Subject: Graphics Cards, Processors | December 8, 2015 - 01:07 PM | Scott Michaud
Tagged: hsa, GCC, amd
Phoronix, the Linux-focused hardware website, highlighted patches for the GNU Compiler Collection (GCC) that implement HSA. This will allow newer APUs, such as AMD's Carrizo, to accelerate chunks of code (mostly loops) that have been tagged with a precompiler flag as valuable to be done on the GPU. While I have done some GPGPU development, many of the low-level specifics of HSA aren't areas that I have too much experience with.
The patches have been managed by Martin Jambor of SUSE Labs. You can see a slideshow presentation of their work on the GNU website. Even though features froze about a month ago, they are apparently hoping that this will make it into the official GCC 6 release. If so, many developers around the world will be able to target HSA-compatible hardware in the first half of 2016. Technically, anyone can do so regardless, but they would need to specifically use the unofficial branch on the GCC Subversion repository. This probably means compiling it themselves, and it might even be behind on a few features in other branches that were accepted into GCC 6.
Subject: Graphics Cards | December 7, 2015 - 04:29 AM | Scott Michaud
Tagged: gigabyte, cooler master, asetek, amd
AMD and Gigabyte have each received cease and desist letters from Asetek, regarding the Radeon Fury X and GeForce GTX 980 Water Force, respectively, for using a Cooler Master-based liquid cooling solution. The Cooler Master Seiden 120M is a self-contained block and water pump, which courts have ruled that it infringes on one of Asetek's patents. Asetek has been awarded 25.375% of Cooler Master's revenue from all affected products since January 1st, 2015.
This issue obviously affects NVIDIA less than AMD, since it applies to a single product from just one AIB partner. On AMD's side, however, it affects all Fury X products, but obviously not the air-cooled Fury and Fury Nano cards. It's also possible that future SKUs could be affected as well, especially since upcoming, top end GPUs will probably be in small packages adjacent HBM 2.0 memory. This dense form-factor lends itself well to direct cooling techniques, like closed-loop water.
Even more interesting is that we believe Asetek was expecting to get the Fury X contract. We reported on an Asetek press release that claimed they received their “Largest Ever Design Win” with an undisclosed OEM. We expected it to be the follow-up to the 290X, which we assumed was called 390X because, I mean, AMD just chose that branding, right? Then the Fury X launched and it contained a Cooler Master pump. I was confused. No other candidate for “Largest Ever Design Win” popped up from Asetek, either. I guess we were right? Question mark? The press release of Asetek's design win came out in August 2014 while Asetek won the patent case in December of that year.
Regardless, this patent war has been ongoing for several months now. If it even affects any future products, I'd hope that they'd have enough warning at this point.
Subject: Graphics Cards, Mobile | December 6, 2015 - 01:10 PM | Scott Michaud
Tagged: nvidia, graphics drivers, 860m
Users of notebooks with the GeForce GTX 860M GPU have apparently been experiencing crashes in many new titles. To remedy these issues, NVIDIA has published GeForce Hotfix Driver 359.12. If you do not have the GeForce GTX 860M, and all of your games work correctly, then you probably shouldn't install this. It has not been tested as much as official releases, by either Microsoft or NVIDIA, so other issues could have been introduced and no-one would know.
If you do have that specific GPU though, and you are having problems running certain titles, then you can install the driver now. Otherwise, you can wait for future, WHQL-certified drivers too. Some users are apparently claiming that the issues were fixed, while others complain about crashes in games like Mad Max and Shadow of Mordor.
Subject: Graphics Cards | December 4, 2015 - 03:34 AM | Sebastian Peak
Tagged: Tonga XT, tonga, Radeon R9 380X, Radeon R9 285, Radeon R9 280X, Radeon R9 280, radeon, amd, 384-bit
While it was reported a year ago that AMD's Tonga XT GPU had a 384-bit memory bus in articles sourcing the same PC Watch report, when the Radeon R9 380X was released last month we saw a Tonga XT GPU with a 256-bit memory interface.
The full Tonga core features a 384-bit GDDR5 memory bus (Credit: PC Watch)
Reports of the upcoming card had consistently referenced the wider 384-bit bus, and tonight we are able to officially confirm that Tonga (not just Tonga XT) has been 384-bit capable all along, though this was never enabled by AMD. The reason? The company never found the right price/performance combination.
AMD confirms 384-bit bus available on Tonga, just not enabled on any product, including 380X. Didn't find a perfect perf/$ slot.
— Ryan Shrout (@ryanshrout) December 4, 2015
AMD's Raja Koduri confirmed Tonga's 384-bit bus tonight, and our own Ryan Shrout broke the news on Twitter.
So does this mean an upcoming Tonga GPU could offer this wider memory bus? Tonga itself was a follow-up to Tahiti (R9 280/280X), which did have a 384-bit bus, but all along the choice had been made to keep the updated core at 256-bit.
Now more than a year after the launch of Tonga a new part featuring a fully enabled memory bus doesn't seem realistic, but it's still interesting to know that significantly more memory bandwidth is locked away from owners of these cards.
Subject: Graphics Cards | November 30, 2015 - 08:48 PM | Sebastian Peak
Tagged: rumor, report, price cut, nvidia, GTX 980, GTX 970, gtx 960, geforce
A report published by TechPowerUp suggests NVIDIA will soon be cutting prices across their existing GeForce lineup, with potential price changes reaching consumers in time for the holiday shopping season.
So what does this report suggest? The GTX 980 drops to $449, the GTX 970 goes to $299, and the GTX 960 goes to $179. These are pretty consistent with some of the sale or post-rebate prices we’ve seen of late, and such a move would certainly even things up somewhat between AMD and NVIDIA with regard to cost. Of course, we could see an answer from AMD in the form of a price reduction from their R9 300-series or Fury/Nano. We can only hope!
We’ve already seen prices come down during various black Friday sales on several GPUs, but the potential for a permanent price cut makes for interesting speculation if nothing else. Not to disparage the source, but no substantive evidence exists to directly point to a plan by NVIDIA to lower their GPU prices for some 900-series cards, but it would make sense given their competition from AMD at various price points.
Here’s to lower prices going forward.
Subject: Graphics Cards | November 29, 2015 - 10:52 PM | Scott Michaud
Tagged: nvidia, cancer research, gpgpu
The University of Toronto has just received a $200,000 grant from the NVIDIA Foundation for research in identifying genetic links to cancer. The institution uses GPUs to learn and identify mutations that cause the disease, which is hoped to eventually help diagnose the attributes of cancer for a specific patient and provide exact treatments. Their “next step” is comparing their technology with data from patients.
I am not too informed on cancer research, so I will point to the article and its sources for specifics. The team state that the libraries they create will be freely available for other biomedical researchers. They don't mention specific licenses or anything, but the article is not really an appropriate venue for that sort of discussion.
Subject: Graphics Cards | November 29, 2015 - 10:25 PM | Scott Michaud
Tagged: amd, graphics driver, radeon, crimson
Users have been reporting that the latest AMD graphics driver, Radeon Software Crimson Edition, has been incorrectly setting fan speeds. Some users report that the driver spins up fans to 100% and others report that they slow down to 30% regardless of load.
Over the weekend, AMD acknowledged the issue and claim that a fix is intended for Monday.
Some users also claim that the card will stick with that fan setting until it cooks itself. This seems odd to me, since GPUs (and CPUs of course) are now designed to down-volt if temperatures reach unsafe levels, and even cut power entirely if heat cannot be managed. We haven't really seen reports of graphics cards cooking themselves since the Radeon HD 5000 series implemented hardware in response to Furmark and OCCT. That said, the driver bug might some how override these hardware protections.
In the mean time, you'll either want to keep an eye on your fan settings and reset them as necessary, or roll back to the previous driver. AMD didn't comment on the high fan speed issue that some were complaining about, so I'm not sure if this fix will address both issues.
Subject: Graphics Cards | November 27, 2015 - 02:42 AM | Scott Michaud
Tagged: asus, 980 Ti
As most of our readers are well aware, the graphics market is dominated by two GPU vendors, both of which sell chips and reference designs to add-in board partners. ASUS is one of the oldest add-in board partners. According to their anniversary website, ASUS even produced a card that was based on the S3 ViRGE/DX graphics chipset all the way back in 1996.
To celebrate their 20th anniversary, although they don't exactly state when they start counting, ASUS has released a few high-end versions of Maxwell-based graphics cards. This one is the ASUS 20th Anniversary Gold Edition GTX 980 Ti. It comes with a base clock speed of 1266 MHz, which boosts up to 1367 MHz as needed. This is quite high, considering the reference card is clocked at 1000 MHz and boosts to 1189 MHz, although overclocking the 980 Ti is not too difficult to begin with. Ryan got up to 1465 MHz with a reference card. The Gold Edition GTX 980 Ti might go even higher with its enhanced cooling and power delivery, and it's designed for liquid nitrogen if you're that type of enthusiast.
Speaking of liquid nitrogen features, the card advertises a “Memory Defroster” feature that looks rather extreme. I can't say that I've ever seen a graphics card get covered in a visible layer of ice, but I've also never attached it to a reservoir of liquid with a temperature that's easier to write in Kelvin than Celcius.
Is this a legit problem? Or does this seem like “anti-dust shield” to everyone else too?
The ASUS 20th Anniversary Gold Edition GTX 980 Ti ships this month.
Subject: Graphics Cards | November 26, 2015 - 08:09 PM | Scott Michaud
Tagged: amd, graphics drivers, GCN, terascale
The Graphics Core Next (GCN) architecture is now a minimum requirement for upcoming AMD graphics drivers. If your graphics card (or APU) uses the TeraScale family of microarchitectures, then your last expected WHQL driver is AMD Catalyst 15.7.1 for Windows 7, 8.x, and 10. You aren't entirely left out of Radeon Software Crimson Edition, however. The latest Crimson Edition Beta driver is compatible with TeraScale, but the upcoming certified one will not be.
GCN was introduced with the AMD Radeon HD 7000 series, although it was only used in the Radeon HD 7700 series GPUs and above. The language doesn't seem to rule out an emergency driver release, such as if Microsoft breaks something in a Windows 10 update that causes bluescreens and fire on older hardware, but they also don't say that they will either. NVIDIA made a similar decision to deprecate pre-Fermi architectures back in March of 2014, which applied to the release of GeForce 343 Drivers in September of that year. Extended support for NVIDIA's old cards end on April 1st, 2016.
I wonder why AMD chose a beta driver to stop with, though. If AMD intended to support TeraScale with Crimson, then why wouldn't they keep it supported until at the first WHQL-certified version? If they didn't intend to support TeraScale, then why go through the effort of supporting it with the beta driver? This implies that AMD reached a hurdle with TeraScale that they didn't want to overcome. That may not be the case, but it's the first thing that comes to my mind none-the-less. Probably the best way to tell is to see how people with Radeon HD 6000-series (or lower-end 7000/8000-series) cards work with Radeon Software Crimson Beta.
Likely the last drivers that users with Radeon HD 6000-series graphics need are 15.7.1 or Radeon Software Crimson Edition Beta. We will soon learn which of the two will be best long-term.
Or, of course, you can buy a newer GPU / APU when you get a chance.
Subject: Graphics Cards | November 25, 2015 - 07:54 PM | Jeremy Hellstrom
Tagged: radeon, r9 nano, mini ITX, amd, obsidian 250d, corsair
When Ryan tested out how the R9 Nano performs in tiny cases he chose the Cooler Master Elite 110, the Raijintek Metis, the Lian Li PC-Q33BL and their PC-Q30X. The card did slow down somewhat because of a lack of airflow in the case but that was quickly remedied with a drill press and we saw vast improvements in the in-game frequencies. [H]ard|OCP performed a similar experiment with the Cooler Master Elite 110 as well and found similar results.
They are now back at it again, this time testing in a Corsair Obsidian Series 250D Mini ITX case, which is large enough to accommodate a full sized GPU and provide improved airflow. They tested the Nano against a GTX 980 Ti and a R9 Fury X as they cost a similar amount to the tiny little Nano. They tested the cards at both 1440p and 4K resolutions and as you might reasonably expect the Nano fell behind, especially at 4K. If you have a case which can fit a full sized GPU then the Nano does not make sense to purchase, however in cases in which the larger cards will not fit then the Nano's performance is unmatched.
"Our second installment covering our AMD Radeon R9 Nano in a Small Form Factor chassis is finally done. We will upgrade the case to a Corsair Obsidian Series 250D Mini ITX PC Case and compare the R9 Nano to price competitive video cards that can be installed. We game at 1440p and 4K for the ultimate small form factor experience."
Here are some more Graphics Card articles from around the web:
- AMD Radeon Software Crimson Edition Is A Letdown On Linux @ Phoronix
- Radeon Software Crimson; The Review @ Hardware Canucks
- AMD Radeon Software Crimson Editon Detailed Briefing @ TechARP