nVidia GTX 260m put to the test with Sager NP8662 - comparision Core 2 Duo vs Quad

The intent of this thread is to post my benchmark results from my new NP8662 notebook PC with nVidia GTX 260m and Core 2 Duo P9700 @ 2.8GHz. In particular I would like this to be a comparison to all the benchmarking that LaptopNut has performed with his similarly equipped Sager, except the uses the Quad Core Q9000 @ 2.0GHz compared with my Dual Core @ 2.8GHz.

LaptopNut used XP SP3 and DOX drivers in his initial trial. We will be using Windows 7 Ultimate RTM with 186.81 WHQL from nVidia's site for both machines (Dual and Quad) in this comparison. In my opinion, using WHQL offers a good baseline. I can always trial other drivers later to review benefit of performance. I also prefer to run built-in benchmarks due to their repeatability, as opposed to in-game benchmarking that can be very subjective, although it does give a warm indication of general performance.

Before I get into the details, I'd like to make a few comments on the NP8662 in general. I have heard great things about Sager in the past and decided to go this route for my laptop this time around. So far I am glad I did. I ordered this machine from XoticPC.com and they processed and delivered my machine very quickly. The machine itself definitely exudes high class and quality. While the machine doesn't scream at you aesthetically, and lacks any bling that most other performance notebooks offer (i.e. Alienware), I love the classic look of the brushed aluminum. The fit and finish is excellent (gaps between panels are small and consistent), and overall quality is readily apparent. It is also very silent. I was shocked and wondered if the fans were even connected while using the Windows desktop, but running games the fan was audible, but still quiet compared with other laptops. Of importance as well is that the screen has zero backlight bleed and no stuck or dead pixels. Overall the machine feels very solid.

Benchmarks will be added as we go along, since I figure I'd like to get out there what I've completed in a timely fashion, than wait another week (or more) to complete everything. This thread will be a work in progress for a while, so bear with us.

Also, where you see a hyperlink on the game settings description, it will link you to a screenshot of the settings if interested.

Here's the pertient specifications of the NP8662 being compared:

Intel Core 2 Duo P9700 @ 2.8 GHz (25W) OR Intel Core 2 Quad Q9000 @ 2.0GHz
nVidia GTX 260m with 1024MB DDR3
15.4" 1680x1050 Glossy Widescreen (16:10 FTW!)
500GB 7200RPM Seagate Momentus Hard Drive
4GB DDR3 RAM
DVD+/-RW
Windows 7 Ultimate RTM

I did a fresh install of Windows 7 RTM and downloaded appropriate updates thanks to @nthony in his thread. mtness is running Vista with a same spec machine as the P9700 for validation.

neilnat offered up a comparison with the ATI HD 4850. Great comparison!

Notes: Screen resolution is set at the native 1680 x 1050 unless otherwise specified. Drivers are nVidia 186.81 WHQL. NO Overclocking is done unless otherwise specified. This is for comparison use only. GPU temperatures so far have not exceeded 72C. CPU temperatures so far have not exceeded 62C.

Following are the benchmark results:

(1) 3DMark Vantage 1.01
P9700: P5332, GPU 4256, CPU 22058
Q9000: P5462, GPU 4369, CPU 21891

(2) 3DMark06 1.10
P9700: 10451, SM2 4890, SM3 4525, CPU 2558
Q9000: 10064, SM2 4320, SM3 4388, CPU 2933

(3) World In Conflict Demo Benchmark
P9700:
1680 x 1050 @ Very High - Min: 2, Max: 41, Avg: 22
1680 x 1050 @ High - Min: 18, Max: 54, Avg: 30
1280 x 800 @ Very High - Min: 11, Max: 54, Avg: 28
1280 x 800 @ High - Min: 17, Max: 76, Avg: 36

Q9000:
1680 x 1050 @ Very High - Min: 11, Max: 41, Avg: 23
1680 x 1050 @ High - Min: 16, Max: 54, Avg: 31
1680 x 1050 @ Medium - Min: 32, Max: 108, Avg: 55
1280 x 800 @ Very High - Min: 12, Max: 55, Avg: 29
1280 x 800 @ High - Min: 16, Max: 77, Avg 38

(4) Crysis 1.2.1 Built-in Benchmarks 64-bit DirectX 10 using Crysis Benchmarking Tool 1.05.

