Massman

It's quite difficult to find the exact reason. Do the points add up correctly?

It's not thát much work . Oceania region is up and running. I noticed some countries are located in Oceania, but belong to other countries. Not sure what rule to apply here; I don't think it will be a big issue either.

Ideally, it would get included in GPU-Z. Sadly, W1zzard is not up for it.

Achievements are currently broken. We've added it to the backlog, but other things have higher priority at the moment.

Yeah . Interesting tracking idea. Can it be flash, and can it be integrated at hwbot

Looking into this, sounds like we might have missed something in the R4 logic.

Sadly enough, no. We make use of two team_id entries in our database: one in the user database (to link a user to a team) and one in the result database (to link a result to the team). Normally, the engine should update all the team_id fields in the result table when the user changes team, but that is not the case at the moment. The team_id key in the user database is working. For performance reasons, we make use of the team_id key of the result table to find all results of the team rather than going through the user database. So, to fix your problem, I have to manually update the result table with the user's new team. It's a small bug, so it should be that difficult to solve.

This subforum is to keep track of the different application that are currently in development.

In this thread you can post your suggestions for new HWBOT applications. To keep it easy for the developers, please include the following topics in your post: - description: what is the application supposed to do - sketch: quick sketch on how you visualize the application - language: no, not english, but what coding language - already started developing? (yes/no) - what kind of help you need (no $$) Thanks for the input!

Ooh, I remembered how I fixed it the last time . I fixed both of your team-mates' scores. The points should be adding right now.

Marcolecorse is now visibly contributing to the team?

Check PM

Of course official background is not needed for CPU-Z validation based stages

Impressive! Well done, Matt!

You don't ever lose points by submitting more results.

There's a 'best of team' filter on every ranking page. Using that option, you can see all TPP scores and rankings.

No Access or Origin. Google charts APIs might be interesting for adding this kind of graphs to the frontpage, though. Need to look into that. Here's a more pertinent chart: what brand of mainboard people use for Xtreme overclocking. XOC defined as CPU or VGA on subzero cooling. 2007/2008 might be a tad erratic, 2009-2011 data is okay. Check out Evga's position in 2009!

Expanding in vertical direction? I could add more brands, but then the chart would become very messy, I think.

Hey all, I've been having a bit of fun with the database and pulled this chart from it. The unofficial mark means that I didn't put a lot of effort in taking out the false positives (eg: user indicated wrong motherboard), but these should be quite minimal. This chart shows the amount of unique overclockers that use a certain mainboard brand over time. Note: one person can use multiple brands! We can see that: - 1/2 overclockers use ASUS - 1/3 also use GIGABYTE This is pretty interesting, because it kind of points to the situation where a lot of people are convinced of the quality of an ASUS mainboard (well, 50% does not use one!), but more and more people will test a GIGABYTE mainboard too. Translated into a current situation: people may use M4E first, but then also use GIGABYTE for good 3DMark01.

People rightfully say that "hwboints are not everything".

Yes, 30 minutes after we started testing internally, we were already able to submit hacked results. It should have been fixed, though. When we reported the issue to FM, they released a patch version that fixed the way we could cheat it. Can't remember if we spend more time on it after that. Maybe we should ...

if you use gigabyte, msi and asrock, you're part of each group.

You submitting 1 result with an MSI board or 10000 doesn't make a difference in this chart. These are # unique users per brand.

Posted this also at XS. Unofficial means I didn't put a lot of effort in excluding false results (eg: someone linking P67 board to 1366 cpu). Like I said on XS: people use different brands and different boards.