der8auer

damn great!

Thanks corrected

First post updated

So I just spent 3 hours checking every single posting up to page 57. So far about 75 different batches checked and it seems like there is a big variety even if you have the same batches. Keep going guys!

L315B5xx is usually good on air.

Well now there is one

That's exactly the problem. You rather loose the records than sharing the information with pro who volunteered to be PCM05 moderator and doesn't submit scores anymore himself. PCM05 is the only benchmark where we have to check almost every top score because there is some new and obscure tweak. So another top overclocker reports the score and we have to check it which means that we have to ask for the tweak to test and approve it. For all other benchmarks it's pretty easy to say whether a score is out of the line and if there is a tweak it has to be shared with the Staff. If not the result will be blocked. The wrapper will not fix the moderating issues. It will just validate the subtests and maybe exclude some of the illegal tweaks but I'm sure that the wrapper will not detect all hardware modifications or unknown tweaks. @ Schmuckley: I'm really open for good solutions but like I said, this time a wrapper will not fix the problems. The result moderating is the problem. We'd need more people like pro who step back from submitting scores but stay up to date and check every single new tweak. Like sofos said I don't want to remove the benchmark. But disabling global points would lower the moderating pressure a lot.

whoops edited

BTW I just added up the first 20 pages in the 1st posting. Will do more tomorrow.

And then you end up with a CB at -90°C and a max clock of 5,8 haha

I think you still need low voltage on air but it has to scale on cold and high voltage. My first 4670K did 1,15V 4500 MHz primestable which was a very good one. 6,3 GHz max valid and 6,1 for 32m. I had a 4770K which did 4500 MHz at 1,175 Volt which is just a little bit worse but it did not scale on cold at all. 5,9 GHz max on LN2... But the L313B996 ones which only did 4300 MHz at 1,275 V primestable were 5,5 to 5,6 max valid on LN2.

Which batch is this one?

insane memory

Should be good on air. But I think it won't scale on cold.

I didn't do anything Probably just took a bit to calculate the submission.

Can you check again? Looks good to me.

Interesting. I had about 10 of the same batch and all were crap.

Nice work!

Yep. I'm just saying we have to think about the case that we have nobody anymore to moderate those results. Of course with a time frame of e.g. 4 months or something like that. I also thought of a wrapper but the problem is, that the wrapper does not fix the issue here. The wrapper just approves the subtests and settings. But I'm also 100% sure that you can still use some illegal tweaks to push the score. And again then there will be nobody to approve or block those scores.

Yes I know that but the effort to keep this benchmark alive is just too high.

Holy crap. That's an insane score! btw can you post more details about the mod on your board?

Well. I'd support that aswell. Hardware-Points are fine to me but globals should be removed. Reasons: - It's 99% about tweaks and not about real hardware performance -> this means the PCMark does not create comparable results and the ranking is nonsense - The moderators have big problems to remove the moderated submissions. At the moment only pro has enough knowledge to aproof a result. Of course he can't be online 24/7 checking PCM05 results so there are a lot of results which stay reported for weeks and I can understand that the reported user is complaining about that. - I'm completely sure that no other moderator will take the time to go deeper into PCMark. So whenever pro stops moderating those scores, we have to disable the points anyway. - Due to the complex rules almost nobody submits scores to PCMark anymore. There are a few left like dhenzjhen, mikecdm, SteveRo and Moose83 but that's about it. I know it's this discussion again but at some point it just doesn't make sense anymore to keep this benchmark alive.

Well if you ran at high clocks and got the result you can switch to slow mode to open GPU-Z, CPU-Z and all that stuff. Switch off slow mode to take the screenshot and turn it on again. This will prevent your system to crash if you open CPU-Z or GPU-Z which sometimes happens.

For screenshots you should show the real CPU-Frequency.