EIGHT "different" GPUs differentiated exclusively by clock speeds and memory configuration? With ranges of clock speeds on some of those parts? Is there some requirement that GPU naming conventions must be actively consumer-hostile?
I agree that it is over the top, but at least the different clock speeds / bus widths are separated into different models (with the exception of the 8750 :/ ). I'd prefer that to having the same model number and a lucky dip on what clockspeeds/bus width/gddr type you get. Especially when most laptop vendors don't provide that much detail on what you are getting.
Having this many variants is obviously necessary to meet different cost points and TDP envelopes.
Yes it's important to have consumer-hostile naming conventions in all graphics cards, just as it's necessary to have tit and arse graphics on GPU cards and boxes. How do you expect to stimulate sales from the 13 year olds?
"Notably, NVIDIA’s mobile parts do not feature NVIDIA’s equivalent technology (GPU Boost) despite the fact that they introduced the technology on the desktop first, so AMD is ahead of NVIDIA in this regard."
I'm pretty sure the Kepler-based Mobile cards have Boost; I have a Geforce GT645M and Boost is enabled by default.
That's what I get when I do some googling as well and in the Alienware M17x-R4 review it says of the GTX680M: "The core clock now runs at only 719MHz with a boost clock of 758MHz[...]"
...the lack of any news on the 8900M series is disappointing.
I am concerned that AMD usually leaves their parts ROP-limited, though here, with such a narrow memory interface, I suppose it's not really going to be a bottleneck.
Hopefully these cards end up coming with drivers. I've learned not to trust AMD on that point, so I'll be waiting for others to test these cards first.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
15 Comments
Back to Article
lowlymarine - Tuesday, January 8, 2013 - link
EIGHT "different" GPUs differentiated exclusively by clock speeds and memory configuration? With ranges of clock speeds on some of those parts? Is there some requirement that GPU naming conventions must be actively consumer-hostile?scottjames_12 - Tuesday, January 8, 2013 - link
I agree that it is over the top, but at least the different clock speeds / bus widths are separated into different models (with the exception of the 8750 :/ ). I'd prefer that to having the same model number and a lucky dip on what clockspeeds/bus width/gddr type you get. Especially when most laptop vendors don't provide that much detail on what you are getting.Having this many variants is obviously necessary to meet different cost points and TDP envelopes.
Beenthere - Tuesday, January 8, 2013 - link
Yes it's important to have consumer-hostile naming conventions in all graphics cards, just as it's necessary to have tit and arse graphics on GPU cards and boxes. How do you expect to stimulate sales from the 13 year olds?CeriseCogburn - Monday, January 14, 2013 - link
AMD is guilty of the sex box sales. Not nVidia.bstulic - Thursday, February 21, 2013 - link
http://todayswall.com/files/imagecache/1680x1050/d...usteg - Tuesday, January 8, 2013 - link
"Notably, NVIDIA’s mobile parts do not feature NVIDIA’s equivalent technology (GPU Boost) despite the fact that they introduced the technology on the desktop first, so AMD is ahead of NVIDIA in this regard."I'm pretty sure the Kepler-based Mobile cards have Boost; I have a Geforce GT645M and Boost is enabled by default.
Death666Angel - Tuesday, January 8, 2013 - link
That's what I get when I do some googling as well and in the Alienware M17x-R4 review it says of the GTX680M: "The core clock now runs at only 719MHz with a boost clock of 758MHz[...]"CeriseCogburn - Monday, January 14, 2013 - link
Ahh, the gigantic article error that will NEVER be corrected because of course it's a lie against nVidia.I don't see the fixed thanks response here we usually see.
Nope, no amd bias here. I was wrong all along.
roll eyes
silverblue - Tuesday, January 8, 2013 - link
...the lack of any news on the 8900M series is disappointing.I am concerned that AMD usually leaves their parts ROP-limited, though here, with such a narrow memory interface, I suppose it's not really going to be a bottleneck.
silverblue - Tuesday, January 8, 2013 - link
They currently read "4.5Hz GDDR5"; missing a G.iMacmatician - Tuesday, January 8, 2013 - link
From the spec PDF, its memory should be DDR3, not GDDR3.chucknelson - Tuesday, January 8, 2013 - link
Are any of these new 8000 mobile chips rumored to be in the next-gen consoles from Sony or Microsoft?A5 - Tuesday, January 8, 2013 - link
Some people think a modified 8700M part is the basis for the next XBox. The rumors around the PS4 seem to be pointing at a 6750-class part.silverblue - Wednesday, January 9, 2013 - link
Mars looks for all the world like the GCN replacement for the VLIW4 parts in Trinity.ZippityD - Saturday, January 12, 2013 - link
Hopefully these cards end up coming with drivers. I've learned not to trust AMD on that point, so I'll be waiting for others to test these cards first.