This site may earn chapter commissions from the links on this page. Terms of employ.

Skylake1

At IDF this week, Intel shared some of the architectural details for Skylake that were missing from the initial launch. While not a consummate disclosure, we now have a amend film of what the CPU is capable of, and where Intel focused to improve performance and power consumption. The briefings at IDF covered both the CPU and GPU; we'll cover CPU details in a time to come story. For now, here's what we've learned about Intel's new on-die GPU.

When AMD launched its Kaveri back in early 2022, it focused on the fact that the GCN GPU accounted for more than half the die. The argument was straightforward — the GPU was going to be ever more important to hitting high performance targets and deserved a larger slice of the pie. Intel, it seems, heard this line of reasoning and agreed. Skylake's GPU slices can be trimmed or expanded depending on which market segments the company wants to hit, simply the full Skylake GPU takes upwardly more die infinite than whatsoever previous bit.

Intel-GPU1

Skylake is also Intel's first GPU with explicit DirectX 12 back up. Information technology's not articulate how the company is handling DX12 on previous chips, we've seen early demos of games running DX12 on Haswell and Broadwell, but Intel continues to officially list DirectX 11.ii as the API for its earlier chips.

Intel hasn't released whatsoever hard data on what the part numbers on hereafter Skylake GPUs will be, or disclosed exactly which chips will receive which cores, but nosotros can infer some bones relationships. Typically, Intel has launched high-end desktop SKUs with what it calls a GT2 configuration. GT3 and GT3e (with EDRAM) configurations tend to be reserved for mobile chips or special desktop processors, while GT1 ships on lower-end chips in the Pentium and Celeron families.

Skylake will shift this somewhat past adopting a new GT4 design (more on that in a moment), but the basic structure remains the same.. Similar Intel's previous GPUs, Skylake's graphics core is divided into slices, sub-slices, and unslices. Each slice contains three sub-slices, and each sub-piece contains 8 execution units (EUs). In that location are 24 EUs per slice, and Intel can stack up to three slices in a GPU configuration, for a total of 72 EUs.

Gen9GPU

Skylake'southward new GPU has more back-stop fill charge per unit capabilities, ameliorate performance with MSAA enabled, and larger caches to feed the GPU. Like AMD and Nvidia, Intel has increased its ain lossless colour compression, saving bandwidth and power in the process. The gains here are highly dependent on the championship, but Intel claims that Arkham Origins is 11% faster, BioShock Infinite picked upwardly 4%, and Tomb Raider gained three%. While 3% operation isn't much in and of itself, information technology still matters when rolled into a group of more extensive improvements.

GT4, multimedia, and EDRAM

In the past, Intel offered ane or two slices in its GT1/GT2 and GT3 architectures. With Skylake, certain SKUs volition also include a GT4 GPU that adds a third slice. This will further improve EU throughput, Z/Stencil and pixel operations, and full GPU performance. While the benefits of adding another slice typically aren't linear, we should still meet significant improvements for GT4 over and above GT2 and GT3.

GT4

Skylake's video decoder will back up HEVC and VP8 encode and decode in hardware, and the new EDRAM equipped SKUs should significantly improve integrated performance. Intel is also adding support for hardware-accelerated encode and decode for boosted formats, as shown in the table below:

Gen9Codec

We don't know which chips will conduct eDRAM, but Skylake volition deploy it in both 64MB and 128MB configurations. Past benchmarks have shown that EDRAM equipped Broadwell processors are significantly faster than AMD's Kaveri or whatsoever other integrated graphics solution. That'southward not really surprising, since Kaveri is bandwidth constrained by its dual-channel DDR3, merely it could make life even more difficult for AMD if Intel pushes a 64MB Skylake processor into APU territory.

Intel is serious about its graphics performance

From 2000 – 2022, Intel's integrated graphics performance was barely worthy of being called a joke. Even beginning generation Atoms, which weren't exactly known for stellar functioning, improved noticeably when paired with an Nvidia Ion GPU. And by "improved," we mean that even browsing and desktop navigation was noticeably faster. Intel was a CPU visitor, not a GPU company, and it seemed as though this would never change, especially after Larrabee was canceled.

Memplat

Sandy Span was arguably the first Intel GPU that could reasonably exist described as such without provoking a round of snickering. Each generation of Intel processor that'southward followed it has improved performance, additional game compatibility, and delivered a improve all-in-i solution. Today, Skylake is capable of holding its own with a $lxx GPU, and that'southward in a GT2 configuration without the benefit of EDRAM. Intel has signaled that information technology intends to bring EDRAM to more SKUs with Skylake, and that'south only going to improve low-end performance farther. The fact that then many OEMs have ditched AMD and Nvidia GPUs in high-end mobile hardware is just another example of the trend.

If you're a gamer, this is a unilaterally good thing. Developers take to consider the hardware configurations of everyone who might play their game, which means they've had to assume that at least some players would be using Intel integrated solutions. Equally Intel raises its ain bar for acceptable GPU operation, it lifts all players who desire a game that doesn't look awful. Integrated graphics volition never kill the enthusiast loftier-end, considering the laws of physics dictate that a split up GPU with dedicated cooling and power will always be able to striking college performance targets than an integrated solution.

Chipzilla won't win any points for sprinting across the finish line in this regard — the visitor is but starting to coil out its EDRAM solution to a wider marketplace now, over two years later information technology first shipped. Slowly merely surely, however, the "skilful enough" bar is rise. Today, desktop Skylake is capable of driving modern games at 720-1080p with low to medium detail at >30 FPS. Higher end Skylake will probable offer 1080p as a given, with medium or even high item depending on the game.

AMD will reply with new APUs and a new GPU architecture on 14/16nm, but not for some time. Zen CPUs are expected to transport next year, but Zen APUs may not launch until 2022. HBM will solve the bandwidth constraints that plagued Kaveri, only I doubtable AMD will confront a tough fight by the fourth dimension it's gear up to ship its next generation of integrated hardware. Intel may have started off equally a joke, but it'south evolved its GPUs into genuinely decent solutions — even if they'd never exist the first choice of whatsoever hardcore gamer.

Thank you to Matthew Murray, who is on location at IDF this calendar week and contributed to this report.