knopflerbruce

True story

Score looks fine for the frequency, and was most likely OK at the time it was posted.

Score looks fine for the frequency, and was most likely OK at the time it was posted. No reason to block. Remember: rules were different back then, this is explained in the rules section.

I will. But, I don't want to remove my scores, so they must remain in the wrong ranking until there is a fix.

Not my fault that ES are unlocked but I will buy retails alter, for the models I can't find ES Thanks!

You know M.Beier has a valid point when he's posting something not related to asian chicks

Maybe I'm not great at programming, but... to me this sounds as if the score is determined by the settings directly, that there is no real measurement done.

Well, if I cared, I could buy some chips and do a 4+8+12+16 core Interlagos or Abu Dhabi setup 40 cores

Bla bla all cpus must eventually be added. Mine are engineering samples, though.

I'd say the one with the most computing power should be the "leading" chip. I guess it's not possible to mix entirely different chips, so an easy way to figure that out would be # of threads * frequency for all chips, and choose the one that gets the higher "performance number"

Ticket ID: 1722 Priority: Low Please add:\r\n\r\nOpteron 6170\r\nOpteron 6262 HE\r\nOpteron 6266 HE\r\n\r\nI dont have CPUZ validations, as they\'re \"corrupted\", even at stock - using the current version of CPUZ.\r\n\r\nThe 6170 isn\'t listed at CPU-World, but it\'s the same as 6168, but runs at 2 GHz. http://www.cpu-world.com/CPUs/K10/AMD-Opteron%206168%20-%20OS6168WKTCEGO%20%28OS6168WKTCEGOWOF%29.html\r\n\r\n6262 HE: http://www.cpu-world.com/CPUs/Bulldozer/AMD-Opteron%206262%20HE%20%28OS6262VATGGGU%29.html\r\n\r\n6266 HE: http://www.cpu-world.com/CPUs/Bulldozer/AMD-Opteron%206266%20HE.html

I got that, it's a bit complicated at times I'm crossing some fingers and toes for this to be fixed...

6300 series are not OK for points in XOL. 6200 chips are one generation old, similar to Ivy Bridge if you like, so they're OK.

No, we don't do it that way - for example, 4870 + 4870x2 goes into the 3x4870 category (instead of 4x, as if 2 4870x2's were used - which happens if your logic applies). The best ranking for my subs would be a 20 core Opteron 6170 ranking (total number of cores of best performing CPU - in my case an ES with Opteron 6170 specs - but sadly no such ranking exists. I've added a helpcenter ticket for Turrican to add generic 10 and 20 core CPU rankings. That should solve everything, I guess. Hopefully he doesn't think it's too weird - then we're stuck with the current situation.

Ticket ID: 1721 Priority: Low As there are a few complaints about some scores using two CPUs with two different core counts (i.e. 6+4), please make generic categories where they could be placed. If you can\'t leave specification fields blank (except core count of course), just set some reasonable values. 3 GHz stock speed, 125w TDP and so on. Brand is also not important. Please make one for 10 core and one for 20, I think there are no other # of cores that needs to be added at the moment.\r\n\r\nThis is a neat little fix, so just do it it can easily be undone if needed. If not I have to let the scores stay where they are, making a bunch of people angry.

I wouldn't be surprised if I can, by using 4 and 6 core C32 Opterons Nonetheless, brand is irrelevant. My scores are not in the right category, but it's not worse than posting 1366 Xeons in a 2011 category either. They're also misplaced, an apparently acceptable solution to a weird issue - as shown in links posted in an older post. I won't respond to questions regarding the category of the scores. The logic and precedence has been explained already.

The GPU case is a bit different. I benched an Opteron 6128 together with a 6132. Those scores will be posted as 2x Opteron 6132, due to the rule you refer to. However, say you bench a 4870 and a 4870x2, then the current way to deal with it is to use the 3x 4870 ranking, as it's matches the correct # of GPUs. It's pretty much the same thing I'm doing here, except I don't have a fitting category to put them in. I've posted a few suggestions in a classifieds thread... let's see if there will be a quick fix then.

Click the links I provided... look at the screenshots... closely. Don't tell me that e5620 + x56XX = e7 2870 either. Too hard to understand, perhaps?

http://hwbot.org/submission/2323466_dhenzjhen_wprime___32m_xeon_e7_2860_2sec_562ms and http://hwbot.org/submission/2271587_stevero_wprime___32m_xeon_e7_2860_2sec_624ms This was an issue a year ago, and was mentioned here: http://forum.hwbot.org/showthread.php?t=43606 ...and also here: http://forum.hwbot.org/showthread.php?p=170425&highlight=e5620#post170425 There's no indication that this practise is not OK. I do have some possible solutions in my head, but for now this is "legit enough" (I could move my scores to e7 2870 instead, if that helps).

Yep, that screenshot is crystal clear.

Weird, HWBot prime told me it didn't exist when I tried to upload a result. Or, maybe it was Multi core PI/LINPACK.

PCMark removed, as it wasn't suppsoed to be there. Score is bad because I used a 1st gen 30gb SSD, and the CPU speed is very low.

Depends. HWBot doesn't have a system that takes properly care of CPU subs with different # of cores. I can't select one 8 core chip and a 12 core chip, so posting my scores in this category to get my well earned ( ) points is the easiest fix. Also, as my rig is always slower than these servers, they don't lose any HW points here I think. There are s few similar SR2 subs in the 1x 8870 rankings, using one quad and one hexa, so this issue is well known already. Maybe time for a fix?

I could overclock if I used a different board

Ticket ID: 1719 Priority: Low http://valid.canardpc.com/show_oc.php?id=2843698