Projectors Reviews, price and comparision

Both Patriot Memory and Super Talent have launched new SSDs using SandForce's SF-2200 controller.

On Monday Patriot Memory said that it's now shipping the new series of Wildfire SSDs featuring the SandForce SF-2200 controller. This provides the new line with up to 555 MB/s sequential read and up to 520 MB/s sequential write speeds when connected to a SATA 6 Gb/s interface.

"The Patriot Wildfire is the fastest SSD drive we have ever released," says William Lai, Patriot Memory's Product Manager. "Although we are a bit late to market, we wanted to ensure that the technology was rock-solid. We are confident with the latest firmware, Patriot’s reputation for quality and performance stand second-to-none."

According to Patriot, 120 GB and 240 GB capacities are available now with a 480 GB coming soon. Other features include up to 85K 4K random write IOPS, TRIM support, Intelligent "Recycling," DuraClass and DuraWrite technology, and a 2.5-inch form factor. The drives will ship with the latest v3.1.9 firmware and a 2.5-inch to 3.5-inch mounting bracket for installing in standard hard drive bays.

In related news. Super Talent announced its new TeraDrive PT3 series of SSDs which also utilize the SandForce SF-2200 controller. Using a SATA 6 Gbp/s connection, the SSD line features Double Data Rate ONFi flash to reportedly give it the upper hand over competing SSDs released just months ago. A benchmark test, seen below, shows that they're capable of read speeds up to 558 MB/s and write speeds up to 526 MB/s.

"The union of these 3 new technologies (the SF-2200, SATA3, DDR ONFi Flash) has forever changed the storage landscape," the company said. "Right out of the gate, this new generation SSD stretches well within the new found bandwidth promised by the SATA 3.0 bus and breaks the 500MB/s barrier in both Read and Write scores. The comparisons between SSDs and HDDs just became a little harder to justify."

The new TeraDrive PT3 line is now available in 60 GB, 120 GB, 240 GB and 480 GB capacities although they're currently not available on Amazon as of this writing.

Super Talent TeraDrive PT3 Benchmark

OCZ has officially launched its RevoDrive 3 and RevoDrive 3 X2 PCI-Express SSDs.

Wednesday OCZ launched its new "workstation class" RevoDrive 3 and RevoDrive 3 X2 lineup of PCI Express-based SSDs. These "drives" offer users monster performance, designed with the company's proprietary Virtualized Controller Architecture (VCA) technology to deliver maximum throughput in multithreaded applications.

"OCZ RevoDrive 3 PCIe SSDs use our VCA 2.0 technology to deliver superior performance and functionality for everything from scientific computing to high availability clustering," said Daryl Lang, Vice President of Product Management of OCZ Technology. "This exciting new workstation-class storage product accelerates application performance and takes full advantage of today's multithreaded processors and software, providing customers with greater throughput in an easy-to-deploy, single card solution."

According to the specs, the RevoDrive 3 provides sequential read and write speeds of up to 1 GB/s and up to 925 MB/s respectively, and random write (4K) performance of up to 130,000 IOPS. The specs of the RevoDrive 3 X2 are a bit meatier, with sequential read and write speeds of up to 1.5 GB/s and up to 1.25 GB/s respectively, and random write (4K) performance of up to 230,000 IOPS. Both models arrive in 240 and 480 GB capacities, but the RevoDrive 3 X2 also sports an additional 960 GB model.

So far the drives aren't available on the OCZ website or listed on online vendors like Amazon and Newegg, but you can check out our preview of the RevoDrive 3 X2 by heading here until the products finally go live. "The RevoDrive 3 X2 continues OCZ's tradition of delivering innovative technology able to serve up blistering performance to enthusiasts," Andrew said in his hands-on article. "It's not a business-class product. It's for the power user who is able to tax it using the right workload. If you're not one of those folks, the RevoDrive 3 X2 is seriously overkill."

Pricing for the X2 version includes $699.99 for the 240 GB model, $1699.99 for the 480 GB model, and an insane (but justified) $3199.99 for the 960 GB model.

Thanks to the acquisition of U.S.-based iGware, Acer plans to launch a cloud service in 2012.

Reuters reports that the world's second largest PC maker, Acer Inc., plans to purchase Silicon Valley-based cloud computing firm iGware for $320 million in cash and stock. Acer will also shell out an additional $75 million performance-based payout to iGware as part of the deal. This acquisition will reportedly be the fifth-largest Taiwanese buyout of a U.S. company ever, trailing behind Acer's acquisition of Gateway for $761.5 million back in 2007.

