DigiTimes, citing "sources from the upstream supply chain", is reporting an expected decrease in graphics card pricing for July. This move comes as a way for suppliers to reduce the inventory previously piled in expectation of continued demand from cryptocurrency miners and gamers in general. It's the economic system at work, with its strengths and weaknesses: now that demand has waned, somewhat speculative price increases of yore are being axed by suppliers to spur demand. This also acts as a countermeasure to an eventual flow of graphics cards from ceasing-to-be miners to the second-hand market, which would further place a negative stress on retailers' products.

Alongside this expected 20% retail price drop for graphics cards, revenue estimates for major semiconductor manufacturer TSMC and its partners is being revised towards lower than previously-projected values, as demand for graphics and ASIC chips is further reduced. DigiTimes' sources say that the worldwide graphics card market now has an inventory of several million units that is being found hard to move (perhaps because the products are already ancient in the usual hardware tech timeframes), and that Nvidia has around a million GPUs still pending logistical distribution. Almost as an afterthought, DigiTimes also adds that NVIDIA has decided to postpone launch of their next-gen products (both 12 nm and then, forcibly, 7 nm) until supply returns to safe levels.

As the mining craze seems to have hit a steaming wall alongside the current contraction in the crypto market (which has almost all cryptocurrencies redlining), Gigabyte is revising its graphics card shipment expectations for 2Q18. This isn't a sudden move, mind you: the "cryptocurrency mining accelerator market" has been slowing its ludicrous demand for some time now. However, Gigabyte expects the slowdown to continue and maybe even become steeper: a 20% reduction in its overall shipment expectations for 2Q18, from 1.2M units down to 1M, and a 10% reduction in ASP (Average Selling Price) do speak to this decline in demand. As a result of this expected decrease, Gigabyte will once again turn its marketing efforts towards gaming products and usage scenarios for their graphics cards, diverting funds that had been allocated to mining.

Don't worry though: Gigabyte is doing great. 1Q18 saw the company post record profits higher than the first half of 2017 - 1Q18 profits rose 91% sequentially and skyrocketed five-fold YoY to NT$1.61 billion (US$52.75 million). The company's revenue for graphics cards hit an all-time high of 49% per graphics card sold (a result of increased ASP). the company's motherboard business should see the same results as the previous year - a fault of Intel's increased delays in launching a new, compelling product line-up. Who would have thunk - Intel, the company that's always launching new platforms and chipsets and ending motherboard support for new CPUs.

According to DigiTimes, the entire AIB partner and graphics card supply channel is gearing up to an expected demand decrease for graphics cards in the second half of 2018. This marks an expectation on the continuation of the downward trend since December 2017, a time where Bitcoin (and as such, alternate cryptocurrencies) were at all-time highs. As profits decrease, difficulty increases, and mining players offload their graphics cards to still-interested buyers of their hardware, the market's ability to trade existing graphics cards and absorb new inventory is dwindling. Naturally, this reduced demand means that prices for new graphics cards have also been decreasing and somewhat stabilizing towards pre-mining boom prices.

However, producers of graphics cards obviously don't want to give away their record-high profits in their entirety; and they're showing some reluctance, some "pricing memory" on their graphics cards, maintaining gross margins in the 20% area, double that of pre-mining pricing. As such, graphics card makers are again abandoning the mining boom as a source of stable revenue, looking to other solutions (such as servers, datacenter acceleration and such, DigiTimes reports in the case of TUL). Another thing that would certainly help graphics card manufacturers in keeping up high demand and profits, of course, would be the impending release of a new NVIDIA architecture... At least for those that have AIB status with the company.

Job postings on LinkedIn reveal that Samsung is looking to hire a ton of graphics chip engineers to bring forward their own GPU design. In the past the company has licensed GPU IP from companies like ARM and Imagination Technologies, but these designs come with cost, low performance and low flexibility. With Samsung needing graphics IP for a large range of products like phones, tablets and exploring options in markets like automotive, machine learning and AI, it's not surprising that the company is now looking into rolling their own GPU - from scratch as indicated by a recruiter's posting:

"This is Samsung's proprietary IP. We will define the ISA, the architecture, the SW, the entire solution."

Matrox Graphics Inc. is pleased to announce the 20-year anniversary of its Matrox G200 graphics chip. The milestone celebrates two decades of dedicated, active software development and support for multiple technology nodes across countless Microsoft Windows and Linux operating systems. Developed for 2D, 3D, and video acceleration, the G200 powered a number of industry-first, graphics and multi-monitor-based product lines that delivered unprecedented image quality across one or more displays. Today, G200 is and remains the trusted and preeminent integrated graphics solution of choice for the majority of baseboard management controllers used in servers worldwide.

ATOMONTAGE today announced that they're officially a company, and that they're looking to leapfrog graphics rendering by doing away with regular polygon-based graphics. The idea isn't new - some have tried it before, and we've even seen this technology being used in no other than No Man's Sky to some success.

However, there have been some technical hurdles that needed to be dealt with (mainly data management concerns, alongside rigid-body physics, soft-body deformations, standard PBR shaders, and efficient playback of multiple volumetric video streams). It appears that ATOMONTAGE now feels they've been able to bring the technology to a level that's enough to do actual development with, using today's powerful graphics processing units.

Intel's next generation "Ice Lake" processor could integrate a significantly faster integrated graphics solution (iGPU), if a SiSoft SANDRA online database entry is to be believed. A prototype "Ice Lake" chip was benchmarked, with its iGPU being described by the database as "Intel UHD Graphics" based on the company's Gen 11 graphics architecture, which succeeds the current Gen 9.5 architecture implemented on "Coffee Lake" and "Kaby Lake." This iGPU is endowed with 48 execution units (EUs), which work out to 384 unified shaders; against 24 EUs and 192 shaders on Intel UHD 620. SANDRA also describes the iGPU as being able to share up to 6 GB of memory from the system memory; and featuring 768 KB of dedicated cache. Its reference clock is 600 MHz, double that of the UHD 620, although its boost clock remains a mystery. "Ice Lake" is being built on Intel's new 10 nm+ silicon fabrication process, so it's understandable for the company to significantly enlarge its iGPU.

Sapphire took to CES 2018 to showcase one of the most elusive products in recent times: a custom variant of AMD's RX Vega graphics cards. Sapphire went to great lengths to keep the entire affair under as many wraps as possible, even going so far as to book an entire room to showcase their RX Vega Nitro+, laying still, like a mirage, on top of its anti-static plastic wrap.

There's something eerily beautiful about this graphics card: not on its exquisite back and faceplate designs alone, nor on the custom Sapphire-cut exhaust port; it's really for the condition of a somewhat "unicorn" type of product. Some people believe they're out there, but sightings in their natural environment (read, correct MSRP) are so rare that they are brought up as mass deliriums. The most interesting tech product on CES 2018 based on this story alone, surely.

It was a banner year for graphics, both in terms of market strength and technology advancements. Gaming, virtual reality, crypto mining, and artificial intelligence fueled demand for GPUs in 2017. The market responded with a wide array of products: high-performance discrete PC graphics cards that let gamers run multiple 4K displays; game consoles and VR headsets; and workstation-class GPUs that can build the stunning effects we have all come to expect. And since these products are full of our GDDR5 or G5X memory, it was an exciting year for Micron's graphics team too. We had a record-breaking year in GDDR5 shipments and further solidified Micron's industry leadership in graphics memory with the launch of our 12 Gb/s G5X, the highest-performance mass production GDDR memory.

EVGA is feeling both generous and nostalgic, and has announced a contest that's as welcome as it is original. Looking back on its roots as an innovative designer of cooling systems for graphics cards, allowing gamers everywhere to play their favorite games with the best temperatures and least noise possible, EVGA are looking for users that still keep their GeForce 4 ACS and ACS2-powered graphics cards. ACS and ACS 2 are effectively the precursors to EVGA's latest cooling innovations, even if these types of cooling designs are now the norm rather than the exception.

If you have a GeForce4 MX 440 or a GeForce4 4600 Ti, both launched way back in 2002, EVGA is asking that you submit your information to them, and the company will gladly pay you 1,000 EVGA bucks - its digital currency that is equivalent to $ in a 1:1 ratio, and would allow you to buy a great 1080 Ti from the EVGA store. Not such a bad deal for an old GeForce 4 graphics card, eh? Like Sonic would say, you "gotta go fast", though: the contest is available for a maximum of 3 ACS owners and 3 ACS2 owners.

A mysterious AMD APU showed up on SiSoft SANDRA online database, featuring a massive integrated graphics. The chip reports itself to SANDRA as "AMD Fenghuang Raven," and is likely a semi-custom chip being tested by an AMD engineer in the course of its development. SANDRA reports the integrated graphics component as "AMD 15FF Graphics," featuring 1,792 stream processors across 28 compute units, 555 MHz engine clock, and 2 GB of video memory with 182.15 GB/s memory bandwidth. The result doesn't put out too many details about the CPU component, except its 2.40 GHz clock speed. The iGPU scored 98 points on SANDRA graphics tests with Direct3D 11 API, and 39.99 GB/s observed score.

