knopflerbruce

I'll flash to 6-7-5 later

Picture added. Wasn't "really" Ln2, just had to get the base temp down from 45 to 20. Will do full pot later Just wait till I get a 4P board

Yep. Avoiding silly stuff like screenshot mixups is a good thing. Happened before, and is always a huge pain in the ass - for everyone involved. Perhaps those who do alot of session should watermark their scores. Then there is no doubt (unless you watermark the wrong result, but then you're an idiot )

Yep. Like, a link to the original thread, and then perhaps a list of items for sale (with or without prices). If people would abuse that and do trades per PM it would obviously be behind our backs, and nothing we would care about.

With sandbagging you make sure people do their best as they know that there is a huge chance that the best score from their competitors is not posted yet. If you "had" to post straight away it would be a HUGE advantage for those who chose to bench later. Sandbagging just makes the playing field level. IMO: Personally i don't find it that fun, but sometimes it's necessary. I try to do a bit of both, save some scores for the last day, and post new ones every now and then to stay #1 while the stage is running.

We could do it this way: Everyone uses it for advertizing only, like, make a topic, and post link to the FS thread(s). That way users here can search for items, but without involving HWBot directly.

I've barely been to that site so I have no idea, was just a thumbs up to what he wrote:p

Very good points. "+rep" as they would've said at OCN:p

I feel it's a good idea, but we'd have to know how to do it. If we simply go "buy/sell at your own risk" I don't think it could work, butwe have to babysit and actively solve cases, I'm against it. Too much extra work.

It does indeed sound bugged, yes.

Grats

Areca has onboard cache, usually as a stick of RAM. I don't see any difference, really. The Areca cache and the controller communicate on the same PCB, same as this controller - the only real difference is that the cache of the onboard raid controller is also used by other apps, as that cache also happens to be the system memory.

This is not an ideal situation, but still - major screwup to use ES in a competition. I can't remember ANY HWBot competition where you could use engineering samples, at least not the old monthly challenges. ES scores can't be allowed here, period. In my opinion.

RAM is weird stuff... I sld a few sticks just before Cebit, with screens that proved they were working - still DOA.

I suggested the following set: 30-20-15-13-11-10-8-7-6-5-4-3-2-1-0.5 for 1st-15th for categories with less than or equal to... 30 subs would solve some if the issues... and then go for the regular distribution above that. Top scores are top scores and should be rewarded to some extent regardless of the competition. 30 is still not much at all, you can easily get alot of low skill points in the other categories, too - take a random 3770k, go full pot, and you'll see what I mean.

Popularity already decides the amount of points given? The time factor could be removed, though - atm single core wprime scores are worth 60p, although there is an insane amount of scores in those categories. Dropping because time passes by is not a good idea. There's no good reason to reward 1x and 4x differently because the 4x category has more recent subs.

Popular ones could work, categories with 100 subs or more should be enough. Then you don't really have the problem with outliers anymore. Perhaps you can even "ignore" the ranking when calculating the boints then, award cups by ranking, but points by score relative to the average one, independent of the actual rank (for example two people get 10.000s in 1m in a popullar category, one is 1st because he submits first. He gets a gold cup, the other one a silver cup, but they get the same amount of points as relative to the average score they're equally good).

I suggested this, but there is a valid problem: outliers. In a ranking with one chip ran at stock, and one overclocked 100% - there will be rewarded an insane amount of points for no reason.

wprime1024m scales with more than 12 threads. I'm 99% sure of that, but if you can "prove" me wrong, feel free to do so I think 12 is also too low for 32m, but there is indeed a limitation because of the thread execution time (time needed to execute thread X is higher than time needed for thread 1 to finish it's part of the total computation). This also applies to 1024m, but the critical thread count will be much higher, as the computation pr thread is 32 time bigger. On older systems the optimal thread count for both of these will differ, as the computation time increases - so while there may not be a point in having a wprime32m ranking for modern 24 core systems, maybe the oldest rigs with this configuration will benefit from all 24 threads. THis also means that later, the optimized thread count will drop, so we'd have to remove points as tech gets better - which is indeed a problem. PCMarks do in fact scale beyond the "official" thread count, but not by that much. You can verify this for pcmark04 for example, use a thuban set to 2,3,4,5,6 cores or whatever. Not that much, but there is a difference. S_A_V did some testing on this I believe. 05 does the same thing. Got no clue about 3D, so can't comment on those - but: Note that what's the case today doesn't have to be the case in the future. We don't know if Heaven won't scale to 4x with the next generation of GPUs. I'm pretty sure that what you say about 01 is correct, can't imagine there being any 4-way scaling in that bench ever I'm also thinking that the optimal thread count stuff applies to 3D, as in optimal GPU count. AM3 does scale with 2x GPU on older cards, but not any more (in other words we do at least have to keep the hardware rankings for 2x GPU forever). Perhaps all 3D benchmarks will go that route as time goes by, 1x is the best option to get the highest score. Then we again would have to revise these rankings every now and then, which is not a good thing IMO.

I think we should be consistent. We don't question text edit and TW scores that are crazy high, we just "assume" they're OK based on trust (it looks that way to me, at least). Vapor's word is just as good as Pro, CherV, Stevero's - or the rest of that bunch of crazy guys, so if we can trust them when they say their tweaks are legit (without proof), then we have to trust Vapor as well.

I don't know where to place TEC. Personally i feel it's extreme, but could be that we've decided otherwise before. A good tec can give temps where you need insulation to avoid condensation - that kinda speaks for itself IMO. I'd probably throw in an exception for the V10 air cooler, though:p

I'm not THAT sure about this, but here is what I THINK we decided: If a compressor is used, for example as a part of a water chiller, then it's thought if as a type of phase change unit, and therefore goes into the extreme cooling category. AC units are exceptions to that rule... I thiiiiiink They're usually just used to cool the room to decent ambients anyway, and we have to draw the line somewhere. Having an AC unit in a room is a very passive way of using it, but of course, if you can use them to get -20c air blown directly on a heatsink... it's a bit different. We could to not have the enthusiast league at all... We can't verify all cooling solutions or prove what's correct and not, unless it's REALLY obvious. Common sense and trust are key in the enthusiast league

You keep talking about livestream, what do you think a livestream session will prove? Nothing we know already - it's been CONFIRMED that they have the hardware needed to get those sets of results. In fact they proved that part before there even was a case. Can't remember if it was Christian Ney or Massman (or someone else) who confirmed, but since I THINK (I could be wrong here) that only massman watched your livestream session, the word from just one moderator should be enough in this case, too. Right? Why "do it alone"? There's no rule against helping others if that's what happened. There's nothing strange in having two people running the same setup these days. Of course just one can post results from that particular setup (CPU/GPU to be precise), but that's it. The accusations against you and Gnidarol were just as valid as these, whether you like it or not. It sucks about your holiday, but what can i say - I spend what could be the last summer days we have here helping you guys:o

Only if you're poor

We're still working on it, if we find solid proof punishment will be given. We've found truth in weird excuses before, like stolen USB sticks and god knows what... we'll see what happens when we've got all the facts straight:)