What video card wont be bottlenecked by my 2.4ghz E660

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

AzN

Banned
Nov 26, 2001
4,112
2
0
6.jpg


Same spot with 4xAA 1080P @ 3.36ghz. I'll lower the CPU settings I'll be back.


7.jpg


Here's 1080p with no AA
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
now I can take a screenshot with it at a much higher framerate too. I was taking my shots only when the framerate dipped to its lowest. in other words I can be at 40-45 and then hit the 30s with my cpu at 3.16 or mid 20s with it at 2.5 or so and thats where I was grabbing my screenshots.


see heres one while its not dipping down. right when this lady walked by it dropped without me even moving so it must be the npc in this area.

1080 no AA cpu at 3.16
 
Last edited:

AzN

Banned
Nov 26, 2001
4,112
2
0
I'm not taking screenshots with much higher frame rate. I'm just going to that same spot and taking screen shots as the frame rates is steady and not jumping back and forth from movement. I ran around at these settings. I'm hovering again 30-45fps at these clock settings. @3.36ghz I was hovering 32-50fps at this location.

8.jpg


1080P no AA @ 2.8 ghz.

9.jpg


Here's 1080p with 4xAA
 
Last edited:

AzN

Banned
Nov 26, 2001
4,112
2
0
This is where I was running around in the citadel with my CPU @ 2.8ghz with no AA.

FPS
32
37
38
36
33
31
37
38
40
43
41
44
46
45
43
35
34
33
31
34
36
35
37
35
32
31
33

AVG 37fps. You said this is the worst part of the game where it dips hardest.


FPS
1
20
33
38
35
34
34
39
40
42
42
45
46
48
45
40
34
34
32
32
32
36
36
36
38
32
28
33

AVG 35fps. This is with 4xAA @ 2.8ghz. Never mind the first part where the fps dip 1fps as it's part of fraps doing when recording.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
well I can do that too.

FPS
35
35
40
40
43
32
32
41
44
38
31
40
44
47
44
30
47
47
48
44
47
46
42
44
41
39
36
44
46
46
49
48
50
48
50
45
29
32
44
47
46
48
49
50
53
49
38
47
45
43
42
41
43
42
42

now according to that there are no low framerates but here is the actual benchmark during that time.

Frames, Time (ms), Min, Max, Avg
2369, 55394, 20, 54, 42.766

there are certain spots that make it dip every time almost like its doing a quick load. it does this on 7 and Vista and even does it on my other pc with an 8600gt running medium settings at 1024x768.
 

AzN

Banned
Nov 26, 2001
4,112
2
0
And what CPU settings would that be? Strange your lowest fps is 29fps but your min says 20.

noaa.png


Same thing with no AA @ 3.36ghz

So all in all 1fps difference between 2.8ghz to 3.36ghz. Although the bench was a short one.

http://www.pcgameshardware.de/aid,645004/PCGH-Performance-Review-Mass-Effect/Mass-Effect/Test/

ME_CPU.PNG


According to these benches even a 1.86ghz E6320 can play the game fine but 2.33ghz would be better. Quad makes minimal impact in this game. it's more like cache difference to show discrepancies as core 2 duo has 4meg cache and Quad has 2sets of 4meg cache between all the cores.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
And what CPU settings would that be? Strange your lowest fps is 29fps but your min says 20.
that was actually at 3.16 because I found the one spot that dipped down the most so i kept running by it to see how low I could make it dip. at 2.5 I hit 16fps right there during the bench.

I know thats why I posted both the readout and the min/max/average. I didnt see 20fps even come on the screen either.

I just went and set max prerendered frames to 0 in the CP and the min is now 25fps in the benchmark no mater how many times I run by that spot. also the average went up to 47 and now hit 60 for a max quite a bit. that setting certainly helped because I am repeating the same loop over and over.

I have played the game for about 12 hours or so and that spot in the presidium is the only place I have seen that type of behavior and I really have to have fraps on to see how bad because its not awful feeling. I have checked other parts using fraps and it just doest seem to do that anywhere else that I have checked so far.
 
Last edited:

AzN

Banned
Nov 26, 2001
4,112
2
0
that was actually at 3.16 because I found the one spot that dipped down the most so i kept running by it to see how low I could make it dip. at 2.5 I hit 16fps right there during the bench.

So you are intentionally looking for dips? A spike of fps dipping can be caused by anything like bad current, hard drive seek, bad coding. It is no way an indication of your CPU was the main factor.

I know thats why I posted both the readout and the min/max/average. I didnt see 20fps even come on the screen either.

Perhaps it's another set of benches you got mixed up with.

I just went and set max prerendered frames to 0 in the CP and the min is now 25fps in the benchmark no mater how many times I run by that spot. also the average went up to 47 and now hit 60 for a max quite a bit. that setting certainly helped because I am repeating the same loop over and over.

I have played the game for about 12 hours or so and that spot in the presidium is the only place I have seen that type of behavior and I really have to have fraps on to see how bad because its not awful feeling. I have checked other parts using fraps and it just doest seem to do that anywhere else that I have checked so far.

So you are intentionally trying to make the fps dip below 25fps? How ironic that you do this in your spare time to tell yourself and others that you need a faster CPU. Just play the game and stop worrying about your fps. You don't need fraps to tell you that you can't play a game.

Mass effect is a 3rd person shooter role playing game. You should be fine with 35fps average.

If you haven't looked at mass effect 2 thread I posted with benches you should head over there. You get linear scaling with faster GPU. 5770 getting 36fps while 5870 is getting 66fps @ 1920x1200 4xAA. that's a huge difference. Something your E8400 @ 4ghz can't muster vs my E6300 @ 3.36ghz.
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
So you are intentionally looking for dips? A spike of fps dipping can be caused by anything like bad current, hard drive seek, bad coding. It is no way an indication of your CPU was the main factor.



Perhaps it's another set of benches you got mixed up with.



So you are intentionally trying to make the fps dip below 25fps? How ironic that you do this in your spare time to tell yourself and others that you need a faster CPU. Just play the game and stop worrying about your fps. You don't need fraps to tell you that you can't play a game.

Mass effect is a 3rd person shooter role playing game. You should be fine with 35fps average.

I was looking for the lowest dips in the SAME spots with DIFFERENT cpu speeds. I was also looking at average fps too but the minimum was what I was most concerned with.

no I didnt mix them up because they are right next to each other with the time and date right there.

I already explained that I came across this when I had left my cpu at 2.4. I just ran around a small area and knew something was wrong so I turned on fraps to check. I then looked at where the dips were and checked it again after I put the oc back on my cpu. it made a huge difference there so I never bothered to check anywhere else at that time.

if I would have started up the game in another area I probably would not have even noticed. but remember that I never even checked the framerate during any action though so maybe that a faster cpu would help there too.

overall the faster cpu certainly helps with the averages and minimums around the presidium even in that one crazy part. I will certainly put my cpu on at least its stock speed but most other parts of the game seem to be fine with it around 2.4 or 2.5. I am pegged at nearly 100% the whole time so a multi tasker thats crazy enough to do other things while gaming will likely want a quad.
 
Last edited:

evolucion8

Platinum Member
Jun 17, 2005
2,867
3
81
toyota, you are taking the CPU bottleneck too serious, in every thread that somebody asks for advice for a card with a midrange CPU, you always posts the same thing, that he will be severely CPU bottlenecked and stuff. The E6400 isn't the best CPU right now, but at least it will be enough for even Mass Effect 2, I doubt that a 25fps spike will make a difference in playability if the game has a higher average, even the Xbox 360 and PS3 has the same issues in some games and yet I don't see nobody complaining of slow or unplayable Fps.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
toyota, you are taking the CPU bottleneck too serious, in every thread that somebody asks for advice for a card with a midrange CPU, you always posts the same thing, that he will be severely CPU bottlenecked and stuff. The E6400 isn't the best CPU right now, but at least it will be enough for even Mass Effect 2, I doubt that a 25fps spike will make a difference in playability if the game has a higher average, even the Xbox 360 and PS3 has the same issues in some games and yet I don't see nobody complaining of slow or unplayable Fps.
no where in this thread did I say he would be severely bottlenecked because he wont.
 

AzN

Banned
Nov 26, 2001
4,112
2
0
overall the faster cpu certainly helps with the averages and minimums around the presidium even in that one crazy part. I will certainly put my cpu on at least its stock speed but most other parts of the game seem to be fine with it around 2.4 or 2.5. I am pegged at nearly 100% the whole time so a multi tasker thats crazy enough to do other things while gaming will likely want a quad.

Yes it does but GPU makes the biggest difference in 99% of games out there. Your crazy part is the result of some bad current spike it seems. A faster CPU would show less of that spike in gaming as the dips are more severe with slower CPU. A slower clocked quad will not cure this if anything it will probably be worse. You are really nitpicking at that point however.

Your CPU is pegged @ 100% because the games are dual core optimized. Did you really expect anything less with your dual core or were you unaware of this 100% CPU usage when gaming? I don't know why you got your system recently and got a E8500. I know I would Qxx00 over E8500 any day. Just because you are having doubts about it doesn't mean all dual core users are doubting as I got my CPU long long time ago. Although I would like a quad it just isn't worth it right now for me at least. 4 or 5 games that is quad utilized isn't end of the world nor do I have trouble playing those games because I have a dual core. I just accept the performance or upgrade when that day comes.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
Yes it does but GPU makes the biggest difference in 99% of games out there. Your crazy part is the result of some bad current spike it seems. A faster CPU would show less of that spike in gaming as the dips are more severe with slower CPU. A slower clocked quad will not cure this if anything it will probably be worse. You are really nitpicking at that point however.

Your CPU is pegged @ 100% because the games are dual core optimized. Did you really expect anything less with your dual core or were you unaware of this 100% CPU usage when gaming? I don't know why you got your system recently and got a E8500. I know I would Qxx00 over E8500 any day. Just because you are having doubts about it doesn't mean all dual core users are doubting as I got my CPU long long time ago. Although I would like a quad it just isn't worth it right now for me at least. 4 or 5 games that is quad utilized isn't end of the world nor do I have trouble playing those games because I have a dual core. I just accept the performance or upgrade when that day comes.
very few games get to 100% on the cpu so it is rare from what i have seen. I built this system 16 months ago and the quads were very expensive except for the Q6600. I didnt want that cpu because I knew I didnt want to oc at first and very few games were making use of more than 2 threads. really if you look at benchmarks the E8500 has been the better overall cpu at realistic settings. sure if I was to build a pc now it would be i5 750. I havent really made the jump because I am like you in that it isnt worth it just for a few games that I can push through with my current cpu.
 
Last edited:

AzN

Banned
Nov 26, 2001
4,112
2
0
I've been PC gaming since late 80's. My first computer a 286 whopping 8mhz. lol.. Long as I remember CPU was pegged 100% in games to give max frame rates. I remember some games were too fast on a fast computer. Computer had to run on slower speeds to slow the games down. It's just that recently multiple cores is where the CPU wasn't pegged 100% when dealing with single threaded games. I would have still gotten a q6600 over E8500. Get a good Q6600 and you would be @ 3.6ghz over a 4ghz dual core. You knew multi-threaded applications were already here but you didn't think games would be for how long you would be holding your system? Doesn't make much sense.

What you can do is sell your E8500 as it does retain some value and get a Q9550 at microcenter. Shouldn't be no more than $60-$70 for the upgrade. As for me it's more financially sound or performance point of view to get a i5-750 than Q9550. It would literally cost 20 more for i5-750 but it still isn't worth it for the performance it gives over what I have now other than GTA4. What I did though is buy E5200 to hope for 4ghz and hold on to my core system long as I can which cost me $10 total after I sell my CPU. My new CPU should be here tomorrow. :)
 
Last edited:

toyota

Lifer
Apr 15, 2001
12,957
1
0
I've been PC gaming since late 80's. My first computer a 286 whopping 8mhz. lol.. Long as I remember CPU was pegged 100% in games to give max frame rates. I remember some games were too fast on a fast computer. Computer had to run on slower speeds to slow the games down. It's just that recently multiple cores is where the CPU wasn't pegged 100% when dealing with single threaded games. I would have still gotten a q6600 over E8500. Get a good Q6600 and you would be @ 3.6ghz over a 4ghz dual core. You knew multi-threaded applications were already here but you didn't think games would be for how long you would be holding your system? Doesn't make much sense.

What you can do is sell your E8500 as it does retain some value and get a Q9550 at microcenter. Shouldn't be no more than $60-$70 for the upgrade. As for me it's more financially sound or performance point of view to get a i5-750 than Q9550. It would literally cost 20 more for i5-750 but it still isn't worth it for the performance it gives over what I have now other than GTA4. What I did though is buy E5200 to hope for 4ghz and hold on to my core system long as I can which cost me $10 total after I sell my CPU. My new CPU should be here tomorrow. :)
well you can just take a look at the task manager and see that plenty of games are not using 100% cpu. strangely there are games that are faster on a quad that dont even use more than 80% of dual core. really every game is different but it only takes a second to check it out.

I dont live close to a microcenter so thats not really an option. if I did I would probably just get the deal they had on the i5 750 since by the time I sold my other stuff it would not be much more than going with the Q9550 anyway. oh well, good luck on the E5200.
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
I'm trying to play new to newish games at med high settings at 1680x1050 for under 150

Try looking for a used or new HD4870 1 GB on closeout. I bought a XFX (New in the box) from someone here at forums for only $120 shipped.

Originally I didn't want a card with high idle power consumption like HD4xxxx, but some of these deals are too hard to pass up at the moment. In comparison, A $129.99 HD5750 was around $150 shipped including tax at Newegg.
 

AzN

Banned
Nov 26, 2001
4,112
2
0
well you can just take a look at the task manager and see that plenty of games are not using 100% cpu. strangely there are games that are faster on a quad that dont even use more than 80% of dual core.

Like single core optimized game on a dual core. Quad is the the same way. 80% dual core because it's being spread out between cores.

Now when you get a dual core optimized game your cpu gets pegged 100% and same will be with quad once these games get here.

Like I'll play battlefield 2 that's single core optimzed and it's only 60% on 1 core 40% on the other.
 

BFG10K

Lifer
Aug 14, 2000
22,709
3,007
126
I'm on an i5 750 now and surprise surprise, there's almost no performance gain over my E6850 in my games because my GTX285 bottlenecks me by almost 100% (an article is coming shortly). I especially know this because my i5 750 always turbo-boosts all four cores to 3.2 GHz.

I'm also not seeing any smoother multi-tasking, faster game load times, or less hitching over a dual-core that some people with quad-cores reported.

Putting money into the GPU is the better investement because you'll get the most gains in gaming, by far. I'd take an E6600 with an 5870 over an i7 4 @ GHz with a 5770. The former will game far better than the latter, no contest.
 

AzN

Banned
Nov 26, 2001
4,112
2
0
Ha you are on a 30" though. LOL.

At slightly lower resolution you would probably see a little bit of performance gain with your new i750 but yeah that's why I haven't upgraded yet. So little performance for amount paid isn't worth it. At least not until more quad optimized games surface.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
I'm on an i5 750 now and surprise surprise, there's almost no performance gain over my E6850 in my games because my GTX285 bottlenecks me by almost 100% (an article is coming shortly). I especially know this because my i5 750 always turbo-boosts all four cores to 3.2 GHz.

I'm also not seeing any smoother multi-tasking, faster game load times, or less hitching over a dual-core that some people with quad-cores reported.

Putting money into the GPU is the better investement because you'll get the most gains in gaming, by far. I'd take an E6600 with an 5870 over an i7 4 @ GHz with a 5770. The former will game far better than the latter, no contest.
overall sure but if you compare the min framerates you would get some pretty bad results with a 2.4 E6600 compared to the i7 if both were using the 5870. Ghostbusters, Red Faction Guerrilla, and GTA 4 are some examples that would be significantly held back and even gameplay would be affected. now if that E6600 was overclocked then the i7 750 would offer very little value over it but it would still be faster.
 
Last edited:

AzN

Banned
Nov 26, 2001
4,112
2
0
I think he was comparing it to his 6850 over i7 stock for stock. Which is understandable as there's really not many games that are quad optimized. Those few games you mentioned yea. I don't know if any of those games you mention are quad optimized except for GTA4 but with his resolution he would be GPU limited anyway in most situations.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
I think he was comparing it to his 6850 over i7 stock for stock. Which is understandable as there's really not many games that are quad optimized. Those few games you mentioned yea. I don't know if any of those games you mention are quad optimized except for GTA4 but with his resolution he would be GPU limited anyway in most situations.


yeah between an stock E6850 and stock i5 750 there probably is very little difference in most situations with a single 5870. Ghostbusters and Red Faction Guerrilla are very much optimized for quads though. its the physics in both of those games that is a killer at times. now those with weaker cpus can turn down some of the settings but maxed out both of those games need a fast quad for the best playability during heavy scenes.
 

xboxist

Diamond Member
Jun 25, 2002
3,017
1
81
OP, is there any particular reason why your E6600 is OC'ed? 3.2GHz is relatively easy to get and 3.6GHz is achievable with a litte finessing.
As far as bottlenecking, I've been happily using a 5850 with my E6600 (@ 3.6GHz on a 965P-DS3 Rev 1.0).

Would you mind sharing your OC settings? I have that CPU, board (also rev 1.0) and would be nice to have a frame of reference to start when I try to OC mine. Also, what cooler?

Thanks much!