2 hours ago, GTI-R said:

If 9900k is too advantageous

Current 1st place is held by a i5-6500

[Broken APEX49]

Participants need to understand the rules more

[Leeghoofd]

57 minutes ago, GTI-R said:

Participants need to understand the rules more

not sure what you mean by the above

[keeph8n]

2 hours ago, GTI-R said:

Participants need to understand the rules more

everywhere that this contest has been posted I've been monitoring so that I can help anyone and everyone who has questions.

 

If there are people who don't understand, please send them my way

[Nwalm]

First i want to thanks keeph8n for this competition open to non extreme cooling.

But there is an obvious issue with the score calculation. Righ now high core count cpu get double bonus in 2/5 stages (GPUPI and WPrime, where lower scores are better and multithreaded, when divised by the number of cores this increase the initial lower time advantage of high core count CPU). If the point is to remove the core count from the competition then this formula will work only for cb15 (highter score is better so you can divise the score). For GPUPI and WPrime you need to multiply the time by the number of cores instead.

[Leeghoofd]

Correct observation, thanks for the clarification. Not sure if we can continue with the current format for wprime and gpupi.

To make it easy on everyone we will limit the wprime and GPUPI to 4 threads & no hyperthreading  (disabling cores thus allowed)

[Nwalm]

Replacing the diviser with a multiplier wasn't possible ?

[Leeghoofd]

Nope, sadly not an option in the competition creation tool. But something that needs to be added for sure

[Nwalm]

I am not here for long but its strange that this is the first occurence of this issue. Thanks anyway

[Leeghoofd]

12 minutes ago, Nwalm said:

I am not here for long but its strange that this is the first occurence of this issue. Thanks anyway

this is probably one of the reason we never did per core for these particular benchmarks. Thx for the input!

 

UPDATE TO THE COMPETITION FORMAT:

Due to a limitation in the competition engine we have to adjust the rules for the GPUPI and WPrime stage. Current format was incorrect as spotted by Nwalm.

 

Therefore for the WPRIME and GPUPI stage a 4 Core max rule is being imposed and no Hyperthreading allowed. Disabling cores is allowed

[Nwalm]

Fast ^^

Missiontroll scores where made on an i5 6500 so valids even with the new rules, i dont know if that can be salvaged or not.

[Leeghoofd]

27 minutes ago, Nwalm said:

Fast ^^

Missiontroll scores where made on an i5 6500 so valids even with the new rules, i dont know if that can be salvaged or not.

sadly he didn't use the competition background and took pictures of the screen, so he has to rebench

[Arni90]

The competition states that temperatures monitored by benchmate cannot exceed 15C.

If I use chilled water at 5C, but idle core temps are still above 15C, is that allowed?

[Leeghoofd]

The processor lowest  temps (idle) read out by Benchmate must be 15C or higher... Not talking bout fluid temps. So if you set water temps at 5 and the benchmate program reads out 15 your are fine