One of the biggest changes DX12 brings to the table is the increased reliance on developers to properly optimize their code for GPUs. Unlike DX11, there will have fewer levers to tweak in the GPU driver, with more work being needed in the game engine and the game itself. To address this, AMD has announced a partnership with multiple game engine and game developers to implement DX12.
To kick start the effort, AMD is headlining 5 games and engines they are partnering with to ensure DX12 works smoothly with Radeon GPUs with the software. These are Ashes of the Singularity by Stardock and Oxide Games, Total War: WARHAMMER by Creative Assembly, Battlezone VR by Rebellion, Deus Ex: Mankind Divided by Eidos-Montréal and the Nitrous Engine by Oxide Games. These titles span a wide range from RTS to RPG and FPS which gives a sense that AMD is trying to cast as wide a net as possible.
In addition to this, AMD will also be working with EA and Dice to get the Frostbite 3 engine to enable DX12. This engine is of particular importance due to the many AAA EA and other titles using it. AMD is also hoping to push Asynchronous Compute and to make sure games will squeeze the most out of GCN using DX12.
In the few days after AMD first demoed Polaris 10 to us at Capsaicin, more details about the upcoming graphics cards have been revealed. Set to be the big brother to the smaller Polaris 11, the better performing chip will drop sometime after June this year.
First off, we’re now able to bring you more information about the settings Hitman was running at during the demo. At Ultra Settings and 1440p, Polaris 10 was able to keep to a constant 60FPS, with VSync being possible. This means the minimum FPS did not drop below 60 at any point. This puts the card at least above the R9 390X and on par if no better than the Fury and Fury X. Of course, the demo was done with DX12 but the boost is only about 10% in Hitman.
Another detail we have uncovered is the maximum length of the engineering sample. Based on the Cooler Master Elite 110 case used, the maximum card length is 210mm or 8.3 inches. In comparison, the Nano is 6 inches and the Fury X 7.64 inches. Given the small size, one can expect Polaris 10 to be as power efficient as Polaris 11 and potentially be using HBM. Given that Vega will be the cards to debut HBM2, Polaris 10 may be limited to 4GB of VRAM. Finally, display connectivity is provided by 3x DP 1.3, 1x HDMI 2.0 and 1 DVI-D Dual Link though OEMs may change this come launch unless AMD locks it down.
2016 may well go down as the year VR finally takes off for real. Sony and Microsoft have both been making progress towards VR and augmented reality while Oculus and HTC are set to launch the Rift and Vive respectively. Given the efforts and lengths AMD has gone to push VR, it should come to no surprise that a report has revealed that the company has a massive 83% lead in providing the hardware for VR capable systems.
Hardware wise, it is not surprising to see the lead over Nvidia. While PC hardware is a large segment of the VR market, only higher end systems are capable of producing the frames necessary for VR at 90fps and enough resolution for both eyes. Because of this, the PS4 is a viable candidate for VR adoption and with the APU inside it being AMD, Nvidia stands no chance in terms of sheer hardware market share for VR.
As noted many times during the Capsaicin event, AMD has been working with many developers in both gaming and other forms of media with LiquidVR and GPUOpen. AMD has also been on the forefront with developments like VR cafes and partnering with Oculus and HTC to ensure that the Rift and Vive work seamlessly with Radeon. There is even a Radeon VR Ready Premium program to ensure consumers are informed.
With the VR market still in it’s growing stages, AMD has seen an opportunity to get in before it’s competitors have a chance and secure a bastion of developer support and integration. Considering the price of VR capable hardware, AMD stands a good chance to reap a windfall when VR takes off. This can only bode well for AMD as for once they are ahead and hopefully will be able to leverage their position to help the rest of their business grow.
Even as Polaris approaches us quickly within 3 months, the planning for its successor has long been in the works. At their Capsaicin event, AMD took off the wraps for their upcoming GPU plans with a roadmap detailing the planned releases up till 2019. In keeping with the star nomenclature that started with Polaris and ditching the islands, we will have Vega and then Navi following Polaris.
Starting off with Polaris later this year, AMD’s main selling point it seems is the 2.5x performance per watt the new GCN architecture will bring. This is no doubt due to the combination of improved hardware itself, the new 14nm LPP process and DX12 finally making use of the previously wasted hardware resources like asynchronous controllers and shaders.
Moving along, we have Vega to release in what looks to be early 2017. The biggest change it seems is the use of HBM2, replacing GDDR5(X) and HBM1 no doubt. This means we can no doubt expected all Vega releases to utilize HBM2. While this may suggest Polaris won’t be using HBM2, it could also mean that only certain Polaris chips, likely only the high-end ones, will use HBM2.