Jonsbo has introduced a somewhat strange product to its lineup - a companion cooler to your graphics card. This cooler isn't meant to be installed directly on your graphics card, as some other aftermarket coolers like the Arctic Cooling Accelero series are; instead, this one is meant to be installed in an expansion slot (and a PCIe x16 at that, of all things) below your graphics card, providing additional cooling to your card's reference or custom cooler.

The VF-1 has an anodized aluminum magnesium alloy enclosure, which features a LED lighting accent towards your side window, and features 3x 80 mm cooling fans that blow air towards your graphics card.The VF-1 apparently operates at 18 dBA with the fans at their maximum 1500 RPM, and uses a SATA power connector instead of a fan header. The VF-1 measures 134 x 270 mm and is 20 mm thick, while the aluminum shroud material itself is 1.2 mm thick. It weighs 372 grams, and for all this, Jonsbo is quoting a 5ºC improvement in your graphics card's operating temperatures, which is... pretty underwhelming.

At SIGGRAPH 2017, NVIDIA introduced the latest version of their AI-based, GPU-enabled ray-tracing OptiX API. The company has been at the forefront of GPU-powered AI endeavors in a number of areas, including facial animation, anti-aliasing, denoising, and light transport. OptiX 5.0 brings a renewed focus on AI-based denoising.

AI training is still a brute-force scenario with finesse applied at the end: basically, NVIDIA took tens of thousands of image pairs of rendered images with one sample per pixel and a companion image of the same render with 4,000 rays per pixel, and used that to train the AI to predict what a denoised image looks like. Basically (and picking up the numbers NVIDIA used for its AI training), this means that in theory, users deploying OptiX 5.0 only need to render one sample per pixel of a given image, instead of the 4,000 rays per pixel that would be needed for its final presentation. Based on its learning, the AI will then be able to fill in the blanks towards finalizing the image, saving the need to render all that extra data. NVIDIA quotes a 157x improvement in render time using a DGX station with Optix 5.0 deployed against the same render on a CPU-based platform (2 x E5-2699 v4 @ 2.20GHz). The Optix 5.0 release also includes provisions for GPU-accelerated motion blur, which should do away with the need to render a frame multiple times and then applying a blur filter through a collage of the different frames. NVIDIA said OptiX 5.0 will be available in November. Check the press release after the break.

Ah mining. The revival of an old craze. Who doesn't want to make their room's temperature increase to insane levels over the summer in order to cash in on the mining wagon? Who doesn't want to pull their hardware by the ankles and wrists, stretching it in utilization so as to maintain the PoW (proof of Work) cryptographic security in cryptocurrencies? Apparently, a not insignificant number of users and would-be miners does want that. That has, in turn, placed a whole lot of pressure on the graphics card market from both AMD and NVIDIA, with prices climbing and skyrocketing for graphics cards in the $200-$400 price ranges, as you know. It remains to be seen whether the flow of new miners decreases somewhat now, considering the recent market correction (read: dip) in the cryptocurrency market value (down around 42% from the all-time high of 357€ [~$400] of June 12th.)

After ASUS, it would seem like it's Sapphire's time to try and sway miners from their consumer-oriented, gaming graphics cards, through the launch of five different graphics cards models especially geared for mining. These are currently available for pre-order on Overclockers UK, and there are five different products in total, one based of RX 560 silicon, and four different takes on the RX 470 silicon (no, that's not a typo; it really is the 400 series.)

GPU-accelerated crypto-currency mining poses a threat to the consumer graphics industry, yet the revenues it brings to GPU manufacturers are hard to turn away. The more graphics cards are bought up by crypto-currency miners, the fewer there are left for gamers and the actual target-audience of graphics cards. This is particularly bad for AMD, as fewer gamers have Radeon graphics cards as opposed to miners; which means game developers no longer see AMD GPU market-share as an amorphous trigger to allocate developer resources in optimizing their games to AMD architectures.

To combat this, both AMD and NVIDIA are innovating graphics cards designed specifically for crypto-currency mining. These cards are built to a cost, lack display outputs, and have electrical and cooling mechanisms designed for 24/7 operation, even if not living up to the durability standards of real enterprise-segment graphics cards, such as Radeon Pro series or Quadro. NVIDIA's "Pascal" GPU architecture is inherently weaker than AMD's "Polaris" and older Graphics CoreNext architectures at Ethereum mining, owing in part to Pascal's lack of industry-standard asynchronous compute. This didn't deter NVIDIA from innovating a lineup of crypto-mining SKUs based on its existing "Pascal" GPUs. These include the NVIDIA P104 series based on the "GP104" silicon (on which the GTX 1080 and GTX 1070 are based); and P106 series based on the "GP106" silicon (GTX 1060 series is based on this chip). NVIDIA didn't tap into its larger "GP102" or smaller "GP107" chips, yet.

