Maybe looking to steal a bit of thunder from Nvidia's upcoming GeForce GTX 275 release, there's a chance AMD will release its ATI HD 4890 on April 2nd, a week ahead of schedule, says VR-Zone. The reviews and news outlet doesn't cite any sources, but did say that both Asus and Gigabyte have already begun selling the HD 4980 in Taiwan and Hong Kong for HK$2,280 (US$297) and HK$1,999 (US$258), respectively.
As previously reported, reference specs for the RV790-based HD 4890 include 1GB of GDDR5 memory running at 3,900MHz and a core clockspeed of 850MHz. No official release date (that we know of) has been given, but if AMD does introduce the new card on April 2nd, it will have beat Nvidia to the punch by a full week if Nvidia sticks to its April 9 release date for its upcoming GeForce GTX 275. A likely scenario if, as VR-Zone claims, "their GTX 275 isn't ready yet and the clocks aren't even finalized."
News site DailyTech has gotten its paws on AMD's upcoming ATI Radeon HD 4890 videocard from an undisclosed source based in Taiwan, and has thus been able to confirm rumored details of the new card's spec sheet.
Built around the RV790 core, the Radeon HD 4890, as received by DailyTech, comes with a core clock of 850MHz . The site also reports 1GB of GDDR5 memory running at 3,900MHz resulting in 124.8 GB/s of bandwidth. Final memory specs could change, however, as the card DailyTech received came with Qimonda chips, which declared insolvency (went bankrupt) back in January 2009.
No word yet on a projected price point or release date, but not to be outdone, Nvidia plans to go head-to-head with ATI's 4890 with its GeForce GTX 275. According to reports, the upcoming GTX 275 is being built around the G200b GPU core with 240 shader processors chugging along at 1,404MHz. Other specs include 80 texture units, 30 render back ends, and a 448-bit memory interface. GPU and memory clockspeeds are expected to debut at 633MHz and 2,322MHz, respectively.
Look for the GTX 275 to launch on April 9, 2009 for somewhere between $230 and $280.
As a general rule, our belief is that pairing two slow-performing cards using SLI or CrossFire is a bad idea—you’re usually better off running a single faster card. However, the Radeon 4850 X2 delivers astounding performance compared to the single-GPU boards in its price range, spanking the Radeon 4870 and the GeForce GTX 280, with none of the pitfalls that have plagued dual-GPU boards in the past.
At the heart of the board is a pair of ATI’s RV770 GPUs running at 625MHz, just like the single-GPU in the 4850 boards. Each GPU features a full complement of 800 stream processors, which are connected to identical 1GB GDDR3 frame buffers running at 993MHz on a 256-bit bus. Although X2 boards are labeled as featuring 2GB of memory, because the contents of each GPU’s frame buffer must be mirrored, applications can utilize only 1GB of video memory.
Windows 7 is already feeling the love from both graphics camps. Earlier this month, Nvidia released a specialized Forceware driver for the beta OS along with the promise of regular updates, and now AMD has followed suit with its new ATI Catalyst 9.3 driver this week. However, the new Catalyst driver rolls both Windows 7 and Vista support into a single download, allowing the company to lay claim as offering the "first unified driver installation package to incorporate Windows 7 support." AMD says future Catalyst releases will also be unified for both the current and upcoming Windows OSes.
"AMD's expertise in visual computing shines through in the combined experience of Windows 7 and ATI Radeon graphics," said Anantha Kancherla, group manager responsible for Windows graphics, Microsoft. "With today's release of a Windows 7 unified driver, AMD once again demonstrates its ability to deliver perfromance and cutting-edge driver support."
Hit the jump to find out what else to expect from the new Catalyst 9.3 unified driver.
According to Nvidia's Form 10-K filing for the fiscal year ended January 25, 2009, the graphics chip maker has spent $43.6 million to cover warranty and product replacement claims for GPUs suffering from a "weak die/packaging material set."
Nvidia had original set aside a one-time charge of $196 million when it was discovered some of its notebook graphics were failing at an "abnormal rate."
"The previous generation MCP and GPU products that are impacted were included in a number of notebook products that were shipped and sold in significant quantities," Nvidia said in the filing. "Certain notebook configurations of these MCP and GPU products are failing in the field at higher than normal rates. While we have not been able to determine a root cause for these failures, testing suggests a weak material set of die/package combination, system thermal management designs, and customer use patterns are contributing factors."
Given that only $43.6 million -- or 22 percent of the original amount -- has been spent so far, you can take this in one of two ways. Either the problem isn't as widespread as originally thought and the remaining $152.4 million will be more than enough to cover future claims, or there are a lot of mobile GPUs in the wild still to fail from the packaging defect.
You have to admire when a company doesn't just talk the talk, but walks the walk. EVGA, one of three Nvidia add-in board (AIB) partners to offer lifetime warranties for its videocards, not only allows end-users to overclock their cards without voiding the warranty, but arguably encourages the behavior by serving up OCing tools. Registered owners can download the company's GPU Voltage Tuner utility, and anyone can download EVGA's Precision overclocking software whether they own an EVGA-brand videocard or not.
In the case of the latter, a new version is now available, v1.5.1. Several updates have been added to the 1.5.x refresh (the majority were included in v1.5.0 released last Friday), including multi-GPU support. The overclocking tool now gives users independent access to thermal monitoring, overclocking, and fan control settings of all installed GPUs whether configured to run in SLI or not.
