Bookmark and Share


Advanced Micro Devices has been rumored to prepare an improved version of its Radeon HD 7970 graphics card in order to better compete against Nvidia Corp.'s GeForce GTX 680. What has not been known so far is that the new version of AMD's top-of-the-range single-chip graphics solution will also feature a new chip revision with higher frequency amid lower voltage.

Known under Tahiti XT2 code-name, the new flagship graphics processor from AMD will get 1.10GHz (1100MHz) clock speed, a 18% increase from the default 925MHz; as well as lowered core voltage of 1.020V, a 14% decrease from 1.175V implemented on the Radeon HD 7970 graphics card. Thanks to lowered power consumption and increased clock-speeds, AMD will probably use the code-named Tahiti XT2 chip for its flagship dual-chip Radeon HD 7990 solution, reports Ocaholic web-site.

In case the unofficial information about the Tahiti XT2 is correct, then AMD will not only be able to significantly boost - between 14% and 18% - its yield of Tahiti chips in general, but also offer about 20% of performance by increasing clock-stand alone. Given that AMD already supplies three free AAA-type video games with the HD 7900-series, the product family with higher performance promises to be successful.

AMD did not comment on the news-story.

Tags: AMD, Radeon, Southern Islands, ATI, cgn, 28nm, Tahiti


Comments currently: 9
Discussion started: 06/18/12 06:12:38 AM
Latest comment: 06/20/12 12:01:33 AM
Expand all threads | Collapse all threads


Of course they are since I just bought one three weeks ago.
3 1 [Posted by: Hawkeye666  | Date: 06/18/12 06:12:38 AM]
- collapse thread

It's the same card just do what they did.

Lower voltage, increase clock.
2 0 [Posted by: seronx  | Date: 06/18/12 06:55:54 AM]
It may not be as simple as that. There is a chance the TSMC 28nm process has changed, and a better implementation was derived for this design.

18% overclock and 14% lower voltage will make a huge difference in the final chip. The power consumption will go down approximately 10%, and the performance would still be 18% higher! If you over-volt the chip to 1.175V then the overclock should be insane.

Also the article claims that AMD boosts their yield, which can only mean an improved implementation.
2 0 [Posted by: redhavoc  | Date: 06/18/12 08:58:25 AM]
Same card but the chip is not exactly the same. Latest chip was a tweaked one to improve performance and reduce power consumption.
0 0 [Posted by: pogsnet  | Date: 06/19/12 11:50:02 PM]
Don't worry even those who buy this card won't have the best AMD card for long. I doubt this new revision will overclock much more than 100mhz over the current model, if that. HD8000 series has already tapped out and is rumoured to enter production for Q4 2012, ready to launch for Q1 2013.
1 1 [Posted by: BestJinjo  | Date: 06/19/12 08:13:09 AM]

A big blow to Nvidia fanboys........
2 0 [Posted by: tks  | Date: 06/19/12 03:53:57 AM]
- collapse thread

The hardcore NV fanboys have already purchased GTX670/680 anyway. And those who haven't still are probably waiting for GTX660 until August. Also, chances are the 'fanboys' of either brand would buy their preferred brand even if their brand's card was 10-15% slower because of brand specific features:

GTX680 - PhysX, CUDA, better 3D support/3D vision surround for NV, TXAA, Adaptive Vsync, Ambient Occlusion in the drivers, workable H.264 video decoding that's completely broken on 7900, better power consumption


HD7970 - 3GB of VRAM for GTA V and SKYRIM mods, playing at higher resolutions (2560x1440/1600), better single and double precision compute performance, Bitcoin mining $, 3 free games coupon.

Overall, on air an overclocked 680 and an overclocked 7970 are tied as has been shown @ Xbitlabs:

The performance difference between an overclocked 7970 and 680 is so slim, it 100% comes down to features and if you care about the extra power consumption of the 7970 or if you are watercooling the 7970 or willing to voltage unlock a 680.

The only round where things are crystal clear is HD7850 vs. GTX570 (7850 wins) and HD7950 vs. GTX670 (670 wins). At the $500 price level, it's too close to call a winner.
1 1 [Posted by: BestJinjo  | Date: 06/19/12 08:45:26 AM]
Regarding HD7970: You Forgot!

Better X-Fire Scaling, Better Overclocking, Can be Overvolted manually, Eyefinity X6 Screens, Better Availability & cheaper pretty much everywhere but the USA.

PS The other thing i want to mention is; People waiting for the GTX660 are better of buying the HD7870 or HD7850, because; by the time GTX660 is released, it would be too close to the next round of GPU`s (HD8XXX & GTX7XX series). A max of 3 months before the GTX660 is outdated & duly stomped on by both camps (Thats if the GTX660 even exists!)
2 0 [Posted by: keysplayer  | Date: 06/19/12 05:38:38 PM]
Yes you're right. I heard HD 8000 hardware already existed now and is already on testing phase. Dubbed GCN2 with 1.3-1.5x faster performance than previous generation and much better HSA this time. Nvidia still can't figure out what's the problem with their chips why they can't mass produce it, until they figure it out that will be the time GTX 660/650 will be out. For now keep dreaming.
0 0 [Posted by: pogsnet  | Date: 06/20/12 12:01:33 AM]


Add your Comment

Related news

Latest News

Wednesday, November 5, 2014

10:48 pm | LG’s Unique Ultra-Wide Curved 34” Display Finally Hits the Market. LG 34UC97 Available in the U.S. and the U.K.

Wednesday, October 8, 2014

12:52 pm | Lisa Su Appointed as New CEO of Advanced Micro Devices. Rory Read Steps Down, Lisa Su Becomes New CEO of AMD

Thursday, August 28, 2014

4:22 am | AMD Has No Plans to Reconsider Recommended Prices of Radeon R9 Graphics Cards. AMD Will Not Lower Recommended Prices of Radeon R9 Graphics Solutions

Wednesday, August 27, 2014

1:09 pm | Samsung Begins to Produce 2.13GHz 64GB DDR4 Memory Modules. Samsung Uses TSV DRAMs for 64GB DDR4 RDIMMs

Tuesday, August 26, 2014

10:41 am | AMD Quietly Reveals Third Iteration of GCN Architecture with Tonga GPU. AMD Unleashes Radeon R9 285 Graphics Cards, Tonga GPU, GCN 1.2 Architecture