System Builder Marathon, Sept. 2011: $2000 Performance PC

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
[citation][nom]cangelini[/nom]Check out Scott Wasson's excellent story about Crysis 2 on The Tech Report. Tons of artificial geometry to favor Nvidia's tessellation-emphasizing architecture. Not sure that's something we want to fold into our benchmark suite. We are looking at new games, but bear in mind that as soon as we ditch these, we kill the comparison points from the previous quarter. That's why you don't see the SBMs shift benchmark suites as often![/citation]
I think you should at least include a couple, like Witcher 2 and i dont see why not Crysis 2, it's an extremely popular game and you can test it with or without tesselation, most people are over playing the original Crysis. Just include a couple new games but dont publish them if you think it will spoil a comparison, and save them for the next months comparison, or compare them between your low, mid and high end builds.
 
Why does Tom's never swap parts?
For instance if the motherboard proves to be an obstacle, send it back and get a different one. In your story you could mention the problem with the first motherboard and how you overcame it with the second motherboard. I would think that an enthusiast able to spec and build a computer like this would be prone to send a motherboard back for a replacement.
 
[citation][nom]cadder[/nom]Why does Tom's never swap parts?For instance if the motherboard proves to be an obstacle, send it back and get a different one. In your story you could mention the problem with the first motherboard and how you overcame it with the second motherboard. I would think that an enthusiast able to spec and build a computer like this would be prone to send a motherboard back for a replacement.[/citation]Author's guess that the MSI competing model might be able to support higher continuous current, but if that didn't pan out then what, a third board? The boards known to O/C better are all out-of-budget.
[citation][nom]nevertell[/nom]But you wouldn't be able to store your boatload of apps and games on a 120 gb ssd, would you ? I wouldn't, as I would have the ssd split in half for dualbooting and I have games. Otherwise, a great build.[/citation]The system needs around 90GB just to run these tests. You'll need around 20GB for the normal accumulation of files, assuming you put all your "junk" on the other drive.
 
I don't understand how on one-hand you guys are adamant a high-end build should have a Bluray drive (it shouldn't and you finally realized it), but you willfully cheap out on the case. A coolermaster HAF 922 with an optional side fan attached will offer worlds better cooling performance for about $50 more than the antec. The 200mm side fan is almost large enough to actively cool the entire motherboard not to mention cable management would improve as well.
 
[citation][nom]kriswone[/nom]I guess i don't understand this statement:"Remember those tests where both systems were too slow at 2560x1600 to play smoothly, but the new build was the choppier of the two? Even though the performance delta was meaningless at frame rates that low, these results add high-end gaming to the new build’s list of value deficits...Can someone explain this?[/citation]Easiest answer: If one does 16 FPS and the other does 4, neither are playable. Ergo, the performance difference therefore doesn't matter. Even though the performance difference between THOSE TWO scores doesn't matter, the lower score still hurts its machine when added to the "average performance" calculations.
[citation][nom]soloburrito[/nom]I don't understand how on one-hand you guys are adamant a high-end build should have a Bluray drive (it shouldn't and you finally realized it), but you willfully cheap out on the case. A coolermaster HAF 922 with an optional side fan attached will offer worlds better cooling performance for about $50 more than the antec. The 200mm side fan is almost large enough to actively cool the entire motherboard not to mention cable management would improve as well.[/citation]Hmm, what part of "no heat issues with this build" isn't clear to anyone?

Let me use the BTX form factor as an example: The more closely the CPU area resembles a "duct", the better Tom's Hardware tests have found a tower-style CPU cooler to work. Narrow cases typically have the lowest temperatures in Tom's Hardware comparison, whether tall or short, likely because more of the exhaust air is forced to first go through the CPU cooler.

The Antec Three Hundred is cheap and has good airflow. It's been used in many SBM builds because of the low CPU temperatures and price. Size only played a major role in one of those builds, because the graphics cards in that build were internally vented and all their heat had to go through the CPU cooler as well. So Tom's Hardware quit pairing internally vented graphics cards with the Antec Three Hundred, and the problem went away.

It's really that simple. Look up the Hyper 212 Plus review to see how bad a cooler it is, then take a look at the temperatures in this article. Remember that you need to get into the high 90's before Sandy Bridge overclocking is significantly hampered.
 
can do ! systems !Hope it works better than the original defy
17.gif
 
I'm a believer in premium parts, but the one place I would say people do tend to "over-spend" most (I include myself in this category) is on a new case. There are simply nice cases like this one that do the job well, and will last a decade or more.

That being said, the "expensive" cases from Antec and other high-quality manufacturers, across the lifetime of case that, properly purchased, will last you through several builds even if you only build every few years, is very low in the lifetime of the unit. Why not get what pleases you in overall features and looks?

Tomshardware has a "hard cap" of a specific amount, and most of us can fudge a bit if we are approaching the upper limit for a build so have more flexibility in our choices. Still, case choice, bottom line, usually (not always of course) ends up mostly about aesthetics, not performance or quality.

😉

 
I wonder why this build is tested at 1280x1024 and 1680x1050 resolutions. I'd assume such graphics focused equipment would be normally be used on big or multi-monitors using 1920x1080 and higher.
Is this done on purpose to get a comparison ground against the other builds on the marathon?
 
[citation][nom]Crashman[/nom]No, because the only serious issue with the build was the motherboard VRM not outputting enough current to go to 1.38V [/citation]

But if you read the conclusion on the Tom's motherboard roundup for the Gigabyte motherboard you find that the MSI one was considered better for overclocking. I realize it lacked some of the Gigabyte's features (that were deemed necessary for this build), but I want to point out that you started with a motherboard that had iffy overclocking (from the review) and you got exactly that. I think you could have easily foreseen overclocking problems with this board just from reading your own in-house review and then rejected it. Skimping $$ on a motherboard shouldn't be recommended at this level of the SBM.