Other changes and fixes include better support for GeForce 9 and older videocards, a pair of new skins, settings no longer reset at Windows startup on multi-GPU systems, fixed memory clock monitoring for G98 GPUs, and Precision now displays the graphics processor codename in the system info.
Four-monitor support in a single-slot, low-profile, half-length videocard? You betcha. That's exactly what AMD's offering up with the release of its professional ATI FirePro 2450 multi-view graphics card.
"The ATI FirePro 2450 offers the reliability professionals expect and the efficiency IT departments require. IT managers want a card they can test once, deploy virtually anywhere, and count on to run reliably,” said Janet Matsuda, senior director, AMD Professional Graphics. “The low power consumption enables cool, energy-efficient operation, as well as superior reliability and longevity. The compact form factor can be deployed in nearly any system. This accelerator is perfect for customers who need more than two displays, such as in financial services and process control."
The FirePro 2450 comes equipped with 512MB of GDDR3 memory, variable speed fan sink, maximum resolution of 1920 x 1200, DirectX 10.1 and OpenGL 2.1 support, native PCI-E x16 and x1, and support for up to four independent DVI or VGA monitors.
According to AMD, the new card consumes less than 18W on average for 2D business usage, and no more than 32W when fully stressed.
The ATI FirePro 2450 is available now with an MSRP of $499.
Nvidia has said on more than one occasion that it wants to simplify its product line to make it easier for consumers who might not know the difference between, say, an 8800 GTS 256MB, 8800 GTS 512MB, and 8800 GTS 640MB, and why the 512MB trumps them both. Then there's the 9800GTX+, which is a supercharged 9800 GTX, which is really a supercharged G92-based 8800 GTS, which is confusing as all get-out.
It gets better. Meet the GeForce GTS 250, the videocard formerly known as the 9800 GTX+. The rebranded videocard still uses the 55nm G92b GPU, however in a more mature yielding chip in GTS 250 trim. Available in both 512MB and 1GB configurations, the latter includes a new board design noticeably smaller than the 9800 GTX+ by about an inch and a half.
Other specs include a 738MHz core clockspeed, 1100MHz memory clockspeed, 256-bit memory interface for a 70.4 GB/s of total memory bandwidth, 16 ROPs, 64 texture filtring units, and 128 processor cores. The GTS 250 carries a TDP of 150 watts, and according to Anandtech's testing, both idle and load power consumption runs about 30W less than the 9800 GTX+.
The 512MB and 1GB versions will run $130 and $150 respectively, with widespread availability expected next week.
The official release of Windows 7 might still be several months away, but that isn't stopping Nvidia from preparing for Vista's successor with new graphics drivers aimed at Windows 7 beta users. The new drivers are available now, and Nividia promises this is just the start of a regular driver release schedule. Remember that shortly after Vista debuted, Microsoft blamed buggy Nvidia drivers for giving the OS a bad rap.
"Since its release last month, the Windows 7 Beta has been eagerly tested by hundreds of thousands of NVIDIA GeForce owners, who are excited about the many graphical improvements Microsoft has added into the upcoming operating system," said Ujesh Desai, vice president of GeForce desktop business at NVIDIA.
Nvidia says it has been working closely with Microsoft so that its new drivers will take full advantage of the additional features and functionality Windows 7 brings to the table. Kicking off with v181.71, Nvidia's graphics drivers support the new Windows Display Driver Model (WDDM) v1.1. The drivers also support SLI on DX9, 10, and OpenGL applications, PhysX, CUDA, and Direct3D, Direct2D, and DirectWrite.
ATI today announced a pair of mobility chips -- ATI Mobility Radeon HD 4860 and HD 4830 -- built off of AMD's new 40nm manufacturing process, purportedly making them the first-ever 40nm notebook GPUs.
Both the HD 4860 and HD 4830 come with 640 stream processor, 826 million transistors, support for DirectX 10.1, dual integrated DisplayPort, HDMI with 7.1 surround sound, and are CrossFireX ready. The 128-bit HD 4860 boasts a 650MHz core clockspeed and GDDR5 memory clocked at up to 4Gbps. The HD 4830 (also 128-bit), meanwhile, ships with a core clockspeed ranging from 450MHz to 600MHz, and GDDR3 clocked between 800MHz to 900MHz.
"It's not just 40nm process technology that makes these chips so potent, they are based on the same award-winning TeraScale engine of our ATI Radeon™ HD 4800 desktop series," Rick Bergmen, senior VP and GM for AMD's graphics products group, wrote in a blog post. "Combining this gaming power with our ATI Avivo™ HD technology and Unified Video Decoder will keep all your HD content humming along at full 1080p resolution with bright colors and seamless playback on your HD display. We've also packed in our power-saving technologies like ATI PowerPlayTM, ATI PowerXpressTM, and ATI Switchable Graphics™ technologies so that you can keep gaming, watching and surfing a little longer."
Bergmen went on to say that 40nm desktop parts are "coming soon," with at least one site having already posted benchmarks of what's believed will be ATI's first 40nm-based desktop graphics release.
As for the 40nm Mobility parts, Asus is scheduled to ship laptops using the new processors in the second half of 2009.
Update 3/4/09 AMD has has posted more pictures of its new Mobility chips, along with a video showing the Mobility Radeon HD 4860 running a on a desktop system uing the MXM modules (no notebooks are currently shipping with the part).