All FPS are rounded to the nearest whole number, AA is off, Vsync is Off

P9700:
GPU Bench Very High Settings 1680 x 1050 - Min: 11, Max: 17, Avg: 15
GPU Bench Very High Settings 1280 x 720 - Min: 14, Max: 29, Avg: 23
CPU Bench Very High Settings 1680 x 1050 - Min: 9, Max: 20, Avg: 15
CPU Bench Very High Settings 1280 x 720 - Min: 16, Max: 29, Avg: 23

GPU Bench High Settings 1680 x 1050 - Min: 18, Max: 31, Avg: 26
GPU Bench High Settings 1280 x 720 - Min: 20, Max: 51, Avg: 35
CPU Bench High Settings 1680 x 1050 - Min: 14, Max: 32, Avg: 24
CPU Bench High Settings 1280 x 720 - Min: 22, Max: 44, Avg: 35

Q9000 (32-bit mode):
GPU Bench Very High Settings 1680 x 1050 - Min: 10, Max: 17, Avg: 15
GPU Bench Very High Settings 1280 x 720 - Min: 12, Max: 28, Avg: 22
CPU Bench Very High Settings 1680 x 1050 - Min: 5, Max: 20, Avg: 15
CPU Bench Very High Settings 1280 x 720 - Min: 11, Max: 29, Avg: 23

GPU Bench High Settings 1680 x 1050 - Min: 17, Max: 30, Avg: 25
GPU Bench High Settings 1280 x 720 - Min: 8, Max: 43, Avg: 34
CPU Bench High Settings 1680 x 1050 - Min: 4, Max: 33, Avg: 32
CPU Bench High Settings 1280 x 720 - Min: 20, Max: 44, Avg: 32

LaptopNut Configuration:
Texture Quality High
Objects Quality High
Shadows Quality Medium
Physics Quality High
Shaders Quality Medium
Volumetric effects quality High
Game effects quality High
Post processing quality Medium
Particles Quality High
Water Quality Medium
Sound Quality Medium
Motion Blur High

P9700:
GPU Bench 1680 x 1050 - Min: 23, Max: 60, Avg: 40
GPU Bench 1280 x 720 - Min: 28, Max: 62, Avg: 48
CPU Bench 1680 x 1050 - Min: 22, Max: 47, Avg: 38
CPU Bench 1280 x 720 - Min: 22, Max: 64, Avg: 41

Q9000:
GPU Bench 1680 x 1050 - Min: 23, Max: 58, Avg: 45
GPU Bench 1280 x 720 - Min: 24, Max: 67, Avg: 46
CPU Bench 1680 x 1050 - Min: 25, Max: 51, Avg: 42
CPU Bench 1280 x 720 - Min: 15, Max: 67, Avg: 44

Playing at LaptopNut's recommended settings not only looks good, but also executes smooth frames. Averaging in the low to mid-40's throughout the first level at least.

(5) Grand Theft Auto IV (GTA IV) Benchmark
Graphics Settings
Clip capture is disabled
Aspect ratio Auto
1680 X 1050 resolutions
Texture Quality - High
Reflection Resolution - High
Water Quality - High
Shadow Quality - High
Texture Filter Quality - High
View Distance - 25
Detail Distance - 24
Vehicle Density - 31
Shadow Density - 1
Definition - ON
Vsync OFF

P9700:
Avg FPS: 37.77
Duration: 37.14 sec
CPU Usage: 96%
System Memory Usage: 59%
Video Memory Usage: 76%

Q9000:
Average FPS: 40.20
Duration: 37.27 sec
CPU Usage: 88%
System memory usage: 64%
Video memory usage: 82%