"This is the right direction for Acer," said Tracy Tsai, an analyst at IT research company Gartner. "Companies can no longer rely only on hardware; they have to bring new values to customers through providing applications and software services, and by that to increase their margin."

Although its simple website doesn't reveal anything about what goes on under its roof, iGware provides cloud software and infrastructure tools that support more than 100 million consumer devices worldwide, including Nintendo game consoles. Acer is expected to integrate iGware into its cloud software and platform as Acer Cloud Technology Co. after completion of the deal by late September. An actual Acer Cloud product is slated to launch sometime during 2012.

But because Nintendo is one of iGware's major clients, both parties are supposedly in talks over a potential cooperation after the acquisition. Nintendo is supposedly in favor of the deal and plans to shell out a $20-30 million service fee to Acer every year after the deal is signed. Yet some analysts aren't thrilled with the acquisition or Acer's potential involvement with the Japanese gaming company.

"Acer is wasting its money. It's spending almost $400 million on a small software company," said Vincent Chen of Yuanta Securities. "Why does it need a client like Nintendo, which doesn't have a cloud or data center? Acer has been wanting to do online gaming and server business, but it doesn't have a clear vision in the cloud business."

Earlier in March of this year, Acer's CEO Gianfranco Lanci stepped down from his position, thus pushing the company to name Jim Wong as the new corporate president a month later. Then in June Acer revealed that it will book a $150 million charge to earnings so that it could write-off disputed accounts in Europe regarding overall inventory management and related "abnormalities."

But despite some setbacks in 2011 alone, Acer has been focusing its energies on tablets and smartphones to drive future growth, following other tech companies like Microsoft, Apple, Google and HP who are also offering mobile devices and cloud-based services. According to market research firm IDC, the cloud is the future, with the sector expected to grow 27-percent annually over the next five years.

"Acer Cloud will be the firm's key differentiator for the next 10 years," said Henry Wang, deputy spokesman of Acer, at a news conference Thursday.

"As a mid- to long-term investment objective, the valuable core technology and capabilities will help create uniqueness for the Acer brand," the company said in a separate statement.

It's an SSD trapped inside a little USB 3.0 body.

On June 15, Super Talent began shipping a device that looks like a standard USB stick on the outside, but is really an SSD on the inside thanks to a SandForce SF-1222 controller and four NAND devices (eight channels). Called the USB 3.0 Express RC8, this new external storage device includes an advanced multi-channel SSD architecture, virtual caching system and an advanced wear leveling algorithm – all crammed in a USB-stick form factor.

"Featuring an 8 channel architecture and virtual caching system, the USB 3.0 Express RC8 was built for speed," the company said. "True SSD performance comes from reading and writing to multiple flash channels simultaneously. The new RC8 has 8 such channels of flash, just like a full sized SSD. A full-sized SSD also includes a data cache to elevate “real world” performance and the RC8 is no different.  Now true SSD performance can be integrated into a portable device that is both small and convenient."

The flash drive measures 93.5 x 25 x 8 mm and is enclosed in sleek aluminum accented by a translucent blue plastic trim. It also requires no separate cable and is fully backward compatible with USB 2.0 ports, at USB 2.0 speeds. The SandForce SSD processor includes DuraClass and DuraWrite technology which enhances both flash durability and data integrity. The specs claim that the drive reaches read speeds up to 270 MB/s and write speeds up to 240 MB/s.

“Super Talent’s new RC8 blurs the line between flash drives and SSDs," said Super Talent’s Director of Product Marketing, Peter Carcione. "Now SSD reliability, performance and capacities can be had in a portable USB 3.0 flash drive."

Super Talent is now offering the USB 3.0 Express RC8 in 25 GB, 50 GB and 100 GB capacities. Prices aren't available as of this writing, but the 50 GB version is believed to retail for around $110 USD.

Super Talent USB 3.0 Express RC8

Kingston has officially launched its HyperX SSD line featuring the SandForce SF-2281 controller and a SATA 3.0 6 Gb/s connection.

On Monday Kingston revealed the shipment of its first-ever SATA 3.0 6 Gb/s solid-state drive, the HyperX SSD. It features a SandForce SF-2281 controller cranking out sequential read/write throughput of 555 / 510 MB/s and premium Intel 25-nm Compute NAND (P/E 5K).

The 2.5-inch drive hits the market in two capacities, 120 GB and 240 GB, and as a stand-alone device with an included HyperX-branded 2.5-inch to 3.5-inch desktop mounting plate, or bundled with a HyperX Upgrade Kit for easy installation. This kit includes Acronis True Image HD migration software, a desktop mounting plate, an external drive bay, a multi-head screwdriver and a SATA data cable.

