I'm asking this question again, would a dual core E7500 bottleneck a 5870?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

toyota

Lifer
Apr 15, 2001
12,957
1
0
You'll be more than fine with a 5770. Once you start playing at 1680x1050 (or above), there will be no difference between your CPU and an i7.
that is absolutely not true. even at 1920 his cpu would limit a 5770 in the more cpu intensive games. his 2.13 cpu would be like me running my E8500 at 2.0 and I have tested numerous games at 1680 and 1920. a 5770 is just as strong as my 192sp gtx260 and I can tell you the difference in cpu speed is quite large in many games. at just 1280 he would easily give up 40-50% of what a 5770 can do in cpu intensive games. a 5750 512mb would make much more sense at just 1280 if he isnt going to oc that cpu.
 

thebomb

Member
Feb 16, 2010
101
0
0
I plan to game at 1600 x 1200 with all settings maxed full AA and AF and so on.

So will a core 2 duo E7500 bottleneck a 5870 or would I be better off with something lower?

Note: My tastes in gaming varies widely from FPS to RPG to RTS. There is no game I wouldn't play and I love eye candy!

You have a pretty good CPU - it won't bottleneck your 5870 in any noticeable way (<5 FPS).
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
You have a pretty good CPU - it won't bottleneck your 5870 in any noticeable way (<5 FPS).
sorry but it will make much more than 5fps difference. sure it wont be noticeable in most cases but neither will selecting the 5870 over the 5850 while using the E7500 at just 1600.
 

thebomb

Member
Feb 16, 2010
101
0
0
that is absolutely not true. even at 1920 his cpu would limit a 5770 in the more cpu intensive games. his 2.13 cpu would be like me running my E8500 at 2.0 and I have tested numerous games at 1680 and 1920. a 5770 is just as strong as my 192sp gtx260 and I can tell you the difference in cpu speed is quite large in many games. at just 1280 he would easily give up 40-50% of what a 5770 can do in cpu intensive games. a 5750 512mb would make much more sense at just 1280 if he isnt going to oc that cpu.

Look here: http://www.tomshardware.com/reviews/build-balanced-platform,2469-8.html

If we take Crysis as an example then we see that a 5770 (which is comparable to a GTX260) paired with an E6300 or an i7 makes virtually no difference.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Look here: http://www.tomshardware.com/reviews/build-balanced-platform,2469-8.html

If we take Crysis as an example then we see that a 5770 (which is comparable to a GTX260) paired with an E6300 or an i7 makes virtually no difference.
are you kidding me? thats your example? you seriously dont know what you are talking about. a 2.13 Core 2 at just 1280 would bottleneck the shit out of a 5770 in many games. I have tested many modern games at 1680 and 1920 and the difference with a cpu like that is quite large in many games. PCGH also has tone of cpu benchmarks that show a cpu of that speed is a very noticeable bottleneck at 1680 nevermind just 1280.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
OP said he's gaming at 1600x1200 not 1280
reread the comments. you replied to the person with a 2.13 playing at 1280 asking about the 5770 and then I replied to you about that. you got his res confused with the OP that was asking about a 5870.
 

thebomb

Member
Feb 16, 2010
101
0
0
are you kidding me? thats your example? you seriously dont know what you are talking about. a 2.13 Core 2 at just 1280 would bottleneck the shit out of a 5770 in many games. I have tested many modern games at 1680 and 1920 and the difference with a cpu like that is quite large in many games. PCGH also has tone of cpu benchmarks that show a cpu of that speed is a very noticeable bottleneck at 1680 nevermind just 1280.

Do you have any solid numbers for your tests? If we look at the analysis done at tom'shardware http://www.tomshardware.com/reviews/build-balanced-platform,2469-16.html we see that anyone buying anything over an E8400 because they think their CPU is bottlenecking their GPU is wasting money.
 

thebomb

Member
Feb 16, 2010
101
0
0
reread the comments. you replied to the person with a 2.13 playing at 1280 asking about the 5770 and then I replied to you about that. you got his res confused with the OP that was asking about a 5870.