Tremendous VALUE is achieved through overclocking and I believe that is the spirit of the SBM.
 
And might I point out that since you have determined that the VR was overheating at 1.38v, running at 1.35v is still clearly pushing the limits of that circuitry. It might be stable now but how long do you think that system will remain stable while pushing the thermal limits of its VR? A year, two? Don't get me wrong, I totally appreciate the SBM series and I think you guys are awesome, but I certainly wouldn't base this system on that motherboard.
 
I need an expert opinion here guys. Here are my specs:

Asrock extreme4 gen3
EVGA GTX 580
Core i7 2600k
G.Skill DDR3-1866 CAS 9
Thermaltake level 10 GT case
SeaSonic X750 Gold 750W ATX12V V2.3/EPS 12V V2.91 SLI Ready 80 PLUS GOLD

I don't plan on doing ANY Overclocking. I want to add another GTX 580 for SLI. I'm just wondering whether or not I need to upgrade my PSU, and if so, what would you guys recommend.
 
[citation][nom]gmp23[/nom]I need an expert opinion here guys. Here are my specs:Asrock extreme4 gen3EVGA GTX 580Core i7 2600kG.Skill DDR3-1866 CAS 9Thermaltake level 10 GT case SeaSonic X750 Gold 750W ATX12V V2.3/EPS 12V V2.91 SLI Ready 80 PLUS GOLDI don't plan on doing ANY Overclocking. I want to add another GTX 580 for SLI. I'm just wondering whether or not I need to upgrade my PSU, and if so, what would you guys recommend.[/citation]
No, you do not need to upgrade your power supply. Look at the INPUT power numbers here and multiply by 0.89 (89% efficiency) to see what the system actually needed for power.
 
[citation][nom]gmp23[/nom]I need an expert opinion here guys. Here are my specs:Asrock extreme4 gen3EVGA GTX 580Core i7 2600kG.Skill DDR3-1866 CAS 9Thermaltake level 10 GT case SeaSonic X750 Gold 750W ATX12V V2.3/EPS 12V V2.91 SLI Ready 80 PLUS GOLDI don't plan on doing ANY Overclocking. I want to add another GTX 580 for SLI. I'm just wondering whether or not I need to upgrade my PSU, and if so, what would you guys recommend.[/citation]

While you will probably be OK, NVIDIA does not recommend anything less than a 900 watt supply for 2 x 580. They are certainly being conservative, but that is their recommendation nonetheless. Source: http://www.slizone.com/object/slizone_build_psu.html

On a side note: You have a 2600k. It is begging you to overclock it. That's why there is a 'K' in its name - it has an unlocked multiplier. They O/C so easily and so well that it would be a shame not to. Especially if you are driving 2 580s which will thrive on the O/C.
 
[citation][nom]Crashman[/nom]No, you do not need to upgrade your power supply. Look at the INPUT power numbers here and multiply by 0.89 (89% efficiency) to see what the system actually needed for power.[/citation]

Efficiency really isn't a factor here unless the guy is concerned about his electric bill (And 2 x 580s probably means he isn't worried). A quality 750 watt PS rated at 89% efficiency still outputs a (rated) maximum of 750 watts (to the components) while drawing 843 watts from the wall (actually more since efficiency is reduced above about 80% of rated maximum).
 
Nice article! It's crazy that 580 GTX SLI still can't run Metro 2033 at 2560x1600. Have you tried Crysis 2 DX11? I'm a bit surprised that it's not part of your benchmarks.

Personally, if I was buying an enthusiast machine around this price, I would definitely spend $500-600 more and get a big 8-slot case, 1200W PSU, and dual 590 GTX SLI.

However, since I don't have an immediate need for a new computer, I think that right now it's be a good idea to wait until Q1 2012 or so for IvyBridge (for it's potentially better OC specs due to 22nm), X79, and the new high-end dual-GPU cards. Then the only thing that's left is to hunt for a deal on a 30" screen :)
 
[citation][nom]larkspur[/nom]Efficiency really isn't a factor here unless the guy is concerned about his electric bill (And 2 x 580s probably means he isn't worried). A quality 750 watt PS rated at 89% efficiency still outputs a (rated) maximum of 750 watts (to the components) while drawing 843 watts from the wall (actually more since efficiency is reduced above about 80% of rated maximum).[/citation]Efficiency is a factor if you look at the rest of what I said. I told him to look at the power numbers from this article. It shows 676W from the wall.

IN order to know how much power the system drew from the power supply, you have to know how efficient it was. I believe I remember it being roughly 89% efficient, so that's only 602W. If it's less efficient, say 85%, the system is drawing only 575W.

So you're looking at 575 to 600W to power these parts. He has a 750W power supply from a company that's known to output at least 100% of its rating, therefor he does not need to upgrade his power supply to upgrade from a single GTX 580 (his configuration) to dual GTX 580s (this configuration).

There's a few other items different in his system, but not enough to account for more than a 20W increase and it looks like he has roughly 150W to spare.
 
[citation][nom]Crashman[/nom]Efficiency is a factor if you look at the rest of what I said.[/citation]

I didn't understand you originally so I apologize. I didn't get your reference to the article. All other considerations aside, his PSU will probably do it. Don't forget that in this article they did not measure peak power, they measured power at full load. I'm not purporting to be some PSU expert, but I know that peak power can be a great deal more than load power (definitely in motors, probably not so much in electronics). And I usually don't suggest that others go against NVIDIA's own recommendations.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom