40nm
Latest
TSMC begins construction of new $9.3b foundry, wants to sate our constant hunger for chips
TSMC might not necessarily be a household name, but the product of its labors tends to be all over home electronics. Aiming to keep that trend going, the Taiwanese chipmaker has just broken ground on its third 300mm wafer plant, located in Taichung's Central Taiwan Science Park. The new Fab 15 will have a capacity of over 100,000 wafers per month -- earning it the prestige of being described as a Gigafab -- and once operational it'll create 8,000 new skilled jobs in the area. Semiconductors built there will also be suitably modern, with 40nm and 28nm production facilities being installed, and lest you worry about such trivial things as the environment, TSMC says it's doing a few things to minimize the foundry's energy usage and greenhouse gas emission. Then again, if you're going to spend nearly $10 billion on something, would you expect anything less?
NVIDIA GeForce GTX 460 becomes everyone's favorite midrange graphics card
It's rare to come across a universally lauded product nowadays, but NVIDIA's fresh new GTX 460 is just that sort of exceptional creation. Contrary to its GTX 465 elder brother, the 460 isn't a chopped-down top-tier part and is instead built on the new GF104 core. This smaller core, designed from the start to perform humbler functions, has ameliorated the famed power inefficiency that has been a Fermi signature so far, and has resulted in AnandTech describing the new card as "the $200 king." You'll get 768MB of onboard RAM at that point, but we'd splurge an extra $30 to make that a round gigabyte and enjoy some extra L2 cache and ROPs on the card. Either way, the GTX 460 seems to have completely killed off the market for the 465 and is stepping all over ATI's toes with its competitive pricing and, for once, decent heat and power metrics. Oh, and apparently it "overclocks like a monster" too -- hit the links below for the full reviews. Read - AnandTech Read - HardOCP Read - Hot Hardware Read - Tom's Hardware Read - PC Perspective Read - Bit-tech Read - Guru 3D Read - Legit Reviews
NVIDIA GeForce GTX 480M reviewed: fastest mobile GPU to date
It's one thing to have a product called the world's fastest on paper, but it's another thing entirely to have the benchmarks confirm it. NVIDIA's GeForce GTX 480M has been swooned over for months now, but it just recently hit the hardcore review sites in a big way. Frankly, there's not a whole lot to say about the thing at this point: it's simply the fastest mobile GPU to date, with Hot Hardware finding it to be "significantly faster in nearly all gaming benchmarks," with just one title showing the Mobility Radeon HD 5870 as the champ by only a few frames. If you've been searching for the fastest mobile GPU in town, you're wasting your time looking any harder; 'course, all of that power consumes an insane level of energy, so true road warriors will certainly want to look elsewhere. Critics pointed out that energy consumption and excess heat were real issues, though both of those are easily overlooked when you're able to take a beastly laptop to a LAN party rather than your desktop. Give those links below a visit if you still need proof.
NVIDIA GeForce GTX 465 rounds up mostly positive reviews
Well, it's not quite June 1, but the GeForce GTX 465 reviews have come flooding out all the same. The official specs are exactly as a recent leak indicated: 352 CUDA cores running at 1,215MHz, a 607MHz graphics clock, and 1GB of GDDR5 memory operating at a 3.2GHz effective rate and exploiting a 256 bit-wide interface. With an MSRP of $279.99, this Fermi-lite GPU scored plenty of admiration for the value it offers, with one reviewer going so far as to call it "quite possibly the most powerful DirectX 11 graphics card for under $300." Others weren't so enthusiastic, citing the far cheaper HD 5830 from ATI as a better choice, but it's true enough that the next best GPU, the HD 5850, tends to be at least $30 more expensive than the 465, depending on brand. You'll want to delve into the game benchmark numbers in order to make up your mind about which card might make for the best bit, but be warned that NVIDIA's 465 retains the GTX tradition of ravenous power consumption -- something to consider if you're rolling along with an old school 400W PSU in your rig. Read - Hot Hardware Read - PC Perspective Read - TweakTown Read - Legit Reviews Read - Tom's Hardware Read - Guru 3D Read - techPowerUp
NVIDIA GeForce GTX 480M: 'world's fastest' mobile GPU now official, landing in June
NVIDIA has just announced that the GTX 480M, the mobile re-spin of its extravagantly overpowered Fermi desktop parts, will be arriving in the middle of next month. Touted as the world's fastest mobile graphics processor, this chip will bring 352 CUDA cores and a 256-bit memory interface to up to 2GB of GDDR5 RAM. These are clear and convincing advances over the incumbent Green Team leader, the 360M, but things start to look a little worrying when we check the 480M's clock speeds. The processor speed is nearly halved from the desktop GTX 480, at 850MHz, the memory does only 1,200MHz, and the graphics run at 425MHz -- we didn't know anything worthwhile even operated below 500MHz these days. Either way, you're getting a computing powerhouse, with the 480M's 897 gigaflops comfortably dwarfing its predecessor's 413 and promising almighty tessellation performance. What it all means with regard to keeping your frame rates up while traversing the Terminus Systems, we can't yet say. We'll let the benchmarking gurus figure it out -- go past the break for the full press release and spec sheet.
