MMO benchmarking: the problem no one wants to address

Hello all,


I have been watching the Tek for the longest but just recently joined the forums.  I've brought this up in other forums as was soundly shot down.  I am hoping for a more adventurous group here.


I mean Logan did dare to say the FX-8350 was actually really solid in heavily threaded apps.

Anyway the actually question.  I play MMO's.  I enjoy MMO's, and not just MMORPG's like WoW, Rift, SWTOR but shooters as well like Planetside 2 and BF4.  I am in the market again for some upgrades as I want to replace my girlfriends aging Phenom 955 with my CPU and upgrade my existing 3570K.  Here's the thing most major tech sites like nice, clean repeatable, scientific benchmarks.  So when they benchmark MMO's the do stupid stuff like I don't know benchmark WoW in a flight path in Cata on a empty server in a empty zone.

Yes it creates nice, clean repeatable, scientific numbers.......

Numbers that are worthless.

As the test of equipment in an MMO is not in the middle of no where staring at your feet but balls deep in a 25 man raid with every spell effect in the game going off or a 200+ person battle in Planetside 2.  

When I pointed out the worth of the numbers the shiny, happy useless test provided it turns out I was the same weight as a duck and burnt as a witch.

I let it go as I bought a 3570K and overclocked it and was fairly happy with it but here I am again attempting to buy a new CPU possibly motherboard combo and guessing is the best I can do for information.

I am trying to ready our nerd his/her gamer household for Battlefield 4(yes EA is the devil) and The Elder Scrolls Online.  I would like to make an informed decision when spending $500 on an equipment refresh but the information is lacking.

There has got to GOT to be a way for a more open minded community to produce some more useful number on the subject. What really prompted this was the BF 4 beta. Turn out "most game only use 2 cores", yeah the Frostbite engine didn't get that memo.  Running the BF4 on my 3570K is like running Handbrake, 4 core at 75-80% the entire time and no huge CPUs frame rate related dip.  I was extremely impressed by this as say Plantside 2 I all ways CPU bound when the combat gets heavy (even with the 3570K at 4.4Ghz).

So that little shocker spun off my idea again.  What is the next gen MMO's going to be able to use CPU wise core v clock; cores v Hyperthreading; Hyperthreading v AMDs module approach.


If you look closely I never did ask a question... so here goes.  So could a community like this pull off some realistic MMO benching?  Like I don't know.  A private server full of Tek Syndicate forums members being asked to run a set set of actions to try and limit variables while Wendell and Logan bench and record...

It would be so useful to the MMO community to have meaningful information while purchasing equipment.

Thank you,



I can't answer all your questions but I can tell you hyper threading isn't anything that resembles physically cores. It's software that creates virtual cores. Number crunching generally takes advantage of it better such as physics. Nonetheless benchmarks or I gtfo.