Based on "out-of-box" performance with IOMeter08, the 120 GB model offers a max 4KB random read/write of 95,000 / 70,000 IOPS. Kingston did not provide numbers for the 240 GB model, but its Amazon listing states that it has 525 MB/sec read and 480 MB/sec write speeds (as does the 120 GB listing despite what Kingston reports).

Pricing for the HyperX SSD doesn't come cheap: $269.99 for the stand-alone 120 GB drive and $284.99 for the bundled version. The 240 GB model costs $519.99 by itself and $539.99 for the bundled version. All four are now available on Amazon.com, Buy.com, Newegg.com, and Tigerdirect.com.

Gigabyte G1.Sniper and GA-X58-OC: LGA1366 Mainboards for Gamers and Enthusiasts [08/04/2011 12:05 PM | Mainboards]
by Doors4ever The terms “gaming mainboard” and “overclocking mainboard” are very popular these days, but they somehow remain pretty vague and unclear. It turned out Gigabyte have their own idea of what the real gaming and overclocking mainboards should be like.

First LGA1366 mainboards appeared back in 2008 and since then they have already become somewhat obsolete. That is why mainboard makers initiated a second wave of Intel X58 Express based products. The major difference between the new boards and the older models became the support of such new interfaces as USB 3.0 and SATA 6 Gbps. Gigabyte did exactly the same thing and rolled out an entire family of new LGA1366 mainboards starting with the UD3R model and going up to UD9. However, they didn’t stop there and besides the common mainboards, began offering products in the new gaming series called “G1-Killer” as well as a special overclocking board.

In fact, the terms “gaming mainboard” and “overclocking mainboard” are very popular these days, but they somehow remain pretty vague and unclear. This is why they may often be applied to almost any contemporary mainboard. Today any mainboard with at least two graphics card slots supporting ATI CrossFire and Nvidia SLI configurations is positioned as a gaming product, even though most gamers do perfectly fine with just one graphics accelerator. The boards that were specifically designed with gamers in mind are Asus RoG (Republic of Gamers) and MSI Big Bang. As for overclocking, even some entry level mainboards have overclocking-friendly features and all manufacturers always stress their products’ unprecedented overclocking potential. It turned out Gigabyte have their own unique ideas of true gaming and overclocking mainboard. Well, it is going to be even more interesting to see what these ideas are with the help of a gaming Gigabyte G1.Sniper mainboard from the G1-Killer series and Gigabyte GA-X58A-OC mainboard for overclocking fans.

The box with Gigabyte GA-X58A-OC inside is of pretty standard size, but its design is dramatically different from what we saw on regular mainboards packaging. No bright multi-color images, no numerous logo icons, just the model name and a modest but significant description: “Overclocking Motherboard”.

There is a mainboard photo on the back of the box with a brief description of its selected peculiarities, which we are going to dwell on later in our article.

The accessories bundle is quite ordinary. Here is the detailed list of everything inside:

Four SATA cables with metal connector locks, two with L-shaped locks and another two with straight ones;A flexible bridge for two-way SLI graphics configurations;A hard bridge for three-way SLI graphics configurations;A flexible bridge for two-way CrossFireX graphics configurations;A set of adapter-cables for manual voltage monitoring with a voltmeter;I/O Shield for the back panel;User manual;A booklet with brief assembly instructions in 18 languages;DVD disk with software and drivers;“Dolby Home Theater” and “Gigabyte” logo stickers for the system case.

The only unique thing is the 2-Way CrossFireX bridge and a set of adapter-cables for manual voltage monitoring with the help of a voltmeter. Special bridges for ATI CrossFire graphics configurations rarely come even with graphics and hardly ever with mainboards. As for the sables, the board has a row of control spots, which allow users to manually take important voltage readings off them at any time. If we need to know the current voltage in any of the primary knots, then the best way to obtain the reading would be with a voltmeter probe. However, if you need to constantly monitor the voltages, then it is easier to use adapters, to free your hands.

 

  Tweet Comments currently: 1
Discussion started: 08/04/11 08:06:07 PM
Latest comment: 08/04/11 08:06:07 PM

View comments

OCZ Introduces One of the Fastest Solid-State Storage Systems on the Planet.

OCZ Unleashes Enterprise-Class PCIe SSDs with 2.8GB/s Speeds, 500K IOPS

[08/04/2011 06:52 AM]
by Anton Shilov

OCZ Technology Group this week introduced what can easily become the world's highest-performing PCI Express-based solid-state storage device with extreme - up to 2.8GB/s (2800MB/s) - maximum read and write speeds as well as up to 500 thousand input/output operations per second (IOPS). OCZ's Z-Drive R4 PCIe SSDs are designed for enterprise users and are available only directly from OCZ.

