Sager NP5796 Benchmarks

w/ Intel X9100, 4GB (2GB x2), nV GF 9800M GT w/512MB, 200GB (72).

Vista Ultimate x64.

Configured with the updated parts to the NP5793 review unit NBR did here differences are just the HDD size and vista edition.

 

Very nice. When will there be a benchmark of the new Sager NP8660?

 

Not bad......similar to a X90 + 88M?

 

That's not much of a performance difference compared to the 8800M GTX, but I think (hopefully I'm right) that it's cheaper than a 8800M GTX.

 

Hey justin, turn off some of those procceses (sidebar) and stuff and run it again?

 

It wouldn't give you 1000 points more.

 

According to the Bench i would consider to keep my RU and change to 8800M instead of buying a new TU.....

 

I got P4380 2 days ago

 

Thats why I was wondering if all the extra procceses like sidebar were turned off what the score would be like? Maybe a little higher? It does seem pretty low - it must be the gfx card isnt quite as good as the 8800M GTX like you say.
I think I get like 9300 pts with my CPU at 2.6ghz.

 

Methinks bwhxeon makes a valid point worthy of investigation.

 

What score can a similar system with 8800GTX get?

 

I have tested 3DMark on my system and even if I disable every useless process I don't get more than 100 points extra.
If Justin shows us that score, then it means that it's the right one.

 

Kind of a crazy day, will try to post some OC'd results up ASAP.

 

at 2.8Ghz I get 9800 3d Marks..
Look at my system config... Pretty similar I think?

 

>.> it would be easier for us if you could just list the specs on the 9800m gt. gpuz shows the same clocks as a 8800m gtx. Also if you overclocked the cpu to 3.2ghz, I bet the score goes up to atleast 9800. That would match my score for my cpu + you are running vista which is a little slower. That could throw of your score by 200 points compared to mine running in xp.

 

Surely the 9800 would show much better results in a few months time when better drivers are released though...right?

 

As I understand it, Nvidia uses the same drivers for its cards (probably oversimplifying here). The same driver would work on the 9800M as it would for the 8800M. Certain drivers may work better for certain cards.

 

That happens, it's all up to nVidia.
I remember that when DOOM 3 came out the right card to have was a 6800GTX, but after a while you could do with a 6600GT (the drivers were great).

 

>.> well I've come to another conclusion, which some people buying the system
may not want to hear....my conclusion is that the montevina cpu's are nothing but slightly adjusted current penryn cpu's. They have a faster fsb and a little lower tdp, "not the extreme processor though". Other than that there seems to be almost zero benefit. Justins score can be explained by using vista and not running the cpu overclocked. His cpu at 3.0 ghz is only slightly off from mine at 3.2 ghz.
At this time, it might be in peoples best interest to wait till next year for Calpella when there will be real improvements. >.> this knowledge in hand Justin it would be really interesting to see if you could get sager to offer some kind of upgrade program again for the 9800m gtx.

 

As Hauser explains in this post, the 9800m GT is not based on new architecture. It's based off the G94b chipset which apart from the die size should be more or less identical to the desktop 9600 GT (although the clocks seem to be much slower) and so should already be well supported by existing drivers.

 

That is well known, isn't it? But not everyone can wait until next year, whenever Intel will decide to offer the new platform.

 

The biggest benifit of MV platform for me is the support of mobile QX, but the price.......

 

They might not be able to wait, but they should at least know that a montevina system isn't anything special. Easier to justify buying a cheap santa rosa system now and upgrading to much better laptop next year. There are going to be a lot of sales in the next few weeks which might make it even more worth waiting just a tad longer to buy a laptop. >.> I'm looking at this from the best value for your dollar.

 

The biggest benefit is that it's a little bit faster and that it's a little bit down on power consumption and that the laptops are a little bit better overall.

bwhxeon, yes, Montevina doesn't offer much, but if you want better than the usual (current) Santa Rosa generation, then you have to get a Montevina laptop. Sure, the next generation is going to be great, but it's a long way until 09 (maybe AMD will have a surprise, though I highly doubt it).

 

I'm not sure a the momemt if it is really the 55nm version, but the score indicates that the G94 GPU is used for the 9800M GT and so all rumors were correct. It is the same GPU as in the desktop 9600 GT. The G94 is a "physical" halfed G92, which means that the architecture is very same to the 8800M GTX.

The clocks are very conservative, but there will come a 9800M GTS with the same GPU an quantity of active units, which will have some higher clocks.

But the good thing: this baby will be also be built into 15,4" devices.

 

What I'm seeing from this benchmark is not that there's anything wrong or deficient with the new CPU's, just that the 9800m GT might be a tad less powerful than the 8800m GTX (although I'm eager to see further benches).

In fact if you check Donald's posts here or the Nvidia site it seems that Nvidia have defied their own naming convention on this one and the 9800m GTS is in fact the weaker out of it and the 9800m GT.

 

>.> I didn't even think about AMD. There griffin system seems very promising.

Edit: Gophn or Justin do you have any info about the new AMD clevo units supporting the XGP external graphics card system?

 

I don't think that there was a single person on this planet to think that the 8800M GTX would suck compared to the 9800M GT (suck = underperform).
From my point of view this benchmark is as valid as it can be (as it was done by a reliable source and because it's not the fastest card in the new range).
Stop worrying about this 9800M GT. At least it's good.

 

>.> you got it backwards the concern was that the 9800m gt wasn't = to the 8800m gtx

 

Have I? I've only read a few posts about it before this thread.
Anyway, as you may know, 3DMark is not very reliable, other than getting a bigger e-*****. Let's wait for some game benchmarks, as those tend to reflect reality. I think that despite the score, it might underperform in some games (I have a hunch).

 

^^^agreed (to short)

 

Well, you have to remember the 9800m GT is a new card and the drivers are pre-mature. Wait a month and you should see scores jump at least 250.

 

use it in XP and it should gain a few hundred more.

 

gophn do you know anything about clevo's amd laptops supporting amd's external graphics card system?

 

not sure about that yet.

We wont know about the external videocards until they are available for testing.

Mobile gaming gets more and more interesting as the days go by.

 

Here are some scores OC'd. Will try to get to the others tomorrow.

 

The scores are quite low..

 

3.5G, quite good. And seems to be BIOS OC is still working in 570TU (11.5 --> 13).

 

I was referring to the 3dmark score.

 

Thanks for these benchmarks. The 9800mGT does seem to be the bottleneck. I'd be interested in seeing some FPS comparisons with various games.

 

yeeeshh... yea not impressed... like i say my t7700 @ 2.8 gets the same score... gotta be the gfx then.

 

Justin what driver version are you using?

 

Please throw in crysis and let it lose or even COD4. Am not not much on 3dmarks accept to see how stable my system is after OC'

 

There is a GPUZ Screen shot in my POST #1 showing the driver version

 

Is it possible you can try a driver like 175.95 or 177.39?

 

Justin,

When will you get any new puma based laptops?

cheers
Hawk

 

How does 3d Mark Vantage's scoring system compare to 06 or 05? I don't exactly remember a letter in the final result...

 

p is for performance. but because there aren't alot of reviews done with vantage yet its harder to see exactly what the score means.

 

BTW Justin, since you've already had a go at overclocking the X9100 could you let us know how it responded? ie. did it create any noticeable problems with heat, etc.?

 

Mark over at Killernotebooks.com is pushing the edge, as usual. He offers the Executioner SRX version of the Clevo M570TU with the Penryn QX9300 2.53 Quad Core. For those of you with $$$$ to burn for bragging rights check it out:

http://www.killernotebooks.com/executioner_sr/executionersr_order_page.html