GPUs


AMD Releases Radeon Software Adrenalin Edition 18.2.2 & Inaugural Ryzen Desktop APU WHQL Drivers

AMD Releases Radeon Software Adrenalin Edition 18.2.2 & Inaugural Ryzen Desktop APU WHQL Drivers

This week, AMD released Radeon Software Adrenalin Edition 18.2.2, a smaller patch bringing support for the just-launched Kingdom Come: Deliverance, as well as performance optimizations for Fortnite and PlayerUnknown’s Battlegrounds (PUBG). Purely game-focused, 18.2.2 documents no new bugfixes or issues. And alongside Monday’s launch of AMD Ryzen 5 2400G and Ryzen 3 2200G, AMD has put out the inaugural Windows 10 WHQL drivers specific to those two new APUs.

Released just yesterday, Kingdom Come: Deliverance is a CryEngine-powered first-person single player historical RPG, thematically reminiscent of Mount and Blade and The Witcher. The developers, Warhorse Studios, have stated their desire to focus on realism, story, open-world freedom, and hardcore combat mechanics, the latter of which is not too surprising considering the Operation Flashpoint and ARMA pedigree of Warhorse Studio’s team members.

For Kingdom Come: Deliverance, AMD cites their 18.2.2 testing to show up to 3% faster 1440p performance for the Radeon RX Vega 56 and up to 4% faster 1080p performance for the Radeon RX 580 compared to Radeon Software Adrenalin Edition 18.2.1. As it so happens, Warhorse Studios marks down the Radeon RX 580 as the recommended AMD GPU requirement, with the Radeon HD 7870 as the minimum.

As for Fortnite and PUBG, AMD compares Radeon Software Adrenalin Edition 17.12.1 performance to 18.2.2, claiming 1440p improvements for the Radeon RX Vega 56 up to 3% faster for Fortnite and up to 5% faster for PUBG. Meanwhile, for the Radeon RX 580 at 1080p, AMD cites up to 6% faster performance in Fortnite and up to 7% faster performance in PUBG.

Wrapping things up on the 18.2.2 side, there has been no documented bugfixes and the list of open issues remain identical to 18.2.1:

  • FreeSync intermittently engages during Chrome video playback incorrectly, resulting in playback flicker.
  • Radeon Overlay hotkey fails to bring up the overlay or causes a Radeon Host Application crash intermittently on a limited number of gaming titles.
  • FreeSync may rapidly change between min and max range when enabled causing stutter in fullscreen games on multi display system configurations.
  • When Enhanced Sync is enabled on some FreeSync connected displays, flickering occurs with the performance metrics overlay.
  • Water textures appear to be missing in World of Final Fantasy.
  • A random system hang may be experienced after extended periods of use on system configurations using 12 GPU’s for compute workloads.
  • The GPU Workload feature may cause a system hang when switching to Compute while CrossFire is enabled. A workaround is to disable CrossFire before switching the toggle to Compute workloads.

Radeon Software for New Ryzen Desktop APUs

Released as “Radeon Software for Ryzen Desktop Processors with Radeon Vega Graphics,” the inaugural public graphics drivers are WHQL certified and are documented as version 17.40.3701 (Windows Driver Store Version 23.20.827.0). The update applies only to the Ryzen 5 2400G and Ryzen 3 2200G, and notes the following:

  • RAID drivers are not included in display driver packages. Users wishing to use RAID should navigate to the amd.com APU chipset driver page to find and install RAID drivers.
  • AMD has noted a potential system crash while running certain OpenCL applications like Linpack-DGEMM & Indigo benchmark. AMD is currently testing a solution to this issue and an updated driver will be released very shortly. Please visit amd.com for updates.
  • 3DMark Firestrike may experience an application hang during GT2 test.

Additionally, AMD has put up a support page for issues with system boot-up failure on configurations with some 2nd generation Ryzen desktop processors (CPUs & APUs) and AM4 motherboards. Unsurprisingly, AMD notes that the likely cause in this scenario is a motherboard that has not been updated to the latest BIOS with APU support, but offers a boot kit solution under warranty if this is not the root cause.

The updated drivers for AMD’s desktop, mobile, and integrated GPUs are available through the Radeon Settings tab or online at the AMD driver download page. More information on these updates and further issues can be found in the respective Radeon Software Adrenalin Edition 18.2.2 release notes and Radeon Software for Ryzen Desktop Processors with Radeon Vega Graphics release notes.