My bad - what I meant to say was if he got a 5770 then he wouldn't need to play at 1280 - he could play at 1680 and eliminate any bottlenecking.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
look at modern game reviews at 1600 and you will see that his cpu is giving up 20% or more even with much slower cards than a 5870. hell I can find benchmarks where he would actually give up 50% such as GTA 4, Prototype and some others. cranking the AA because your cpu is too slow to come close to utilizing your gpu is a stupid excuse to spend 100 bucks more. a 5850 will provide the same playable experience as a 5870 at 1600 with his cpu. the only game there would likely be any advantage would be the STALKER games and Metro 2033 and thats if you run the games at settings the card is barely going to handle anyway.

i write modern game reviews at 1680 and 1920 :p

You can't even link to anything that supports your extreme views except at ridiculously low resolutions.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
i write modern game reviews at 1680 and 1920 :p

You can't even link to anything that supports your extreme views except at ridiculously low resolutions.
the op is at 1600 and you are living in a fantasyland if you think that an e7500 isnt going to limit a 5870. he would get the SAME playable performance in most games with a 5850 because he is already giving up more performance with his cpu than the 5850 gives up to the 5870 with a cpu like that at just 1600.


http://www.legionhardware.com/articles_pages/battlefield_bad_company_2_tuning_guide,7.html

http://www.pcgameshardware.com/aid,...System-Requirements-and-Screenshots/Practice/

http://www.pcgameshardware.com/aid,...ead-of-Core-2-Quad-in-CPU-benchmarks/Reviews/

http://www.pcgameshardware.com/aid,...arked-in-Anno-1404-Dawn-of-Discovery/Reviews/


and most of those are with way slower gpus than a 5870 so the gap would be even larger.
 
Last edited:

thebomb

Member
Feb 16, 2010
101
0
0
the op is at 1600 and you are living in a fantasyland if you think that an e7500 isnt going to limit a 5870. he would get the SAME playable performance in most games with a 5850 because he is already giving up more performance with his cpu than the 5850 gives up to the 5870 with a cpu like that at just 1600.


http://www.legionhardware.com/articles_pages/battlefield_bad_company_2_tuning_guide,7.html

http://www.pcgameshardware.com/aid,...System-Requirements-and-Screenshots/Practice/

http://www.pcgameshardware.com/aid,...ead-of-Core-2-Quad-in-CPU-benchmarks/Reviews/

http://www.pcgameshardware.com/aid,...arked-in-Anno-1404-Dawn-of-Discovery/Reviews/


and most of those are with way slower gpus than a 5870 so the gap would be even larger.

Dawn of Discovery and GTA 4 are pretty CPU intensive games, so yeah, the difference between CPUs will be more pronounced there. I'd agree with you though that it would be best if the OP got a 5850 instead of a 5870.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
the op is at 1600 and you are living in a fantasyland if you think that an e7500 isnt going to limit a 5870. he would get the SAME playable performance in most games with a 5850 because he is already giving up more performance with his cpu than the 5850 gives up to the 5870 with a cpu like that at just 1600.


http://www.legionhardware.com/articles_pages/battlefield_bad_company_2_tuning_guide,7.html

http://www.pcgameshardware.com/aid,...System-Requirements-and-Screenshots/Practice/

http://www.pcgameshardware.com/aid,...ead-of-Core-2-Quad-in-CPU-benchmarks/Reviews/

http://www.pcgameshardware.com/aid,...arked-in-Anno-1404-Dawn-of-Discovery/Reviews/


and most of those are with way slower gpus than a 5870 so the gap would be even larger.

As i said you have *nothing* but low resolution tests to show :p
--Let's see some real testing. Not the 12x10 "No AA.No AF" BS you keep posting over and over in your 2 or 3 cherry picked benchmarks
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Dawn of Discovery and GTA 4 are pretty CPU intensive games, so yeah, the difference between CPUs will be more pronounced there. I'd agree with you though that it would be best if the OP got a 5850 instead of a 5870.
yeah but you see in cpu intensive games is already giving up about 80-100% to a an i7. so in other games the difference would still be like 25% or higher at 1680.

just remember the 5870 is about 20-25% faster than a 5850 when both are ran at stock clocks. he will be giving up that in all games and even way more in the cpu intensive games so he will never really see the advantage the 5870 has over the 5850 at just 1600 with his cpu. plus he could easily oc the 5850 and eliminate any advantage that a 5870 might have with his current setup.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
As i said you have *nothing* but low resolution tests to show :p
--Let's see some real testing. Not the 12x10 "No AA.No AF" BS you keep posting over and over in your 2 or 3 cherry picked benchmarks
READ because those are at 1600 just like the op is.

and I knew you would pull the AA BS out. so do you think that turning on some AA is going to make up for a cpu getting 90-100% more performance? of course not. he will still be giving up at least 25% in most games and even more in modern cpu intensive games at just 1600 with his cpu. and again that wipes out any advantage that a5870 would provide over a 5850.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
READ because those are at 1600 just like the op is.

and I knew you would pull the AA BS out. so do you think that turning on some AA is going to make up for a cpu getting 90-100% more performance? of course not. he will still be giving up at least 25% in most games and even more in modern cpu intensive games at just 1600 with his cpu. and again that wipes out any advantage that a5870 would provide over a 5850.
no they aren't :p

GTA4 is tested at 12x10 no AA; useless testing - quit posting that crap!