OCZ's Z-Drive R4-series solid state drives are available in half-height PCI Express form-factor for 1U machines in capacities of 300GB, 600GB and 1.2TB (RM84) as well as in full-height PCIe form-factor for 2U and other systems in capacities of 800GB, 1.6TB and 3.2TB (RM88). Both types of solid-state storage systems utilize OCZ's proprietary SuperScale controller with virtualized controller architecture 2.0 (VCA 2.0) technology; the half-height RM84 version sports four SandForce SF-2582 NAND flash controllers, the full-height RM88 version features eight SandForce SF-2582 controllers.

OCZ Z-Drive R4 RM84 provides maximum sequential read and write speeds of up to 2GB/s, 250 thousand random write 4KB IOPS, 160 thousand random write 8KB IOPS and up to 260 thousand maximum IOPS in total.

OCZ Z-Drive R4 RM88 provides maximum sequential read and write speeds of up to 2.5GB/s, 410 thousand random write 4KB IOPS, 275 thousand random write 8KB IOPS and up to 500 thousand maximum IOPS in total.

VCA 2.0 is a multifaceted virtualization layer and interfaces with the OCZ SuperScale Storage Controller which combines processing and full DMA (direct memory access) cores, as well as internal PCIe, SATA and SAS physical interfaces. When coupled with VCA 2.0 technology, the SuperScale Controller manages critical internal functions such as OCZ’s intelligent Complex Command Queuing Structure (CCQS) with unique queue balancing algorithms. This combination leads to superior performance aggregation and results in significantly higher throughput and reduced burden on the host CPU.

With the VCA 2.0-based virtualization layer, the Z-Drive R4 offers an advanced set of capabilities designed to solve problems inherent to pure software solutions. The Z-Drive R4 features power fail protection, differentiating the drive from software-based solutions where in-flight data resides in host DRAM and the flash translation layer, thus resulting in loss of data during power interruption. In addition, VCA enables user-selectable data recovery and non-stop modes with IntegRecover technology ensuring enterprise-class data integrity. Furthermore, the Z-Drive R4 installation in enterprise computing environments is seamless because OCZ VCA 2.0 supports the industry-standard SCSI command set over a PCIe interface. Other features include consolidated SMART support, increased NAND flash life due to OCZ’s lower write amplification and TRIM/SCSI Unmap Support.

"The demand for higher performing PCIe-attached solid state storage continues to grow, and our new Z-Drive R4 represents an increased focus from OCZ to meet the needs of its enterprise and data center clients; it also represents a turning point in the evolution of PCIe SSDs. The Z-Drive R4 provides the best of both worlds for storage architects by merging the industry-leading feature-sets of both the pure hardware and pure software based approaches to data management, while maintaining user-configurable flexibility not found in other devices," said Ryan Petersen, chief executive officer of OCZ Technology Group.

OCZ's Z-Drive R4 RM84 and RM88 are available in R-series and C-series configurations. Both families feature TRIM/SCSI unmap, 128-bit and 256-bit AES-compliant encryption, ECC recovery, SMART support and feature 2 million hours meantime between failures (MTBF). The more advanced R-series adds power fail protection with DataWrite Assurance technology as well as SMART with enterprise attributes.

All Z-Drive R4 configurations support single-level cell (SLC) or multi-level cell (MLC) NAND flash memory (installed by default) with reduced write amplification to extend drive life more than three times the manufacturer’s rating of the NAND. All Z-Drive R4 are compatible with Windows 7, Windows Server 2008, Linux Red Hat Enterprise 6.1 operating systems.

As with all OCZ enterprise products, customer-specific configurations and functionality are available upon request.

Tags: OCZ, SSD, NAND, Flash

Tweet Comments currently: 1
Discussion started: 08/04/11 08:05:06 AM
Latest comment: 08/04/11 08:05:06 AM
it's 2.8GB/sec not 2.8TB/sec ^^

maybe more "affordable" than fusionio "ioDrive Octal" :-D (100k$)

FSP Aurum Power Supply Unit Series: in Chase of Gold [08/04/2011 04:28 PM | Cases/PSU]
by Oleg Artamonov FSP Company, which has fallen a little behind competition in the PSU field lately, decided to catch up in one powerful dash and introduced an entire new series of power supply units built on a completely new platform.

Gold is highly popular among power supply makers. On one hand, the precious metal has been attracting people for centuries and, on the other hand, every PSU maker deems it a matter of personal pride to reach the highest grade of the 80+PLUS certification which has been split up into Bronze, Silver and Gold categories.

