Intel's Architecture Day 2018: The Future of Core, Intel GPUs, 10nm, and Hybrid x86
by Dr. Ian Cutress on December 12, 2018 9:00 AM EST- Posted in
- CPUs
- Memory
- Intel
- GPUs
- DRAM
- Architecture
- Microarchitecture
- Xe
The Next Generation Gen11 Graphics: Playable Games and Adaptive Sync!
Some of the first words out of the mouth of Raja Koduri about graphics is that Intel has a duty to its one billion customers with integrated graphics to give them something that is useful, and that it is time for Intel to provide graphics which people can actually play games on. Given his expertise on the matter, it shouldn’t sound too far-fetched: more people play games than ever before, and these users want to play no matter what their hardware. To that end, Raja stated that Gen11 graphics is the first step in a new graphics policy to provide the performance and features to let gamers play the most popular games, no matter what implementation.
Gen11: Intel’s first GT2 TFLOPS Graphics
In 2015, Intel launched the Skylake processor with Gen9 integrated graphics. Rather than moving straight to Gen10 the next time around, we were given Gen 9.5 in both Kaby Lake and Coffee Lake, which supposedly draw features from what would have been Gen 10. Actually, the graphics for Intel’s failed 10nm Cannon Lake chip were meant to be called Gen10, however Intel never released a Cannon Lake processor with working integrated graphics, and because Gen11 goes above and beyond what Gen10 would have been, we’ve gone straight to Gen11. Make sense? Well Intel didn’t even bother to acknowledge Gen10 in its history graph:
We will see Gen11 graphics being paired with Sunny Cove cores on 10nm sometime in 2019 according to the roadmaps. However rather than give a detailed architecture layout for the new product, we instead were given a rather high level diagram.
From here we can deduce a few things. We were told that this configuration is the GT2 config, which will have 64 execution units, up from 24 in Gen9.5. These 64 EUs are split into four slices, with each slice being made of two sub-slices of 8 EUs a piece. Each sub-slice will have an instruction cache and a 3D sampler, while the bigger slice gets two media samplers, a PixelFE, and additional load/store hardware. Intel lists Gen11 targeting efficiency, performance, advanced 3D and media capabilities, and a better gaming experience.
Intel didn’t go into too much detail regarding how the EUs are at higher performance, however the company did say that the FPU interfaces inside the EU are redesigned and it still has support for fast (2x) FP16 performance as seen in Gen9.5. Each EU will support seven threads as before, which means that the entire GT2 design will essentially have 512 concurrent pipelines. In order to help feed these pipes, Intel states that it has redesigned the memory interface, as well as increasing the L3 cache of the GPU to 3 MB, a 4x increase over Gen9.5, and it is now a separate block in the unslice section of the GPU.
Other features include tile-based rendering, which Intel stated the graphics hardware will be able to enable/disable on a render pass basis. This will make Intel the final member of the PC GPU vendor community to implement this, following NVIDIA in 2014 and AMD in 2017. While not a panacea to all performance woes, a good tile rendering setup plays well to the bandwidth limitations of an integrated GPU. Meanwhile Intel's lossless memory compression has also improved, with Intel listing a best case performance boost of 10% or a geometric mean boost of 4%. The GTI interface now supports 64 bytes per clock read and write to increase throughput, which works with the better memory interface.
Coarse Pixel Shading, Intel's implementation of multi-rate shading and similar in scope to NVIDIA’s own Variable Pixel Shading, is also supported. This allows the GPU to reduce the amount of total shading work required by shading some pixels on a less than 1:1 basis. Intel showed two demos for CPS, where pixel shading was reduced either as a function of object distance from camera (so you do less work when things are further away), or reduced as a function of how close the object is to the center of the screen, designed to help features like foveated rendering for VR. With a 2x2 pixel stencil applied – meaning only one pixel shading operation was done per block of 4 pixels – Intel stated a ~30% increase in frame rates in supported games. Unfortunately this needs to be applied on a game-by-game basis in order to prevent significant image quality losses, so the performance gains won't be immediate or universal.
For the media block, Intel says that the Gen11 design includes a ground up HEVC encoder design, with high quality encode and decode support. Intel cited the fact that its media fixed function units are already used in the datacenter for video processing, and home users can take advantage of the same hardware. Intel also stated that by using parallel decoders it can either support concurrent video streams or they can be combined to support a single large stream, and this scalable design will allow future hardware to push the peak resolutions up to 8K and beyond.
