NVIDIA GeForce GTX 480M, Fastest Notebook GPU Yet

In March of this year, NVIDIA launched their GeForce GTX 480, aka Fermi desktop graphics card and though this new killer GPU is both big and hot, there was little question it offered record-breaking performance.  Almost shockingly, only a calendar quarter later, they're now ready with that card's notebook-targeted variant, the GeForce GTX 480M.  We have one of the very first notebooks to hit the market in-house with the new GeForce GTX 480M under its hood.  Fittingly, NVIDIA turned to Clevo to ODM a machine that would house their new mobile gaming frame-rate crusher and it took residence in the Clevo D900F quite comfortably...

NVIDIA GeForce GTX 480M, Fastest Notebook GPU Yet

Via:  HotHardware
Comments
la_guy_10 4 years ago

Great review Marco hopefully we will see a 460 derivative that will hit the "sweet spot" with regard to performance and heat. Good to see Nvidia back anyhow.

ChrisMorley 4 years ago

This design is getting so long in the tooth - Clevo needs to get some new material already!!

silvery_wing 4 years ago

O my! No way!


We've got a NVIDIA GeForce GTX 480M paired with a desktop i7-940 with 130W TDP and $600 pricetag quad core CPU outperforming an HD 5870 with the slowest mobile i7-720QM with 45W TDP in the G73. That's fairly impressive. Also, look at the vast array of games tested! And all with nVidia's PhysX. The GTX480M has shown its true muscles by beating out its 6 month old rival in such a fair and unbiased test. This is a major breakthrough in GPU technology. I hope we see more like this soon.

 

[/sarcasm]

FYI, in the TDP required for this one 100W card, you could have fit two HD5870s in Crossfire which would totally demolish this thing. And still cost less. This is $1000? The 5870s are ~400 apiece.


I wonder how much Nvidia paid for this review.

Dave_HH 4 years ago

Hey Silvery, I don't appreciate your accusations here at all, though you're entitled to your opinion. The Core i7 720 in the G73 has a turbo mode speed of 2.8GHz and all of the game tests we performed are highly GPU limited, so CPU bandwidth isn't going to matter much, if at all. We stated that clearly as well. Finally, the only game test that used NVIDIA PhysX at all was 3DMark Vantage and we noted that clearly too.

We also boldly and clearly called NVIDIA out for it's 100 Watt TDP and the fact that two 5870s (which are 50 Watt TDP actually, you're wrong there) could be utilized in the same thermal envelope.

Do your homework and actually READ the article before casting wild accusations.

sheltem 4 years ago

I'm sorry but I have to agree with silvery_wing here, this review is really really awful.  If were to play along with your logic, 1920x1080 is still too low of a resolution to bottleneck the GPU.  Why did you even include the lower resolution tests, which are heavily biased towards CPU?  You should have used an external monitor for higher resolutions to minimize the CPU differences.  

And that is besides the point.  The whole concept of this review is flawed to begin with.  The Asus G73 and Clevo machines are two entirely different machines.  Isn't the whole point of the scientific method to test only one variable?  

One enterprising user from notebookreview modded a Mobility 5870 and put it in his Clevo D900F.   According to his numbers, the Mobility 5870 scored 9277 in 3dmark vantage, while the Geforce GTX 480M scored 9277 according to notebookcheck.net.  Granted the modding process is not for the faint of the heart but this doesn't excuse you from posting this disingenuous review. 

I got my numbers from the following links:

http://forum.notebookreview.com/sager-clevo/491999-first-benchmarks-d900f-running-mobility-radeon-5870-a.html

http://www.notebookcheck.com/Test-NVIDIA-GeForce-GTX-480M-Fermi-im-Clevo-D901F.32155.0.html

Dave_HH 4 years ago

On a side note (because rants like this always have a way of getting my goat), I just re-tested the Clevo machine with Turbo Mode AND hyperthreading disabled on the CPU. I tested HAWX (the least disparity in test results) and Far Cry 2 (the widest spread in test results), neither of which employ PhysX of course, and my results didn't change even a single FPS. I literally got the same score, with in tenths of a FPS. See, in high-end graphics testing, at a certain system performance threshold, a few extra CPU clock cycles mean nothing or damn near nothing.

bob_on_the_cob 4 years ago

I love you Dave Big Smile

Also Nice review. That is some heck of a lot of power draw. Less than an hour on a 12 watt battery. But hey if you are in a college dorm or something and want a desktop replacement then I see nothing wrong with this. Just don't expect it to be something that it's not. Which is a laptop you bring everywhere with you. I was not expecting Nvidia to bring the 400 cards to laptops for a while at least.

InfinityzeN 4 years ago

As much performance as that thing has, I don't think I could buy one for myself.  Seriously, 100watt TDP in a laptop?! Gives whole new meaning to 'Hot Hardware'.  How hot/loud was it Dave?

dlim783 4 years ago

Yeah, that gaming laptop with gtx 480M looks really nice! But can't release potential performance!

Post a Comment
or Register to comment