(6) Company of Heroes Built-in Benchmark
Settings: Ultra = Maximum settings, no AA, no Vsync, Post Processing ON, Max Model Detail
P9700: Min: 12, Max: 130, Avg: 50
Q9000: Min: 22, Max: 173, Avg: 62

Settings: mtness settings - All High
P9700: Min: 48, Max: 265, Avg: 122
Q9000: Min: 103, Max: 512, Avg: 208

Settings: Medium (The only settings allowed to set them to medium were Texture detail, shadows, Physics, Tree Quality, Terrain Detail, Effects fidelity, Effects Density. All other settings were left on high)
Q9000: Min: 36.3, Max: 226.3, Avg: 97

(7) Far Cry 2
Common settings: no AA, no Vsync, Bloom & HDR Enabled

Built-in Benchmark Ranch Small
DirectX 9
P9700:
High - Min: 34, Max: 60, Avg: 43
Very High - Min: 28, Max: 49, Avg: 34
Ultra High - Min: 25, Max: 40, Avg: 29

Q9000:
High - Min: 34, Max: 63, Avg: 44
Very High - Min: 29, Max: 52, Avg: 36
Ultra High - Min: 27, Max: 46, Avg: 32

DirectX 10
P9700:
High - Min: 32, Max: 61, Avg: 42
Very High - Min: 31, Max: 57, Avg: 40
Ultra High - Min: 28, Max: 46, Avg: 34

Q9000:
High - Min: 34, Max: 62, Avg: 44
Very High - Min: 35, Max: 59, Avg: 42
Ultra High - Min: 29, Max: 47, Avg: 35

Built-in Benchmark Ranch Long
DirectX 9
P9700:
High - Min: 31, Max: 74, Avg: 50
Very High - Min: 24, Max: 61, Avg: 37
Ultra High - Min: 21, Max: 52, Avg: 32

Q9000:
High - Min: 29, Max: 76, Avg: 50
Very High - Min: 23, Max: 69, Avg: 39
Ultra High - Min: 20, Max: 57, Avg: 34

DirectX 10
P9700:
High - Min: 32, Max: 79, Avg: 48
Very High - Min: 28, Max: 77, Avg: 45
Ultra High - Min: 24, Max: 63, Avg: 37

Q9000:
High - Min: 32, Max: 78, Avg: 49
Very High - Min: 28, Max: 77, Avg: 46
Ultra High - Min: 23, Max: 62, Avg: 38

(8) Supreme Commander Built-In Benchmark
Settings: Everything "ON", no AA, no Vsync
P9700: Min: 12.75, Max: 81.49, Avg: 36.867*
Q9000: Sim 8699, Render 7846, Composite 16545 / Min: 11.26, Max: 84.24, Avg: 41.89
* = different laptop

Settings: All Medium
Q9000: Sim 8658, Render 7945, Composite 16603 / Min: 10.96, Max: 93.89, Avg: 45.65

(9) Street Fighter IV Benchmark
Settings: All high, no AA, no Vsync, Parallel rendering ON, Texture Filter 16x, Extra Touch OFF
P9700: Score 11114, Avg 100.19fps
Q9000: Score 10930, Avg 97.53 fps

(10) ArmA 2 Demo Built-in Benchmark
Note on Q9000 - even at very high settings, all of the CPU usage on all cores were never anywhere near maxed out and rarely passed 50-60% usage. There is very little difference in relation to settings and performance and the game seems poorly optimised.
P9700: Very High - 21 fps, High - 25 fps, Normal (post process low) - 29 fps, Low (shadow normal) - 35 fps
Q9000: Very High - 22 fps, High - 26 fps, Normal (post process low) - 31 fps, Low (shadow normal) - 33 fps

(11) Last Remnant Demo Benchmark Tool
P9700 - Avg 57.6 fps, CPU is max on both cores.
Q9000 - Avg: 57.8 fps, CPU usage is high on most cores

(12) Resident Evil 5 Benchmark
Settings: 1680x1050, All High, No AA, Motion Blur On
P9700:
Fixed Benchmark DirectX 9 Avg 42.2 fps
Variable Benchmark DirectX 9 Avg 48.9 fps
Fixed Benchmark DirectX 10 Avg 44.9 fps
Variable Benchmark DirectX 10 Avg 48.7 fps

Q9000:
Fixed Benchmark DirectX 9 Avg 47.5 fps
Variable Benchmark DirectX 9 Avg 49.4 fps
Fixed Benchmark DirectX 10 48.0 fps
Variable Benchmark DirectX 10 46.6 fps

(13) Flight Simulator X FSXMark07 Benchmark Procedure Here
Settings:
Graphics: Custom (click customize at bottom):
Target Frame rate: Unlimited (slider all the way to right)
Full Screen Resolution: 1680x1050x32
Filtering: Anisotropic
Anti-Aliasing: ON
Global Texture Resolution: High
Lens Flare: ON
Light Bloom: OFF
Advanced Animations: ON
Aircraft: Medium High
Scenery: Medium High
Weather: Medium High
Traffic: Medium High

P9700: Min: 28, Max: 66, Avg: 45
Q9000: Min: 21, Max: 56, Avg: 37

(14) Batman Arkham Asylum Built-in Benchmark
Settings:
Fullscreen: Yes
Resolution: 1680 X 1050
Vsync: NO
Multi sample Anti-Aliasing: Disabled
Nvidia(TM) Stereoscope 3D: Disabled
Detail Level: High
Bloom: Yes
Dynamic Shadows: Yes
Motion Blur: Yes
Distortion: Yes
Fog Values: Yes
Spherical Harmonic Lighting: Yes
Ambient Occlusion: No

Physx Normal:
P9700: Min 21, Max 59, Avg 41
Q9000: Min 21, Max 88, Avg 40

Physx High:
P9700: Min 18, Max 74, Avg 32
Q9000: Min 18, Max 62, Avg 33

Physx Disabled:
P9700: Min 44, Max 121, Avg 85
Q9000: Min 43, Max 132, Avg 88

 

The problem is that LaptopNut used an arbitrary mixture of settings in his benchmarks, so it's impossible to do a true comparison. I don't know why he didn't just use standard High/Very High settings...

 

Nice start mate, i got near exact benchmarks in 3dmzrk06 as you using same drivers, interesting results it looks as though the quadcore performs better than the higher clocked C2D, which surprises me alittle, anyway keep them coming mate.

 

Yeah, that's what I realized as I was looking at his list. It's good for a "warm fuzzy" but for repeatability, it's near impossible. I am hoping to use built-in benchmarks or set up a repeatable pattern that can be used by anyone else if they so choose.

Good to hear we're at least similar! In some situations the quad core will outperform, but in general, for the time being at least, I still believe a fast dual core is the best solution. In a couple years that will most likely change. But considering most of what I play is older, I'm best to stick with dual core. My intention is to add a GTX 280m (hopefully modified with higher clocks by then) and a fast quad core in about two years time.

 

that's some godly cooling system. even better than some desktops.

 

I'm glad you decided to use standardized settings, as well as 64-bit Windows 7. I think it's important to benchmark in an environment that will be commonly used for the next few years. Benchmarking in 32-bit XP with arbitrary graphics settings just isn't very helpful in my opinion. Sure it might give you a general idea of what games will run at what settings, but it gives little to no info as to how that particular laptop configuration compares to others running a modern OS, which is what REALLY matters (to me, at least).

+1

 

i thought i might as well add my benchmarks here as well i have the same set up as htwingnut apart from i have the 320gb hard drive and im running vista64 sp2

Intel Core 2 Duo P9700 @ 2.8 GHz (25W)
nVidia GTX 260m with 1024MB DDR3
15.4" 1680x1050 Glossy Widescreen
320GB 7200RPM Seagate Momentus Hard Drive
4GB DDR3 RAM
DVD+/-RW
Windows vista home premium 64 sp2
Drivers are nVidia 186.81 WHQL.
3dmark06 10432

(1) so heres a bench mark from COMPANY OF HEROES i used the performance test that is supplied with the game, i ran the benchmark all on high settings with AA off ( i usually have it on x4) and at 1680x1050 & 1280x800 screen resolutions.
heres the results:

1680x1050



Min: 27.8 Max: 146.7 Avg: 63.6



1280x800



Min:40.5 Max: 223.5 Avg: 139.2



*NOTE* im wondering if there is some sort of fps cap with the performance test it seems strange that i got an average of 59 and a max of 60 but maybe that the way it is.
*NOTE#2* ok so i had vsync on and ive now updated the results with it turned off, makes a huge difference.

(2) ok next up WORLD IN CONFLICT

it ran the in game performance test at 1680x1050 screen resolution.
heres the results

1680x1050

all settings on high, AA off Vsync off, UI texture quality uncompressed, unit track length long, water reflection size 768 and all the extra boxes ticked on.

Min: 5
Max: 66
Avg: 30

all settings on medium, AA off, Vsync off, texture quality uncompressed, unit track length medium, water reflection size 768, if there was a choice of high or low i keep it at HIGH, all the extra boxes ticked on.

Min: 17
Max: 71
Avg: 35

so thats it for World in Conflict great thing the game still looks fantastic on medium settings.

(3) ok next up TOMBRAIDER UNDERWORLD

i did this benchmark by playing through the game on one of the later levels for half an hour and noting the MIN, MAX and AVERAGE FPS in 1680x1050 screen resolution.

all settings on highest with AA4 and Vsync off



Min: 47
Max: 93
Avg: 66

and this one i changed the settings to medium with no AA Vsync off



Min: 66
Max: 136
Avg: 89

great looking game and very stable

(4) ok next up is THE LAST REMNANT
i ran the benchmark for this game using the demo benchmark and had the screen resolution at 1680x1050



although the benchmark only gives the average FPS i noted after running the benchmark three times the highest and also the lowest fps as well.

Min: 21
Max: 101
Avg: 57.92



also note that i believe i get much more stable and higher fps with the real copy of the game probably averaging in the high 60's.

(5) ok heres MASS EFFECT as per requested by Pavel961

there isnt a in game benchmark/performance test so a ran the game for around 20 mins and noted the MIN, MAX and AVG FPS especially during battles using rivatuner
i ran the game at max settings with Vsync off and at the highest screen resolution allowed 1280x1024



Min: 41
Max: 74
Avg: 56

next FALLOUT3 also requested by Pavel961

again this doesnt come with a benchmark/performance test so i noted the MIN, MAX and AVG fps over two 15 in sessions using rivatuner, i ran two tests both at 1680x1050 screen res and the first at ultra settings and the second at high both with vsync off.



Min: 21
Max: 51
Avg: 42



Min: 42
Max: 71
Avg: 60

the game ran very smooth and only with alot of action did i find the frames going down to the min and only very briefly also they didn't really go very high either but rather hovered around the avg for most of the time.

 

mtness, seems like vsync is turned on, this would cap your fps. And by all means throw your benchmarks in here. More info the better. And with Vista is a good comparison.

 

ok wheres the option to turn vsync off mate?..is it in nVidia control panel? i couldnt see that option in CoH!

NOTE ok so i found where to disable it, it makes a huge difference, so i have to ask what actually does vsync do and should i have it disabled during normal gameplay?

 

Vsync basically paces the GPU to the screen refresh or rewrite interval. So it will not max past your displays maximum "refresh rate", which usually means 60fps. If you have what's called "texture tearing", usually is noticed by strange lines on your screen while playing, or sections of the screen are offset from where they need to be, setting vsync will eliminate this. Vsync isn't a bad thing, however, it usually maxes out at 60 fps, and if your machine can't manage 60fps, then the frame rate is cut in half to 30fps, then 15, etc.

 

For many of the games I benched, I did use all high settings or maxed out. The reason why there are a mixture is simply because I took a note of frame rates during casual game play to give people an idea of how well the Q9000 performs in different games. Those happened to be the settings I actually play at and I did not set out or plan to do any benchnmarks specifically. That is why you don't see any standard settings.

It was more a case of sharing what ever frame rates I happened to see during game play because I always have Core Usage and fps on display. So after hearing so many comments on how bad the Q9000 was I decided to give an overview.


However, now I am also on Windows 7 X64, I am running exactly the same 186.81 official drivers at the same resolutions so it is easy for me to re run any benchmarks that have been done.

I see no point in even considering XP SP3 anymore and I did say that I would bench on the new OS and in the other multicore thread, that's what I have done.

 

I think LaptopNut put in a lot of effort for benchmarking, and I don't discount them at all. I would like to come up with a set number of benchmarks and run them as common as possible, preferably with games that have built-in benchmarks. Or run a certain level or part of a level if needed. Problem with that is you have to get to that level before you can use it!

It would just be nice to see a good comparison. I think for newer titles you'll see benefit from the quad core, but for older games I think the dual core may pull ahead by a slim margin.

Either way, we can come to some concensus I'm sure. I recommend (and what I already started):

- 3DMark Vantage
- 3DMark 06
- World in Conflict Demo Bench
- Crysis built-in bench (GPU & CPU)
- Far Cry 2 built-in Bench
- GTA IV
- Company of Heroes built-in bench
- FSX
- Arma 2
- Batman Arkham Asylum
- Red Faction: Guerrilla
I'll think of more...

Let's PM and figure something out if that's ok with you?

 

yeap its a great idea and i really appreciate what LaptopNut and htwingnut are offering to do for the community, its really precious Ive got world in conflict and ill do the benchmarks for that as long as the real game still has them ill go check now, ill update to my original benchmark post when im done.

 

@htwingnut


devil may cry 4
Resident Evil 5
Street Fighter IV

 

That's a great idea, I have sent you a pm. I was already in the process of doing those benchmarks with identical settings and resolutions, I have already done World In Conflict and Crysis.

 

ok updated with world in conflict benchmarks for resolutions of 1680x1050 and 1280x800, still one of the most demaining games of played to date.

 

To simplify things, you guys should bench exclusively @ 1680x1050.

 

That's what I would have done originally but I am assuming they started benching at a lower resolutions as well because it is supposed to place more load on the CPU than the higher resolution would.

 

I can understand that thought process, but... I think the point of these comparisons should be to prove that the Q9000 (or any other lower clocked quad) can hold it's own against the C2Ds at every day gaming resolutions .

At the end of the day, which CPU has better performance at WSXGA+ is all that matters to us, yes?

edit: btw, I'm willing to fill in any gaps, like the Dawn Of War II benchmark.

 

Not looking to highjack this very useful thread, but I will be at some point collecting some benchmarks from here and using them to compare the GTX 260m to my HD 4850. We've done some runs v/s the GTX 280m, but I'm still curious for a direct comparison to the 260. I'll post the results in a different thread when I've collected a good number and link to it from here. Thanks for this thread. So many people throw around things as fact that have very few real tests to prove.

 

No such thing as hijacking this thread, the more results and different comparisons the better. My results are almost done and I will be doing the same benchs again when I am on the RTM version of Windows 7 X64 as well.

 

No, that would be great. In the end I'll be happy to tabulate all the data for an easier to read result. This would definitely benefit everyone looking for a new gaming notebook.

 

Fallout 3
Mass Effect
Neverwinter Nights 2

would all be good

For some reason the first two run ok on my laptop (Acer 8920) but NWN2 is unplayable. I'm due a change though hence the request.

 

Ok i own all three of those games, i dont think they have in game performace tests so ill have to give them based solely on in game fps, we have decided to stick with defult screen res of 1680x1050, as soon as power comes on ill bench the last two on the list and fallout3 ill have to reinstall first.

 

I own Fallout 3 and Mass Effect. I am on the very early levels near the start on both of those titles so I will test them later. It is a shame they don't have any in game benchmarks in those games.


I just benched Supreme Commander at high settings using the built in benchmark. I think this is a must as far as gaming benches go. I also found a game called ''The Last Remnant'' has a benchmark demo download so I will test that later. I have no idea about the actual game though.

http://tinyurl.com/ngnute

 

LaptopNut, you're on FIRE! I gotta catch up to you! Will do so as quickly as possible.

 

i have the last remnant and i cant remember seeing a benchmark tool, ill go take a look now, if it has one ill post the results in my benchmark post.
*NOTE* just checked 'The last Remnant' and couldn't see a benchmark/performance tool maybe its only on the demo.

 

Nice HTwing, plus rep

 

ok my benchmark post has been updated with tombraider underworld in game benchmarks, there was no tool for this so i played the game on a later stage for around half an hour noting down the MAX, MIN and AVG fps.
im downloading the last remnant benchmark demo and will post the results when im finished.

 

Rep laptopnut. He's done all the Quad Core (Q9000) benchmarking, I have to finish the P9700 stuff for comparison. I've been so busy with my kids I haven't had a chance!

 

If anyone knows of any other games that have a built in benchmark that are worthy of a comparison bench test, feel free to let us know. Once the W860CU is out, hopefully some one can do the same benchs and add those too.

 

Yes, please recommend other games to benchmark. I am going to figure out a way to do FSX repeatedly between machines.

 

Hi guys, good job! . Maybe I will be the one making the benchmarks with the W860CU , I have to take some decisions based in my personal research and the very valuable information found in this and other sources to finally choose the winner laptop for me.

Htwingnut could you add to your list Xplane 9 ? , Laptopnut made for me some tests with that simulator, would be great to compare.

 

Xplane 9 does not have a built in benchmark and due to its annoying dialog boxes reminding me that it was a demo and its disabling of the controller after 10 mins, it was very difficult to get a real idea of performance.

 

Regarding W860CU, personally, if you can wait a few months before getting your new notebook, then it'd be worth seeing what comes of it. However, keep in mind it will be new components and new architecture which can mean bugs. I'll take my NP8662 with proven track record, and put it up for sale in 18-24 months to supplement payment for my next notebook PC.

@redstone. I will put X-plane 9 on my list.

 

Is that Q9000 OC'ed? So far it looks like the results are very similar. Which games listed in OP utilize quad core technology and which dual core? I wonder what the results will be if the game is a cpu resource hog, that will at the same time not support quad core. In theory a higher clock on p9700 should make it run better. I've just ordered mine np8662 with P9700, this thread is beginning to make me doubt my decision. Q9000 is only 5 dollars more

 

Well, the whole point is to validate whether then Q9000 @ 2.0GHz can perform as good as a dual core at 2.8GHz.

 

My Q9000 was not overclocked during any of the benchs and overclocking is pretty much out of the question on the NP8662 since it is locked in the Bios. Set fsb doesn't work either, usually crashes.

Everything in those tests are at stock clocks for accuracy and a fair comparison, exact same GPU drivers are installed too.

I might add some benchs with an Overclocked GPU later on at some point though.

I think the reason why you see similar results is sometimes due to GPU dependency where even a 2.0 Ghz Core2Duo is enough to drive the GTX 260M in the vast majority of games.

The tested games that make the most use of Quads are Supreme Commander, GTA IV and Fsx.

As most of you know, GTA IV was designed with Quad in mind but Fsx was patched to take advantage of Quads in its SP1 update but originally only using 1-2 Cores. After that Fsx SP1 patch, my frame rates and performance in Fsx almost doubled, however, they spent a lot of time optimising and really taking advantage of the parallel processing but many Devs will not spend this amount of time to do the same.

If I have not mentioned other games that are multicore scalable or Quad optimised, feel free to post them up. I know some of the source engines that many games are based on will scale to multicores.

Other games such as World In Conflict, Company Of Heroes, far Cry 2 and Arma 2 will scale to multicore but in my experience, that does not always make much difference, if any at all.

 

Well I'd like to thank you both for doing this test. My other interest is the various mundane tasks that you perform in windows. I'm not talking about multitasking in this case, because it's clear that Quad will work better. However at one challenging task such as converting a large video file from one format to another, installing some large software program, rendering some big photo, etc. Will Quad Core perform that stand alone task better than dual core in addition to being better at multitasking? Will HDD speed be a bottleneck and they would perform the same? Is there any way to test it objectively?

Although it's a very premature observation, but I have a feeling that Q9000 will be just as good as P9700 in most of the games and better at some specific games and perhaps some future ones. I might even change my order specs to Quad because of this thread, so yet again thank you for doing this.

 

You are correct as far as I know.

Thanks in advance for running the tests to all, a general idea of fps would be fine for my purposes. For comparison I'll run some tests on the Acer as this might help show what improvement can be expected. I'll use FRAPS for consistency

 

^^^^ you brought up a good point mate, Laptopnut and htwingnut what program are you using to record in game fps, fraps or rivertuner, i personally prefer rivertuner as it doesnt seem to be as much of a resource hog but im happy to use either.
As soon as power returns ill run the last remnant demo bench as well as mass effect.

 

Right now I'm using the in-game benchmarks to report FPS. If given the choice, I'll use the in-game FPS if it exists. Then I'll use FRAPS. I registered FRAPS YEARS ago, and keep getting free updates. Nice.

And FRAPS does NOT reduce your frame rate. I still don't understand where that myth came from. **Recording Video** with FRAPS obviously will because it's hogging resources, but using it for viewing or recording FPS has no deficit to performance, and I really don't feel like going through the testing I did in the past to prove it (sorry no links or resources for that - it's been a couple years).

 

Ok mate well i think we should use the same program so ill use FRAPS as well
Now just need to power to return its been off for nearly 24 hours lol

 

I use Rivatuner and it is constantly running on my system showing me Core usage, HDD temps, fps in all games etc. It was also running during all of the benchmarks allowing me to note Quad Core usage, or lack of it.

 

Interesting updates, I think my prediction is coming true. Q9000 is on par with P9700 at most games and much better at quad core capable ones. So you haven't really answered my question about the general windows software applications, will there be any sort of difference in their performance?

 

Ok well as im already using rivatuner illl stick with that, at the end of the day there wont be a noticable difference in fps between the two anyway.

 

I am not 100% sure either way on this one but what I can tell you is that I think the HDD and RAM will be the bottleneck more than the CPU in any software installations. As long as you have a reasonable Core2Duo it should be more dependent on HDD efficiency and RAM but I am just guessing here.

One of the mundane tasks I do a lot on Windows is decompressing lots of zip files of varying sizes using power archiver 2007 and I noticed that even this uses all Cores, the 2010 update is specifically multicore optimised so I will have to update to that soon. The 2007 version must simply scale to extra Cores.

It seems like quite a few apps take advantage of the extra cores or are updated to do so, sometimes without even advertising because they were just designed like this.

Since Windows 7 is very recent, it will do its best to make use of the Cores, for example I notice Windows uses the more idle Cores for its OS tasks during gaming.

I do think it will do many mundane tasks very quickly. Maybe we can come up with some way of testing this out Core2Duo Vs Quad with either a benching tool or some other method to do something with a file of a large / set size. Even something as simple as a batch file that does something with a file would give some insight into this.

 

What did you get on this one at 1680 X 1050?

Maybe the benchmark is only on the Demo, that was also the case with the Arma 2 game.

 

that was the case mate the demo was the only one that had the benchmark but i downloaded it last night and have updated my original post to reflect the latest benchmark..... average of 57.92

@ htwingnut really nice job with the benchmarks mate looks really tidy and easy to read thanks to you and laptopnut for your hard work keep it up.

 

Interesting result, that game seems to be pretty intensive. If you have GTA IV, that would be another very interesting benchmark to do at the same settings in the first post that I used. Those are my actual gaming settings and a good compromise between performance and eye candy.

Don't forget to patch to 1.040 if you do a test though. Actually, a bench test before and after would also be useful. That is one of my favourite games. I think I will do a benchmark with my Realism IV 6.2 modded folder too.