A report from HWBattle is making the rounds claiming that new information has surfaced on AMD's upcoming high-performance, consumer versions of the Vega architecture. According to these reports, Vega graphics cards will (at least initially) come in two different performance tiers. A top of the line GPU, Vega 10 (being identified as Vega XT), and a cut-down version of it, based on Vega 11 (which is being called Vega Pro). Graphics chips for graphics card integration are supposedly being shipped to partners as of this week.

HWBattle goes on to say that there will be a myriad of approaches to AMD's AIB partner designs around the Vega graphics chips, with multiple cooling solutions being worked on (which isn't surprising, really; graphics cards nowadays can see upwards of 4 different cooling designs for the same GPU, according to the use case the company is designing it for. HWBattle is also saying that Vega will be faster than the GTX 1080, though there's no information on whether this only applies to the top-tier GPU or no. Other details are scant, scarce, or nonexistent; it would seem that the launch delay from AMD has sapped some of the interest surrounding Vega.

Colorful Technology Company Limited, professional manufacturer of graphics cards and motherboards, announces its latest addition to the Colorful family of graphics cards with the COLORFUL GT1030 2G graphics card suited for office/home use.

The Colorful GT1030 2G ships with 384 Streaming Multiprocessors (SMs), and runs at a base clock frequency of 1227 MHz while the GPU Boost clock speed is rated for 1468 MHz and is based on the Pascal GP108 silicone. This card has been outfitted with GDDR5 memory at 2GB capacity and is wired to a 64-bit bus. The GPU has 24 TMUs, 8 ROPs and has a rated draw of 30 watts power consumption making the card a highly efficient choice. With an all-solid-state capacitor power delivery design, COLORFUL provides customers better and more stability while maintaining excellent performance. It has a 90mm fan and a solid cooler featuring compact thermal fins keeping the card cool even under extreme loads. No pricing information was disclosed at the time.

At its Computex 2017 showcase, Sapphire showcased (ahem) an upcoming e-GFX enclosure. The unit delivers its display signal through a Thunderbolt 3 port, which is being hailed as "The USB-C that does it all", with up to 40 Gb/s connections, dual 4K @ 60Hz support, charging capabilities, and support for the USB and DisplayPort interfaces.

The unit makes use of what appears to be a SFF PSU (it isn't clear whether or not a PSU comes bundled with the chassis), and space for a dual-slot high performance graphics card. The side of the enclosure presents an airflow-enabling grill, so that your graphics card can pull in fresh air from the exterior. There's also what seems to be an Ethernet port and 2x USB 3.0 ports, which means this enclosure can be used as a hub, or as an extra charger for your other pieces of tech. Aesthetics-wise, this unit uses the tried and true black and white contrast, with a black design language on the front and back, and a smooth white contour on the top and sides of the enclosure. I enjoy the understated design here, with no extremely hard and "gaming" screaming angles, and the neutral color scheme and LED design doesn't look half bad. It remains to be seen whether that blue LED lighting is too strong so as to be distracting, but if it's not, it stands as a good complement to the overall design. Expect this unit to be released sooner rather than later, though no details on pricing or specifications were at hand.

NVIDIA today released an updated version of their GeForce graphics drivers, version 382.23 WHQL, which brings with it support for the company's GT 1030 entry-level graphics cards. Strangely, the drivers' supported products only register as the GT 1030 itself, which is a little unusual, to say the least.

Follow the link here to grab the drivers directly from NVIDIA. Note that since these drivers ONLY support NVIDIA's new GT 1030, you shouldn't upgrade to these if you have any other NVIDIA card. To download the drivers, you'll have to choose the GT 1030 from the drop-down menu on NVIDIA's site, as well as the proper OS version for your use case.

VisionTek, a company whose last graphics card announcement we've posted here was on June 2015 with its Fury X and R9 300 line of graphics cards, has announced its vision for the RX 500 series graphics cards. This one seems a throwback to the RX 400 series, though, with Visiontek's "Overclocked Edition" GPUs carrying clockspeeds more in line with the prevous AMD series: 1,366 MHz for the RX 580, and 1264 MHz for the RX 570, respectively.

