Friday, December 31st 2010

AMD Mobility Radeon HD 7000 Series Reality by Q4-2011

Shortly after finishing off the top-order of its Radeon HD 6000 series desktop discrete graphics processors (GPUs), AMD is letting out information about its first line of GPUs built on the 28 nanometer fabrication process. The first products carrying the Mobility Radeon HD 7000 series brand identifiers are notebook GPUs in the MXM 3.0 form-factor. AMD is planning an entire lineup top-to-bottom to address almost all mobile computing market segments. At the very low end of course would be its Fusion accelerated processing units with GPUs embedded. On top of that is what follows.

The lineup begins with "Wimbledon", an ultra high-end mGPU. It has a 256-bit wide high-speed GDDR5 memory interface, 2-4 GB of dedicated memory, and 65W TDP. The DirectX 11 GPU will be about 25% faster than "Blackcomb", the Mobility HD 6000 series flagship. This is slated for Q2-2012. Next up is the high-end "Heathrow" mGPU, which has a 192-bit or 128-bit (selectable between variants) GDDR5 memory interface, 1-3 GB of dedicated memory, up to 45W TDP, and 30% higher performance compared to "Chelsea". This is slated for Q4-2011 (this should tell you that Radeon HD 7000 series will be in existence towards the end of 2011).

Going down, there's "Chelsea" itself, with its 128-bit GDDR5 memory interface, 1-2 GB memory, 20-30W TDP, performance 30% higher than "Whistler", production starting in Q4-2011. Lastly, there's "Thames". This mainstream mGPU will have 128-bit GDDR5 with option of GDDR3, 1 GB memory, 15-20W TDP, and 100% higher performance than "Seymour", Radeon HD 6000 series' mainstream mGPU. Production for this starts in Q4-2011, as well.
Source: DonanimHaber
Add your own comment

37 Comments on AMD Mobility Radeon HD 7000 Series Reality by Q4-2011

#1
Over_Lord
News Editor
Gosh, they have info on mobility series before desktop series?
Posted on Reply
#2
wolf
Performance Enthusiast
I havent even heard or seen anything about new Mobilty 6000 series chips, and were already hearing how select 7000 series GPU's will be X amount faster? faster than something that doesnt exist yet?

am I the only one confised, maybe I missed a news report about the Mobility 6000's...
Posted on Reply
#3
entropy13
If you know your football (not handegg, or "American" football), Wimbledon being better than Chelsea...LOL (although that's not really impossible considering Chel$ki's form atm).
wolfI havent even heard or seen anything about new Mobilty 6000 series chips, and were already hearing how select 7000 series GPU's will be X amount faster? faster than something that doesnt exist yet?

am I the only one confised, maybe I missed a news report about the Mobility 6000's...
6000 series (Blackcomb) is 1Q 2011. They've been reported (or "rumored" about) already as early as September.
Posted on Reply
#4
Steevo
If any of you believe that AMD/ATi does not have samples of 7000 series by now your newb is showing.


It takes 2 years to plan, design, simulate, test, order samples, take delivery of samples, test more, refine, reorder, etc... for a GPU or CPU architecture change. Look how long we have been waiting on bulldozer.
Posted on Reply
#5
wolf
Performance Enthusiast
holy cripes my noob is showing :o
Posted on Reply
#6
largon
I don't think they have samples of 7000 series yet. Late 2Q11 to early 3Q11 would be my estimate.
Posted on Reply
#7
MrAlex
SteevoIf any of you believe that AMD/ATi does not have samples of 7000 series by now your newb is showing.


It takes 2 years to plan, design, simulate, test, order samples, take delivery of samples, test more, refine, reorder, etc... for a GPU or CPU architecture change. Look how long we have been waiting on bulldozer.
Erm they don't. They'll begin sampling in about end of 1H 2011 / beginning of 2H 2011. And Bulldozer is a CPU, not a GPU. And it was delayed on purpose. It was meant to be released in 2009 on the 45nm fab process, but then got delayed to 2011 to be on 32nm fab process because AMD believe that that would be the best time for it to be released. The same with Fusion. AMD were too late to the 45nm party, and they made the better choice moving on to 32nm.
Posted on Reply
#8
pantherx12
I think they have samples considering their desktop 28nm parts come out next year at the end of q2/ start of q3.

