How are Add-In GPUs Selling Anyway?

Subject: General Tech, Graphics Cards | August 19, 2013 - 03:04 PM |
Tagged: jpr, Matrox, s3, amd, nvidia

Well, according to John Peddie Research (JPR), not too good if you are Matrox or S3. The total market for add-in boards decreased 5.4% from last quarter. 14.0 million were shipped across the entire industry. Neither company accounted for a thousandth of that value leaving them with a maximum 7000 units shipped, best case scenario. This industry is, basically, a two horse race.

GPU suppliers, market share
  This Quarter Prev. Quarter Last Year
AMD 38.0% 35.7% 40.3%
Matrox 0.0% 0.0% 0.3%
NVIDIA 62.0% 64.3% 59.3%
S3 0.0% 0.0% 0.1%
Total 100.0% 100.0% 100.0%

Two horses unless you count the Intel Xeon Phi. While technically not a graphics processor despite hardware design, 48,000 of these coprocessors were sold, already, for the Tianhe-2 supercomputer. This is at least seven-fold more than an entire quarter for Matrox. Unfortunately JPR does not report on Intel add-in cards despite its overlap with the GPU add-in market. These numbers could get even more interesting as years progress.

As for the two big players, AMD and NVIDIA, both hold very dominant positions. Almost spiting the 750,000 unit industry decline, AMD experienced a total increase of 0.8% quarter-over-quarter. Their market share gained 2.3% as a result of this growth. NVIDIA experienced a total decrease of 8.9%.

In all, AMD has been doing better than the industry average. They are fighting the slight decline in the graphics industry while simultaneously helping GPUs hold off against larger declines in PC systems.

Source: JPR

New Matrox DS1 Dock For Macbooks Uses Thunderbolt

Subject: General Tech, Mobile | June 3, 2012 - 11:08 PM |
Tagged: computex, thunderbolt, Matrox, macbook, dock, apple

Matrox has recently launched a new Thunderbolt dock aimed at adding desktop peripherals to Apple Macbooks and Ultrabooks. The dock connects via a single Thunderbolt cable (it does require a separate power source as well) and provides one USB 3.0 port, two USB 2.0 ports, a DVI video output, audio in/out jacks, and a Gigabit LAN port. It will be available for purchase in September with an MSRP of $249 USD.

Matrox has released a new laptop dock called the Matrox DS1 that is designed to pair with Thunderbolt-equipped notebooks and provide several additional connectivity options. The aluminum chassis is reminiscent of a slimmer WD My Book drive because of the book like shape. The front of the DS1 dock is a Thunderbolt input and status LED. On the back of the dock is a DVI output, three USB ports (one USB 3.0, two USB 2.0), microphone input, headphone output, and a Gigabit LAN port. To the far right is a DC power input which means that the dock is a bit less portable than I would like but it is not clear how big of a “wall wart” it will come with.

Matrox_DS1_Thunderbolt_Docking_Station.jpg

The company has stated that the aluminum case should protect the dock in just about every use case, and the additional IO certainly adds much needed connectivity to Ultrabooks where available ports are at a premium. Senior Director of Sales and Marketing for Matrox, Alberto Cieri, has been quoted by Apple Insider in stating “The new Matrox DS1 docking station easily enables the creation of an ergonomic workspace and brings much-needed expandability for printers, scanners, storage, smartphones, optical drives, cameras, flash drives, and other peripherals.”

Matrox_DS1_Thunderbolt_Docking_Station_Connections.jpg

The Matrox DS1 will be shown off at Computex 2012 this week in Taipei, Taiwan at Intel’s booth (M0410 in the Nangang Exhibition Hall). After that, it will be shown off at WWDC in San Fransico and Infocomm in Las Vegas on June 12th and June 13th to June 15th respectively. In September of this year it will be available for purchase with an MSRP of $249 USD.

Situations like this are where Thunderbolt really shines, and I would not be surprised to see companies outfitting employees with Ultrabooks for mobile use and a larger monitor and peripherals for in-office use (eschewing a separate desktop machine altogether). The price, especially considering Thunderbolt cables themselves are expensive is going to be the most limiting factor for docks like these despite their usefulness.

Editor's Note: Although Tim didn't mention it, one kind of interesting drawback is that this device does NOT include a Thunderbolt pass through, basically preventing users from taking advantage of the daisy-chain capability TB can offer via a single port / connections on the laptop or computer.

Intel MLAA: Matrox had the right idea, wrong everything else

Subject: Editorial, General Tech, Graphics Cards, Processors | July 22, 2011 - 08:20 PM |
Tagged: MLAA, Matrox, Intel

Antialiasing is a difficult task for a computer to accomplish in terms of performance and many efforts have been made over the years to minimize the impact while still keeping as much of the visual appeal as possible. The problem with aliasing is that while pixels are the smallest unit of display on a computer monitor, it is large enough for our eye to see it as a distinct unit. You may however have two objects of two different colors partially occupy the same pixel, who wins? In real life, our eye would see the light from both objects hit the same retina nerve (that is not really how it biologically works but close enough) and it would see some blend between the two colors. Intel has released a whitepaper for their attempt at this problem and it resembles a method that Matrox used almost a decade ago.

MatroxAA.jpg

Matrox's antialiasing method.

(Image from Tom's Hardware)

Looking at the problem of antialiasing, you wish to have multiple bits of information dictate the color of a pixel in the event that two objects of different colors both partially occupy the same pixel. The simplest method of doing that is dividing the pixel up into smaller pixels and then crushing them together to an average which is called Super Sampling. This means you are rendering an image 2x, 4x, or even 16x the resolution you are running at. More methods were discovered including just flagging the edges for antialiasing since that is where aliasing occurs. In the early 2000s, Matrox looked at the problem from an entirely different angle: since the edge is what really matters, we can find the shape of the various edges and see how much area of a pixel gets divided up between each object giving an effect they say is equivalent to 16x MSAA for very little cost. The problem with Matrox’s method: it failed with many cases of shadowing and pixelshaders… and came out in the DirectX 9 era. Suffices to say it did not save Matrox as an elite gaming GPU company.

37399.png

37400.png

Look familiar?

(Both images from Intel Blog)

Intel’s method of antialiasing again looks at the geometry of the image but instead breaks the edges into L shapes to determine the area they enclose. To keep the performance up they do pipelining between the CPU and GPU which keeps the CPU and GPU constantly filled with the target or neighboring frames. In other words, as the GPU lets the CPU perform MLAA, the GPU is busy preparing and drawing the next frame. Of course when I see technology like this I think two things: will this work on architectures with discrete GPUs and will this introduce extra latency between the rendering code and the gameplay code? I would expect that it must as the frame is not even finished let alone drawn to monitor before you fetch the next set of states to be rendered. The question still exists if that effect will be drowned in the rest of the latencies experienced between synchronizing.

AMD and NVIDIA both have their variants of MLAA, the latter of which being called FXAA by NVIDIA's marketing team. Unlike AMD's method, NVIDIA's method must be programmed into the game engine by the development team requiring a little bit of extra work on the developer's part. That said, FXAA found its way into Duke Nukem Forever as well as the upcoming Battlefield 3 among other games so support is there and older games should be easy enough to just compute properly.

37407.png

The flat line is how much time spent on MLAA itself, just a few milliseconds and constant.

(Image from Intel Blog)

Performance-wise the Intel solution performs ridiculously faster than MSAA, is pretty much scene-independent, and should produce results near the 16x mark due to the precision possible with calculating areas. Speculation about latency between render and game loops aside the implementation looks quite sound and allows users with on-processor graphics to not need to waste precious cycles (especially on GPUs that you would see on-processor) with antialiasing and instead use it more on raising other settings including resolution itself while still avoiding jaggies. Conversely, both AMD and NVIDIA's method run on the GPU which should make a little more sense for them as a discrete GPU should not require as much help as a GPU packed into a CPU.

Could Matrox’s last gasp from the gaming market be Intel’s battle cry?

(Registration not required for commenting)

Source: Intel Blog