Overclocked Q6600 v E8400 Compared. (Benchmarks included)

[winkgood]

There has been much debate on this issue of which of these two CPUs would make the better upgrade. Arguments have gone back and forth. The consensus seemed to be that the E8400 would be better for gaming whereas the Q6600 would excel in video editing.

I originally decided to go with the E8400 as an upgrade from my E6400. I figured that it would provide more advantage in most applications if I could get a 4.2 Ghz overclock over the Q6600 @ 3.6 Ghz. After receiving the E8400, I wasn't all that impressed with the performance increase over my overclocked E6400 so I decided to go with the Q6600 and return the E8400. Before sending it back however, I decided to do some benchmarking to make sure I was going with the better processor. Hopefully, this helps others who are also debating which processor to go with.

My test platform for the gaming benchmarks was an Asus P5E-VM HDMI, 2 gigs of Crucial Balistix 4-4-4-12 with Windows Vista Ultimate so I could take advantage of DX10.

The test platform for the video encoding and DVDShrink was my old P5B Deluxe Wifi/AP, and 2 gigs of Corsair Dominator 4-4-4-12.

I used an 8800 GT Superclocked for all the tests.

From the Crysis results, its apparent that CPUs don't really have a significant impact in performance but that the bottleneck still lies on GPU performance. In other words, for modern games, these two CPUs are going to be comparable with an exception of games like Supreme commander that take advantage of four cores.

 

[winkgood]

I tried using the SSE4 optimization with the DivX 6.8 Codec. Since it actually decreased performance, I will assume that this is still a WIP for them and that we'll need to wait to see benefit in this area.

 

[Shadow703793]

Nice findings. Did you use Vista x64 or x86, because that some times makes a difference in some benchmarks.

 

[vagetaqtd]

The Q6600 overclocked obviously has more usage flexibility than the E8400 overclocked. Sure E8400 might do well with gaming, but so does the Q6600. And at the currently price over $240 for E8400, it is certainly not worth it at all. Maybe $200 would be okay.

As for the lowere temps and less voltage of the new 45nm, its a positive if the price were fair. Because for that price, performance to price ratio isn't that high.

 

[jimmysmitty]

Gaming wise it looks very close and that at higher resolutions both are about the same. As for why your encoding results are slower is beyond me.

I think the encoding could be biased though towards a quad core so even with SSE4.1 optimizations a dual core may not keep up with a quad core since there are more threads to do the work on. I know that the QX9650 did the DivX encoding in less than half the time compared to a QX6850.

I think for SSE4.1 to really shine you need to pit it against a E6850 to see the real difference it makes and wiil make.

 

[winkgood]

Nice findings. Did you use Vista x64 or x86, because that some times makes a difference in some benchmarks.

I used 32 bit operating systems for all tests. True, it might have made a difference but I wanted to use a platform that I think the majority are still using. Though I'll admit that 64 bit Windows is becoming more common. If I didn't have to go to work I'd probably have done more in depth analysis. As it is, I started to get lazy and ended up not using all the processors/settings for all the results as you can tell. Is 64 bit more optimized for quads?

One thing I did conclude though is that the quad is neck and neck with the dual in games even though the dual is clocked much higher. Based on my findings, I would say that people would be better off going with the Q6600 than the E8400 for its multitasking abilities and potential for multi core optimized apps in the future. Plus games have really hit their bottleneck with GPUs, so any high end processor is going to be sufficient for games as long as you have a good graphics card.

Just as a note, the way I conducted the gaming tests was four runs for each setting in the Crysis benchmark program and five for UT3. For UT3 I excluded the highest result from each test as oftentimes, I found that if the bot died a lot, the FPS count would be increased significantly.

 

[bc4]

good post. Very informative and very good to know

 

[babybudha]

Great review on a very popular question!

 

[cisco]

Nice post and even though it doesn't make much difference in games, I would still pay the extra to get the Q6600. Applications and games will continue to improve their utilization of multi-cores, thus opting for more cores will only extent the performance gain for your system.

 

[Reynod]

Excellent work and thanks to the OP.

We need more threads like this and it is a pleasure to read.

I have been debating this very issue and with the newer games now starting to make more use of the quad I think I will now settle for the G0 Q6600.

Thankyou

 

[drfaraz]

well i wanted to put different aspect in order to choose b/w E8400 and q6600, is temperature yeah i know many people are using q6600 on air cooling @3.6 ghz BUT what if the climate is very hot , i am from pakistan and summer time is really very hot . est 42-45 degree calcius outside at 32-37 calcius room temperature .guys can you really help me to choose what is best for me plz......?

