News articles search results by Chris Ledenican
Automatically adjusts settings for best experience
Along with the GTX 690 being released today, Nvidia has also created a new GeForce Experience feature that allow gamers to play their games at the maximum settings possible or their GPU and CPU configuration automatically. The GeForce experience works by connecting to a super computer that has stored settings for virtually any combination of Nvidia graphics cards and processors.
To connect to the super computer all that is needed it to hit a simple optimize button, which sends the request to the super computer allowing it to find the best settings for your desktop or laptop. After it finds the best settings for the game, the program takes care of the rest, allowing the gamer to simply enjoy the game without having to dig through a complicated control panel.
The GeForce Experience was first introduced by Jen-Hsun Huang, president and CEO at Nvidia, during the GeForce LAN event in Shanghai. If you missed it we have included the full introduction to the GeForce Experience below.
Traditionally, graphics cards launches follow a set order. Usually the flagship single GPU solution is released and then followed by the more affordable graphics cards in the stack. However, Nvidia is turning this on its head by releasing the GTX 690 just a month after the GTX 680 launched.
Making today’s announcement even more exciting is the GTX 690 is the most powerful and refined graphics card Nvidia has built to date. This is clear just for the exterior of the graphics card, as it doesn’t use plastic casing or generic materials. Instead it is utilizes a thermal solution constructed of cast aluminum and magnesium alloy with polycarbonate windows and nickel plating, making it the most expensive and solidly constructed graphics card Nvidia has ever built. Beyond the exterior though the GTX 690 is also built on a 10 layer 2oz PCB, has ample 10 phase power supply and two GK-104 graphics processors.
By now most of us are familiar with the GTX 680, and the new features such as Nvidia GPU Boost technology, which is a major factor in the performance of the GTX 690. According to Nvidia the GTX 690 is able to perform nearly 2x faster than the single GPU based GTX 680. This is due to the power efficiency of the Kepler architecture, which allows Nvidia to keep the Boost speed of the dual cores above 1GHz, while staying in an acceptable power envelope. The power efficiency, along with the robust thermal solution also allows the GTX 690 to run at a low acoustic level of only 47dBA, making it as quiet as many case fans.
The GTX 690 also utilizes binned graphics processors that were handpicked for their low leakage and power efficiency. Essentially, this means Nvidia has selected the best of the best graphics processors, which allowed them to keep the base and boost clock at 915MHz and 1019MHz respectively. But like the GTX 680 the Boost clock is just an estimate and it should run at faster than 1015MHz during gaming. In addition the GTX 690 has dual Kepler cores that internally feature 3072 CUDA cores, 64 ROPs, 128 Texture units and a 4GB GDDR5 frame buffer running on a 512-bit interface. So, Nvidia didn’t scale anything back, as the GTX 690 packs in two full GK-104 graphics processors.
So, what is this beast going to cost? Well... at $999 it is double the cost of a single GTX 680. The tradeoff though is it is a single PCB solution that is quieter than dual GTX 680s and comes with massive bragging rights.
Along with the GTX 690, Nvidia also introduced a new program called the GeForce Experience. Essentially, GeForce Experience has been a long term program where Nvidia ran programs and games through a super computer for it to find the best settings for the GPU and CPU configuration being used. When being used the program allows a host system to connect to the super computer allowing gamers that traditionally don’t change the in game settings to enjoy the maximum eye candy available by automatically allowing the program to adjust the settings. There is some user interaction still, but it is only limited to hitting an optimize button, and GeForce Experience takes care of the rest.
Be sure to check back May 3rd for our full launch review.
We have been sent plenty of odd things over the years, but a crowbar for killing zombies might take the cake.
The crowbar is an obvious tease for an upcoming GeForce launch, which is widely expected to be a dual GPU GTX 690. The latest obscure tease comes on the heels of a message post last week on the GeForce Facebook page stating, “Something is coming." The crowbar shown below is inscribed with: “For use in case of Zombies or…”
The "or" could be one of many possibilities, such as being the means to open something GeForce related; that or Nvidia actually plans on starting the zombie apocalypse and just wanted to make sure some of us were prepared. Whatever it is though, we will be here covering it... unless we have already turned.
Mystery image pops up on social site
Nvidia released a teaser image of an upcoming product today via their Facebook page. The image consist only of a vague close up of a mystery product with the following message: “It’s coming."
Vague, much? Speculation of what the mystery product is has been all over the map, with guesses ranging from it being a dual GPU GTX 690 to a GK-100 based graphics card, or even a new case. Whatever it is though, we should know within the next few weeks, and rest assured we will have an in-depth article of this yet to be disclosed product.
In the mean time, feel free to indulge in some rumors.
Lower prices and three free games
The graphics segment is perhaps the most competitive in the PC market, and today things are really starting to heat up.
Just three weeks ago, Nvidia launched the GTX 680 to wide spread acclaim, as it was both faster and more affordable than the HD 7970. This of course meant AMD would have to drop the prices of their Southern Islands graphics cards to counter the Kepler launch, but no official word has been released until today.
According to an email sent to us over the weekend, AMD is ready to revamp the pricing structure for most of their current gen graphics cards, effective immediately. However, they are also unveiling a new promotion called the “Three for Free” Program.
Starting with the new pricing structure, the Radeon HD 7970 will now retail for as low as $479, while the HD 7950 has an MSRP of $399. The HD 7770 is also getting an updated price tag, as it now can be found for as low as $139. There was no mention of the HD 7870 or HD 7850, so we expect those graphics cards will remain at $349 and $249, respectively.
The “Three for Free” program is a partnership between Codemasters, THQ, and AMD, where three free games will be bundled with AMD’s high-end graphics cards. The games include upcoming titles DiRT Showdown and Nexuiz, as well as the widely popular Deus Ex: Human Revolution and its DLC expansion, The Missing Link.
28nm production to pick up in Q3 2012
It has been no secret TSMC -- the world's largest dedicated independent semiconductor foundry -- has not been able to deliver when it comes to producing high yields for their latest 28nm parts.
This shortage, though affecting other companies, is predominately hurting AMD and Nvidia, as neither company can ship their latest 28nm graphics cards in high-volume. In the case of Nvidia, it has caused them to delay the launch of more 28nm models. What this means for you is it could be some time before the GTX 680 becomes readily available and for AMD to reduce the prices of their Southern Islands graphics line to Kepler prices.
The good news is production is expected to pick up in Q3 2012, at which point we should see a dramatic increase is the volume of both AMD and Nvidia products.
NVIDIA's new GeForce GTX 680 has high overclocking potential, as even under standard conditions we were able to increase the GPU clock speed to nearly 1300MHz. So imagine what could be possible if the VRM was modified for high voltage AND the core was chilled to -150°C. This is exactly what a team of overclockers from EVGA did.
The team was led by the "Kingp|n" himself, renowned in the enthusiast circle for his overclocking abilities. To achieve the highest clocks possible, the team used an add-on PWM board with additional power connectors and chilled the core using LN2. The end result was a GTX 680 running at an astonishing 1842MHz.
At 1842MHz, the system was capable of scoring 14912 points in the 3DMark 11 performance test. This is over 4000 marks higher than what we achieved in our own labs. However, the real kicker is the 14912 score was achieved with tessellation set to the maximum level. Taking this into account, the actual score in comparison to our results would have been between 5000 and 6000 marks.
Along with the modified GTX 680, the test system utilized an Intel Core i7-3930K processor overclocked to 5.5GHz and memory running at 2400MHz.
Traditional displays have not changed much over the years, but there have been some advancements such as stereoscopic 3D and improved resolutions. Researchers at Microsoft’s Applied Sciences division are looking to change this however, as they have developed the first see-through 3D display.
The clear display works by utilizing an internal 3D grid to create a virtual desktop while depth cameras detect the user’s interactions, allowing them to manipulate the images on the screen. Additionally, the screen tracks and captures the head motion of the user to giving them the most appropriate perspective of the images on the display.
Of course it will be quite some time before displays like this are available in the market, but the video above does give us an idea of where display technology is headed.