So, we’ve got such products as Seasonic X Gold (reviewed already and soon to be reexamined in our labs) or Cooler Master Pro Gold (it will hopefully reach us soon). In the name of the product I am going to review today the English word is just replaced with its Latin equivalent Aurum. Of course, there are a lot of references to gold on the product packaging and in the promotional materials. After all, some users may just not know Latin or the chemical element Au.

Well, my job is not to refresh your school chemistry course but to test new power supplies from FSP. So, let’s get started.

Click the following link for a description of our testing methodology and equipment and a brief explanation of what the specified and tested parameters of power supplies mean: X-bit Labs Presents: Power Supply Units Testing Methodology. If you feel overwhelmed with the numbers and terms this review abounds in, refer to the Methodology.

You can also go to our Cases/PSU section to check out reviews of all other PSU models we have tested in our labs.

We will mark the actual power consumption of three system configurations (discussed in our article PC Power Consumption: How Many Watts Do We Need?) in the cross-load diagrams. This will help you see if the tested PSU can meet the requirements of a real-life PC.

The Aurum family is split in two subseries depending on whether the PSU has detachable or fixed cables. The Aurum products with fixed cables range from a 400W model to this 700W one.

The PSU comes in a nice-looking gold-and-black box. FSP prints a special wrapper for each model, so the wattage rating and product name can be found all over the packaging rather than on a sticker somewhere in a corner.

The product’s numerous advantages are detailed on the back of the box. Besides enumerating the PSU’s connectors, the length of each cable is also indicated, which may be important for many users.

Here are the advantages listed on the box:

MIA IC is FSP’s very own exclusive chipset. Its description you can read here seems to have been compiled with the single purpose of confusing the buyer with an abundance of abstruse terms because, for example, it’s hard to grasp the importance of “maximum protection to Capacitor” in the PFC device. I’ve never seen a PSU with a ruptured PFC capacitor. Besides, it is more important for the user that his computer rather than a 2-dollar jar of electrolyte in his PSU be protected against overvoltage.The Arrow Flow vent grid makes it easier for the hot air to leave the PSU. The arrow-shaped grid is indeed pretty but, frankly speaking, air flows better when there are fewer obstacles in its way. So, if the arrows were transformed into, for example, squares by cutting off the extra metal, the PSU would be cooled even better.The Hybrid Synergy 12V Rail Design is supposed to ensure maximum compatibility with graphics cards. People at FSP don't seem to understand the meaning of this innovation themselves because this feature goes without a translation or explanation on the company's Russian website. In practice, although the PSU is declared to have and really has (at least, there are appropriate shunts on the PCB) four "virtual" +12V lines with a max current of 18A each, I could not trigger the protection. Any single line could work at a continuous load of 35 amperes as well as cope with almost instantaneous spikes from 0 to 35 amperes.Active PFC, Japan-made capacitors rated for 105°C, a fan with fluid dynamic bearing, and a set of protection mechanisms are typical features of any high-quality PSU of the 2011 model year.The efficiency is specified to be over 90%. I will check this out in my tests.The noise level is 25 to 40 dB. That’s quite a lot. For example, Seasonic promises that its X Gold series makes only 16 dB of noise in quiet mode at loads up to 50% from the maximum. Cooler Master promises 20 dBA with its Pro Gold series under the same conditions.

The accessories are rather scanty.

We've got the following things here: an installation guide in multiple languages, four screws, three cable straps, a Power by FSP sticker, and a power cable. Well, what else could they have put with a PSU?

The PSU itself is even prettier than its packaging. It’s got a black case with a rough surface, a number of vent grids, including the Arrow Flow grid in the side panel, and a beige (this must be meant to be golden) plastic piece on the fan.

The back panel is blank except for the dozen vent grids shaped as arrows. Take note of the consistency of design: the plastic rim of the opening for cables follows the overall style with its angular shape and beige color.

The interior of the PSU makes it clear that it has nothing in common with the popular Epsilon series.

It’s quite roomy inside. The medium-sized black heatsink decorates the power components of the PFC device and main transformer whereas the tiny silvery one (higher in the photo) cools the transistors of the +12V synchronous rectifier. Well, I do believe in the efficiency of over 90% after seeing such heatsinks.

  Tweet Comments currently: 1
Discussion started: 08/04/11 11:04:50 PM
Latest comment: 08/04/11 11:04:50 PM

View comments

Elpida Begins DRAM Manufacturing at 25nm Process Technology.

Elpida Uses Industry’s Most Advanced Process Tech in the Industry

[08/03/2011 12:21 PM]
by Anton Shilov

Elpida Memory, a leading global supplier of dynamic random access memory, has announced that as of the end of July it had started sample shipments of the smallest chip now available in the DRAM semiconductor memory industry. Elpida completed development of its 25nm  manufacturing process in May and has begun to apply the process to commercial production.