 

[T8RR8R]

I think if you're a gamer dual core CPU's will be the way to go for a least another year or 2. Even modern games see very little boost from having 4 or more cores. So if you can get a super fast dual core for the same price as a decent quad, you're probably better off getting the dual. 2 cores would also be a little cooler too. If you're doing alot of Video editing and things of that nature, maybe waiting for the q9300 or so would be best.

 

[night_wolf_in]

great work. u just made me happier that i bought my q6600 3 monthes ago. n u proved my decision was right. yayayaya

 

[grieve]

I think if you're a gamer dual core CPU's will be the way to go for a least another year or 2. Even modern games see very little boost from having 4 or more cores. So if you can get a super fast dual core for the same price as a decent quad, you're probably better off getting the dual. 2 cores would also be a little cooler too. If you're doing alot of Video editing and things of that nature, maybe waiting for the q9300 or so would be best.

I disagree…

I am surprised how well the Q6600 held its own in these tests; I see no reason to purchase a higher clocked dual core duo when the Q6600 is so cheap, and gets pretty much the same frames… PLUS you get the quad core for apps that can utilize it.

 

[T8RR8R]

I disagree…

I am surprised how well the Q6600 held its own in these tests; I see no reason to purchase a higher clocked dual core duo when the Q6600 is so cheap, and gets pretty much the same frames… PLUS you get the quad core for apps that can utilize it.

An equally clocked dual core will do very well compared to a quad core. Not that the quad isn't better, but for the price and heat of most quads a dual will do just fine and there would be little benefit to spend more for a quad just for gaming at this point in time.

Here's page with charts that show demanding games and how some CPU's compare.
http://www.hardforum.com/showthread.php?t=1271109&page=2

 

[Grimmy]

An equally clocked dual core will do very well compared to a quad core. Not that the quad isn't better, but for the price and heat of most quads a dual will do just fine and there would be little benefit to spend more for a quad just for gaming at this point in time.

Here's page with charts that show demanding games and how some CPU's compare.
http://www.hardforum.com/showthread.php?t=1271109&page=2

I noticed some people tend to post benchmarks that have different MB/RAM and Video cards.

However both those are similar hardware test beds:

Quad side:
Intel platform:

* CPUs:
o Intel Core 2 Duo Q9300 (LGA775, 2,5GHz, 1333MHz FSB, 2 x 3MB L2, Yorkfield);
o Intel Core 2 Duo Q6600 (LGA775, 2.4GHz, 1066MHz FSB, 2 x 4MB L2, Kentsfield).
* Mainboard: ASUS P5E (LGA775, Intel X38, DDR2 SDRAM).
* Memory: 2GB DDR2-1066 with 5-5-5-15 timings (Corsair Dominator TWIN2X2048-10000C5DF).
* Graphics card: OCZ GeForce 8800GTX (PCI-E x16).
* HDD: Western Digital WD1500AHFD (SATA150).
* OS: Microsoft Windows Vista x86.

Dual Core Side:
* Mainboards
o ASUS P5E (LGA775, Intel X38, DDR2 SDRAM);
o DFI LANParty UT 790FX-M2R (Socket AM2+, AMD 790FX).
* Memory: 2GB DDR2 SDRAM (Corsair Dominator TWIN2X2048-10000C5DF).
* Graphics card: OCZ GeForce 8800GTX (PCI-E x16).
* HDD: Western Digital WD1500AHFD (SATA150).
* OS: Microsoft Windows Vista x86

I'm glad those are basically the same, since I do recall someone posting benchmarks that showed different scores simply because the GPU was not the same.

Those are good comparisons but you also have to remember they are running at stock speeds, and they are different die sizes. As far as the heat goes, as long as you have the right cooling, you shouldn't have to worry much. The Q6600 G0 is 95W, which should run cooler then a Pre-hott 3.0 single core, and they survived the heat. :lol:. o O (I thought it was another way of looking at it)

But all in all, everyone should understand their needs... if they don't... whelp, theres a 50/50 chance that they will buy the wrong chip, and MB. 😗 . o O(Will this Q6600 work on this AM2 MB? Edit: Or was that AM2+?)

 

[T8RR8R]

I just buy what I "need" to do the things that I want. When my e6400 starts to crap out at some newer game that "I" like then I'll change stuff, but 'till then I don't need a quad core or a 45nm CPU. However if I get a nice tax return and find a good deal on some stuff then maybe I'll change up, but I doubt it.

 

[grieve]

An equally clocked dual core will do very well compared to a quad core. Not that the quad isn't better, but for the price and heat of most quads a dual will do just fine and there would be little benefit to spend more for a quad just for gaming at this point in time.

Here's page with charts that show demanding games and how some CPU's compare.
http://www.hardforum.com/showthread.php?t=1271109&page=2

I do Agree with you however this review consisted of an E8400 $239 and a Q6600 $275 (just checked prices, newegg.com)

The Q6600 had almost the same frames in everything PLUS it has quad core for later use when apps take advantage. In my personal opinion the $35 difference is well spent.

 

[T8RR8R]

lol if this were an Nvidia Vs ATI thread 12 fps would be such a huge difference. Anyway all I'm saying is that 4 cores come with the extra cost of heat, power consumption, CPU price. Although they do perform slightly better I just don't think they are worth it at this time since most game can't use more than 2 cores and so many don't even use more than 1.

I guess if you want the best of the best a quad would make sense though for the extra 4 or so FPS you'll gain on average.

 

[Grimmy]

Whelp.. when you run a game that takes advantage of 2 cores... what happens when something else in the back ground decided to do something? No more cores to handle anything in the background.

Hmmm... Mentioned it before, I guess I'll say it again:

$239/2 = $119.5 Per core
$275/4= $68.75 Per core

And on top of that... 40-50 bucks off in April.

As far as the heat/power consumption, no different then owning a Prescott 3.0 single core... Except more performance.

 

[winkgood]

I noticed some people tend to post benchmarks that have different MB/RAM and Video cards.

However both those are similar hardware test beds:

For the above tests I tried to keep the platforms as similar as possible. The reason I switched to the P5E for the gaming tests from the P5B is because I wanted to be able to use my main machine for other stuff while conducting the tests. Switching platforms shouldn't affect benchmarks since I was comparing apples to apples (e8400, q6600, and e6400 all used the same platform for the gaming tests).

 

[hockyis1]

Something else to think about...

I bought the e8400 when it first came out for $209 - obviously for someone buying now they're out of luck...

It overclocked on the stock cooler to 4Ghz in about 5 minutes worth of my time.

There is a *very* good chance that any Q6600 you get won't even break 3.0Ghz as noted in many threads - some batches are just crap, why take the chance? Will you need an aftermarket cooler to get a Q6600 to 3.6? More than likely you will - so add another 30-60 bucks onto that Q6600 cost.

I game only and do zero video stuff - so I will never see any difference in performance over a quad core and I spent almost $100 less if you consider I'd have to buy an aftermarket cooler for the quad. *shrug*...

 

[Grimmy]

For the above tests I tried to keep the platforms as similar as possible. The reason I switched to the P5E for the gaming tests from the P5B is because I wanted to be able to use my main machine for other stuff while conducting the tests. Switching platforms shouldn't affect benchmarks since I was comparing apples to apples (e8400, q6600, and e6400 all used the same platform for the gaming tests).

I understand that.. I was referring to the links from the other post. I read another thread that showed comparisons, but they were not based off the same system/gpu. So it made the Quad look even worse.

But what you did, and the other post is great. I wasn't saying anything wrong with your benchmarks.


Sorry if I was confusing it. 😳

 

[Grimmy]

Something else to think about...

I bought the e8400 when it first came out for $209 - obviously for someone buying now they're out of luck...

It overclocked on the stock cooler to 4Ghz in about 5 minutes worth of my time.

There is a *very* good chance that any Q6600 you get won't even break 3.0Ghz as noted in many threads - some batches are just crap, why take the chance? Will you need an aftermarket cooler to get a Q6600 to 3.6? More than likely you will - so add another 30-60 bucks onto that Q6600 cost.

I game only and do zero video stuff - so I will never see any difference in performance over a quad core and I spent almost $100 less if you consider I'd have to buy an aftermarket cooler for the quad. *shrug*...

Well.. as long as you get what you think you need. Then again, some of us like to make what we have better.

But as far as I'm concerned... we are in an expensive hobby. When we see what comes around next years corner, will perhaps make us blubber. . o O (I've only had this system for 5 months 12hours 13mins and 45 sec's)

 

[T8RR8R]

Well.. as long as you get what you think you need. Then again, some of us like to make what we have better.

But as far as I'm concerned... we are in an expensive hobby. When we see what comes around next years corner, will perhaps make us blubber. . o O (I've only had this system for 5 months 12hours 13mins and 45 sec's)

Get a performance car and try to double it's power, that's not an expensive hobby?