(basically when ever they actually get bulldozer out the door)
Posted on Reply
#9
jpierce55
I don't believe that can start mfg within 3 months of designing the final product. That would be fast to work out mfg bugs, get the lines setup, all the components stocked. I fully imagine they have the samples several months ahead.
Posted on Reply
#11
Unregistered
hayder.masterWTF another damn 256 bit wide memory
its non issue, because its laptop so it need to be as efficient as it can be
#13
bear jesus
I can't wait to see what AMD and Nvidia will be doing with the 28nm fabrication process :D
hayder.masterWTF another damn 256 bit wide memory
If it uses 6GHz effective GDDR5 would you really expect that to be a bottleneck on a mobile part?

I'm not sure how to do the math but I'm sure that's around/over 200gb/s but also Hynix has made GDDR5 chips that can run at 7GHz effective speed, how many hundreds of gb/s bandwidth is enough? :p
Posted on Reply
#14
Unregistered
MXM 3.0 form factor?!?!? Is it me, or currently there are only very few laptops that use that interface?
And besided MXM SUCKS!! It can only provide a max of 75W to the interface....They should make revision C asap.
#15
majestic12
Red_MachineWow, the Radeon 7000 series again. XD
I can't believe they're already cycling over -it seems like only a little while ago I was using a Radeon 7500.
Posted on Reply
#16
KainXS
there are actually a couple that use the MXM3.0 form factor its just that most companies, like asus and their GTX460M cards etc are using proprietary versions of it with pins changed or upside down PCB's etc so you can't swap with a normal MXM 3 card
Posted on Reply
#17
cdawall
where the hell are my stars
hayder.masterWTF another damn 256 bit wide memory
man its terrible 256 bit GDDR5 is so slow in comparison to what 512 bit GDDR3? oh w2ait no those are the same. current cards are not having bandwidth issues esp on the notebook side of things...
Posted on Reply
#18
_JP_
TAViXMXM 3.0 form factor?!?!? Is it me, or currently there are only very few laptops that use that interface?
Acer and Cevo (you know them as Eurocoms) laptops mostly...
TAViXAnd besided MXM SUCKS!! It can only provide a max of 75W to the interface....They should make revision C asap.
Yeah, it sucks. It sucks because companies are more interested in launching a new laptop every year with little improvement over the last one and with no room for upgrade. Why is that...oh right, PROFIT!
If companies saw the MXM form factor as a way of making profit out of laptops, don't you think it would be more evolved than it is now?
I got a pseudo-MXM card on my toshiba and I'm very upset about not being able to replace it with something better.
Posted on Reply
#19
OneCool
It makes some since to me.


Test the waters with low power samples before they go with the higher clocked and power hungry desktop series.... why not
Posted on Reply
#20
Rexter
Hey guys! I have exclusive new photos of the 7000 series!

Posted on Reply
#21
entropy13
No "HD" though. That's why they thought of that new prefix first starting with the 2xxx series, for this inevitability.
Posted on Reply
#22
largon
I wonder what will they do after HD9000...
Posted on Reply
#23
Zubasa
largonI wonder what will they do after HD9000...
Something like HD X800? :laugh:
Posted on Reply
#24
bmwm3gtr
Why AMD doesn't increase their bus width on gpu???
they always stick to 256-bit~
at least use like they produce in 2900XT which has 512-bit bus witdh~
with the incriment of the bus witdh i think ATI can beat nvidia gpu's~
Posted on Reply
#25
entropy13
bmwm3gtrWhy AMD doesn't increase their bus width on gpu???
they always stick to 256-bit~
at least use like they produce in 2900XT which has 512-bit bus witdh~
with the incriment of the bus witdh i think ATI can beat nvidia gpu's~
These are Mobility Radeon we're talking about here. They will go to laptops. We are not talking about discrete GPUs for desktops.
Posted on Reply
Add your own comment
Apr 25th, 2024 03:13 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts