System Builder Marathon, March 2011: $1000 Enthusiast 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]Crashman[/nom]This has happened before: Right after Tom's Hardware gave the cooler an award for it's $30 value, the price went up to $40. That put it against the better-performing Scythe Mugen 2 Rev. B[/citation]

Actually it went up into the $50's, albeit briefly, while I went to MicroCenter and bought one on special for $19.99. A little later I bought a Scythe Mugen 2 Rev. B, also not from NewEgg. I now view NE as I do Sears- never buy unless marked down massively on sale or closeout.
 
[citation][nom]luis1[/nom]Just pulled the trigger on this system.. trumps your 1K system..For 820.00 shipped..http://www.ibuypower.com/Store/Special-B/w/62031[/citation]

With a Phenom II 840?

In your dreams, bro... 😀
 
[citation][nom]flyinfinni[/nom]"We suspect the gaming difference would be even wider if these systems were using identical graphics cards." WHAT? This makes no sense. If they were using identical graphics cards, they would have been much CLOSER in gaming performance.[/citation]

Ah, but this is GeForce vs. Radeon and we're subject to game engine preferences.

Using the same cards would allow the 2500K a little more room to up the ante: I suspect some of these titles work a little better with GeForces.
 


I understand the game engine preferences, but a 6950 beats a 470 hands down either with an i5 760 or and i5 2500k. I would bet you that even with the nvidia bias of some of these engines putting the 470 with the 2500k would still be slower than the stock 6950. In fact- I bet if you swapped the GPUs and put the 6950 with the 760 and the 470 with the 2500k, the 6950 would STILL win.

Also- I bet you'd find that the 760 is not held back by the 470, but in a majority of games the 470 will be the limiting factor.
 


They don't work harder, they work about the same, but really, its the Graphics cards are not working as hard at all, which make the CPU the limiting factor. The low res frames are so easy for the graphics card that they can just pump them out like crazy and it becomes difficult for the CPU to keep up with getting the data to the GPU's to render.
 

Ok then that makes sense :) So the other way around when the GPU is having to do more work and less FPS the CPU is able to keep up with the 40FPS better than the 120 FPS right?
 


In general 🙂
 

So I know this is very general but lets say I crossfire 6950's. I know my CPU is not as good as say a 2600k sandybridge. So Lets say im pulling 50-60 FPS on crysis/Metro 2033 with max settings with my CPU and crossfired 6950's. What would be a good guess as to what FPS improvements I might get swapping out my cpu/mobo to that 2600k sandybridge I was just talking about :) lol I would not spend $600ish for anything less than around 15 FPS you know what I mean.
 
OMG, that case is so fugly that I'm surprised it's not too embarrassed to come out of the cardboard box it shipped in!

If I win this PC in the contest, I will be sure to let everyone know how well the parts fit into a different case.
 
[citation][nom]flyinfinni[/nom]I understand the game engine preferences, but a 6950 beats a 470 hands down either with an i5 760 or and i5 2500k.[/citation]

[edit] OK, I see what you're saying--I don't necessarily agree though. Certain games like StarCraft 2 may show the 2500K take a strong lead.

Without performing that specific test (760 vs 2500K) I have my doubts.
 


Yes, some prefer one architecture or the other, however, even in games that prefer the Nvidia architecture, the 6950 STILL tends to be faster. Here are some comparisons of the 6950 to the 470 in the particular games you reviewed-
From the Toms review of the 6950 - Dirt 2 and Just cause 2:
http://www.tomshardware.com/reviews/radeon-hd-6970-radeon-hd-6950-cayman,2818-13.html
http://www.tomshardware.com/reviews/radeon-hd-6970-radeon-hd-6950-cayman,2818-14.html

From Hardware Canucks - F1 2010:
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/38899-amd-radeon-hd-6970-hd-6950-review-18.html

From Techpowerup - Crysis and Stalker (clearsky not COP, but still)
http://www.techpowerup.com/reviews/HIS/Radeon_HD_6950/12.html
http://www.techpowerup.com/reviews/HIS/Radeon_HD_6950/20.html

From Guru3d - COD MW2
http://www.guru3d.com/article/radeon-6950-6970-review/15

The 6950 beats the 470 in ALL of these (minus a couple of low res virtual ties). How is it possible then that you could get a LARGER difference in the gaming benchmarks if you used the same graphics card in both situations? A 470 matched with a 2500K is going to be slower in your benchmarks than a 6950 matched with a 2500k. A 6950 matched with a 760 is going to be faster than a 470 with a 760. In any case where you have the same card with both CPU setups, you are going to have CLOSER results than you did with the 470/760 vs 6950/2500k. I don't understand how you can even argue that.

