Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!
Discussion on Article:
Nvidia GeForce GTX 480 - Our First Date
I thought the whole point to a new GPU architecture was to totally best the competition's ultimate part? The GTX 480 doesn't remotely hold a candle to the HD5970 and barely tries to do much with the lower end HD5870. This is the Nvidia 5800 fiasco all over again..
What Nvidia always has going for it is stable Forceware drivers. ATi Catalyst has always been unreliable especially in video applications. ie. have two friends who work for Autodesk and they have so many horror stories with Catalyst users. Forceware.. never an issue of note. If anything, people want reliable hardware and drivers it talks to.
Also i dont like to much the batman physx on test on radeons. If the whole game wouldnt run without it than i say nothing, but for some fog and particles.
Without the useles physx u can enjoy the same good game on radeons too. Its quite misleading.
No, just in 3DMark 2006 the point whas high on 1600 scores (2560x1600). The ocerclocking results will be in some next articles about 480.
We test it early:
The way I look at it the Fermi and 5 series card both have their pro's and con's so it boils down to what the user cares about most, but that's how it always ends up being.
All the other results seem consistent with other review sites.
Try to read more attentively:
Just Cause 2 (DirectX 11) - version 18.104.22.168, maximum quality settings, Background Blur and GPU Water Simulation off, a double sequential ruin of the Dark Tower demo.
Not many reviewers have explicitly stated that they used the latest profile updates. HardOCP, Anandtech and Techreport did state it though. HardOCP: "Catalyst 10.3a Preview + Profile Update 3/19/2010" Anandtech: "For the AMD cards, we used AMD’s Catalyst 10.3a drivers along with the latest driver profile update." Tom's Hardware, Guru3D and Xbitlabs only mention the driver. Did they use the latest profiles?
According to Terry Makedon, the ATI Catalyst 10.3 CrossfireX Application Profiles (available since 19/3) added these new profiles to the Catalyst 10.3/10.3a Preview:
•Heaven – Crossfire profile
•Dark Void – Crossfire profile
•Singularity – Crossfire profile
•Unigine Heaven Demo – Crossfire profile
•Just Cause 2 -> Single GPU and Crossfire profile
•Battlefield Bad Company 2 -> Crossfire profile
•AvP -> Single GPU and Crossfire profile
In Xbitlab's test suite they would have affected performance only in Just Cause 2 and maybe Unigine, so no major impact if they were not used, but still, when it's not stated it makes me wonder if it's a minor oversight or (worse) an unawareness of this new Catalyst system with separate application profile updates?
Did Xbitlab's test use the newest profiles?
Otherwise, great review- way to get games across the spectrum- not just the "latest and greatest."
Where is the bench in the Bad Company 2?
In this game there is no possibility to record and play demos for the correct benching.
Yeah, one could see from the graphs that a 40nm GTX 480 is almost even less efficient "per watt of consumption" than a GTX 295 in terms of performance, let alone a 5970!
This suite is heavily optimised for CUDA.
And it launches tomorrow - April the 12th!
CS5 supports OpenCL - Yes. But it's been heavily optimised to perform well with CUDA.
Check out Nvidia website. As of today, they are showing off the exclusive support from Adobe.
Same # stream processors.
Similar heat or worse.
Similar power consumption.
Noise? Don't ask.
Such a deal!
Add your Comment
Enter your username and e-mail address. Password will be sent to you.