NVIDIA Announces Record Q4 2018 Results

NVIDIA Announces Record Q4 2018 Results

This afternoon NVIDIA announced yet another record quarter, with revenues of $2.91 billion, which is up 34% from a year ago. Gross margin was 61.9%, and operating income was up 46% to $1.07 billion. Net income for the quarter was $1.12 billion, up 71% from a year ago, which provided diluted earnings per share of $1.78.

NVIDIA Q4 2018 Financial Results (GAAP)
  Q4’2018 Q3’2018 Q4’2017 Q/Q Y/Y
Revenue $2911M $2636M $2173M +10% +34%
Gross Margin 61.9% 59.5% 60.0% +2.4% +1.9%
Operating Income $1073M $895M $733M +20% +46%
Net Income $1118M $838M $655M +33% +71%
EPS $1.78 $1.33 $0.99 +34% +80%

For their 2018 fiscal year, NVIDIA racked up $9.71 billion in revenue, which is up 41% from a year ago. Earnings per share for the year were $4.82, up 88%.

NVIDIA has kind of hit a magic sweet spot, where they had been diversifying into growing markets, but have found themselves also swept up in the GPU fed craze of cryptocurrency, feeding their core business as well.

Gaming, which is their GeForce lineup, had revenues for the quarter of $1.74 billion, up almost $400 million from a year ago. But for the entire fiscal year, they haven’t even refreshed their lineup, although the Pascal series of GPUs have certainly been strong in terms of performance and efficiency. But you’d be lucky to find any in stock thanks to the copious number of GPUs purchased for mining coins. It’s not been ideal for gamers, but for NVIDIA’s R&D, and their investors, it’s been welcome news.

Professional Visualization was also up about 13% to $254 million. Growth here was attributed to the ultra-high-end and high-end desktop workstations, such as the Quadro GP100 launched earlier in the fiscal year.

NVIDIA’s foray into the datacenter is also paying off with substantial growth. Datacenter includes Tesla, GRID, and DGX systems, and for the quarter, NVIDIA saw revenues of $606 million, which are up 105% from a year ago.

Automotive was up a more modest 3% to $132 million, and we’ve seen NVIDIA get some big design wins with large automotive companies with their DRIVE platforms.

Automotive is powered by Tegra, and including the automotive sales, Tegra brought in $450 million this quarter, and the very popular Nintendo Switch is likely a nice chunk of that.

OEM and IP was $180 million for the quarter, which is up about 2.2%, and NVIDIA stated their licensing agreement with Intel concluded in the first quarter of fiscal year 2018.

NVIDIA Quarterly Revenue Comparison (GAAP)
($ in millions)
In millions Q4’2018 Q3’2018 Q4’2017 Q/Q Y/Y
Gaming $1739 $1561 $1348 +11% +29%
Professional Visualization $254 $239 $225 +6% +13%
Datacenter $606 $501 $296 +21% +105%
Automotive $132 $144 $128 -8% +3%
OEM & IP $180 $191 $176 -6% +2%

For Q1 2019, NVIDIA is forecasting revenues of $2.9 billion, plus or minus 2%.

Source: NVIDIA Investor Relations

The EVGA GeForce GTX 1070 Ti FTW2 Review: iCX Brings the Lights and Sensors

Today, we are taking a look at the EVGA GeForce GTX 1070 Ti FTW2, EVGA’s top GTX 1070 Ti model featuring the iCX temperature sensor and cooling system. Launched alongside the GeForce GTX 1070 Ti Founders Edition and other partner boards in November 2017, the EVGA GTX 1070 Ti FTW2 similarly targets the competing Radeon RX Vega 56 with price and performance in between the GeForce GTX 1080 and 1070.

To avoid sales cannibalization, all partner GTX 1070 Ti cards, including the FTW2, adhere to the reference clockspeeds. To differentiate itself, the EVGA GeForce GTX 1070 Ti FTW2 looks to the power, temperature, and noise characteristics due to iCX cooler and adjusted power system, in addition to manual overclocking via a GTX 1070 Ti specific autoscan feature in EVGA’s Precision XOC overclock utility. A key piece of the puzzle, pricing, has been unfortunately been subject due to the extraordinarily intense cryptomining demand that has drastically inflated prices and depleted stocks of virtually all graphics cards at the time of publication.

