websmile

I agree, still we have the problem that some boards are preferred this way, Asrock reports 4007 for example and even higher as 4,00ghz and Asus as 4,01Ghz already. Well, I guess we have to live with this^^ - what is clear for future is that we want monitoring enabled and visible

On the "average of 3 best" stages we want three different members with different hardware, this is because we want to emphasize this is a team competition. On the dogpile, feel free to submit as much different cpu/vgas as you personally have, it is unlimited amount of subs per bencher as long as it is different hardware

OK, we talked about this and as long as it is at same multiplier, the 4,51ghz cache readout is OK, especially because we refer to the common sense but did not write it explicitly in the rules. As we decided to handle it this way, I also make an exception for the result with hidden settings and monitoring, but only once. To make this clear, hiding settings at an XTU low clock and disable monitoring is no go in the future

Ja, unglücklicherweise gehen wir davon aus dass die Leute die Regeln kennen - aber wie immer sind halt die anderen schuld wenn man trickst und erwischt wird^^

I made a fast check at the results of xtu 4,5ghz today, and I would recommend top 3 results rebench, top 2 have cache above limit and #3 disabled monitoring so I cannot see cache frequency - unless stated otherwise cache is always equal or less to core, and if you disable monitoring and I cannot check this, result is most likely invalid, I have to check with head of moderation and sent out warning

Yes, two vgas cores are OK - Crossfire includes 1x 3870X2 or 3850X2 card

You can oc with multiplier at 6700k, if you have decent cooling you can go for 45x and set vcore to around 1,3v, then do some stability tests with prime, Linx or xtu for the start if you can enter windows at these settings. If it works, lower vcore and try again until you found lowest possible vcore stable. I guess this would be a good start, no need to mess with memory settings at first for example

I hope it helps - this is a problem that can also hit GPUPI http://forum.hwbot.org/showthread.php?t=154332 If solution does not work, you might have to re install aquamark or even OS

Did you disable system info at the runs like it is recommended? Make a package and send the files to christian ney or genieben to check, maybe they can fix if runs are not bugged or something else is wrong with system.

Apart from the fact that Samsung ICs can take much tighter timings at same volts than Hynix, which is most likely one reason for your results, LN2 BIOS might have higher power target so the card runs frequency continuously, on lower power target it might drop a few mhz during longer load. Did you monitor the frequencies during the run to check this?

awesome score, well done

Do you think a couple of days before start make a difference on a competition that runs for 13 weeks?^^ P.S. Please take in that we need to implement some stuff and fix some hardwarelistings at database to make all work perfectly at start, so it is only partially trolling

I guess the site will eventually go online a day before start earliest

Seems that the dummy site still is full of misinformation

I have no idea what you found, but this is no team cup stage^^ - in general, when we need three results for a stage, always three different cpu or vga models are required to make an overall score, for your example this would mean indeed 6700k, 6600k, 4670k or whatever other cpu

Nope

Seems you once again mastered the beast

There is an option in the BIOS to reflash/recover MEI

No restrictions planned so far, neither on cooling nor do we plan frequency caps

Logistics, organizing hardware and coordinating within the team is always one of the biggest challenges at team cup, we know this. We opted for max of three submissions at multi sub stages to give chances also to smaller teams, often this ends in sessions where they bench together to get the results needed. I really hope for your understanding that we made these multi-sub stages because we want to emphasize the team spirit, and from my experience in the past as member of awardfabrik, which has very few active benchers last years but always did very good, I think smaller groups often do better because they are better organized

good job as usual, always a pleasure to watch you bench those legacy cards

Don´t worry, if I remember correctly it is not user limited (means submit as many cpus as you have personally) and we put weight on points in a direction that this is for sure not over present. I bit my tongue a bit on comments above, maybe people should see that 75% of hardware needed is 4 years or older, 50% 6-8 years and so on. Even on new gen stage you can use 3-4 year old stuff at some stages. We can not omit new gen benchers completely, for us everyone is welcome at team cup

Good job, Stu

Maybe we should make a one man cup in the future - this is extremely poor, if you are not able to get three!!! different guys to submit at certain stages I wonder why you are in a team anyway... I am always open for legit criticism, but you should simply remember what this is called - team cup and not one man ego trip cup

Last bump for the Impact, all other items are sold Closed, thanks for the space, hwbot