CPU Bottleneck? Q9550

P

PornFlake

Limp Gawd
Joined
Feb 24, 2005
Messages
270
I have read a few discussions on this and I still don’t think I have a clear answer.
In my situation, what real world benefits would I see going from a Q9550 clocked at 3.6 to a 2500k?

I am wondering if the Q9550 is bottlenecking my GTX460 1GB @ 900/2000, 1920x1200. with as much eye candy turned up that results in a suitable frame rate

I tend to play with as much eye candy turned up that results in a suitable frame rate

I do mostly FPS gaming, COD, BF2, Crysis, etc.

-Flake
 
I went from a Q9550 at 4.0 to a 2600K at 4.6 (with a GTX 480) and didn't notice any astronomical gains in games. About 10% or so, if I remember correctly (I did some tests when I first upgraded).

It is quite a bit faster at video encoding though.
 
I switched from a 9550 to an I7 950. The 9550 was @ 3.8GHz. The 950 at stock clocks (3.06GHz) ran about the same as the OC'd 9550. When I OC'd it to 4.2GHz, CPU intensive games shot up 15+ FPS with my Crossfired system.

As to bottleneck, no, not really with that hardware. The 9550 is still a very capable chip. Now, if you had 2 460's, I'd say it might, might start to struggle a bit depending on the game.

This is just my opinion based off my personal experiences though, not a lot of in-depth reading on the subject.
 
Last edited:
Not as bad as my 9450. THe 460 isn't quite fast enough to be bottlenecked by that cpu. Pretty balanced. BFBC2 appears to like the 2500K more than the older quads though.
 
PornFlake said:
I have read a few discussions on this and I still don’t think I have a clear answer.
In my situation, what real world benefits would I see going from a Q9550 clocked at 3.6 to a 2500k?

I am wondering if the Q9550 is bottlenecking my GTX460 1GB @ 900/2000, 1920x1200. with as much eye candy turned up that results in a suitable frame rate

I tend to play with as much eye candy turned up that results in a suitable frame rate

I do mostly FPS gaming, COD, BF2, Crysis, etc.

-Flake
Click to expand...

I just went from a Q9550 @ 4.0 to a 2600k @ 4.6 and the difference is amazing!

That said I don't believe the Q9550 @ 4.0 was limiting me in any way, I had just had it long enough where I wanted to upgrade.
 
If your mobo can overclock quadcores well, (unfortunately my P35 based doesnt) then get a good cooler and take that chip to 4ghz level, perhaps above. Its definetly not a bottleneck there. Q9550 is very capable CPU. While new sandybridges are very fast, in real-life perfomance only relatively few games (known CPU dependant ones) make real use of it. This will propably change in future, but right now unless you have suffer from serious upgraditis I would wait and see what future brings.
 
Last edited:
I doubt u will see any bottleneck in gaming unless u went to sli or xfire. Once u turn all the eye candy up along with AA its mostly on the gpu at that point. Sounds like u just have an itch to upgrade.... Me too!
 
I seriously doubt you'd see any type of bottlenecking..

2qlskki.jpg
 
I noticed a big difference in smoothness in BC2 going from a Q9650 at 4.0ghz to a 2500k at 5.0ghz.
 
you really noticed a difference going from 4ghz to 5ghz?

damn man we are just spoiled by today's tech..
 
IndianScout said:
4 cores at 4g, I really find that hard to believe you were bottle necked

that you'll have to show me..
Click to expand...

Sure.

http://hardforum.com/showthread.php?t=1592732

I can also confirm that the approxiamate CPU usage in his thread is in line with what ive experienced.

My results were based on a GTX 260. Time to face the facts that your Q6600 is old.

Id show you evidence that even a 2600k makes a difference in BC2 over a 2500k, but based on your responses in that thread you probably wont care.
 
Last edited:
With your current CPU at it's current clock speeds combined with your GPU at the resolution you play at, I think by in large you will notice very little difference. There may be a small handful of exceptions and even then, the difference won't be huge IMO. The link posted above was a user running a more powerful GPU with a lower resolution display.
 
Last edited:
nissanztt90 said:
Sure.

http://hardforum.com/showthread.php?t=1592732

I can also confirm that the approxiamate CPU usage in his thread is in line with what ive experienced.

My results were based on a GTX 260. Time to face the facts that your Q6600 is old.

Id show you evidence that even a 2600k makes a difference in BC2 over a 2500k, but based on your responses in that thread you probably wont care.
Click to expand...

please post the full system specs you had your bottleneck on.
By full I mean all the onboard devices that you had in use like sound, lan, hdd setup etc etc.
There's users with crappier specs that don't have issues with it so I don't believe its the cpu being too slow no way. It might be on the cpu side of the code but not the cpu to be precise
 
I bought Crysis 2 to test if my 2.5+ year old system can still handle it. To my amazement, it still can! I'm even playing at 1920x1200 extreme settings with 50 fps average. ATM with my current rig, I can still run my other games at max settings. This is the reason why I'm still holding off upgrading.

If you really want to upgrade, then upgrade to LGA2011 this coming Q3, not now. Heck you can even wait until Ivy Bridge. :)
 
nissanztt90 said:
Sure.

http://hardforum.com/showthread.php?t=1592732

I can also confirm that the approxiamate CPU usage in his thread is in line with what ive experienced.

My results were based on a GTX 260. Time to face the facts that your Q6600 is old.

Id show you evidence that even a 2600k makes a difference in BC2 over a 2500k, but based on your responses in that thread you probably wont care.
Click to expand...

I use the Q6600 because I want too, I have others, but still that doesn't prove your point, I've disproved your findings with my 6600 and 560ti,

post your full specs..
 
It doesnt have a whole lot to do with raw FPS numbers but spiking CPU usage to 100% during intense action scenes (which still happens even on an OCed 2500k), you havent disproved anything. I have absolutely zero interest in proving that a 3 month old CPU is less of a bottleneck than a 3 year old CPU.
 
Q9x50 to 2500/2600k at same mhz shows about a 30-35% raw CPU horsepower improvement. This is for single threaded applications. No idea about multithreaded. Then, throw on the bonus of being able to overclock to 4.5-5 ghz.
 
nissanztt90 said:
Sure.

http://hardforum.com/showthread.php?t=1592732
My results were based on a GTX 260. Time to face the facts that your Q6600 is old.
Click to expand...

here's a little evidence that you were going to show me..

3dmark11

Core i5 2500k @ 5gkz

P5190

Q6600 @ 3.6ghz

P4755

both using a 560ti card..

so where is the double performance, that's why I have a problem with someone claiming double the FPS in gaming with a cpu change, and this is with a newer card, not a 260..

now going from a CoreDuo I can see the double in FPS, I watched a friend of mine do it personally, but not quad vs quad..

granted the Q6600 is getting dated, sure, but it can still game with the best of cpu's

that's my point..
 
Last edited:
no tweaks no nothing, no LOD changes or affiliation..

just plugged it in and ran it..

another example..

155jh51.jpg


I too was going to jump on the SB band wagon, but at this point I think I'm going to wait until the price comes down on them, I'm perfectly happy with the way my system runs right now..

my specs are in my sig, nothing special other than a ram change from the previous runs,

:cool:
 
I recently upgraded from the q9550 to the i7-2600k; didn't notice much of a difference (if any) while gaming, but it was a totally different story once I started photoshop. That was definitely a noticable difference :)
 
Neb said:
I recently upgraded from the q9550 to the i7-2600k; didn't notice much of a difference (if any) while gaming, but it was a totally different story once I started photoshop. That was definitely a noticable difference :)
Click to expand...

Do you have BC2 by any chance to benchmark?
 
Falkentyne said:
Q9x50 to 2500/2600k at same mhz shows about a 30-35% raw CPU horsepower improvement. This is for single threaded applications. No idea about multithreaded. Then, throw on the bonus of being able to overclock to 4.5-5 ghz.
Click to expand...

yet it can't knock the E8600 off 3dmark 01. Its design might give it a advantage. Especially the IMC vs the fsb when the data is not in the caches
 
Neb said:
I recently upgraded from the q9550 to the i7-2600k; didn't notice much of a difference (if any) while gaming, but it was a totally different story once I started photoshop. That was definitely a noticable difference :)
Click to expand...

Lmoa looool
 
IndianScout said:
here's a little evidence that you were going to show me..

3dmark11

Core i5 2500k @ 5gkz

P5190

Q6600 @ 3.6ghz

P4755

both using a 560ti card..

so where is the double performance, that's why I have a problem with someone claiming double the FPS in gaming with a cpu change, and this is with a newer card, not a 260..

now going from a CoreDuo I can see the double in FPS, I watched a friend of mine do it personally, but not quad vs quad..

granted the Q6600 is getting dated, sure, but it can still game with the best of cpu's

that's my point..
Click to expand...

processor tests/benchmarks

PCMark
2004
PCMark
2005
PCMark Vantage

PiFast

SisoftSandra

SuperPi

SuperPi 32m

wPrime 1024m

wPrime 32m

Video card tests/benchmarks

3DMark 99 Max

3DMark Vantage - Performance

3DMark03

3DMark05

3DMark06

3DMark11 - Entry

3DMark11 - Extreme

3DMark11 - Performance

3DMark2000

3DMark2001
SE
Aquamark

Unigine Heaven - Basic Preset (DX9)

Unigine Heaven - Xtreme Preset (DX11)
 
IndianScout said:
so where is the double performance, that's why I have a problem with someone claiming double the FPS in gaming with a cpu change, and this is with a newer card, not a 260..

now going from a CoreDuo I can see the double in FPS, I watched a friend of mine do it personally, but not quad vs quad..

granted the Q6600 is getting dated, sure, but it can still game with the best of cpu's

that's my point..
Click to expand...

i doubled my framerate in BC2 going from a Q6600 @ 3.0 to an i5-2500K, using a single GTX 460. what nisanztt90 is saying is absolutely correct, in my experience.
 
Porter_ said:
i doubled my framerate in BC2 going from a Q6600 @ 3.0 to an i5-2500K, using a single GTX 460. what nisanztt90 is saying is absolutely correct, in my experience.
Click to expand...

did you had the exact same setup minus the cpu and mobo or did you add more?
 
IndianScout said:
so where is the double performance, that's why I have a problem with someone claiming double the FPS in gaming with a cpu change, and this is with a newer card, not a 260..

now going from a CoreDuo I can see the double in FPS, I watched a friend of mine do it personally, but not quad vs quad..

granted the Q6600 is getting dated, sure, but it can still game with the best of cpu's

that's my point..
Click to expand...

And my point was that a 2500k ran BC2 smoother than a Q9650, because the Q9650 becomes the choke point at certain times. Nowhere did i personally claim it doubled FPS. The link i posted was simply readily availably evidence that the Q9650 can easily become a bottleneck in BC2. Im not so sure there was even a significant improvement in FPS, the FPS simply didnt drop anywhere near as much when the CPU spiked to 100%, thus creating smoother gameplay compared to when the Q9650 became a bottleneck, which was what i initially said. BTW, 3dmark is a joke as far as benchmarking is concerned, let alone actual gaming.
 
I went from an Q9550- 4 gigabytes of ram- dual 1GB 460GTX to a 2500k-8 gigs of ram-dual 1GB 460GTX. I benchmarked resident Evil 5 fixed benchmark, my frames with the Q9550 was 27.3 fps average. With the sandy bridge at 4.2 they were 143 fps.

Crysis was unplayable for my old setup, it runs 60fps at high with sandy bridge.

Bad Company I'm consistently getting 100+ fps vs 30-50 fps with Q9550.

A Q9550 is not limiting a single 460GTX, but it is limiting an SLI'd 460GTX system. The gains were astronomical for me.
 
Spooony said:
processor tests/benchmarks
Click to expand...


you're missing the point with the 3dmark scores I put up,

if the Q6600 was bottlenecking the 560ti, it would of showed in the scores a great deal, but it doesn't

but anyway here's a couple more numbers then I'm thru with this..

29aqkqq.jpg


2v3kl91.jpg


IndianScout - Futuremark..
 
Porter_ said:
i doubled my framerate in BC2 going from a Q6600 @ 3.0 to an i5-2500K, using a single GTX 460. what nisanztt90 is saying is absolutely correct, in my experience.
Click to expand...

BC2 is probably an outlier here though, as most games don't show that kind of performance increase.
 
You must log in or register to reply here.
Back
Top