AMD to Ramp up GPU Production, But RAM a Limiting Factor

AMD to Ramp up GPU Production, But RAM a Limiting Factor

One of the more tricky issues revolving around the GPU shortages of the past several months has been the matter of how to address the problem on the GPU supply side of matters. While the crux of the problem has been a massive shift in demand driven by a spike in cryptocurrency prices, demand has also not tapered off like many of us would have hoped. And while I hesitate to define the current situation as the new normal, if demand isn’t going to wane then bringing video card prices back down to reasonable levels is going to require a supply-side solution.

This of course sounds a lot easier than it actually is. Ignoring for the moment that GPU orders take months to process – there are a lot of steps in making a 16nm/14nm FinFET wafer – the bigger risk is that cryptocurrency-induced GPU demand is not stable. Ramping up GPU production means gambling that demand will stay high enough long enough to absorb the additional GPUs, and then not immediately contract and have the market flooded with used video cards. The latter being an important point that AMD got burnt on the last time this happened, when the collapse of the cryptocurrency-prices and the resulting demand for video cards resulted in the market becoming flooded with used Hawaii (290/390 series) cards.

Getting to the heart of matters then, in yesterday’s Q&A session for their Q4’2017 earnings call, an analyst asked AMD about the current GPU supply situation and whether AMD would be ramping up GPU production. The answer, much to my surprise, was yes. But with a catch.

Q: I just had a question on crypto, I mean if I look at the amount of hash compute being added to Ethereum in January I mean it’s more than the whole of Q4, so we have seen a big start to the Q1. […] And is there any sort of acute shortages here, I man can your foundry partners do they have the capacity to support you with a ramp of GPUs at the moment and is there enough HBM2 DRAM to source as well?

A: Relative to just where we are in the market today, for sure the GPU channel is lower than we would like it to be, so we are ramping up our production. At this point we are not limited by silicon per se, so our foundry partners are supplying us, there are shortages in memory and I think that is true across the board, whether you are talking about GDDR5, or you’re talking about high bandwidth memory. We continue to work through that, with our memory partners and that will be certainly one of the key factors as we go through 2018.

So yes, AMD is ramping up GPU production. Which is a surprising move since they were burnt the last time they did this. At the same time however, while cryptocurrency demand has hit both major GPU manufacturers, AMD has been uniquely hit as they’re a smaller player less able to absorb rapid changes in demand, and, more importantly, their GPUs are better suited for the task. AMD’s tradition of offering more memory bandwidth and more raw FLOPS than NVIDIA at any competing price point, coupled with some meaningful architectural differences, means that their GPUs are in especially high demand by cryptocurrency miners.

But perhaps the more interesting point here isn’t that AMD is increasing their GPU production, but why they can only increase it by so much. According to the company, they’re actually RAM-limited. They can make more GPUs, but they don’t have enough RAM – be it GDDR5 or HBM2 – to equip all of the cards AMD and board partners would like to make.

This is an interesting revelation, as this is the first time memory shortages have been explicitly identified as an issue in this latest run-up. We’ve known that the memory market is extremely tight due to demand – with multiple manufacturers increasing their RAM prices and diverting GDDR5 production over to DDR4 – but only now is that catching up with video card production to the point that current GDDR5 production levels are no longer “enough”. Of course RAM of all types is still in high demand here at the start of 2018, so while memory manufacturers can reallocate some more production back to GDDR5, GPU and board vendors have to fight with both the server and mobile markets, both of which have their own booms in demand going on, and are willing to pay top dollar for the RAM they need.


GDDR5: The Key To Digital Gold

In a sense the addition of cryptocurrency to the mix of computing workloads has created a perfect storm in an industry that was already dealing with RAM shortages. The RAM market is in the middle of a boom right now – part of its traditional boom/bust cycle – and while it will eventually abate as demand slips and more production gets built, for the moment cryptocurrency mining has just added yet more demand for RAM that isn’t there. Virtually all supply/demand problems can be solved through higher prices – at some point, someone has to give up – but given the trends we’ve seen so far, GPU users are probably the most likely to suffer, as traditionally the GPU market has been built on offering powerful processors paired with plenty of RAM for paltry prices. Put another way, even if the GPU supply situation were resolved tomorrow and there were infinite GPUs for all, RAM prices would be a bottleneck that kept video card prices from coming back down to MSRP.

With all that said, however, AMD’s brief response in their earnings call has been the only statement of substance they’ve made on the matter. So while the company is (thankfully) ramping up GPU production, they haven’t – and are unlikely to ever – disclose just how many more GPUs that is, or for that matter how much RAM they expect they and partners can get for those new GPUs. So while any additional production will at least help the current situation to some extent, I would caution against getting too hopeful about AMD’s ramp-up bringing the video card shortage to an end.