Southern Island’s graphics cards have dominated the news cycles since their release early this year, but according to VR-Zone, Kepler is fast approaching and the architecture looks promising.
According to the site, the rumored specifications list the GTX 680 as having 1536 CUDA cores, which is three times more than the GTX 580. Like the previous generation, the GTX 680 utilizes "hotclocks" and the GPU clock speed is set at 705 MHz, meaning the shaders will operate at 1411 MHz. Additionally, the rumored specifications show the card to have a 2GB of GDDR5 memory clocked at 6000MHz and running on a 256-bit memory interface. With this sub-system the GTX 680 should have a total memory bandwidth rating of 192GB/s.
Currently this is all speculation, but we will keep you informed on any information relating to Kepler just as soon as we have it.
At the Mobile World Congress event held in Barcelona, Microsoft previewed the latest features and design of their upcoming Windows 8 operating system. The new software was showcased on various mobile and desktop devices before it was announced the consumer preview version of the OS would be immediately be available as a free download via their website.
The Windows 8 consumer preview is a follow-up to the Windows 8 Developer Preview, which was released last September. The difference between the two is the consumer preview includes more features such as wall-to-wall web browsing, full touch screen support, and of course the new look of the Windows 8 UI, making the preview closer to final product expected to come out later this year.
Windows 8 represents one of the largest changes we have seen to the operating system since its very first iteration. Hit the source below to check it out for yourself.
Nvidia just released their latest WHQL driver, which comes packed with performance improvements, enhanced ambient occlusion support, and SLI profiles.
The 295.73 driver can boost performance in Skyrim by up to 40%, and the ambient occlusion first added in their 290.36 beta driver has been improved to increase the graphical fidelity of shadows while limiting the impact on the overall performance. Additionally, ambient occlusion support is added for Diablo III, and SLI performance in Mass Effect III is boosted and now scales up to 90% in SLI (not that you'll need it).
Other highlights of the R295 driver include new 3D Vision and SLI profiles for over 50 titles and new PhysX software for titles such as Alice: Madness Returns and Batman: Arkham City.
Hit the source for the full details and download.
Non-reference designs on launch day
With the launch of the HD 7950 getting closer, it is no surprise leaks are coming far and wide, but today we finally almost get the complete picture. According to an engineering sample detailed on Hard Forum, the HD 7950 will launch with a street price of $449 and included 1792 Stream processors, 32 ROPS, and come equipped with a 3GB frame buffer running on a 384-bit interface. Additionally, the GPU and memory clock speeds on this beast are set at 800MHz and 1250MHz (5000MHz QDR), respectively.
However, most models being released are ditching the reference design for a more robust thermal solution and higher clock speeds. Since the launch is still a few weeks out, we are unsure of the exact clock speeds being used, but the pictures speak for themselves when it comes to the aftermarket design.
The HD 7950 is expected to launch January 31, so make sure to check back for the full launch review, and possibly a CrossFire and TriFire review as well.
Looking polished months before launch
The Tahiti XT graphics processor is currently the fastest GPU on the market, but all signs are pointing to this being a short lived victory for AMD. The word is Nvidia’s Kepler architecture is going to handily outperform any Tahiti based graphics cards, and even two to three months prior to launch the GK104 chip, is looking very polished, though there could be some issues with availability at launch.
The news comes way of SemiAccurate, who got their hands on two Kepler based graphics cards. While was sparing on the details, what we do know is the video configuration consists of dual DVI links, one HDMI output, and a single DisplayPort connector. Additionally, the PCB includes eight on-board memory chips, giving the card a 256-bit memory bus; each memory module is 256MB, which equates to 2GB total.
Other than these rumors, little is known about the graphics card, but more details should trickle out as we get closer to the March/April launch.
Intel’s “K” and “X” series processors are designed for overclocking, but even while they are equipped with an unlocked multiplier, there is no guarantee they can withstand high voltages and frequencies on a 24/7 basis. This can potently be problematic because once a processor is overclocked the warranty is void and it can no longer simply be sent in for repair, leaving only one option: shell out for a new processor.
As of today though, Intel is changing this with their new “Performance Tuning Protection Plan”. Essentially this plan gives consumers the option to purchase an overclocking warranty for their “K”, “X”, and LGA2011-socketed boxed processors. The cost of the plans range from $20 to $35 USD depending on the processor, and lasts up to three years, covering one replacement.
This is a bold move by Intel and one that should increase their street cred with the overclocking community. We would also like to see AMD follow suit and offer an overclocking warranty of their own.
While it’s not quite as disastrous as the infamous fake GTX 480 incident, Intel was caught faking their results during a live CES demo of what was supposed to be an Ivy Bridge based ultrabook running F1 2011.
The event took place yesterday afternoon in a crowded room where the media had been gathered to see Intel’s 2012 roadmap. During the event, Intel vice president and PC group general manager Mooly Eden ran a demo of Codemaster’s F1 2011 on an Ivy Bridge based ultrabook. The problem is Mooly Eden clearly acts as if he is behind the wheel and the game is running in real time. However, videos of the event clearly show a VLC media bar at the bottom of the screen on multiple occasions, showing the game was actually a recording and not a live demo.
Below is just one of many videos circulating around the net showing the faked demo; skip ahead to 16 seconds to see the beginning of the F1 2011 failure.
Strangely enough, Ivy Bridge is capable of playing F1 2011, and there is a real demo showing this to be the case. This raises the question of why Intel decided to fake the demo in the first place, as it would have been just as easy (and less controversial) to run the game in real time instead of playing a video clip.
While the second generation SandForce controller has mostly been bug free, there have been a few instances where SSDs using the controller would experience a blue screen during standard use. For the most part these issues were random and addressed with the latest 3.3.2 firmware patch. However, the issues users are experiencing with their 64GB Crucial M4 drives isn't directly linked to controller, but rather the time the SSD is active.
The M4 SSD will experience a x00000f4 error after approximately 5,000 hours, at which time the drive will crash and require a restart to regain function. After this point the drive will crash again after each additional hour, and continue on this way in a loop. Crucial has acknowledged the matter and informed us a fix is on the way.
“We are aware of the issue that is currently affecting a small number of users whereby their Crucial m4 SSD causes their system to require a restart," a spokespoerson said. "This issue occurs after approximately 5,000 hours of actual “on time” use. Following the initial reboot, the system then requires subsequent restarts after each additional hour of use. However, the data on the SSD is unaffected and will not be lost due to this condition."
"Through our investigation, we have determined the root cause of the problem and will be releasing a firmware update that rectifies the situation. We are currently running through our validation and compatibility process. Once this process is complete, the firmware will be made available to our customers. We are currently targeting the week of January 16th, 2012 to publicly release the new firmware update.”
As of now we are unsure of exactly how many drives are affected by this issue, or if it is isolated only to 64GB M4 SSDs. We will continue to reach out to both Crucial and SandForce to see if this is an issue that could potentially affect other drives when they hit 5,000 hours of use. Still, at least we know the issue is being worked on and users can take some comfort in the fact stored data is not affected.
In our review of the HD 7970, we were able to push the GPU clock speed up to 1125Mhz with no additional voltage. However, since the graphics card we reviewed was BIOS limited there was no chance to push it further, so we couldn't tell how much additional headroom was left. According to sources over at Chiphell, the HD 7970 is capable of scaling above 1300MHz.
Sapphire is going to release two SKUs clocked at 1335MHz, and both are going to be part of the Atomic series. The difference between the two models is the RX utilizes an air cooled thermal solution, while the WX is going to be water cooled. Sapphire also has a full lineup for the HD 7970 including reference models and even one with a 6GB GDDR5 memory buffer.
The most interesting aspect of the image though is not what is clearly visible, but what has been crossed out. At the top of the image are two graphics cards listed with a steaming processor count of 2304. Now we all know the HD 7970 has 2048 streaming processors, and we haven't heard anything regarding a Southern Islands GPU with a higher SP count. This caught our eye, so we did a little digging and found it to be just one version of the Tahiti XT GPU AMD was testing prior to launch, and also that (according to sources at PC Perspective) it doesn't include any unlockable SPs. Still, it is interesting to see manufacturers were actively preparing for a HD 7970 with 2304 streaming processors.