The first chips to be made using 25nm fabrication process is 2Gb DDR3 DRAM capable of operating at 1866MHz and supports 1.35V and 1.5V voltages. A 25nm 4Gb DDR3 memory chip is expected to become commercially available by the end of 2011.

Special features of the 25nm SDRAM include lower current usage (roughly 15% less during operation, 20% less when on standby) compared with Elpida's current advanced process 30nm-generation products and one of the highest data transfer rates in the industry.

These 25nm-process products will be used for PC and server applications. The process will also be used to develop applications of Elpida's Mobile RAM for use in various kinds of mobile devices, starting with the growth areas of smart phones and tablet PCs. Additional probable applications include mobile phones, smart phones, tablet devices, ultra-thin notebook PCs, game consoles, flat screen TVs, set-top boxes and other kinds of digital consumer electronics.

Tags: Elpida, 25nm, DRAM, DDR3

Tweet

AMD Readies Opteron "Dublin" and "Macau" Processors for Launch in 2013 - Slide.

AMD to Move Production of Computer Chips to Bulk Process Technologies

[08/03/2011 07:04 PM]
by Anton Shilov

After many years of relying on silicon-on-insulator (SOI) process technology, Advanced Micro Devices plans to start using bulk process technology for its state-of-the-art Opteron processors code-named Dublin and Macau in 2013, according to a slide that resembles those from AMD's roadmaps.

AMD's forthcoming Bulldozer micro-architecture based Opteron "Interlagos" and "Valencia" microprocessors are set to be made using 32nm silicon-on-insulator process technology at Globalfoundries. The code-named Terramar and Sepang chips, which will feature enhanced Bulldozer micro-architecture (also known as Piledriver) and will have up to 20 and 10 cores, respectively, will also be made using the same manufacturing technology in 2012. But the central processing units (CPUs) for servers that will succeed them in 2013 - Dublin and Macau - will be made using 28nm fabrication process with high-k metal gate, according to a slide published by CPU World web-site that resembles those from AMD's documents.

Terramar and Dublin will use the same G2012 server platform code-named Porto, whereas Sepang and Macau will utilize C2012 server platform known under Luxembourg code-name. AMD G2012 infrastructure will support quad-channel DDR3 memory as well as PCI Express 3.0 controller. The C2012 infrastructure will sport triple-channel memory controller and will also support PCIe 3.0.

It is noteworthy that Dublin and Macau are currently not supposed to increase core count to over 10 for one die or 20 for one MCM module. Potentially, this means that the next-generation Bulldozer micro-architecture will sport a significant amount of enhancements that will allow the new chips to tangibly increase performance without boosting the number of x86 cores. According to AMD's longer-term sketch roadmap, the Opteron processors in 2013 will not start to integrate stream processors to accelerate certain tasks. 

Given the fact that at present both Dublin and Macau are in development, AMD can easily change its opinion regarding any technical specifications or design decisions.

AMD did not comment on the news-story.

Tags: AMD, Macau, Dublin, Bulldozer, Terramar, Sepang, 32nm, 28nm, Piledriver

Tweet Comments currently: 12
Discussion started: 08/03/11 07:59:39 PM
Latest comment: 08/06/11 04:39:27 AM
Expand all threads | Collapse all threads 45nm SOI -> 32nm SOI -> 28nm Non-SOI

Not going to happen

45nm SOI -> 32nm SOI -> 22nm FD-SOI

more likely

+ expand thread (8 answers) - collapse thread  sounds about right because I read somewhere that CPUs don't use half nodes (28nm is one) Nope. 28nm Non-SOI is a way to go for AMD.
AMD can't afford to have exclusive tech process for itself alone.
SOI Fab utilization is the biggest problem which AMD is facing now, 32nm and Brazos being made on TSMC will exaggerate it, and it will turn AMD in red very soon.
Thanks to Doug Grose, which was fired for that, this year WSA removes util tax off AMD. Next year will be quite different. Llano and Bulldozer, then Trinity and "Piledriver?" are all on 32nm SOI HKMG they can afford it
and it's GlobalFoundries for APUs the Llano ones and the Trinity ones....not TSMC Look at the slide again. Dublin and Macau will be made using 28nm. AMD has chosen the path - many cores/low freq/low perf per a core. Bulk process is quite good for that path. Will this path be succesive we'll see soon. Again, 2013 Plans for Next-Gen Bulldozer is 22nm