Finally, we come to Navi, which should debut in early 2018. This release will have scalability and use of next-gen memory like Hyper-Memory Cube for instance. The scalability mention suggests either the use of smaller GCN units used to build the chip to better suit the market or a new process node. For now, we are probably better off trying to figure out what Polaris will be
Just yesterday, AMD hosted their Capsaicin live stream event from GDC. While there were some product announcements like the Radeon Pro Duo and the teasing of the upcoming Polaris 10 GPU, most of the time was spent on reiterating past statements. The key to this was VR and AMD spent a lot of the event focusing on this and ragged in a large number of industry insiders to shore up that point. Of course, we also get the usual cringeworthy humour from their engineers.
First off, AMD spoke about their investment in the pixel and HDR. Once again the focus was on improving the information each pixel portrayed to better present the whole image. Of course, AMD also talked about increasing pixel count more and more to get better image quality. The key to this are developments in new APIs such as DX12 and Vulkan as well as AMD’s own solutions in the form of GPUOpen which has been expanded upon with GeometryFX and other additions. One number mentioned was 16, or the 16ms that is allowed for each frame to be computed in order to allow 60FPS.
In order to power these effects, though, AMD is hoping that GPU scaling will continue to improve. This will be due to both improved scaling of multi-GPU (like for the Radeon Pro Duo) due to better support in DX12 and improved process does and architectures. AMD has noted that while GPUs haven’t been keeping track with Moore’s law in terms of performance per dollar, smaller GPUs have and it is important to be able to get 2 smaller GPUs to work together better since that solution would offer better bang for the buck.
In terms of VR, AMD is looking forward to working with developers to get the best performance out of their hardware to get the best experience. In line with this, AMD is pointing out how their hardware is more than ready for with ACE to allow the best performance under DX12. Combined with LiquidVR and their other software libraries, AMD is presenting a comprehensive solution to allow developers to tackle VR. AMD is also offering a certification program for VR ready systems with their hardware to ensure consumers know that the hardware they are getting can handle VR. With this, maybe VR will go mainstream soon enough.
After many months of waiting, AMD has finally unveiled their dual Fiji graphics card. Though not called FuryX2 as we originally expected, the name Radeon Pro Duo is just as fitting. For now, AMD still has not revealed the full specifications but the most important one, price, is a lofty $1,499 USD. For 2 Fiji GPUs and 16TFLOPs of performance, it may well be enough to entice the VR developers AMD is targeting.
As expected of the most powerful single card GPU yet, the power requirements are massive. The 2 GPUs draw power over 3 PCIe 8pin power connectors for up to 525W of power. Memory bandwidth is doubled but the memory remains split as with CFX or dual-GPU cards, with 4GB of HBM1 each over a 4096bit bus. In total, the card has 8192 Shader Cores, 512 TMUs and 128 ROPs with 4DP display connectors. Cooling is provided by a Cooler Master CLC unit with a 120mm extra-thick radiator.
Targetted towards VR and game developers, it makes sense as it offers the performance necessary to the run the most demanding of titles, especially in their unoptimized form. Furthermore, the use of AMD’s affinity multi-GPU, LiquidVR and DX12 will all serve to limit the impact of having 2 separate GPUs. This should allow for better scaling than we usually see with CFX and other solutions.
By targeting VR developers, AMD is able to get away with the hefty price tag, just as Nvidia was able to do the same with their Titan series. This may limit the market though some AMD fans and prosumers won’t mind too much. The price is only $200 more than 2 Fury X’s but it will use up fewer slots and be less of a hassle to arrange the cooling for it. The late launch however, is more of a problematic issue as Polaris and Pascal are fast approaching. It remains to be seen if AMD’s gamble will pay off.
Prepare your wallets for summer 2016 because both AMD and Nvidia are going to release their new GPUs then. Yesterday, we got the first hint about Nvidia’s GTX 1080 which is reportedly launching May 27th. For AMD, the details for Polaris have always been a bit vague, with only mid-2016 being the only hint. Today, a new rumour has popped up with the suggestion that AMD will launch Polaris in June 2016. Furthermore, AMD will be providing a sneak peak of Polaris at Capsaicin next week.
A June launch puts Polaris right into the area of Computex and E3, perfect events to showcase the new GPUs. Launching at the same time as Nvidia also avoids certain issues as AMD has gotten into trouble both launching before and after Nvidia so maybe launching at the same time will be the key. Set to be on the 14nmLPP process, AMD has a good chance to snag some marketshare away from Nvidia.
Next week, we may get a few more details from AMD about what Polaris will look like in the sneak peak. One can only hope the sneak peak will be more than just a picture or another demo but something more substantive. On March 14th, AMD’s Capsaicin webcast from GDC will likely reveal FuryX2 as well as showcase some of their VR developments. With AMD having hit their worst marketshare yet recently, they have started their come back and can only up. Hopefully, Polaris will deliver what is needed.