The VisionTek Radeon RX 580 8 GB comes with a slightly unoriginal take on the AIB partner design, sporting custom dual-fan cooler with a black-nickel aluminum shroud with 2x 8 mm and 2x 6 mm heatpipes for increased cooling capacity. The Radeon 570 4GB makes use of what is basically AMD's RX 480 reference design cooler, with a blower-style fan that ensures the GPU's heated air output exits the back of your case instead of lingering inside your system. VisionTek's Radeon RX 500 Series cards are now available through the company's website, with retailer availability to follow.

In the latest edition of "RTB's Obvious News," I present you with an alarming statistic:

Digital purchases make up most video game sales in the United states at around 74% of purchases, according to a study by the Entertainment Software Association (ESA).

I know, I can hear it now: "Obvious news is obvious, R-T-B. Why'd you post this? Your news sucks!" etc etc. Believe it or not, I have a reason for bringing this seemingly lame study to your attention, and that's because despite the headline being lame, the study actually contains some interesting tidbits about the gaming market in general.

XFX has launched three variants of the RX 550 graphics cards, the tiny GPU that could, which AMD launched so as to bridge the enormous gap between IGP and its previous entry-line RX 460 (now RX 560) series of graphics cards. There are two low-profile versions of the RX 550, packing either 2GB or 4 GB of memory (whose amounts can be justified or not,) both with boost clocks set at 1203 MHz and 7000 MHz GDDR5 memory over a 128-bit bus. There is also a full-profile, dual slot RX 550, dubbed the Core Edition, and another Core Edition, though this one is a full-profile, single-slot solution.

All of these pack the same 1203 MHz boost clocks, so XFX is basically telling you to pick and choose the size of the graphics cards that best fits your use case, with improvements on cooling and sound profile that come with the larger, beefier cooling solutions. Display outputs stand the same among all the different cards, with 1x DVI-I Dual-Link, 1x DisplayPort, and 1x HDMI 2.0.

The folks at Heise online have put forward a report on how AMD's RX 500 series of graphics cards will be little less than direct rebrands of the Polaris 10 and Polaris 11 GPUs that AMD introduced with its RX 400 series of graphics cards. Apparently, a straight rebrand is in order, with the RX 580 entering the fray in the place of the RX 480, the RX 570 substituting the RX 470, and so on. Heise reports that the Polaris 10-based RX 500 should see the light of day as soon as April 4th, with Polaris 11-based solutions coming in a little later, on April 11th.

Videocardz, however, reports that these will be slightly more than a straight rebrand - if you can call a slight bump in clockspeeds as trumping a rebrand. The RX 580 is supposed to ship with base clocks ar 1340 MHz (74 MHz more than the reference RX 480), with the RX 570 carrying a much less significant 38 MHz increase over its RX 470 counterpart. Videocardz also reports on the possibility of AMD introducing a new Polaris 12 GPU with the RX 500 series, which will apparently be an even lower-end part than even Polaris 11.

AMD today released the 17.2.1 WHQL version of its Radeon Software Crimson ReLive. This release features Multi GPU profiles and support for For Honor (with an up to 4% performance improvement on Radeon RX 480) and Sniper Elite 4 (with a 5% performance improvement being registered here). Look after the break for a list of known, fixed issues. Compared to the previous 17.2.1 Beta release of the ReLive driver suite, this one is virtually the same, the only difference being the WHQL signing.

As always, you can grab the drivers right here at TPU, through our revamped downloads section. Just follow the link below.DOWNLOAD: AMD Radeon Software Crimson ReLive Edition 17.2.1 WHQL

To strangely low pomp and fanfare, MSI has further increased its line-up of offerings for the RX 460 family of graphics cards, with the launch of the Radeon RX 460 4GT LP (4GB), as well as a 2GT (2GB) version. Based on AMD's Olaris 11 "Baffin" graphics chip, these graphics cards feature a 75 W TDP, thus eschewing any auxiliary PCIe power connectors. That TDP being tamed by a dual-fan cooling solution and the usage of MIL-STD-810G certified components mean that this is a card where MSI feels comfortable about its lifespan.

Connectivity-wise, the boards offer 1x DL-DVI-D, as well as 1x HDMI 2.0 with HDCP 2.2 support, which makes any of these a relevant choice for an HTPC. MSI has not revealed a MSRP for any of these cards, but considering the segment at which they fit in like a jigsaw puzzle, one can surely expect their pricing to not be that far off from those recommended by AMD ($109 for the 2 GB version, and $139 for the 4 GB one).