Try coming up with something new - something that actually supports your argument.
in all of the *many tests* i run at 16x10 and 19x12 (and the OP is between 16x10 and 19x12), it runs completely contrary to your 3 CHERRY PICKED benches
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
no they aren't :p

GTA4 is tested at 12x10 no AA; useless testing - quit posting that crap!

Try coming up with something new - something that actually supports your argument.
in all of the *many tests* i run at 16x10 and 19x12 (and the OP is between 16x10 and 19x12), it runs completely contrary to your 3 CHERRY PICKED benches
they tested GTA 4 at 1680 too and right under the 1280 test so stop being a jerk and LEARN TO READ before arguing. ALL of those games were at 1600(1680) like the op is using.

the difference between a 5870 and 5850 is 20-25&#37;. are you really going to say that an E7500 does not give up 20-25% to an i7 at 1680 in most games? if so then you need to find a new hobby.
 
Last edited:

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
they tested GTA 4 at 1680 too and right under the 1280 test so stop being a jerk and LEARN TO READ before arguing. ALL of those games were at 1600(1680) like the op is using.

the difference between a 5870 and 5850 is 20-25%. are you really going to say that an E7500 does not give up 20-25% to an i7 at 1680 in most games? if so then you need to find a new hobby.

Maybe you do :p

My hobby has already become my work. i can point to hundreds of benches that show quite the opposite from your 3 cherry-picked benches. if you like, i will be glad to send you a PM with the links.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Maybe you do :p

My hobby has already become my work. i can point to hundreds of benches that show quite the opposite from your 3 cherry-picked benches. if you like, i will be glad to send you a PM with the links.
so I guess you finally realized that page had 1680 results? I see you still cant count though because that is FOUR not three. I have seen you and your buddies site and its pretty funny. he claimed no bottleneck with a Core 2 at 2.0 which was absolute BS.

anyway so again are you really going to still say that an E7500 does not give up 20-25&#37; to an i7 while using a 5870 at 1680 in many games? so why pay 100 bucks more for a card that you will see little to benefit from.

at 1920 I would see your point much better but at 1680 I just dont think its worth it at all. also its a shame he wont oc that cpu because it has a lot more potential with a card like the 5850 or 5870 in many newer games.
 
Last edited:

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
NO ONE in their right mind displays benches at 800x600 to prove a point in gaming

When a cpu is maxing out at 15 frames at 800x600, that's a very relevant benchmark to me. That means its minimums are in the single digits! That means at higher resolution it will be a slideshow regardless of gpu load! Of course no one plays at 800x600, but if a CPU already can't produce even 30 frames steady at this shady resolution, it's not going to be viable at 1600x1200 where a 5870 will easily do 30 frames. In other words the CPU will push 5870 to 15 frames anyways. This is where you can crank to 8AA and get it for "free" because the E6400 cpu is too damn slow.

Take those same Re5 benches and up the filtering

Huh? I am running 8AA/16AF and getting faster frames than Q6600 @ 3.4ghz did with 0AA/16AF at 1680x1050. Where do you want me to up the filtering to ? 8AA in a shooter is very high already and it's 50&#37; faster than a core 2 quad at 3.4ghz regardless.

Sometimes upping the filtering is not an option. Here is an example:

Dirt 2 1680x1050: http://www.pcgameshardware.com/aid,...1-Phenom-doing-well-quad-cores-rule/Practice/

Core i7 3.5ghz = 58 frames MINIMUM (+75% faster)
E6600 2.4ghz = 33 frames minimum (<-- if I wanted 33 frames I would get an Xbox 360!)

You can crank 4AA on a 5870 all you want, you are not going down to 33 frames. For example, my 4890 gets to 38 mins in Dirt 2 Ultra with 4AA. I am not going to accept 38 minimum in a racing game since I can see these dips as slowdowns in the game. So sure you can get 5870 to 33 frames minimum by cranking edge-detect 8AA to the max. That's not real world gaming because you will get a jerky racing game.