2014 is Server APU not 2013

 The 28 nm Non-SOI is only make sense, if they integrate the GCN architecture CUs inside the CPU. They can make CPU-cores in Non-SOI easily, the only downside is the clock speed, which is may not go up to 3 GHz. On the other side ... making CUs on a SOI node is very-very hard. GCN (?? i need to google it --- Graphic Core Next)

omg. DO NOT DISINFORM US, IT'S NOT ARCHITECTURE but just plain old AMDs marketing gimmick As old Fusion they propagate since 2007 is pretty worn'n'torn nowadays

 it'll happen rethink it twice. they already have 28nm line for gpu if they'll test any production in GloFo, and in 2013 they'll need something to fullfill 28nm lineup. They'll probably wont ramp up 22nm so soon as they let us wait more than 30 month (FULL YEAR LONGER) just for 32nm processors hit the stores after first 45nm in Dec2008 (920/940be)Take AMD's roadmaps with a huge helping of salt. A chimp predicting the weather is more accurate than AMD with their estimates.+ expand thread (1 answer) - collapse thread  This is pretty realistic estimate considering how GloFo now works, and how long they'll need to finish that new 22nm factory. It's budgeting projection, not a cocky one as it was usual in the past.GLOBALFOUNDRIES Fab 8, Module 1 in Luther Forest, Saratoga County, NY:
- "Initial production is expected to ramp at the 28nm technology node and move to volume manufacturing on the 22nm node." (GLOBALFOUNDRIES, July 24, 2009)

fab8update.com June 15, 2011:
- risk production in mid-2012
- volume production in early 2013

Company Seeks to Develop Thunderbolt-Based Dock for Graphics Cards.

Village Instruments Proposes Thunderbolt-Based External Graphics Cards for Macs

[08/05/2011 04:47 PM]
by Anton Shilov

Village Instruments, a designer of expansion chassis for notebooks, proposes to develop external docs for graphics cards for Apple Macintosh computers equipped with Thunderbolt port. The success of the endeavor is not cast on stone and the owner of the company now asks Facebook users whether they would actually use such a dock.

Village Instruments develops ViDock docking boxes for add-in cards for notebooks and similar devices. The company has proposed to design an external docking device for graphics boards with Thunderbolt interface [which uses PCI Express protocol] for systems featuring appropriate technology. The idea is definitely interesting, but it does not seem that there is enough market for it.

“We had a lively discussion here on Facebook if there is a large enough market for a Thunderbolt ViDock. I was convinced that there is not. But there have been so many passionate promoters that there will be a large enough Mac Community who would want to buy a Thunderbolt ViDock, that I now leave the final decision to you,” said Hubert Chen, chief executive officer of Village Instruments.

At present the head of the company is asking over at Facebook what kind of system do Mac owners have, how they would to use an external Thunderbolt dock and what kind of implementation would they like.

Village Instruments is not exactly designing a Thunderbolt-based dock for external graphics cards. The company is trying to find out whether it makes sense to develop one. What is clear, though, is that Thunderbolt is capable of handling graphics boards.

Given the fact that Apple Macintosh computers lack of Direct3D/DirectX support, which are application programming interfaces that are underneath of all modern games, Mac owners simply cannot play the latest games. As a result, they do not need truly advanced graphics cards with high performance and world-class feature-set.

Previously known as Light Peak, Thunderbolt technology supports two low-latency communications protocols - PCI Express for data transfer and DisplayPort for displays. Thunderbolt technology works on data streams in both directions, at the same time, so users get the benefit of full bandwidth in both directions, over a single cable. With the two independent channels, a full 10Gb/s of bandwidth can be provided for the first device in the chain of the devices. All Thunderbolt technology devices share a common Mini DisplayPort connector. Intel's Thunderbolt controllers interconnect a PC and other devices, transmitting and receiving packetized traffic for both PCIe and DisplayPort protocols and thus makers need to develop or use additional controllers to make their products compatible with the TB I/O interface.

Not a lot of devices these days can take advantage of Thunderbolt. Only external graphics cards, external solid-state drives as well as RAID-based storage solutions, professional equipment and some other applications need 10Gb/s demand. As a result, USB 3.0, which can theoretically provide up to 5Gb/s bandwidth, will continue to serve the majority of devices that exist today.

Apple did not comment on the news-story.

Tags: Thunderbolt, Apple, Village Instruments, AMD, XGP, ATI, Nvidia, GeFroce, Radeon

Tweet

Nvidia Denies Plans to Release Kepler GPU in 2011.

Nvidia: Kepler-Based Products to Ship in 2012

[08/04/2011 05:54 PM]
by Anton Shilov

Nvidia Corp. on Thursday clarified its plans regarding the next-generation Kepler graphics processing units (GPUs) and their release timeframes. Apparently, the company is on schedule to receive the early silicon of Kepler from its manufacturing partner later in 2011, but the commercial launch of the product is scheduled to occur only in 2012.

