mtzki

Old rankings should never be changed unless we really really have to. We already have a logical reasonable vs. extreme division (or just single ranking where applicable) in all of them. Dividing by cpu sockets would only make some marginal differences in pcmark05 and 3dmark06 iirc (in the latter only the bot AIs in cpu tests get multithreaded extensively and this workload seems to be quite limited), so there was/is no need to do it. The idea would be to never have more than two rankings for a new benchmark, because more than that leads to a confusing situation, and can give too much emphasis for a single bm. So the division (or opting for just a single category) would be made depending on the particular (new) benchmark, looking at how it scales. We could divide by cpu sockets, or by gpus, or by both of them, but would always have at most two rankings for any benchmark.

Voted performance with boints. This bm is just way too important to leave out. IMO there should be two ranking categories, one "reasonable" for single cpu socket + single gpu, and the other "extreme" for all the other setups. Otherwise one would prolly need SkullTrail (or sth crazier) for any competitive scores. Using performance preset for both these categories would be my choice, especially given the screen resolution requirements. Performance preset for the "reasonable" category, and extreme for "extreme" might also do well.

Here's a multicompare link to the iRAM test: http://service.futuremark.com/orb/multicompare_view.jsp?p0=656&n0=iRAM&p1=277&n1=single+WD+RE16&t=iRAM+test&k=18&s=1&l=1&c=-1370763038 The other test was with completely same hw settings, only Vantage installed to the system drive. The runs are definitely not identical, but i'd guess with Vantage the system drive might play a big role in HDD tests even with the bm installed somewhere else. Not much documentation on Vantage out yet so cannot say for sure.

Well just tried to mention not giving boints is always one possibility. Anyways, here are a few reasons why it might actually make sense too now: * no usable free version available * 26 mins or so * no way to tell how well the run is going * one might need ioDRIVE to go high in the charts eventually Vantage looks to me like a great benchmark for system builders, hw reviewers and such...but really not too good for competitive benchmarking. btw did some testing with iRAM yesterday and the score wasn't any better than with a single WD RE16 drive. Looking at the scores so far in ORB it seems obvious HDDs play a large role again, but it might be mostly about burst bw this time. So a large array of HDDs in RAID0 looks like the way to go in this bm until that ioDRIVE. http://www.pcper.com/article.php?aid=224&type=expert&pid=5

Adding Vantage without enabling hwboints is also possible.

Octf Rip

haha we went out with a big bang! The team has split up, OCTF is no more, and the funny name Muropaketti is back for the main part of the old team. At least we reached the top in some sense and it was fun for a long time.

Thanks Mark! Some of the top benchers in the team tend to concentrate pretty much solely on the latest 3DMark, so maybe we'll never be this good all-around. One can always hope.

We had a small campaign for doing this: :woot: Current situation: http://www.hwbot.org/hallOfFame.do?type=team&applicationId=12

http://www.hwbot.org/members.hall.of.fame.do Huge grats mate! It's an amazing accomplishment with the current level of the rankings. SF3D is preparing for some quad-core runs so his ranking is likely to get even better. Two OCTeamFinland members in top3!

Not intentional.

Yah. Maybe the bench profile should also show the best overall position ever. ( weird that i don't remember ever having less than 200 boints. 9th November 2006 list looked like this: http://www.hwbot.org/forums/viewtopic.php?pid=2898#p2898 )

Why now as R600 is out so soon? I thought you were an ATI guy.

Sure. Looking at your list i'm pretty sure you would love them.

Hey that's a useful page if any. Never taken a closer look at it before hehe.