Gagzila's Gaming Rig for 2012

[HarryTaco]

Sweet Looking Rig Gagzila. I like just about everything about it. Grats Man.

My son wants to contract me to build him an Ultimate Gaming Rig soon as he comes home from Afghanistan. He is comming home for the Thanksgiving Holidays. He wants to give me some cash for the build but I told him just today that I would want to wait for the New Intel 2011 socket and the New Nvidia GPUs. Must be nice man.

[Gagzila]

No worries at all Fogel

 

Thanks for putting some time in researching into the matter, will definitely be looking into it more once I get home myself. I will grab the latest drivers as well, I've only updated the firmware but didn't think at the time to do the drivers as well...it was a late night again so brain wasn't 100%

 

Thanks Kwok for the kind words, I really am starting to love the new rig now...had some little battles over the last couple of weeks but pretty much iron them out 1 by 1 and am almost to the point of not having to worry anymore and just get to play on it

 

Your son will love gaming with the new gen of tech about to come out, good luck with the build....make sure to do a work log like I did with lots of pics

 

Cheers,

 

Craig

[HarryTaco]

No worries at all Fogel

 

Thanks for putting some time in researching into the matter, will definitely be looking into it more once I get home myself. I will grab the latest drivers as well, I've only updated the firmware but didn't think at the time to do the drivers as well...it was a late night again so brain wasn't 100%

 

Thanks Kwok for the kind words, I really am starting to love the new rig now...had some little battles over the last couple of weeks but pretty much iron them out 1 by 1 and am almost to the point of not having to worry anymore and just get to play on it

 

Your son will love gaming with the new gen of tech about to come out, good luck with the build....make sure to do a work log like I did with lots of pics

 

Cheers,

 

Craig

Thanks Craig. I will definately do a work log.. it will be my greatest masterpiece.

[OrangeJuice]

Nice rig, call me Jelly.

 

Also who uses 56k these days?

[Gagzila]

Not many people these days and most likely no one on here LOL ... nevertheless I thought I would warn anyone that still is, maybe using a hotel's 56k service or a mobile connection with a limited data plan

 

Stay tuned people, finally doing some game benchmarks. Just finished the first run of rFactor on standard clocks, basically doing 3 two minute FRAPS benchmark runs to get a more averaged result and will try to keep each run as similar to each other as possible if the game doesn't have it's own inbuilt graphic testing feature.

 

Cheers,

 

Craig

[Gagzila]

Hi Everyone,

 

Finally got round to doing those game benchmarks I kept saying I would do, so tired now though as I have been going at it for the last 4-5 hours...so need to hit the sack

 

So basically using the inbuilt FRAPS benchmark tool to output a 120 second Min/Max/AvgFPS excel spreadsheet, I have setup scenarios where I was fairly accurately able to recreate the test bed for running these benchmarks of which I ran 3 times on factory default clock speeds and then another 3 times on full OC settings that I have tested successfully so far. All game settings were maxed out for all tests, the only changing factor was the clock settings. I then decided to do a 3rd run on default clock speeds with just Anti Analysing set back to x1 and in the case of Aliens vs Predator I also turned off the Dx11 Advanced Shadows as they do squat in terms of noticeable image differences when playing or even when stopped and specifically looking for them....the results speak for themselves

 

 

So a quick run down of the scenarios:

 

rFactor

Starting dead last on the grid in a pack of 30 competing cars with all viewable and a sunny clear day to be able to see everything, basically started the benchmark from when the lights went green till 2min ticked over and it stopped itself automatically. I would simply overtake about 3/4 of the field, showing plenty of vehicles across 3 monitors () and then fighting my way up through the rest racing on the Nurburgring GP circuit which then leads into the famous 1920s built, 22+km / 170 corner track (though I don't get far with only 25 or so seconds left by the time I get into it)

 

Aliens vs Predator

Playing the Jungle level from the Marine missions, it goes through a range of indoor and wide outdoor areas with a lot of atmospheric effects with plenty of Aliens trying to kill you (even on easy) which gives the Dx11 tessellation plenty of opportunities to show them off. For the third run I basically disable Advanced Shadows and put AA back to x1 which gives a huge FPS increase while the game still remains looking superb, Tessellation is the only changeable Dx11 extra feature that is worth leaving on as it makes the Aliens head look a lot rounder and smoother and better depth of field. To test I worked out a regular path from the start of the level to trek each time, remembering which way to look, a couple of pan shots to take in all the beautiful scenery and tried to play out the same actions each time. The consistency of the averaged results I think say I was fairly successful

 

Battlefield: Bad Company 2

I went the Snowblind level in this game on easy, as it really stood out visually with the high contrast of the white snow and sun reflecting of it all and making everything glow. Same thing on this one, start benchmarking at the start of the level. Have a look left, right and then set of sprinting after my squad, battle my way through a few engagements using different attacking animations, making explosions and scoping in at times. Was harder here to maintain an exact action replay each time but managed fairly well since it's a pretty heavily scripted game between the actual fights.