[edit] in reply to your edit ;-) I agree that certain games such as starcraft2 which are much more CPU hungry that would be the case, but with all the game benchmarks you actually run in this article, the statement made is ABSOLUTELY not true.
 
[citation][nom]luis1[/nom]Just pulled the trigger on this system.. trumps your 1K system..For 820.00 shipped..http://www.ibuypower.com/Store/Special-B/w/62031[/citation]

bwahahaha, a phenom x4 + 5770
 
[citation][nom]flyinfinni[/nom]Yes, some prefer one architecture or the other, however, even in games that prefer the Nvidia architecture, the 6950 STILL tends to be faster. Here are some comparisons of the 6950 to the 470 in the particular games you reviewed- From the Toms review of the 6950 - Dirt 2 and Just cause 2:http://www.tomshardware.com/review [...] 18-13.htmlhttp://www.tomshardware.com/review [...] 18-14.htmlFrom Hardware Canucks - F1 2010:http://www.hardwarecanucks.com/for [...] ew-18.htmlFrom Techpowerup - Crysis and Stalker (clearsky not COP, but still)http://www.techpowerup.com/reviews [...] 50/12.htmlhttp://www.techpowerup.com/reviews [...] 50/20.htmlFrom Guru3d - COD MW2http://www.guru3d.com/article/rade [...] -review/15The 6950 beats the 470 in ALL of these (minus a couple of low res virtual ties). How is it possible then that you could get a LARGER difference in the gaming benchmarks if you used the same graphics card in both situations? A 470 matched with a 2500K is going to be slower in your benchmarks than a 6950 matched with a 2500k. A 6950 matched with a 760 is going to be faster than a 470 with a 760. In any case where you have the same card with both CPU setups, you are going to have CLOSER results than you did with the 470/760 vs 6950/2500k. I don't understand how you can even argue that.[edit] in reply to your edit ;-) I agree that certain games such as starcraft2 which are much more CPU hungry that would be the case, but with all the game benchmarks you actually run in this article, the statement made is ABSOLUTELY not true.[/citation]

why use a 570 as comparison? the 570 isn't a good card...
 

I guess you would use that as the card with the closest price in Nvidias line up to the price of the 6950 they could go 580 but its ALOT more cash and SLI 460/560 are more as well
 
[citation][nom]luis1[/nom]Just pulled the trigger on this system.. trumps your 1K system..For 820.00 shipped..http://www.ibuypower.com/Store/Special-B/w/62031[/citation]
Not sure where you get your performance data but that system does not trump the build from this article. Also you bought a gaming rig instead of building one yourself? GTFO and take your Fail with you!
 
[citation][nom]flyinfinni[/nom] but with all the game benchmarks you actually run in this article, the statement made is ABSOLUTELY not true.[/citation]

I's neither true nor untrue; it's unproven.

But I understand you disagree. :)
 



Give me the hardware and software and I'll run the tests for you 🙂
In any case though, the statement "We suspect the gaming difference would be even wider if these systems were using identical graphics cards" necessarily implies one of 2 things- that the putting a 470 into the 2500k system would be FASTER than the 6950 in the 2500k system, OR that putting a 6950 into the 760 system would be SLOWER than the 470 in the 760 system. If the 6950 is faster than the 470 in each of the benchmarks when run using an identical system (as shown by the links I posted), then the situations required for that statement to be true are impossible. This is simple logic. I don't know how you can call it neither true nor untrue. You may not have personally run the tests, but I believe this is sufficient data to prove that statement to be untrue.
 

I think the other build they referance had 460's SLId I could be worng. However 2 460's would cost more and give less room for improvement in the future.
 
4.4ghz is a pathetic overclock if you are building a system to do well in benchmarks, try the latest BIOS (if you did, try an older BIOS) to see if that will fix the throttling, and also if you had tried a different case you may get better temps, make sure you're using good TIM as well as making sure you're seating is perfect you should be able to hit 4.6ghz even if you got a bad chip
 

So how much better would the benches have gone with the extra 200mhz? 2-3 FPS max?
 
Pointless to wait for an AMD build we would have to wait until bulldozer based chips and Liano come out to have a proper comparison right now intel has the advantage but bulldozer might be the game changer...the landscape has definitely changed in the computer industry...and i like it
 
[citation][nom]JohnMD1022[/nom]I have to learn to spell 'case'... lolSeagate drive? No way... pure junk.[/citation]Its a Samsung drive for starters... *sigh*
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom