M17X vs M17 Benchmark / Frame Rate Thread

Now that people are finally getting the new M17X I'd like to see some actual benchmark / frame rate comparisons to the M17. Some of the threads I've read are somewhat misleading because some people are overclocking / different processor, etc... and we're not really seeing a true comparison. Can we get people to post benchmarks / frame rates and what settings (computer specs) they are using so we can see how the two really compare.

 

i'm throw some stuff up also from my whitebook to compare when some scores come up

 

From initial reports I've been reading it doesn't seem to me as if the new m17x trumps the current m17 that badly. Granted I've only been reading numbers on the nikon pic thread.
The M17x should get higher marks, but I think the m17 will stay pretty competetive. At least I'm guessing it will.

 

in the right hands it will demolish the m17. a single 280m heavily overclocked and overvolted, is 10 percent slower the the 3870 crossfired. so depending on how sli scales, we will see what's up.

 

I'd be glad to contribute on behalf of the M17. I know it can't beat the M17x, but close is good enough. I can't justify the hit I would take selling my M17 and then buying a similarly configured 17x.

Vantage: P7,989
3DMark06: 16,133

I'll get some FPS stuff up tomorrow.

 

how high can your cpu to??????

 

Btw, correct me if im wrong but doesnt physx boost Vantage results for nvidias ?

 

And Ill add some D900F benchies when I get my system to compare the single (perhaps oced single) 280m vs sli 280m vs crossfire if you want?

 

thats exactly my view point as well

i will keep mine until the next refresh

 

nvidia's vantage will go up if physx is enabled because it will make the gpu do the cpu's job. just see they score and if the cpu passed 14,(depending on the overclock/gpu) tell them to turn off physx.

(my qx9300 highest was 12,500 i think. i'll give them the benefit of alittle higher because of the fsb raise if the bios should bring it up higher then mines.)

 

So does the PhysX have the same effect in real world gaming, or does it just inflate Vantage scores?

 

Real world not much.

 

well i need a game that heavily depends on physx so i can test it.

i am wondering the same thing cause the gpu can render what the cpu can do 4 times faster. but if the gpu is at full load what will happen to what the "physx" will it go back to the cpu or will the gpu be bottle necked worse. i know for desktop you can set what gpu you want to render physx. like ex
2 295 in sli and a 9800 gtx for physx.

but that's not availbe to notebooks.

if someone can provide a game that heavily depends on physx, i can test it and let you and me know the results.

 

i just thought about it but the only real benifit is RTS games. if physx can make the gpu do the work of the cpu then, we can go from 40fps to 90-100fps. but that's from obersavation from vantage.

i'm busy right now but i have world in confilt. it's doesn't support physx but i'll give it a shot later on.

 

From what I've heard, physx is a complete joke. Next to useless in real world gaming.

 

Yep but bumps up vantage over ati becouse of it.

 

At the end of the day, who cares about benchmarking programs? All we REALLY care about is gaming performance. (And some of us care about showing off high 3DMark scores)

 

Had a debate for that in one thread. The reason "we" use benchmarks like 3dmark and vantage its because it easier to compare scores instead of buying games since lets face it not all people own all the different games you can benchmark.

I do agree that benchmarks like 3dmark and Vantage not fully reflect on the gaming performance of the card but its still a decent way to compare systems if you dont own a lot of games. Especially the games people usually compare, Crysis/Far cry 2 for eg.

 

For that reason free demos with built-in benchmarking tools (Crysis, World in Conflict, Devil May Cry 4...) are very useful.

 

does fear2 demo have a benchmark???????????

 

Not all people have the connections or the time to download them though, not to mention isp bandwidth limits etc. Lots of reasons why people prefer 3dmark over benchmarking 10 different demos for people to compare.

 

That's true. But when you can download the demos, it's very useful.

 

Yea in this case it wont be an issue since people in this thread want to contribute in this so they will download the demos or have the games for benchmarks. Im talking why people in general prefer 3dmark

 

I've got a bunch of overclocked benches in my sig.

 

There in lies a whole other issue...all of anothergeek's benchmarks are at 1680x1050. I run my real world benchmarks at 1920x1200.

 

And there lies the benefit of having dual 256 bit GPU's. Have you figured it out yet?

A single 280M will take a larger FPS hit than CF 3870 will going from 1680 x 1050 to 1920 x 1200. If you compare benchmarks at 1680, the 280M looks very impressive against CF 3870. We're talking within 10%. At WUXGA, it's probably 20-25%. I can do 1080p benches, not 1920 x 1200, but the FPS hit is already pretty noticeable. For example, crysis goes from ~32 to ~28 FPS.

280M sli should be able to do ~20% more performance at WUXGA than I will at WSXGA+. 4-8x AA for all but Crysis will run fine.

 

what drivers did you use for crysis anothergeek?

at 1680*1050 the 3870 crossfired and a 280m will be close. but let's throw it up. i can do a single 9800m gtx which is equal to the single 260m. (don't worry sli is coming soon)


i have vantage, 3dmark06, dmc4, farcry2, crysis, warhead, ut3, and wic installed that have built-in benchmarks.

we can do single, sli and crossfire runs but i would like the OP to organize them in the first post if that isn't a problem. i will run stock and overclocked. and i still have some runs of my quad+3870 CF and single dual core. i can throw up.

 

Warhead, 1920x1200, DX9, ENTHUSIAST, 'Call Me Ishmael': Avg: 17.143 - Min: 8 - Max: 25

CPU OC'd to 3,152Mhz.

 

can you add your settings are you stocked or overclocked.

 

Ati cards favor very high, Nvidia cards favor high. Then there's the problem of CF 3870 just not scaling Crysis.

We should use DMC4, FC2, Vantage, and SM 2.0/3.0 scores in 06.

Here are some of my benches... everything is overclocked, 185.85 for FC2 and DMC4, Dox 182.46 for 06, 185.20 for Vantage.

 

are there any results with a sli gtx 260/280 from vantage and crysis? i really need it for my decision.the 280 sli costs 400euro more and i cannot decide.

 

No Vantage or Crysis, but you may find this benchmarks interesting:
http://www.notebookjournal.de/tests/review-cyber-system-qx18-nvidia-gtx280-sli-923/6

 

what's the stock clock of the 260m. i'll run that first and at what resloution.

farcry2
stocked 9800m gtx
[email protected]
1920*1200
ultra high
min 16.91
average 34.03
max 54.15

equal to a underclock 260m with [email protected].


edit gpu overclocked
620/1620/940
min 25.07
average 39.89
max 64.85

 

I can run FEAR 2 demo at max settings 1920X1200 with no probs.

Lol I have the real game but cant even DL it because of steam and my slow net connection.

 

Im Dling 9.6 drivers now to see if they make a differnece in crysis

 

We need some M17x benchmarks folks.

 

it made a big difference in crysis for me 20-30% gone from mid to upper teens on gamer to upper 20's

 

FRAPS 60 second benchmarks.

Crysis Warhead, 1920x1200, DX9, ENTHUSIAST, 'Call Me Ishmael': Avg: 17.143 - Min: 8 - Max: 25
CPU OC'd to 3.15Ghz.

FarCry 2, 1920x1200, 2xAA with everything else MAX: Avg: 37.65 - Min: 20 - Max: 56
CPU OC'd to 3.06Ghz.

Vantage: P7,989
3DMark06: 16,133

 

Your forgetting that on the native res of anothergeeks, his cards will yield better performance and look exactly the same as a full hd notebook, his results are "real time" for his system, just like mine are for mine and yours are for yours.

 

yeah it's getting confusing with different resolutions. but here's my trump card. Lol.
[email protected]. 3870x2 stocked. 9.5 driver. 1920*1200
http://i44.tinypic.com/20usrqr.jpg

 

Well, I've participated in a dozen of these benchmarking threads over the years. They never yield anything but a pissing contest. I agree there has to be a common denominator with screen resolution.

BTW, what game is the 'Trump Card' benchmark from, anyhow?

 

farcry 2. i showed the link.when i say trump card. it is the very good score i got from the whitebook.

 

I posted some DMC4 benchmarks on the benchmark thread in here. SLI 280m is a really powerful setup and may be the best setup as of today.