"Although we will have early silicon this year, Kepler-based products are actually scheduled to go into production in 2012. We wanted to clarify this so people wouldn’t expect product to be available this year," said Ken Brown, a spokesman for Nvidia, in an email statement.

Chris Malachowsky, senior vice president of research and a co-founder of Nvidia said at a recent event that Nvidia would start shipping its next-generation graphics processing units code-named Kepler by the end of the year. The company did not say that the new chips will actually become available commercially though.

Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on.

The new chip is projected to be made using 28nm process technology. Many believe that Taiwan Semiconductor Manufacturing Company, which makes chips for Nvidia, AMD and many others, will not be able to supply enough 28nm products this calendar year.

Nvidia's Kepler family of products, which will likely get GeForce 600-series name in the consumer market segment, will not only power Nvidia's mid-term future products, but will also help Nvidia to boost sales of its desktop discrete graphics cards. In Q2 2011 shipments of discrete graphics boards for desktops were down 15%, according to some analysts.

Tags: Nvidia, Geforce, Kepler, 28nm, Maxwell, Tesla, Quadro

Tweet Comments currently: 20
Discussion started: 08/04/11 07:18:32 PM
Latest comment: 08/06/11 09:38:05 AM
Expand all threads | Collapse all threads I don't believe AMD has anything to stand up against kepler once it comes out+ expand thread (2 answers) - collapse thread  I don't believe Nvidia has anything to stand up against Southern Islands once it comes out.

There, fixed for you

 I don't believe AMD has anything to stand up against kepler once it comes out
Seriously?? Have you even read any of the reports about the AMD Graphics Core Next Compute Units (CU)? This is a significant departure in GPU power from the previous generation of AMD/ATi GPUs. Don't forget you have the technical graphical prowess of ATi mixed in with CPU tech of AMD. If there's a company that's going take GPU computing to the next level it's AMD/ATi. Don't get me wrong, I like nVidia a lot, but times are a changing. This was no surprise, why does NVIDIA need to rush out the next architecture when they still hold the single GPU crown. I think they should focus on SLI for a little while, considering how CF setups are beating SLI even when SLI has better individual cards if ran by themselves vs the CF cards.+ expand thread (3 answers) - collapse thread  Single GPU Crowns are for both

AMD GPUs are heavily Integer Friendly
Nvidia GPUs are heavily Floating Point Friendly

AMD = The Best at Integer(Bitcoin Mining)
Nvidia = The Best at Floating Point(F@H)

 you could buy two radeon 6950s and crossfire them for the price of a single GTX 580Nvidia has single GPU crown and soon will get dual GPU crown. Asus claims that Mars II is 22% faster that GTX 590.

P.S. Charlie Demerjian eat shit.

+ expand thread (1 answer) - collapse thread  ...i think something is wrong with you Bro. calm down is just silicon, metal, plastic and electricity.One day the Senior VP says something, next day they clarify....
seems to me they are having some problems with the architecture and the process!

Just hope Kepler and SI come out before Diablo3 so that I can have a choice!

+ expand thread (1 answer) - collapse thread  Personally I prefer to hope that Diablo 3 will run well on current generation cards.I don't believe Nvidia has anything to stand up against Southern Islands once it comes out.

There, fixed for you
4 2 [Posted by: quasi_accurate [Rating 2.58] | Date: 08/04/11 08:51:39 PM]

nigga, the southern islands is the same exact arch as the one they are using now. amd is known to use the same arch for many years without updating such as with their phenoms. the only difference with the southern islands is the 28nm fab and probably increased clock speeds. nvidia kepler is a whole new arch and they are not rushing it. they create new arch every 2 years just like intel who creates new arch every 18 months.

+ expand thread (1 answer) - collapse thread  GeForce 8000 family came out in 2006, GeForce 400 family in 2010. about 3.5 year difference. So where exactly you get the "every 2 years" is unclear.Oh here we go again, nVidia fanboys vs AMD fanboys.

I wonder when you will stop being like that and just buy whatever you like without feeling the need to dig some shit for the "opposing force" so you can feel better for your purchase.

So far I see both companies to hold great and both have always a GPU as good as the other, something that make them compete more and more, and give us cheaper GPUs with more horsepower.

Now give me thumbs down for pointing out the obvious in your stupid conflict.

+ expand thread (4 answers) No, it's a stupid conflict. But it's also stupid to goad people into voting a thumbs down.

By the way, far as I've seen competition leads to better products but also higher prices.