E8400 and the Q6600, another take

taltamir

Lifer
Mar 21, 2004
13,576
6
76
Some time ago I upgraded from an E8400 to a Q6600.

Ironically, I once argued that the E8400 obsoleted the Q6600: http://forums.anandtech.com/me...=e8400+obsoletes+q6600

The E8400 was faster in all games at the time. faster in every app except half life map/mod compiler, and when you go into video editing, the improvements done on penryn + the SSE4 speed optimizations make it faster than a Q6600 for common encoding types...

Over the past year I have seen things change. I have seen more and more games able to handle 4 cores, the HEAVIEST of games. Games like supreme commander, and UE3 games like mass effect.

MPEG4 type encoding (divx, xvid) became less valuable as x264 took over for high def encoding, and x264 does not benefit at all from SSE4.

I also learned to overclock, which tilts the balance in favor of the quads...

The biggest change, and the one that made me make the switch, was the "save open tabs" feature in chrome and firefox. Firefox first put in "remember open tabs". And I found out that I am a tab whore, I can have hundreds of them open, last week I upgraded from 4GB to 8GB because I was filling up my ram with tabs...
Than chrome came about with opening at least one process per tab, giving at perfect scaling for at least N cores where N=number of tabs from last session, I just had to make the switch. I finally had something where the Q6600 simply demolishes the E8400...

It is interesting to revisit the ideas of the past. To look at what changed. I was wrong to say that the E8400 obsoletes the Q6600, not that it wasn't better in every way, it is simply that it's dominance was transient.
At the time the only thing people could tell me was faster on the Q6600 was the half life compiler and that it will be "future proof"... But future proofing is bunk, when the future arrives its cheaper and better that the "future proof" hardware you buy today.
A few months later I started saying it "depends on your usage" instead of recommending one CPU over the other, and at the end of july I bought myself a Q6600 to replace my E8400 for 185$. I left it lying around until december where I sold the E8400 for 120$ (yea, stupid to wait, I know). So the cost of the switch was 65$. But those 65$ allowed me to have the better processor for the first 6 months, before they switches places...

Today, I cannot recommend the E8400 over the Q6600 for anyone. The future arrived and for where it matters, the Q is better... there are still rare situation where the E rules, but they have become so rare, and it is generally it lacking.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
mostly to 3.6ghz. I have done higher overclocks on both, but eventually lowered them to settle for a 600mhz OC with no overvolting to reduce power consumption.
 

3MDR

Junior Member
Jan 9, 2009
18
0
0
Too bad your Q6600 is already outdated compared to the other quads, so what's the point of making this comparison? Both CPU's are outdated, I wouldn't recommend anyone getting either at this point.
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
The "future" arrived quite a while ago.

I noticed how nice having a quad was pretty much as soon as i got it. You can multitask like no tomorrow, & yeah, your user habits change & benefit from the extra power available.

But it's hard to convince dual fanbois who are so set in their beliefs that there's no need for quads, blah, blah, that to be honest, it's kinda one of those things where i say ignorance is bliss...let them miss out.
 

JMapleton

Diamond Member
Nov 19, 2008
4,179
2
81
Originally posted by: 3MDR
Too bad your Q6600 is already outdated compared to the other quads, so what's the point of making this comparison? Both CPU's are outdated, I wouldn't recommend anyone getting either at this point.

I suspect the op was comparing duals to quads.

I still suspect duals are most useful if you do not game or do graphical editing.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
well, actually it was a specific comparison, the two were the same price originally. I was comparing a 45nm intel C2D named the E8400 to a 65nm Q6600. The last gen part was originally less useful, but later became more useful.
 

cubeless

Diamond Member
Sep 17, 2001
4,295
1
81
and at @199+ ship a q6700 @ compusa makes an easier oc of older mobos... really a great life extender...
 

TantrumusMaximus

Senior member
Dec 27, 2004
515
0
0
meh... this post would make sense if it were E6600 is inferior to the Q6600 but this is an apples to oranges comparison.

I'll take the E8400 with 45nm & the gigahertz it provides.

Most games (yes even now) respond more to freq. vs. # of cores. In my comparison my wife's system has a Q6600 which I gave her and replaced it with my E8400 which runs Warhammer online 200+ player battles WAY better than her and we have similar systems with very comparable video cards.

Anyone that plays Warhammer knows it's very CPU demanding and her system with Q6600 can not keep the same FPS in big battles even with same vid board.

Q9550 vs. E8400 is a better comparison but I def... wouldn't expect a Q6600 to be the better gaming solution over an E8400.

But as the OP said.... based on a particular usage of the user.... in my case I'll take Gigahertz over cores.
 

StinkyPinky

Diamond Member
Jul 6, 2002
6,766
784
126
We keep going in circles about this debate. I've lost count of the number of threads about the E8400 v Q6600. Both are (were?) great CPU's and great value for the power you get. Depending on your needs, one will suit you better than the other. Neither one is "better" than the other.
 

SlyNine

Member
Sep 15, 2003
46
0
0
So basicaly it sounds like you could do with an old FX53 AMD single core CPU. V1001

I've never played Warhammer, but I have compaired CPU's often enough to know that the 20% increase in clock speed from 3ghz to 3.6 ghz amounts to about a 5% increase in game performence unless you turn the settings all the way down so its only CPU limited.

That also assumes you have a video card that can keep up because even if you turn the settings way down you can still be GPU limited if you have an older or weaker GPU.
 

OCGuy

Lifer
Jul 12, 2000
27,227
36
91
Originally posted by: 3MDR
Too bad your Q6600 is already outdated compared to the other quads, so what's the point of making this comparison? Both CPU's are outdated, I wouldn't recommend anyone getting either at this point.

Wolfdales are outdated? :confused:



No.
 

Lonyo

Lifer
Aug 10, 2002
21,939
6
81
I can't think of the words needed to respond to the OP's post.
It just blew my mind.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: n7
The "future" arrived quite a while ago.

I noticed how nice having a quad was pretty much as soon as i got it. You can multitask like no tomorrow, & yeah, your user habits change & benefit from the extra power available.

But it's hard to convince dual fanbois who are so set in their beliefs that there's no need for quads, blah, blah, that to be honest, it's kinda one of those things where i say ignorance is bliss...let them miss out.

i was a dual fan boy when my e8600 was over 4Ghz last year and many quads were not making 3.4 Ghz

My only criteria is gaming .. i have PS, but i do not do the advanced work. Duals are fine for most people.

HOWEVER, it appears the situation is changing in gaming .. so i care

and now that my e8600 only hits about 4.25Ghz solid and quads are overclocking better now, i got a Q9550S which i am benchmarking right now - both my e8600 and my quad are at 3.99Ghz [and i am also testing my e8600 at 4.25GHhz to see if the +260Mhz 'extra' on the core makes up for "lacking" two of them. i am in the middle of running a 13 game benchmark suite right now with my q9550s .. and some games DO benefit.

rose.gif
 

OCGuy

Lifer
Jul 12, 2000
27,227
36
91
Having a quad and thinking it is faster or "smoother" is the power of suggestion for 99% of people. I've had a Q9450, and I did not hesitate to go back to dual. The quad did not do anything for me except synthetic benchmarks.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: SlyNine
well, statistics can be used to prove anything Ocguy. 99% of all people know that.

well i am going to know for myself soon .. a few days at most .. i am done benching with my e8600
[that is why it is for sale; it does ~4.3Ghz if you give it a little voltage boost]

at least with 13 or 14 GAMES ... benched at 19x12 and and 16x10 with everything maxed out

new games .. DX10 mostly .. both e8600 and q9550s at 4Ghz
- with 3 GPUs - 4870/1GB .. 4870-X2 and GTX280


rose.gif


is there ANY reason for me to bench with my Q9550s at stock, 2.83Ghz ... or anything in-between

??

i am ALSO showing my e8600's results with these 13 games at 4.25Ghz



 

Lonyo

Lifer
Aug 10, 2002
21,939
6
81
Originally posted by: apoppin
Originally posted by: n7
The "future" arrived quite a while ago.

I noticed how nice having a quad was pretty much as soon as i got it. You can multitask like no tomorrow, & yeah, your user habits change & benefit from the extra power available.

But it's hard to convince dual fanbois who are so set in their beliefs that there's no need for quads, blah, blah, that to be honest, it's kinda one of those things where i say ignorance is bliss...let them miss out.

i was a dual fan boy when my e8600 was over 4Ghz last year and many quads were not making 3.4 Ghz

My only criteria is gaming .. i have PS, but i do not do the advanced work. Duals are fine for most people.

HOWEVER, it appears the situation is changing in gaming .. so i care

and now that my e8600 only hits about 4.25Ghz solid and quads are overclocking better now, i got a Q9550S which i am benchmarking right now - both my e8600 and my quad are at 3.99Ghz [and i am also testing my e8600 at 4.25GHhz to see if the +260Mhz 'extra' on the core makes up for "lacking" two of them. i am in the middle of running a 13 game benchmark suite right now with my q9550s .. and some games DO benefit.

rose.gif
Make sure that you give us minimum frame rate numbers if you can, because it would be interesting to see the effect on those (when you consider other programs potentially sucking up CPU time when they decide to have a busy few moments). Might mean that there are less drops on a quad vs dual even if a game only uses 2 cores.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
of course .. Min/Av/Max

.. i am benching on a "normal" PC .. not a "clean" test rig; mine has games and all the applications normally installed that i use
- on both identical partitions .. so there is no difference and it will be 'apples to apples'

but i was wondering .. is ANYONE *interested* in seeing these 13 game benches of q9550-s at STOCK speeds?
- then compared to 4Ghz .. to see the scaling ?

Or just go for q9550 and e8600 both at 4 Ghz ?

[of course, i am going to show e8600 at 4.25Ghz also]
rose.gif
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
use fraps, make it create an excel document with all the data (or is it CSV?), then make a formula to calculate to the exact ms it took to render each frame, and look for the longest to render frames... espcially for cases where they are preceeded and followed by short to render frame... for example, on mass effect, I would have a frame that took 38-42ms to render, between two frames that took 7-12ms to render... to do a quick conversion. 7ms is about 143fps and 42ms is about 23 fps. Thats how you can see the microstutter.
But fps is a per second value, it contains a count of how many frames were rendered during that second. in other words, it averages them out over a period of a second. so you can see a min FPS of 40 and not realise that the instantanous FPS is actually 23 at a portion of that second.
 

F1shF4t

Golden Member
Oct 18, 2005
1,583
1
71
Originally posted by: apoppin
of course .. Min/Av/Max

.. i am benching on a "normal" PC .. not a "clean" test rig; mine has games and all the applications normally installed that i use
- on both identical partitions .. so there is no difference and it will be 'apples to apples'

but i was wondering .. is ANYONE *interested* in seeing these 13 game benches of q9550-s at STOCK speeds?
- then compared to 4Ghz .. to see the scaling ?

Or just go for q9550 and e8600 both at 4 Ghz ?

[of course, i am going to show e8600 at 4.25Ghz also]
rose.gif

I would be interested in benches at 3.6 ghz. The diff between that and 3.8 - 4 ghz would be interesting. Would certainly tell me if there is any reason to "unbox" my old watercooling kit.
 

Denithor

Diamond Member
Apr 11, 2004
6,300
23
81
Originally posted by: Dark Cupcake
I would be interested in benches at 3.6 ghz. The diff between that and 3.8 - 4 ghz would be interesting. Would certainly tell me if there is any reason to "unbox" my old watercooling kit.

Actually 3.4GHz would be more interesting to me.

8.5x400 (DDR2-800 limit)
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: Denithor
Originally posted by: Dark Cupcake
I would be interested in benches at 3.6 ghz. The diff between that and 3.8 - 4 ghz would be interesting. Would certainly tell me if there is any reason to "unbox" my old watercooling kit.

Actually 3.4GHz would be more interesting to me.

8.5x400 (DDR2-800 limit)
it does not matter about the DDR2 "limit" .. use your CPU:RAM dividers
- there is no practical performance difference dropping it a bit

what do you think, 3.5Ghz ?

stock is 2.83 .. on mine, the voltage needs a boost around 3.6Ghz and the temps first go up about there

Thats how you can see the microstutter

this review is not going to do anything to do with MS .. it is a CPU review and i want to bench games and see the difference between Quad and Dual core
- at the same clock .. 4Ghz
- at a higher clocked dual core

and then it is cool to see how a quad scales .. i already did dual core awhile back and concluded ~3.8 Ghz is a point of diminishing returns with a fast GPU like X2 or GTX280

so i was asking for *where* .. 2.83Ghz is a given
--and it's max OC is also a given at 4 Ghz [which also gives an exact comparison with my e8600 at that same speed]

. . . but many people cannot get 4Ghz on their quad and i want to relate its scaling to them


3.5ghz or even 3.6 is what i was thinking




i have about 100 pages to finish writing and publishing .. first :p
.. then i will explore microstutter
rose.gif

 

TC91

Golden Member
Jul 9, 2007
1,164
0
0
Originally posted by: apoppin
of course .. Min/Av/Max

.. i am benching on a "normal" PC .. not a "clean" test rig; mine has games and all the applications normally installed that i use
- on both identical partitions .. so there is no difference and it will be 'apples to apples'

but i was wondering .. is ANYONE *interested* in seeing these 13 game benches of q9550-s at STOCK speeds?
- then compared to 4Ghz .. to see the scaling ?

Or just go for q9550 and e8600 both at 4 Ghz ?

[of course, i am going to show e8600 at 4.25Ghz also]
rose.gif

if you had time I would appreciate seeing the stock benches of the q9550 compared.