NVIDIA GTX 465 detailed ahead of June 1 launch, GTX 460 also rumored
Hey, you there, wanna check out some unreleased new hardware from NVIDIA? Donanimhaber, the same site that brought us early (and accurate) specs of the GTX 470 and 480, is back with the above shot of an ASUS-built GTX 465 that is intended to entice the more mainstream crowd into the Fermi party. Essentially a GF100 with its wingtips trimmed, this'll offer the same 607MHz graphics and 1,215MHz processor clock speeds of the 470, but comes with a narrower 256-bit memory interface, a tamer 3.2GHz effective memory clock, 352 rather than 448 CUDA cores, and -- happily -- lower power requirements. This word is also corroborated by Bit-tech, whose scribes expect a launch at the start of Computex. The Turkish site also has news of a GTX 460, which they say will show up in the middle of July with 768MB of GDDR5, while also being "cheaper and slower" than the 465. Hit the Hexus link below for some early benchmarks of the latter card.
Zotac ZBOX HD-ID11 and its Ion 2 innards reviewed
Small form factor? Check. Low power consumption married to 1080p video playback capabilities? Of course. Quiet cooling? Naturally. Those are the basic requirements for, and their fulfillment is the means by which we judge, a good home theater pc. They're also the highlights of Zotac's Ion 2-powered ZBOX HD-ID11 barebone (you have to add your own RAM, storage and OS) nettop, which recently visited AnandTech's labs for some old fashioned review action. It's a highly illuminating read, particularly for those interested in the differences between NVIDIA's Ion generations, which throws up a mixed bag of results. While you'll be quite alright watching Full HD Blu-rays on the ZBOX, Flash hardware acceleration -- yeah, that old nugget again -- is not yet implemented well enough, resulting in a maximum of 480p resolution before Hulu streams started glitching out on the reviewer. A June driver update from NVIDIA should rectify this issue, and we're encouraged to wait it out and see what we might see then. In the mean time, you can just delve into the complete analysis which awaits at the link below. [Thanks, Wowzers]
NVIDIA GTX 480M will bring Fermi to laptops this June, crazy power requirements and all
We had an inkling NVIDIA wouldn't keep the Fermi goodness just to the desktop and here's our first pseudo-official confirmation. Rushing in ahead of any announcements, Eurocom has started listing a GeForce GTX 480M part, replete with 2GB of GDDR5 memory and a $345 markup relative to ATI's Mobility Radeon HD 5870. It's not clear whether the 100W number refers to the TDP or power requirements of NVIDIA's new GPU, but it's safe to expect both to be pretty high. The MXM 3.0b interface provides a 256-bit linkup between the GPU and CPU, lending plenty of bandwidth, but it also demands plenty of PCB real estate. As a result, Eurocom is offering the GTX 480M on its 17-inch Cheetah and Panther and 18.4-inch Leopard desktop replacements, but not on its 15.6-inch Cougar. Man, no love for the Cougars. According to the listing, we're only a month or so away from release. [Thanks, Jacob]
Sony PS3 upgraded with cooler 40-nm RSX graphics chip, profits await (updated)
It's a milestone folks: the PS3 hardware is finally ready to generate a profit. The loss-leading console once estimated to cost Sony more than $800 per (losing between $241 and $307 per console sold back in 2006) has likely turned a corner thanks to a reduction in manufacturing costs. While Sony isn't saying anything on the matter, PocketNews confirms that the latest PS3 SKU -- CECH-2100A spotted in the FCC back in February -- uses an improved RSX graphics chip based on smaller 40-nm processes similar to the PS3 Slim's new 45-nm Cell processor. The result is a 15 percent decrease in console power consumption when compared to the 120GB CECH-2000A PS3 Slim sporting a 65-nm RSX. The cooler running chip allows for a stealthier heat sink and power supply in addition to a smaller cooling unit. Those changes combined with fewer adjoining chips around the shrunken RSX should make the console cheaper to build which is good news to Sony's sagging bottom line. Update: PocketNews has confirmed with Sony that the RSX graphics chip is built using 40-nm processes (not 45-nm). Post updated to reflect the change.
