Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
I think going strictly 2d is a good idea. I found when we did the Super Pi challenge in December for example, that I really worked on the bench and my hardware to get the best time I could.If we're going with these kind of restrictions then I'll go with strictly CPU benches. Maybe a totaal over a fer of the low time benches. Then maybe a high total from others. Use something like HWInfo for clock monitoring. Saves any questions about cheating. Might hurt scores a bit.
When you limit CPU clock then cooling won't affect results much. In most cases you barely see any difference going from air to LN2 when CPU clock is ~4GHz. I just wouldn't worry that someone may "cheat". Really weird results will be easy to see.
On GPU clock limit has no point as I doubt we will make competition on the same GPUs. Simply most of us won't buy graphics card for one competition.
I don't think there has to be any memory clock limit as all have different memory and lowering clock too much is lowering chances of users who have new IC like hynix mfr/bfr etc. ( most newer members will probably use new memory ). Still that IC which can make high clocks like 3000+ is performing as good as IC which can make 2133 but runs at tighter timings.
In this case I think that memory settings would be like tweaks. There is no clear answer what will be best for each memory kit in each benchmark so all will have to try to adjust settings for optimal performance.
Also here it's obvious that noone will run memory @2666+ CL6/7 or 3000+ CL9/10 on air ( can run Hynix DDR4 @3000 CL10 but nothing else ).
I would stick to popular benchmarks but not these "made" by hwbot which require any online submissions. Also newer benchmarks like Cinebench are more popular as I see and many users like them.
Also like Witchy said, Win8 should be allowed as all this clock error is not affecting scores of most tests and in some is barely affecting final score. Still most users who bought PC in last 1-2 years have Win8. Not all get 2-3 different OS for benching.
When you limit CPU clock then cooling won't affect results much. In most cases you barely see any difference going from air to LN2 when CPU clock is ~4GHz. I just wouldn't worry that someone may "cheat". Really weird results will be easy to see.
On GPU clock limit has no point as I doubt we will make competition on the same GPUs. Simply most of us won't buy graphics card for one competition.
I don't think there has to be any memory clock limit as all have different memory and lowering clock too much is lowering chances of users who have new IC like hynix mfr/bfr etc. ( most newer members will probably use new memory ). Still that IC which can make high clocks like 3000+ is performing as good as IC which can make 2133 but runs at tighter timings.
In this case I think that memory settings would be like tweaks. There is no clear answer what will be best for each memory kit in each benchmark so all will have to try to adjust settings for optimal performance.
Also here it's obvious that noone will run memory @2666+ CL6/7 or 3000+ CL9/10 on air ( can run Hynix DDR4 @3000 CL10 ~1.7V but nothing else ).
I would stick to popular benchmarks but not these "made" by hwbot which require any online submissions. Also newer benchmarks like Cinebench are more popular as I see and many users like them.
Also like Witchy said, Win8 should be allowed as all this clock error is not affecting scores of most tests and in some is barely affecting final score. Still most users who bought PC in last 1-2 years have Win8. Not all get 2-3 different OS for benching.
1. Ambient cooling only (air/water, no chilled water)
2. Perhaps set clock limits on CPU benchmarks to promote tweaking among the team.
3. Submissions must follow Hwbot rules, period. This is for new people to get used to the format as I see some new people have trouble with posting correct SSs, etc. Even if they are not participating on the team, having that information up will allow 'judges' to more easily catch someone that may be fudging a result. And it doesn't let someone 'win' that can't figure out how to post a proper screenshot or file, etc.
4. Stick to boint bearing benchmarks at Hwbot
I agree 4.0 is to low. I'd have to under clock my AMD and Intel stuff. LOL
On the single core idea..... AMD does not = Intel
Maybe (4) AMD cores to (2) Intel cores?????
I have no knowledge in this as I have been AMD since Slot A
You should see my warehouse.The Wife says to you guys.... That I'm a treasure trove of ancient hardware. Unless it has died, it does not leave after it has entered.