News
 

Bookmark and Share

(4) 

The eight-core microprocessors from Advanced Micro Devices based on the Bulldozer micro-architecture that feature Orochi design will get massive internal cache systems compared to current-generation of chips. In total, the top-of-the-range single-die processor will have 16MB of cache.

AMD Orochi chip (which is a universal server and client design that will power Valencia, Zambezi and possibly other implementations of first-gen Bulldozer family) will have 8MB unified level-three cache, according to a document seen by X-bit labs. Since eight-core Orochi features four dual-core Bulldozer modules, each of which is believed to have 2MB of shared level-two cache, the whole chip will pack in whopping 16MB of SRAM, a 77% increase from the current six-core microprocessors that have 9MB of cache in total.

Large L2 caches will help AMD's next-generation microprocessors to ensure higher performance in single-threaded applications compared to today's multi-core chips that only have 512KB of L2, whereas massive L3 cache will maximize memory bandwidth. In both cases this will bring notable performance gains compared to today's Stars/Greyhound architecture.

AMD Orochi design is the company's next-generation processor for high-end desktop and server markets. The chip will feature eight processing engines, but since it is based on Bulldozer micro-architecture, those cores will be packed into four modules. Every module which will have two independent integer cores (that will share fetch, decode and L2 functionality) with dedicated schedulers, one floating point unit with two 128-bit FMAC pipes with one FP scheduler. The chip will have shared L3 cache, new dual-channel DDR3 memory controller and will use HyperTransport 3.1 bus. The Orochi chips will use new AM3+ form-factor and will require brand new platforms.

AMD did not comment on the news-story.

Tags: AMD, Bulldozer, Zambezi, Orochi, Interlagos, Valencia, Phenom, Athlon

Discussion

Comments currently: 4
Discussion started: 09/23/10 06:41:17 PM
Latest comment: 09/25/10 09:42:55 PM
Expand all threads | Collapse all threads

[1-3]

1. 
This has been known for going on a year now...

I can't remember whom reported it first, Dresdenboy, hardware-infos, or whomever.
0 0 [Posted by: turtle  | Date: 09/23/10 06:41:17 PM]
Reply

2. 
I was late by a day . I reported it in this posting in January.

And that was based on a posting in RWT's forum.

The cache info itself appeared in the Open64 compiler source. This still lacked L3 cache info. And it was no official information. The 16kB L1 became official at Hot Chips. And the 2MB L2 per module once slipped into JF's Q&A blog. I think the L2 and L3 sizes still have to be made official.

But looking at the Orochi die, as also shown on X-bit labs, one could already see the 2MB L2 per module and four blocks of 2MB L3 on the die.
0 0 [Posted by: Dresdenboy  | Date: 09/24/10 12:09:41 AM]
Reply

3. 
AMD seems to believe that cache makes up for everything. They did so when launching Phenom II and I hope that, this time, they've really improved the IPC and the intrinsic architecture.
0 0 [Posted by: East17  | Date: 09/25/10 04:04:24 AM]
Reply
- collapse thread

 
LOL

They're not Intel to think that way

But rightfully they acquired same sharky business methods as Intel did with their marketing scheme pushing enormous caches back in 2001-2006 when they couldn't make decent chip

It'll be nice to see how much L3 cache Sandy Bridge will sport against Orochi (codename:Bulldozzer) because they suppose to be comparable chips with similar feature sets so we could have better insight how large cache helps them in their number crunching.
0 0 [Posted by: OmegaHuman  | Date: 09/25/10 09:42:55 PM]
Reply

[1-3]

Add your Comment




Related news

Latest News

Wednesday, June 19, 2013

7:58 pm | Toshiba Reveals 7mm Solid-State Hybrid Drives. Toshiba Announces Slim SSHDs

7:28 pm | AMD Unlikely to Follow Nvidia with Licensing of Graphics Technologies. AMD Not Likely to Follow Nvidia, ARM, Imagination with Graphics IP Licensing

6:56 pm | Intel to Launch Core i7 “Ivy Bridge-E” in Early September. Intel to Release New Extreme-Series Products in September

6:35 pm | Microsoft Ditches Xbox One’s DRM and Always-Online Policies Following Critics from Gamers. After Facing Rage from Gamers, Microsoft Reverses Its Xbox One DRM Policies

Tuesday, June 18, 2013

11:56 pm | Super Talent Introduces SSD-HDD Hybrid Drive with PCIe Interface. Super Talent Announces 1TB – 2TB SSD-HDD Hybrid with PCI Express Bus

11:32 pm | AMD A-Series “Richland” Overclocked to 8GHz. Enthusiast Pushes Latest AMD Processor to Nearly Record Clock-Speed

11:29 pm | Sony: We Have No Plans to Cut PlayStation 3 Plans Just Yet. Sony May Not Cut PS3 Pricing Ahead of PlayStation 4 Launch

11:23 pm | Chinese Tianhe-2 Supercomputer Becomes World’s Fastest Supercomputer. Intel Xeon “Ivy Bridge-EP” and Xeon Phi Power World’s Top Supercomputer

11:19 pm | GlobalFoundries to Make Application Processors for RockChip. RockChip to Address Tablet SoC Market with 28nm Chips Made by GlobalFoundries

11:12 pm | Asrock Extends Warranty of Flagship Mainboard to Five Years. Asrock Z87 OC Formula Motherboards Have Five Years Warranty

11:08 pm | Apple: We Received 4 – 5 Thousands Requests from U.S. Law Enforcements, Refused to Fulfill As Many As We Could. Apple Committed to Customer Privacy, Explains How It Works with Law Enforcement Organizations

11:04 pm | SanDisk Enhances Flashsoft Software for Server-Side SSD Caching. FlashSoft 3.2 Software Adds Support for Multiple SSDs and SSD Mirroring

8:15 pm | AMD Unveils Server Strategy and Roadmap. AMD Adds Berlin, Seattle and Warsaw Processors into Roadmap

7:38 pm | Nvidia Set to Radically Change Business Model, License Graphics Cores to Others. Nvidia Takes ARM, Imagination Technologies Route, Intends to License Kepler Graphics Tech