ATI FirePro V8800 takes Cypress core into workstation woods, emerges victorious
ATI has just let loose the first professional tier card based on its Cypress XT core, which in itself is part of the company's highly successful Evergreen series of 40nm chips. Boasting 2GB of GDDR5 memory running at 1.15GHz, 1,600 stream processors and an 825MHz core clock speed, the FirePro V8800 is very much an HD 5870 adapted to the demands of the digital creation world, and as such it's no surprise that it also requires the same pair of 6-pin auxiliary power connectors. It also supports DirectX 11 and the fresh new OpenGL 4.0 standard, something those business crazies seem to appreciate, but its killer feature is most probably the price, which -- although steep by common standards -- is $300 less than the $1,800 FirePro V8750 predecessor. Oh, and this card can drive 4 simultaneous displays, but we're kinda used to that by now with ATI -- the real question is how this pup ranks relative to its forebears and NVIDIA competition. The answer, according to the commonly used Maya, SolidWorks and 3ds Max applications, is that the FirePro V8800 simply destroys the V8750 while also generally outpacing the aging Quadro FX 4800. Feels pretty nice to have a straightforward conclusion, doesn't it? Check out the links below for the full disclosure.
NVIDIA unleashes GeForce GTX 480 and GTX 470 'tessellation monsters'
Let's get the hard data out of the way first: 480 CUDA cores, 700 MHz graphics and 1,401MHz processor clock speeds, plus 1.5GB of onboard GDDR5 memory running at 1,848MHz (for a 3.7GHz effective data rate). Those are the specs upon which Fermi is built, and those are the numbers that will seek to justify a $499 price tag and a spectacular 250W TDP. We attended a presentation by NVIDIA this afternoon, where the above GTX 480 and its lite version, the GTX 470, were detailed. The latter card will come with a humbler 1.2GB of memory plus 607MHz, 1,215MHz and 1,674MHz clocks, while dinging your wallet for $349 and straining your case's cooling with 215W of hotness. NVIDIA's first DirectX 11 parts are betting big on tessellation becoming the way games are rendered in the future, with the entire architecture being geared toward taking duties off the CPU and freeing up its cycles to deliver performance improvements elsewhere. This is perhaps no better evidenced than by the fact that both GTX models scored fewer 3DMarks than the Radeon HD 5870 and HD 5850 that they're competing against, but managed to deliver higher frame rates than their respective competitors in in-game benchmarks from NVIDIA. The final bit of major news here relates to SLI scaling, which is frankly remarkable. NVIDIA claims a consistent 90 percent performance improvement (over a single card) when running GTX 480s in tandem, which is as efficient as any multi-GPU setup we've yet seen. After the break you'll find a pair of tech demos and a roundup of the most cogent reviews.%Gallery-89115%
ATI busts out Radeon HD 5830, covers all bases
Whether you wanna call it the affordable high end or the really supercharged midrange, ATI has decided that there is a space to be filled between its scorching 5850/70 cards and the only somewhat slower 5770 SKU. So let's all meet the $239 HD 5830, shall we? As is par for the Evergreen course, it's a 40nm DirectX 11 card, and while it matches the 2.15 billion transistor count of the 5850, the new contender makes do with only 16 ROPs and 1,120 stream processors. To compensate for that processing loss, the core clock is raised to 800MHz while maintaining the 1GHz memory speed of the elder 5850. If this sounds like the 5830 will basically be composed of 5850 chips that didn't quite make the high grade, good, because that's exactly what's happening. Internet reviewers have already had some time to play around with ATI's latest card, and if we can draw a consensus, it would be that while the performance is pretty much spot on between its nearest siblings, the price is that little bit too steep to make the HD 5830 an easy recommendation. Read - AnandTech Read - HotHardware Read - HardOCP Read - PC Perspective Read - Legit Reviews Read - ExtremeTech
Samsung's 4Gb DDR3 RAM modules could put 8GB in your next netbook
Rhett and Link (and one Bill Gates, too) may assume that 64KB of memory is "plenty," but let's be realistic here -- it's not. Samsung's fully in agreement, and it's looking to make your next DIMM or SODIMM a lot more capacious with a new 40nm-class process technology. As the story goes, volume production has begun on a new 4Gb DDR3 module, which could be squeezed onto existing sticks in order to produce 16GB / 32GB DIMMs (for desktops) and 8GB SODIMMs (for laptops). There's no clear indication of when this whole "mass production" thing will lead to "on sale right now," but we're guessing it'll make the transition between "now" and "really soon." Bonus video after the break.
AnandTech goes behind the scenes of ATI's RV870 / Evergreen GPU development
Anyone familiar with the constantly shifting release dates and delays that characterize GPU refresh cycles will have been impressed by ATI's execution of the Evergreen series release. Starting out at the top with its uber-performance parts, the company kept to an aggressive schedule over the winter and can now boast a fully fleshed out family of DirectX 11 graphics processors built under a 40nm process. The fact that NVIDIA has yet to give us even one DX11 product is testament to the enormity of this feat. But as dedicated geeks we want more than just the achievements, we want to know the ins and outs of ATI's resurgence and the decisions that led to its present position of being the market leader in features and mindshare, if not sales. To sate that curiosity, we have our good friend Anand Shimpi with a frankly unmissable retrospective on the development of the RV870 GPU that was to become the Evergreen chips we know today. He delves into the internal planning changes that took place after the delay of the R5xx series, the balancing of marketing and engineering ambitions, and even a bit of info on features that didn't quite make it into the HD 5xxx range. Hit the source link for all that precious knowledge.
ATI Radeon HD 5570 fills the last remaining gap in DirectX 11 empire
ATI seems to be so enamored with its 40nm DirectX 11 Evergreen chips that at this point it's bringing out new graphics cards just to remind us of how awesome its technology is. Slotting in between the $100 HD 5670 and $50 HD 5450, the new Radeon HD 5570 will predictably retail at around $75 to $80, with up to 1GB of onboard GDDR3, a 650MHz core clock speed, and a thrifty 43W power budget under full load. Reviewers were big fans of its performance relative to the HD 5450 -- nearly doubling it in some cases -- but still struggled to recommend this as a better value for gamers than the only marginally more expensive HD 5670. Then again, the low profile card, single-slot cooler, and minimal power requirements make the 5570 an absolute gem for HTPC or SFF setups, so whether you're in that particular market or just curious, we recommend you check out the full reviews below. Read - HotHardware Read - AnandTech Read - HardOCP Read - Legit Reviews
ATI Radeon HD 5450 focuses on multimedia features, neglects gaming
It's rare to see a rumor -- hell, even a roadmap -- pinpoint the timing of new releases quite so accurately, but our earlier report of ATI refreshing the middle and lower parts of its lineup turned out to be bang on. Following in the footsteps of the HD 5670, we have the Radeon HD 5450, which drags the entry price for DirectX 11 and Eyefinity multi-monitor support all the way down to $50. Course, the processing power inside isn't going to be on par with its elder siblings, but that also means the 5450 will run cool enough to be offered with half-height, passive cooling solutions as seen above. ATI's focus here is on media PCs, with a DisplayPort, um... port, alongside HDMI 1.3a, Dolby TrueHD and DTS-HD Master Audio bitstreaming support. For the money, you really can't argue with all this extra multimedia juice, but if you must have benchmarks to sate your soul, check out the early reviews below -- they're full of bar charts and performance comparisons, don't you know.
