AMD: Volcanic Islands R1100/1200 (8***/9*** series) Speculation/ Rumour Thread

Nemo

Newcomer
“The next generation of graphics cards – Volcanic Islands – is coming this year and shaping up nicely. When you name products after places, it leads to interesting thoughts about where to hold events surrounding that namesake product. Typically places that are cheap to get travel to internationally for a worldwide congregation are preferred, so I’m off to renew my passport on the off chance I’m headed to Reykjavik later this summer. But I sure as hell wouldn’t complain about being sent to Honolulu, either.” Rage3D

This could be huge news if true for AMD fans and could possibly provide a hint at the late launch of the much anticipated Radeon HD 8000 series GPUs. It was hinted by various techsites that the launch wasn’t planned to be late but it was delayed instead. The new VI (Volcanic Islands) GPU’s would be manufactured on a 20nm Gate-Last process through either TSMC or Common Platform Alliance.


Source: http://wccftech.com/amd-reportedly-...lands-hd-8000-series-gpus-2013/#ixzz2ScN9HKnA

20nm in 2013?
 

Wynix

Veteran
He seems to be speculating them being on 20nm, there is no indication from the source that 20nm will be used for the 8000 series.

The only claims of 20nm for 2013 that i have seen have all been based on speculation and wishful thinking.
 
Not confirmed by AMD, maybe on 28nm

You missed this part:

Update: While this news is not confirmed by AMD yet, Sweclockers reports that Volcanic Islands could be coming in the last quarter and feature the Hawaii Core that replaces Tahiti. The Volcanic Islands architecture although not confirmed could stick to 28nm.
Also the WCCF Tech article is getting its info from Rage3D and the quote from Rage3D is from the Author not AMD.

 

jimbo75

Veteran
AMD has stated on more than one occasion that 28nm was going to be a long lasting node. The chances of the 8-series being 20nm is for me, almost non-existent. It's far more likely it'll be bigger dies on 28nm, with 20nm 9-series cards coming at the end of 2014.
 

3dilettante

Legend
Alpha
Where are they going to find the extra TDP range on 28nm? Different substrates?
Not that 20nm has promised a massive improvement in that regard without finFETs.
 

liolio

Aquoiboniste
Legend
Wasn't there "noises" about TSMC jumping to 16nm process? (and GF going for 14/22nm hybrid).
/a serious attempt a putting a dent into Intel process advantage?
 
They should not be jumping but instead only shrinking the time gap between 20 nm and 16 nm, or something like that. Working on and releasing more products on 28 nm doesn't make any sense
 

3dilettante

Legend
Alpha
One of the data points I hope to see clarification with the 20nm metal layers and the 16/14nm finFet hybrids is where that puts them relative to Intel.

For a few nodes, Intel eased back on how quickly it scaled its metal pitch, while the foundries faced significant challenges by keeping more aggressive scaling there.
I'm curious at this point where that leaves the hybrid nodes relative to Intel's nodes in the same node range.
 
So Hawaii probably in Q4 2013 as a Tahiti successor… so Curacao got cancelled or merged with it? Maybe the same thing happened with Hainan too.

Where are they going to find the extra TDP range on 28nm? Different substrates?
Don't they have ~50 W headroom on top of Tahiti until they hit 300 W? I also wouldn't put it past them to go beyond 375 W for the dual-chip card, after all, they did go past 300 W with Antilles. In addition, given how long it took for a dual Tahiti to be released, if a dual Hawaii takes anywhere near as long, they might as well not bother and instead go for a dual card using the initial 20 nm chips (I'm assuming AMD will go 20 nm before the end of 2014).
 
Last edited by a moderator:

3dilettante

Legend
Alpha
Don't they have ~50 W headroom on top of Tahiti until they hit 300 W? I also wouldn't put it past them to go beyond 375 W for the dual-chip card, after all, they did go past 300 W with Antilles.

For a single-chip general product, I don't recall products that didn't leave a margin below 300, which cuts into the total.
Powertune also allows users to bump the 7970's TDP limit up, which eats into the perceived benefit of a power ceiling lift of the new product.

Niche enthusiast dual-GPU cards can do whatever they feel like.
 

3dilettante

Legend
Alpha
This depends on if you want to go by the numeric designation or the code name designation, and what those mean in the context of the latest marketing, rebrand, feature set, and codename shuffles.
 

Helmore

Regular
I wouldn't hold too many hopes for the TDP side and personally would expect for things to get even worse- with next cards even more power savings

Uhm, I'm not entirely sure what exactly you mean. Unless you mean that they actually lower TDP *gasp* with saying "expect things to get even worse", your sentence doesn't make much sense to me. Or maybe I'm just dumb.
 

jimbo75

Veteran
UT is on a constant crusade to make higher performance cards at the expense of TDP. The problem with that is that as the clocks and voltages go higher, the worse the efficiency gets and all you end up with is a space heater for quite small gains.
 

CRoland

Newcomer
What's the latest word on TSMC 20nm schedule? Is this accurate:

As it appears, the company decided to accelerate the launch of Fab 14/phase 5 by around two months so that to be able to produce chips on 20nm node in higher volumes already this year. Previously it was announced that TSMC's 20nm technology is scheduled to begin volume production at TSMC's Fab 12/phase 6 this year.
[...]
The construction of the facility is now [2013-04-04] complete and it will take from six to twelve months to fully equip the fab completely.
 

Zaphod

Remember
Veteran
What's the latest word on TSMC 20nm schedule?
According to themselves:
"20nm planar HKMG [high-k metal gate] technology has already passed risk production with a very high yield and we are preparing for a very steep ramp in two GIGAFABsTM," Sun said.

Sun noted that 20SoC uses "second generation," gate-last HMKG technology and uses 64nm interconnect. Compared to the 28HPM process, it can offer a 20% speed improvement and 30% power reduction, in addition to a 1.9X density increase. Nearly 1,000 TSMC engineers are preparing for a "steep ramp" of this technology.
 
Top