LinkedIn Posts Hint at Radeon R9 380X Features, Stacked Memory

Subject: Graphics Cards | January 13, 2015 - 12:22 PM |
Tagged: rumor, radeon, r9 380x, 380x

Spotted over at TechReport.com this morning and sourced from a post at 3dcenter.org, it appears that some additional information about the future Radeon R9 380X is starting to leak out through AMD employee LinkedIn pages.

Ilana Shternshain is a ASIC physical design engineer at AMD with more than 18 years of experience, 7-8 years of that with AMD. Under the background section is the line "Backend engineer and team leader at Intel and AMD, responsible for taping out state of the art products like Intel Pentium Processor with MMX technology and AMD R9 290X and 380X GPUs." A bit further down is an experience listing of the Playstation 4 APU as well as "AMD R9 380X GPUs (largest in “King of the hill” line of products)."

Interesting - though not entirely enlightening. More interesting were the details found on Linglan Zhang's LinkedIn page (since removed):

Developed the world’s first 300W 2.5D discrete GPU SOC using stacked die High Bandwidth Memory and silicon interposer.

Now we have something to work with! A 300 watt TDP would make the R9 380X more power hungry than the current R9 290X Hawaii GPU. High bandwidth memory likely implies memory located on the substrate of the GPU itself, similar to what exists on the Xbox One APU, though configurations could differ in considerable ways. A bit of research on the silicon interposer reveals it as an implementation method for 2.5D chips:

interposer.jpg

Source: SemiWiki.com

There are two classes of true 3D chips which are being developed today. The first is known as 2½D where a so-called silicon interposer is created. The interposer does not contain any active transistors, only interconnect (and perhaps decoupling capacitors), thus avoiding the issue of threshold shift mentioned above. The chips are attached to the interposer by flipping them so that the active chips do not require any TSVs to be created. True 3D chips have TSVs going through active chips and, in the future, have potential to be stacked several die high (first for low-power memories where the heat and power distribution issues are less critical).

An interposer would allow the GPU and stacked die memory to be built on different process technology, for example, but could also make the chips more fragile during final assembly. Obviously there a lot more questions than answers based on these rumors sourced from LinkedIn, but it's interesting to attempt to gauge where AMD is headed in its continued quest to take back market share from NVIDIA.

Source: 3dcenter.org

CES 2015: EVGA Shows Two New GTX 980 Cards

Subject: Graphics Cards, Shows and Expos | January 8, 2015 - 12:46 AM |
Tagged: video, maxwell, Kingpin, hydro copper, GTX 980, GM204, evga, classified, ces 2015, CES

EVGA posted up in its normal location at CES this year and had its entire lineup of goodies on display. There were a pair of new graphics cards we spotted too including the GTX 980 Hydro Copper and the GTX 980 Classified Kingpin Edition.

evga-1.jpg

Though we have seen EVGA water cooling on the GTX 980 already, the new GTX 980 Hydro Copper uses a self-contained water cooler, built by Asetek, rather than a complete GPU water block. The memory and power delivery is cooled by the rest of the original heatsink and blower fan but because of lowered GPU temperatures, the fan will nearly always spin at its lowest RPM.

evga-2.jpg

Speaking of temperatures, EVGA is saying that GPU load temperatures will be in the 40-50C range, significantly lower than what you have with even the best air coolers on the GTX 980 today. As for users that already have GTX 980 cards, EVGA is planning to sell the water cooler individually so you can upgrade yourself. Pricing isn't set on this but it should be available sometime in February.

evga-3.jpg

Fans of the EVGA Classified Kingpin series will be glad to know that the GTX 980 iteration is nearly ready, also available in February and also without an known MSRP.

evga-4.jpg

EVGA has included an additional 6-pin power connector, rearranged the memory traces and layout for added memory frequency and includes a single-slot bracket for any users that eventually remove the impressive air cooler for a full-on water block.

Coverage of CES 2015 is brought to you by Logitech!

PC Perspective's CES 2015 coverage is sponsored by Logitech.

Follow all of our coverage of the show at http://pcper.com/ces!

NVIDIA Quietly Releases the GeForce GTX 965M Mobile GPU

Subject: Graphics Cards | January 7, 2015 - 10:51 AM |
Tagged: nvidia, notebook, mobile graphics, mobile gpu, GeForce 965M

With zero fanfare NVIDIA has released a new mobile graphics chip today, the GeForce GTX 965M.

geforce-gtx-965m-3qtr.png

Based on the 28nm Maxwell GM204 core and positioned just below the existing GTX 970M, the new GTX 965M has 1024 CUDA cores (compared to the 970M's 1280) and the new 965M has a lower 128-bit memory interface (vs 192-bit with the 970M). The base clock is slightly faster at 944 MHz (plus unspecified Boost headroom).

Compared with the flagship GTX 980M which boasts 1536 CUDA cores and 256-bit GDDR5 this new GTX 965M will be a significantly lower performer, but NVIDIA is marketing it towards 1080p mobile gaming. At a lower cost to OEMs the 965M should help create some less expensive 1080p gaming notebooks as the new GPU is adopted.

geforce-gtx-965m-front.png

The chip features proprietary NVIDIA Optimus and Battery Boost support, and is GameStream, ShadowPlay, and GameWorks ready.


Specs from NVIDIA:

  • CUDA Cores: 1024
  • Base Clock: 944 MHz + Boost
  • Memory Clock: 2500 MHz
  • Memory Interface: GDDR5
  • Memory Interface Width: 128-bit
  • Memory Bandwidth: 80 GB/sec
  • DirectX API: 12
  • OpenGL: 4.4
  • OpenCL: 1.1
  • Display Resolution: Up to 3840x2160

More information on this new mobile GPU can be found via the source link.

Source: NVIDIA

CES 2015: AMD Talks Technical about FreeSync Monitors

Subject: Graphics Cards, Displays, Shows and Expos | January 7, 2015 - 03:11 AM |
Tagged: video, radeon, monitor, g-sync, freesync, ces 2015, CES, amd

It finally happened - later than I had expected - we got to get hands on with nearly-ready FreeSync monitors! That's right, AMD's alternative to G-Sync will bring variable refresh gaming technology to Radeon gamers later this quarter and AMD had the monitors on hand to prove it. On display was an LG 34UM67 running at 2560x1080 on IPS technology, a Samsung UE590 with a 4K resolution and AHVA panel and BenQ XL2730Z 2560x1440 TN screen.

free-1.jpg

The three monitors sampled at the AMD booth showcase the wide array of units that will be available this year using FreeSync, possibly even in this quarter. The LG 34UM67 uses the 21:9 aspect ratio that is growing in popularity, along with solid IPS panel technology and 60 Hz top frequency. However, there is a new specification to be concerned with on FreeSync as well: minimum frequency. This is the refresh rate that monitor needs to maintain to avoid artifacting and flickering that would be visible to the end user. For the LG monitor it was 40 Hz.

free-3.jpg

What happens below that limit and above it differs from what NVIDIA has decided to do. For FreeSync (and the Adaptive Sync standard as a whole), when a game renders at a frame rate above or below this VRR window, the V-Sync setting is enforced. That means on a 60 Hz panel, if your game runs at 70 FPS, then you will have the option to enable or disable V-Sync; you can either force a 60 FPS top limit or allow 70 FPS with screen tearing. If your game runs under the 40 Hz bottom limit, say at 30 FPS, you get the same option: V-Sync on or V-Sync off. With it off, you would get tearing but optimal input/display latency but with it off you would reintroduce frame judder when you cross between V-Sync steps.

There are potential pitfalls to this solution though; what happens when you cross into that top or bottom region can cause issues depending on the specific implementation. We'll be researching this very soon.

free-3.jpg

Notice this screen shows FreeSync Enabled and V-Sync Disabled, and we see a tear.

FreeSync monitors have the benefit of using industry standard scalers and that means they won't be limited to a single DisplayPort input. Expect to see a range of inputs including HDMI and DVI though the VRR technology will only work on DP.

We have much more to learn and much more to experience with FreeSync but we are eager to get one in the office for testing. I know, I know, we say that quite often it seems.

Coverage of CES 2015 is brought to you by Logitech!

PC Perspective's CES 2015 coverage is sponsored by Logitech.

Follow all of our coverage of the show at http://pcper.com/ces!

CES 2015: EVGA Teases New NVIDIA GeForce GTX 980 with AIO Liquid Cooling

Subject: Graphics Cards | January 6, 2015 - 11:30 PM |
Tagged: nvidia, liquid cooler, GTX 980, gpu cooler, gpu, evga, ces 2015, CES, AIO

EVGA has posted a photo on Twitter of a new GTX 980 with an integrated AIO liquid cooler.

GTX_980_HC_AIO.jpg

The pic is captioned "GTX 980 HC AIO", which indicates that it will join the EVGA GTX 980 Hydro Copper (which carries an MSRP as $799) as a liquid-cooled option in their lineup. The big advantage here, however, is that AIO setup dangling off the back of the card. One free (120mm?) fan opening is all you'd need to be up and running without any extra work.

Of course, you could always buy yourself a suitcase full of AIO liquid-cooled GTX 980's for a cool $2999 if you don't want to wait for this EVGA option.

wf-1.jpg

Triple SLI + AIO liquid cooling = suitcase?

We'll post news of this (seemingly) upcoming EVGA product once details are revealed.

Coverage of CES 2015 is brought to you by Logitech!

PC Perspective's CES 2015 coverage is sponsored by Logitech.

Follow all of our coverage of the show at http://pcper.com/ces!

Source: Twitter

Report: NVIDIA Maxwell GM206 Pictured - Leak Claims GTX 960 Core

Subject: Graphics Cards | January 6, 2015 - 09:44 AM |
Tagged: rumor, nvidia, leak, gtx 960, GM206, geforce

VideoCardz.com is reporting that they not only know the upcoming GTX 960 core will be the GM206, but they reportedly have a photo of the unreleased chip.

NVIDIA-Maxwell-GM206-300-GPU.png

Why are reported leaks always slightly out of focus? (Credit: VideoCardz.com)

The chip pictured appears to be a GM206-300, which the site claims will be the exact variant in the GTX 960 when it is released. The post speculates that based on the die size we can expect between 8 - 10 SMM's, or 1080 - 1280 CUDA cores. They further claim that the GTX 960 will have a 128-bit memory bus and reference cards will have a 2GB frame buffer (though naturally we can expect models with 4GB of memory after launch).

ASUS-GTX960-DC2OC-2GD5-PRO-2GB.png

(Credit: VideoCardz.com)

The post goes on to show what appears to be a search result for an ASUS GTX 960 on their site, but if this existed it has since been taken down. More than likely a GTX 960 is in fact close at hand, and the reported specs (and now multiple claimed listings for the card) are not hard to fathom.

We will keep you updated on this alleged new GPU if more details emerge.

CES 2015: Gigabyte GTX 980 WaterForce 3-Way SLI Monster Spotted

Subject: Graphics Cards, Shows and Expos | January 5, 2015 - 07:15 PM |
Tagged: waterforce, GTX 980, gigabyte, ces 2015, CES, 3-way sli

Back in November Gigabyte asked for all your money in exchange for a set of three GeForce GTX 980 cards each running in a circuit of self-contained water cooling. After finally seeing the GTX 980 WaterForce in person I can tell you that it's big, it's expensive and it's damned impressive looking.

wf-2.jpg

With a price tag of $2999, there is a significant mark up over buying just a set of three retail GTX 980 cards, but this design is unique. Each GPU is individually cooled via a 120mm radiator and fan that is mounted inside of a chassis that rests on top of your PC case. On the front you'll find a temperature, fan speed and pump speed indicator along with some knobs and buttons to adjust settings and targets.

wf-3.jpg

Oh, and it ships inside of a suitcase that you can reuse for later travel. Ha! Think we can convince Gigabyte to send us one for testing?

wf-1.jpg

Coverage of CES 2015 is brought to you by Logitech!

PC Perspective's CES 2015 coverage is sponsored by Logitech.

Follow all of our coverage of the show at http://pcper.com/ces!

ASUS updates their popular series with the GTX 980 STRIX DC II OC

Subject: Graphics Cards | January 5, 2015 - 05:31 PM |
Tagged: GTX 980 STRIX DirectCU II OC, strix, asus, nvidia, factory overclocked

ASUS' popular STRIX line was recently updated to include NVIDIA's top card and now [H]ard|OCP has had a chance to benchmark this GTX 980 with custom quiet cooling.  The DirectCU II cooling system can operate at 0dB under all but the heaviest of loads and the 10 phase power design will mean you can go beyond the small factory overclock that the card arrives with.  [H]ard|OCP took the card from a Boost Clock of 1279MHz to 1500MHz and the RAM from 7GHz to 7.9GHz with noticeable performance improvements part of why it received a Gold Award.  If the ~$130 price difference between this card and the R9 290X does not bother you then it is a great choice for a new GPU.

14192008366QHJaRnDQg_1_1.jpg

"Today we delve into the ASUS GTX 980 STRIX DC II OC, which features custom cooling, 0dB fans and high overclocking potential. We'll experiment with this Maxwell GPU by overclocking it to the extreme. It will perform head to head against the ASUS ROG R9 290X MATRIX-P in today's most demanding games, including Far Cry 4."

Here are some more Graphics Card articles from around the web:

Graphics Cards

Source: [H]ard|OCP

CES 2015: MSI Announces GTX 970 Gaming 100ME - 100 millionth GeForce GPU

Subject: Graphics Cards, Shows and Expos | January 4, 2015 - 03:56 PM |
Tagged: msi, GTX 970, gaming, ces 2015, CES, 100me

To celebrate the shipment of 100 million GeForce GPUs, MSI is launching a new revision of the GeForce GTX 970, the Gaming 100ME (millionth edition). The cooler is identical that used in the GTX 970 Gaming 4G but replaces the red color scheme of the MSI Gaming brand with a green very close to that of NVIDIA's.

100me-1.jpg

This will also ship with a "special gift" and will be a limited edition, much like the Golden Edition GTX 970 from earlier this year.

100me-2.jpg

MSI had some other minor updates to its GPU line including the GTX 970 4GD5T OC with a cool looking black and white color scheme and an 8GB version of the Radeon R9 290X.

Coverage of CES 2015 is brought to you by Logitech!

PC Perspective's CES 2015 coverage is sponsored by Logitech.

Follow all of our coverage of the show at http://pcper.com/ces!

GPU Rumors: AMD Plans 20nm but NVIDIA Waits for 16nm

Subject: General Tech, Graphics Cards | December 28, 2014 - 09:47 PM |
Tagged: radeon, nvidia, gtx, geforce, amd

According to an anonymous source of WCCFTech, AMD is preparing a 20nm-based graphics architecture that is expected to release in April or May. Originally, they predicted that the graphics devices, which they call R9 300 series, would be available in February or March. The reason for this “delay” is a massive demand for 20nm production.

nvidia-gtx-vs-amd-gaming-evolved.jpg

The source also claims that NVIDIA will skip 20nm entirely and instead opt for 16nm when that becomes available (which is said to be mid or late 2016). The expectation is that NVIDIA will answer AMD's new graphics devices with a higher-end Maxwell device that is still at 28nm. Earlier rumors, based on a leaked SiSoftware entry, claim 3072 CUDA cores that are clocked between 1.1 GHz and 1.39 GHz. If true, this would give it between 6.75 and 8.54 TeraFLOPs of performance, the higher of which is right around the advertised performance of a GeForce Titan Z (only in a single compute device that does not require distribution of work like what SLI was created to automate).

Will this strategy work in NVIDIA's favor? I don't know. 28nm is a fairly stable process at this point, which will probably allow them to get chips that can be bigger and more aggressively clocked. On the other hand, they pretty much need to rely upon chips that are bigger and more aggressively clocked to be competitive with AMD's slightly more design architecture. Previous rumors also hint that AMD is looking at water-cooling for their reference card, which might place yet another handicap against NVIDIA, although cooling is not an area that NVIDIA struggles in.

Source: WCCFTech