This could also potentially mean AMD have in their possession a stronger GPU than the current flagship Tahiti XT core, which they could release to counter Kepler later this year. If such a GPU exists, it could potentially be up to 25% to 45% faster than the HD 7970, depending on the clock speed -- possibly a HD 7980 or 8970.
It's inevitable: when a new an architecture is coming down the pipe, there will always be leaks; today's once again comes courtesy of China. The website Chiphell has leaked pre-release images of the HD 7770 graphics cards and benchmarks, which show the HD 7770 to be a strong competitor in the ~$150 market.
According to the site, when the HD 7770 is paired with an Ivy Bridge Core i5-3550K processor clocked at 3.3Ghz, it is able to achieve a score of 1077 marks, making it roughly 25% to 30% faster than its predecessor, the HD 6770. Additionally, the HD 7770 is rumored to have a 2GB frame buffer running on a 128-bit memory interface, and include internal specifications consisting of 896 stream processors, 56 texture units, and 16 ROPs. Sources say all of this should be packed into a chips measuring 245 square millimeters, and will be priced at $149.
If the whispers we have been hearing regarding this graphics card are true, look for it to hit the market sometime around February or March. Of course, as always, nothing is set in stone until the official word comes in. But if there is any truth to these rumors, rest assured we will have a full launch review when the new graphics cards are released.
Unfortunately, there is still no news on the HD 7800 series, but at the rate the Southern Islands leaks have been coming, we shouldn't have to wait much longer
4.3 billion transistors, 1792 Steaming processors, 28 Compute Units
Just before Christmas AMD paper launched their flagship HD 7970 graphics card, which is scheduled for a hard release January 9, 2012. This development focused all the attention on the high-end Tahiti XT processor, as there was little to no info released regarding the slightly slower Tahiti Pro GPU. All of this changes today, however, as we get our first look at the rumored specifications for the HD 7950.
In all the Tahiti Pro GPU looks to be nearly identical to the Tahiti XT core, but with 4 fewer compute units, which reduces the total amount of Streaming processors to 1792, in comparison to the 2048 Streaming processors on the HD 7970. The rumored specs also show the HD 7950 has a 3GB GDDR5 frame buffer running on a 384-bit interface, and utilizes the same video output configuration as the HD 7970. If these specifications are right, the HD 7950 should have solid solid gaming performance, and offer all the same expansion options as its older sibling.
The only pieces of the puzzle missing from the specifications sheet are the total TDP, clock speeds, and the cost of the HD 7950. Sources say we should get the missing pieces January 9, but this is not confirmed, so for the time being all this is mere speculation.
Additionally, there is no answer to the burning question, "Does it unlock?".
"Unacceptable", says founder
In most parts of the world the Internet is an open resource, with an endless amount of content, where the words “restricted” or “limited” are not usually taken to kindly. So, when domain registrar Go Daddy announced its support of the Stop Online Piracy Act (SOPA), at least one of their clients took notice and decided to protest with its pocketbook: Wikipedia.
Founder Jimmy Wales noted on Twitter, "I am proud to announce that the Wikipedia domain names will move away from GoDaddy. Their position on SOPA is unacceptable to us."
More than 70,000 domains were also lost in a week, with 21,054 domains transferring just this past Friday alone, though it's not proven how many of these are SOPA-related.
Go Daddy has since changed their stance on SOPA, opposing the bill and noting they're now better aware of its implications, though whether this is genuine or not is a little difficult to say.
Interestingly, not only was Go Daddy a supporter of SOPA, they actually helped craft part of the bill and were even exempt from it. So, while other smaller sites could be targeted by the DOD for pirating, Go Daddy essentially has a get out of jail free card.
Ed. note: this article was edited December 30 to eliminate assumptions and focus on facts. Consider it an extremely rare oversight, one I've already made moves to prevent from happening again.