As a result, I am going to lower anti-aliasing to 2AA to get smooth gameplay experience (i.e., to get 47 mins) because I don't want jerkiness in gameplay at the tradeoff of better image quality from higher AA in a racing game. So now imagine pairing 4890 with E6600 and using 2AA. The MOST the review got was 33 frames minimum with a 5870 and E6400 and 0AA. I already told you that a Core i7 3.9ghz with a 4890 and 2AA gets 47 frames minimum. That means with a 4890, you are going to be below the 47 mins with 2AA no matter how you slice it because you are at 33 frames without any AA and a way faster videocard!

Do you see the CPU argument now? And why saying oh just crank AA more doesn't always make sense? E6600 is already proving a jerky experience before we even talk about enabling AA.


and my findings completely contradict yours

Maybe you missed GTA4, ArmA2, Resident Evil 5, Dragon Age where there is multi-core scaling and Core i7 architecture smokes C2D. You are telling me a 2.0ghz c2D is going to have similar minimum frames as a Core i7 4.0ghz? Is that why 5870 users are using E7400 processors in their rigs?

The CPU is no where as important as the GPU and mostly any C2D at 3.2 or so GHz will not be bottlenecked by GTX 480 class (HD 4870-X3 TriFire) in most games.

I agree, but that doesn't mean there isn't a performance benefit from a Core i7 4.0ghz in a lot of games as well. In a game like Metro 2033 that rapes a GTX480 and 5870 sure a C2D 3.2ghz is not going to be a bottleneck. But a Core i7 4.0ghz will lay waste to a 3.0ghz C2D in Starcraft 2 for example or GTA4. Obviously if you create an unrealistic scenario of 16AA in SC2 on GTX480, then you are GPU bottlenecked. But who uses 16AA regularly?

Here take a look: http://www.legionhardware.com/articles_pages/battlefield_bad_company_2_tuning_guide,7.html

1680x1050 is 92% of the 1600x1200 resolution. They are using High detail settings and even then Core i7 is getting 2x the minimum frames of X2 550. Which system do you want to game on? You probably would agree that 5870 will take at most a 20% hit going from 0AA to 4AA.

This means that a Core i7 system will get you above 60 frames in BF:BC2 with 4AA, while that Athlon X2 550 already can't give you 60 frames without AA! Think about it, you can't get 60 frames without any AA...so you are already not getting a smooth experience. In other words, you can apply AA with a Core i7 and 5870 in BF:BC2 and still get faster frame rates than a X2 550 without AA using the same videocard. So a cpu is not that important?

Again look here: http://www.legionhardware.com/articles_pages/starcraft_ii_wings_of_liberty_beta_performance,6.html

Core i7 3.7ghz is doubling the minimum framerates of Core i7 2.3ghz. So that means I can get 4AA and still be faster than the 2.3ghz core i7 system without AA.

I think you are missing a crucial point toyota is trying to make, and that is minimum frame rates. Just because you are testing 1680x1050 without AA, doesn't mean a faster CPU is worthless. New videocards are very efficient with AA on.

You can't just look at average frame rates and conclude that a C2D is fast enough because if you crank 8AA on 5870, a 3.2ghz C2D will be as fast as a Core i7 4.0ghz. What if I don't want 33 frames minimum and I want 50 or 60? Not everyone is like BFG who loves to crank AA to the sky. I much prefer smooth mins with Vsync on over AA any day, if that AA takes me down to 20s or 30s.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
Why is this a debate instead of a link to this:
http://www.xbitlabs.com/articles/video/display/radeon-hd5870-cpu-scaling.html

Where they specifically tested CPU scaling on a 5870.

Answer is, yes a E7500 will limit you in at least half the games out there with a 5870 and definitely in RTS/Sandbox style games where AI and pathfinding are heavy loads.
yeah I looked at that but there was no 1680 res used. some games at 1280 didnt make much difference though yet others had 15-20&#37; difference at 1920 with AA. also they didnt test many cpu intensive games on there. it really just depends on what games you want to look at to make your argument. still after looking at that and the other games I play that are much more cpu intensive I know I would pick a 5850 for 1680 with an E7500. a good oc on that 5850 and he will be within 5% of a 5870 even in the most gpu intensive games with his setup.
 
Last edited: