New Posts  All Forums:Forum Nav:

New 3dmark03 High for my F2

post #1 of 36
Thread Starter 
post #2 of 36
ok.... you better start explaining cuz I only got 2609!!! and I believe we have about the same setup. btw... what HD drivers are you using because my HD score went down since I installed XP Pro
post #3 of 36
Thread Starter 
Using Cat 4.1 and I turned all my 3D settings to fastest.

I also have my R9600 @ 390/230 (core/mem).

HD drivers? You mean harddrive?

The comparo system in my benchmark is an Asus
post #4 of 36
just curious.... what drivers are listed for your ide controller in device manager, also what driver is listed for the wireless network card. It appears I can use 3 different drivers for each. Since I installed XP Pro, my hd score went down, but not sure if its the drivers, or programs running in the background.
post #5 of 36
Thread Starter 
Intel(R) 82801 DBM Ultra ATA Storage Controller - 24CA

date: 7/2/2001
ver.: 4.0.1001.0

Keep in mind my HDD is the 60G 5400RPM. It also has 16mb Cache. Your 80G is a 5400 w/8mb.
post #6 of 36
XSternMinator, any luck in breaking your 2830 3dmark03 record?
post #7 of 36
Thread Starter 
Nope, haven't had time to mess with it.

Too busy playing CoD
post #8 of 36
Thread Starter 
Yeehaw! 2891!!!

M10 @ 400/235

http://service.futuremark.com/compare?2k3=1974309

Getting very very very close to 2900
post #9 of 36
X, are u running a PM 1.5 OC'd at 1594MHz or do u have a PM1.6?
post #10 of 36
Thread Starter 
PM 1.6

The M10 is OC'd to 400/235. Stock on F2 is 350/210.
post #11 of 36
That is weird...if the stock on the F2 is 350/210, then it would be a M10P rather than a M10
post #12 of 36
X, what other tweaks did u make besides OC'ing the Graphics? Are u using the catalyst drivers or the Omega?
post #13 of 36
Thread Starter 
According to Anandtech, M10 is core with less than 333, and less than 64MB. Pro is >333 with 64 or 128MB. Pro Turbo is 400 with 128mb I think.

OC'd Video, Cat 4.1, defragged and all settings on fastest.

http://www.anandtech.com/mobile/showdoc.html?i=1951&p=5
post #14 of 36
Can you tell me how permenant is the freq. settings modified by ATITool? Does the system ever reset those settings? Is it stored in a ROM somewhere on the video card, or is it only stored in HD and must be loaded each time the system boots?
post #15 of 36
Also, do u have to buy 3dmark03 to use online score reporting?
post #16 of 36
Thread Starter 
3dmark is free, but some advanced features are only found in the Pro version, which is $19.99 I believe. Some of the Pro features are resolution change and AA, Aniso etc...

As for ATITool, you can set the option for it to load during boot, so the frequency mods stay permanent. Another good thing about ATITool is that it lets your set 2 different profiles for 2D and 3D. For instance, I leave my card on default 250/210 on 2D ( figured there is no point accelerating 2D) and whenever I run a 3d prog, such as CoD, the frequency automatically changes to the 3D profile (290/230).
post #17 of 36

ok, I took the plunge

Just couldn't resist running off and downloading ATITools myself. Sounded fairly harmless, and was. Actually scored a 2913 on my first effort, overclocked at 389/237. Really couldn't find any doc on the software, so I was sort of using the SWAG method. You can see the results at http://service.futuremark.com/compare?2k3=1978051
My question is, is there any sort of trade-off or problem you could possibly incur by overclocking your graphics-card? Maybe a silly question, but when things just seem to good to be true you just have to ask.
post #18 of 36
Thread Starter 
Overclocking is an interesting debate and you can find plenty of info on the web that discuss the art.

My personal opinion is that as long as you do it within reason, and leave enough buffer room for just in cases, it should be harmless. If the reported MAX by ATITool is 412/245, then I'd clock my card at least 5% below max just to be safe. Hence why I am running my card at 390/230.

The truth of the matter is that chips are pretty much the same when they roll out the factory. Take for instance, a P4 1.6 and a P4 1.8. Architecturally speaking they are identical, with the only difference being speed. The speed is determined by series of tests on the chip's tolerance. Then Intel proceeds to lock the multipliers.

The idea is that a chip with greater tolerance at higher speeds would be branded as a 1.8, as opposed to 1.6. Technically, they are the same chip.

Hence sometimes you can get a 1.6 that has enough tolerance to run at 1.8, and in that case it doesn't hurt to run it at 1.8 because it is basically the same chip.

There's an Anandtech article on M10 classifications, having M10 w/at least 64mb and core at 333 makes the chip a Pro, which means that the F2 and G1 are at least M10P. The Pro Turbo is at 400/240 w/128mb. technically, same chips but the vendors can choose where they want to brand the M10, at pro or pro turbo etc... Basically if you've got a pro, there's a good chance it will run at pro turbo speeds with no problem.

In my experience, MEM speed is the bottleneck of achieving higher speeds. ATI cores generally have good tolerance, but their MEM speeds are finicky.
post #19 of 36
Without buying the Pro 3dmark03, can u still submit benchmark scores?

Also, you mentioned that you adjusted the settings to highest performance/lowest quality. Did you mean in 3dmark03 or in the display settings? Doesn't 3dmark03 force all their benchmarks to be run at the same settings, in order to keep the results consistant? Thanks.
post #20 of 36
Thread Starter 
The driver still allows different texturing methods (ok this is where I cannot be 100% certain) and so that there's still some margin for performance at the expense of texture quality. The adjustments are done at the driver level.

3dmark forces certain things like XGA resolution, vertex shaders etc... but not texture quality.

And Yes, we el cheapo benchmarkers have been submitting scores without paying
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Other Brand Notebook Reviews