 

Starcraft 2: Wings Of Libery

Wasn't sure straight off how to do this one, then found I had a replay saved of the one and only game I've played of it so far...not a huge battle or anything but with the FPS I saw in testing, I'm not worried at all ...what I did was replay the 19 minute long game and sped it up 8 times so everything was rendering at 8x and the game could get to a stage within the testing period where there was a fair few units on screen. Basically the game never bothered the rig, though you can tell that it relies on the CPU much more because of all the unit multitasking going on. I also played one normal game on the same map which I attempted to setup the base the same as the replay game and have about the same number of units on screen and then ran a normal benchmark over 1min 40sec...on default clock speeds it averaged 10 FPS more than the OC average, basically because it wasn't rendering everything 8 times faster. Edit: I forgot to mention that this game was only run at 1920x1080 as the game locks out eyefinity / multi-monitor resolutions (though it can easily do it) and you have to modify game files to make it work which is a quick way to having your online account banned, meaning I can’t play the game so wasn’t going to risk that. Basically the developers think it gives an unfair advantage over other single monitor players as you can see the whole width of some maps, though I don’t see the advantage as your viewable area before the cloak of war hides everything still only extends the same amount.

 

I found OC definitely helped but not as much as expected in the higher end FPS games (especially AVP, it's such a dog for graphic processing) and I believe this to be that most of the games rely heavily on the GPUs which I could only OC a further 50MHz over the default MSI OC of 850MHz. I do plan to try reflashing my two HD6950s to the HD6970 BIOS to unlock the shaders and clock speeds of that card and if I am successful I will revisit these games again in another full OC round.

 

I will also be benchmarking Battlefield 3 and TES V: Skyrim once they turn up (can't wait ) so keep an eye on this thread for more testing. I also plan to run OCCT and Prime95 on my 5GHz OC to see how she actually holds up over time, will post results when I get round to it

 

That's it for tonight, seey....

Edited November 1, 2011 by Gagzila

[_TheAlexO]

so.many.boxes...so.fng.jealous!

[Waco]

I've updated their firmware to 2.15 to combat the SSD disappearing issue. I thought maybe I needed a tool to align the start of the data better but from what I've seen in before and after benchmarks, it's only a marginal increase in performance, like an extra 50mb/s on say the SEQ test where I'm at half speed compared to the specs. But then do people think the AS SSD results are directly comparable to factory specs? As I remember reading, it's a self made benchmark program by some programmer that misses out on support on a few things and is just good for a general comparison when changing / updating settings?

 

If you're using AS SSD - it uses incompressible data so you'll see the lowest numbers you'll ever see from your drives using it. Usually the incompressible read/write speeds are roughly half of the rated read/write speeds.

[Gagzila]

If you're using AS SSD - it uses incompressible data so you'll see the lowest numbers you'll ever see from your drives using it. Usually the incompressible read/write speeds are roughly half of the rated read/write speeds.

 

Thanks Waco, between a few other reviews, previous comments by others on here and now your own, this really helps confirm they are operating as normal. I thought that they may be and that the benchmarking software was simply a good tool for comparison in testing changes but not against factory specs, this definitely confirms it. I can't say I've been unimpressed with the speed things actually load and transfer when copying / installing...definitely an SSD addict

 

Cheers,

 

Craig

[Stonerboy779]

You know what, the thing in that graph that I am liking most is those min fps. Good to see everything so far is sticking above 30fps with max setting and dx11 lighting effects.

Also couldn't you clock the gpu further only a 50MHz core oc what about shaders and memory?

 

Maxed settings with AA back at 1x easily outperforms overclocked results with no noticeable difference.

 

Yeah it usually gets to be that way. AA and shadows can rally cut down on fps for not noticeable to sometimes only barely noticeable graphical gains.

Something you can clearly see when you went no advanced shadows for your third run of alien vs predator bench.

 

Dump AA to 2x or 4x shouldn't see too much difference in fps from 1x but diagonal lines will be a bit smoother which is always good I usually sit on 4x and tone back shadows more if I need to raise a few more fps from my min fps.

 

 

Edit: Way to put people to shame, with such a detailed run of benches Now you need to further increase the list of games you do it with when/if you flash to 6970

With graphs showing fps and temps side by side. The one thing you where missing temps!!

Edited November 3, 2011 by Stonerboy779

[GoLLuM13]

what the ... wooow your rig is just Awesome

[Wallaby]

that looks amazing!

 

and I see that you are a fan of Iron Maiden , and of... hoovers...