ATI Radeon HD 5770 and 5750 steal away reviewers' hearts
The ATI onslaught of 40nm DirectX 11 parts continues today with two even cheaper variants coming from its Evergreen family of chips. Based on the Juniper core, the HD 5770 and 5750 represent ATI's mainstream value proposition, with compute performance of more than 1 TFLOPS and pricing between $109 and $159. There's a significant dropoff in specs from the world-conquering HD 5870 and similarly potent HD 5850, but reviewers found the latest cards were still up to the task, with the 5770's performance said to be "just shy" of the very recent top dog for ATI, the HD 4890. With low power consumption, competitive pricing and added features like Eyefinity and 7.1 HD audio, the new cards might not push performance boundaries, but they also leave you with little reason to look elsewhere for your next GPU purchase. Hit up the reviews below for more detailed impressions. Read - Hot Hardware review Read - PC Perspective review Read - HardOCP review Read - TweakTown review Read - Techgage review Read - AnandTech review Read - Hexus review
NVIDIA's 40nm GeForce G210 and GeForce GT 220 desktop GPUs emerge
We can't say we're entirely shocked to see 'em, but a new pair of GPUs based on 40 nanometer process technology has surfaced over at NVIDIA's website. Both of the new devices are expected to be sold exclusively to large OEMs for integration into pre-configured machines, and they'll both support DirectX 10.1, OpenGL 3.0, and CUDA. The lower-end GeForce G210 arrives with a 589MHz core clock speed, 512MB of DDR2 RAM and a 64-bit memory interface; meanwhile, the GeForce GT 220 ups the ante with a 615MHz core clock rate, 1GB of GDDR3 RAM and a 128-bit memory interface. As for outputs, the former packs VGA, DisplayPort and DVI, while the latter sticks with VGA, HDMI and DVI. There's no word on when we'll seen them offered in any entry-level desktop rigs, but surely it won't be long now.[Via SlashGear]Read - GeForce G210Read - GeForce GT 220
NVIDIA pops out five new mobile GPUs to fill invisible gaps in its 200M series lineup
NVIDIA is filling in what it presumes to be holes in its next-generation GPU lineup, adding the 40nm G210M, GT 230M, GT 240M and GTS 250M, with GDDR3 memory ranging from 512MB to 1GB, to its existing GTX 280M, GTX 260M and GTS 160M laptop graphics cards. Apparently the new cards sport "double the performance" and "half the power consumption" over the last generation of discrete GPUs they're replacing. The cards are SLI, HybridPower, CUDA, Windows 7 and DirectX 10.1 compatible, and all support PhysX other than the low-end G210M. Of course, with integrated graphics like the 9400M starting to obviate discrete graphics in the mid range -- even including Apple's latest low-end 15-inch MacBook Pro -- we're not sure what we'll do with eight different GPU options, but we suppose NVIDIA's yet-to-be-announced price sheet for these cards will make it all clear in time.
AMD's 40nm ATI Radeon HD 4770 outed, slated for May release?
Ever since we saw the glowing review AMD's ATI Radeon RV740 prototype received, we've been looking forward to the day that the company would make one of these 40nm wonders available. It looks like that day might be close at hand -- according to these purloined slides, a little something called the ATI Radeon HD 4770 is due to make the scene next month in the $99 price point. This handsome lad sports GDDR5 memory, DirectX 10.1 support, a 750 MHz clock speed, a memory clock of 800 MHz using a 128-bit memory bus, a frame buffer size of 512 MB, and much, much more. Curious? Of course you are. Check the slides out below for all of the glorious details.[Via Tom's Hardware]
