All about Build Your Own
 Latest Build Your Own articles
All Build Your Own articles

Newsletters


  • Ask your question about IT issues
  • Post

Partners

The Games selection

violent : More Mindless Violence Basic shooting game, but still so powerful! Use the mouse to take aim and shoot at the little beasties before they get to you. Use Space to reload....
action : Yoyo the Star Yoyo is a young girl who recently graduated and dreams to become a movie star (don't we all). You'll have to guide her on the path to stardom,...
Ads

Sponsored links

Benchmark Results: 3D Games

Previous Next
12:40 AM - 10/31/2008 by Thomas Soderstrom

Crysis has always put a strain on the latest hardware when visual quality or video resolution is increased, and our $4,500 PC was the only one that could smoothly play at resolutions in excess of 1920x1200 pixels. We narrowed our results to more realistic resolutions, but found that our highest-priced system was required for smooth game play even at an ordinary 1680x1050 resolution.

In Supreme Commander: Forged Alliance, our $1,500 PC survives to 1920x1200 pixels resolution with anti-aliasing turned off. Only our best machine could perform smoothly at high resolutions with anti-aliasing on, and the cheapest system couldn’t even cope at our lowest settings. However, some choppiness and/or lower details may be acceptable to many RTS players.

Unreal Tournament 3 players will be happy to see the $500 PC showing excellent performance at all settings.

Our second RTS game is perfectly playable by the $1,500 PC, though once again some choppiness in the $500 system will be acceptable to many players of this gaming genre.

Talkback
Slomo4shO 10/31/2008 6:55 AM
Hide
-3+

Looking forward to the side by side Intel vs AMD build-offs for the $500 bracket(hopefully you start doing this)

Also, in future write ups, can you please provide power consumption charts?

cangelini 10/31/2008 7:17 AM
Hide
-4+

Slomo4shO :
Looking forward to the side by side Intel vs AMD build-offs for the $500 bracket(hopefully you start doing this)Also, in future write ups, can you please provide power consumption charts?



Slo,
I'll toss the idea around with our authors. Don't see it being a problem--just have to get everyone outfit with the same equipment and methodology. Thanks for the suggestions!

dangerous_23 10/31/2008 7:48 AM
Hide
-5+

what about a $750 or $1000 machine - is this not a more realistic price point for most people?

zodiacfml 10/31/2008 8:43 AM
Hide
-0+

based on these systems, a person should have an idea to build his 750 or $1000 dollar machines.
i like most the $500 machine,the best value,simplicity and efficiency,
only upgrading it to a quad core because i encode HD videos to H264 while surfing the net or watching a video.
only games crysis and supreme c. required more than 3Ghz so a quad is not a big loss to duals in gaming.

dangerous_23 10/31/2008 9:38 AM
Hide
-0+

id very much like to see the benchmarks from a machine costing somewhere between the $500 and $1500 builds
i bet it would hit the sweet spot!

boostercorp 10/31/2008 9:40 AM
Hide
-0+

hi tom's could you tell me where you got the

Quote :2x 20 GB Patriot Viper PC2-6400 CAS 4
ram ?
I could use some more then my 8gb i've got now. ;) :p

cangelini 10/31/2008 10:48 AM
Hide
-0+

boostercorp :
hi tom's could you tell me where you got the



You missed it! That was our limited-time $500 super-computer build. ;-)

Slomo4shO 10/31/2008 10:54 AM
Hide
-0+

And you stuck it in the $1500 machine? I knew you were holding out on that $500 build! :P

dirtmountain 10/31/2008 11:11 AM
Hide
-1+

I really enjoy these System Builder Marathons, yeah i'd pick some different components and price brackets, but great stuff anyways. Going with Newegg as a sponsor is a great idea and i sure hope you continue it in the future for other SBM articles. Having a quality retailer like Newegg supply easily available components should really cut down on the logistics of doing these builds and hopefully they can come a bit more often. I'd like to see other SBM brackets e.g. $600 AMD vs. Intel build. Budget quad core builds - AMD 9950 vs. Q6600. Bracket $750 $1,500 $3,000 builds. How about a reader suggested build? Post a bracket, have folks post suggested builds and pick one or a combination of ideas and have your guys put one together. Any way, great job by the staff, good information, brilliant sponsorship by Newegg and a hell of a lot of fun to read, good job.

neiroatopelcc 10/31/2008 12:20 PM
Hide
-1+

I'd suggest you upgrade your next $500 build to $650! or lower the $1500 to $1100 or so.
According to a newsletter I received 4 days ago from one of the leading danish retailers, A basic pc costs $350, a basic gaming pc costs $600, and a 'good' (in their terms) gaming system costs $1000 - they're not selling any base pc with better graphics than an 4850, but it still means that they consider the $1000 to be the mainstream, and $600 to be lowend. Ofcourse the actual component price will be lower, but it's not going to be 25% lower.

jtt283 10/31/2008 12:25 PM
Hide
-0+

I like these too, especially the bottom-dollar build which is probably easiest to tweak to improve a specific area.

luciiacob 10/31/2008 12:49 PM
Hide
-0+

Nice! Congrats on these four articles!

I read all four of them and I must say I've learned a lot. Although the builds were clearly game-orientated the benchmarks covered a wide range of classical applications used today by most of us, besides games. The synthetic benchmark though don't really help much and, personally, I'd prefer if they would've been replaced with power consumption charts.

Also, in future System Build Marathons, why not build 2 systems in the mid-range price (1200$-1500$), one of which would be game-orientated (dual-core processor+high end graphics card) and the other application-orientated (> 4 GHz overclocked new quad-core processor+mid range graphics card) ? I'm not suggesting to do the same thing for more than one price range because the work would be collosal.

Keep up the good work!

neiroatopelcc 10/31/2008 12:58 PM
Hide
-0+

luciiacob :
Nice! Congrats on these four articles!I read all four of them and I must say I've learned a lot. Although the builds were clearly game-orientated the benchmarks covered a wide range of classical applications used today by most of us, besides games. The synthetic benchmark though don't really help much and, personally, I'd prefer if they would've been replaced with power consumption charts.Also, in future System Build Marathons, why not build 2 systems in the mid-range price (1200$-1500$), one of which would be game-orientated (dual-core processor+high end graphics card) and the other application-orientated (> 4 GHz overclocked new quad-core processor+mid range graphics card) ? I'm not suggesting to do the same thing for more than one price range because the work would be collosal.Keep up the good work!


Adding power consumption fine - relevant for some people. But don't cut the synthetics ! They may not be relevant for building a pc, but they are relevant for seeing how the old rig at home stacks up. So you know if your 1½ year old $800 rig is still adequate, or if you should build a new $500 rig.

tomasf 10/31/2008 1:01 PM
Hide
-0+

I´d like to see some other game benchmarks besides crysis. lets face it, not every one likes that game an there are some newer games that don´t need that kind of power. some racing games wold be nice.

BTW. great systems. the $500 its awesome

luciiacob 10/31/2008 1:22 PM
Hide
-0+

neiroatopelcc :



True! Synthetics do help when it comes to that comparison. I didn't think of it. Sorry.

On the other hand, I honestly believe that no 800$ 1½ years old rig can stand up to a today's 500$ rig (so...no need for a synthetic benchmark when it comes to this decision :) ). That's because prices drop at half after 1 year and today's overclocking capabilities are way better that 2 years go. Also, bear in mind that applications and games also need newer technologies to be supported by the hardware, not just higher frequencies and capacities.

Also, are you really going to spend 500$ on a new rig if you've spent 800$ 1½ years ago? Since my first Pentium 1 PC, about 10 years ago, I've always doubled the amount of money that I spend on a new PC every two or three years.

Rick_Criswell 10/31/2008 1:31 PM
Hide
-0+

Go Crashman !!!! It is good to see someone doing these types of articles. Instead of the fluff advertising articles we normally get.

neiroatopelcc 10/31/2008 2:00 PM
Hide
-0+

luciiacob :
True! Synthetics do help when it comes to that comparison. I didn't think of it. Sorry.

On the other hand, I honestly believe that no 800$ 1½ years old rig can stand up to a today's 500$ rig (so...no need for a synthetic benchmark when it comes to this decision ). That's because prices drop at half after 1 year and today's overclocking capabilities are way better that 2 years go. Also, bear in mind that applications and games also need newer technologies to be supported by the hardware, not just higher frequencies and capacities.

Also, are you really going to spend 500$ on a new rig if you've spent 800$ 1½ years ago? Since my first Pentium 1 PC, about 10 years ago, I've always doubled the amount of money that I spend on a new PC every two or three years.


Just under years ago I built an intel system for a friend. It had a ga-965p-ds4 board and an e6400 cpu. That cpu still today runs at 3,4ghz - that's more than the e2180 used here. More cache on the new models doesn't make that much of a difference really, so the old $800 mashine would stand up, if just the graphics would be upgraded. Now his mashine sported an all new 8800gtx back then, but we could've gone with a 7950 back then and compared that to the 8800gt now. Anyhow, my $800 was more of a theoretical number than a real one. Just meant to say that synthetics are key for comparison between their build and ours.

As for your upgrade strategy - sounds sensible, but I've only started from scratch twice in my life. When going from my p200 to an athlon 1000 back in 2001 or so, and when I changed from my athlon 2100 to a northwood 2.8 - and I only started over because poor inhouse electrics had broken pretty much all hardware in the old one.

sublifer 10/31/2008 2:02 PM
Hide
-4+

You left out the stock scores in the charts :( That was a valuable piece as a lot of people like to see what they can get for their money by overclocking and seeing if it makes it to that next tier. e.g.
($1500pc overclocked) = ($4500pc stock)

Is there anyway you can add it to your charts there?

Thanks!

Slomo4shO 10/31/2008 2:03 PM
Hide
-0+

Another topic for a article: Eco-friendly system. I want to see how low powered a system can be without loosing much performance.

Pei-chen 10/31/2008 2:21 PM
Hide
-1+

neiroatopelcc :
Just under years ago I built an intel system for a friend. It had a ga-965p-ds4 board and an e6400 cpu. That cpu still today runs at 3,4ghz - that's more than the e2180 used here. More cache on the new models doesn't make that much of a difference really, so the old $800 mashine would stand up, if just the graphics would be upgraded. Now his mashine sported an all new 8800gtx back then, but we could've gone with a 7950 back then and compared that to the 8800gt now. Anyhow, my $800 was more of a theoretical number than a real one. Just meant to say that synthetics are key for comparison between their build and ours.

As for your upgrade strategy - sounds sensible, but I've only started from scratch twice in my life. When going from my p200 to an athlon 1000 back in 2001 or so, and when I changed from my athlon 2100 to a northwood 2.8 - and I only started over because poor inhouse electrics had broken pretty much all hardware in the old one.


Wow, freaky. I built my current system in Q1 2007 with an E6400, P965, 7950GT and overclock to 3.4GHz @ 1.45v but step back to 2.56GHz @ 1.1v. Two weeks ago I replaced the 7950 with a 9800GT (8800GT refresh) and gaming performance more than doubled.

And yes, you’re right. My $1000 system still beats the value build.


Comments are closed on this page.

Sponsored links