The highlight of the display engine is support for Adaptive Sync technologies. We were told that it was announced back at the launch of Skylake, but now it is finally ready to go into Intel’s integrated graphics. This goes in hand with HDR support due to its high-precision data path.
One thing in this presentation that Intel didn’t mention directly is that Gen11 graphics would appear to have Type-C video output support, potentially indicating that Intel has integrated the necessary mux into the chipset itself, removing another IC from the motherboard design.
Facebook
Twitter
RSS
148 Comments
View All Comments
CajunArson - Wednesday, December 12, 2018 - link
There's nothing whatsoever revolutionary about "chiplets". A 10 year old core 2 quad used exactly the same technology that AMD calls "chiplets" in 2019 and AMD fantards like you even insulted the Core 2 quad for doing it.Maybe you should actually read the article about what a active interposer can do vs. wiring standard hunks of silicon to a PCB in literally the same way it was done in the 1970s before you run around acting like AMD is the only company to ever innovate anything.
sgeocla - Wednesday, December 12, 2018 - link
I've been reading articles about Intel 10nm promises for years now. And then we got failed laptop chips and low power pc boxes to appease 'mass production' status and not get sued by investors for false claims.Forgive me if I abstain my drooling until Intel actually delivers something that does not require industrial chillers. BTY where is that 28 core HEDT chips anyway ?
Targon - Wednesday, December 12, 2018 - link
There is always a point where the WHEN something is used sets a new direction. Multi-CPU in a workstation/server moving to a single processor with multiple cores was a big shift. Moving from two cores linked together when cache coherency was a big problem to a single dual-core without an interposer was a better way to go. It all comes down to if there is a performance boost or degradation as a result of the technology/implementation.With that said, a single CPU with 64 cores is fairly significant, and keeping the performance from being horrible with that many cores is the reason AMD has been praised. Price/performance for the server market and such.
For a long time, Intel was seen as the king when it came to clock speeds and performance, but Intel hasn't had a significant boost to IPC in over three years. Intel has also been promising 10nm for three years, and still no sign of it, with the promise of holidays 2019.
So, Intel still has nothing, they have vague promises of ways they will improve performance, but it remains to be seen if the performance will actually be better if 10nm slips again. On the flip side, AMD clearly has significant performance boosts from Ryzen 3rd generation in 2019(March/April being when many expect it). 7nm from AMD isn't a, "will they?" question, it isn't even a "when?", with CES in one month and with it, the answers. IPC improvements due to design improvements not related to chiplets at all would be good, as well as higher clock speeds. So, there is a potential for 30+ percent higher performance in one generation.
Yes, I don't expect AMD to deliver huge performance jumps again for years, but we may see things such as Gen-Z support, going beyond two memory channels for the mainstream Ryzen chips when the next socket comes out in 2020/2021, and other things that may boost system/platform performance while AMD figures out how to get more CPU performance.
Intel is still trying to do things the same way, just faster. Faster CPU, faster links to individual devices, fabric on a system level will be Intel trying to reinvent what AMD has been working toward.
I will also note again that some things are not always about being new, but are more about presentation and implementation. Palm really popularized the idea of apps that users could install on a small portable device(PDA), but Apple popularized it with the iPhone. In some cases, the implementation really is good, and will get the respect of the industry, in other cases, you see that something is clearly a case of following the lead of another player.
So, in the PC industry, is Intel leading the way with innovations, or is AMD in the drivers seat?
iwod - Thursday, December 13, 2018 - link
No one insulted Core 2 Quad for doing it, and neither did AMD. But Intel did Insult AMD and went on full force bad mouthing AMD.Spunjji - Thursday, December 13, 2018 - link
Using a term like "fantard" straight-up devalues your argument, but the blatantly false statement about the C2Q using "exactly the same technology" seals the deal.Chiplets refers to the CPU being divided into multiple sections (cores and un-core) on a single package using dedicated interconnects. It's not at all the same technology as having two discrete CPUs joined by the FSB on a single package. Both are novel approaches to particular problems, although the C2Q (and Pentium D before it) were criticized for their inefficiency by using the FSB for inter-core communication. We don't know how "chiplets" will pan out yet, so the jury's out.
Bash the fans for talking nonsense all you want, but maybe don't sink to their level.
edzieba - Wednesday, December 12, 2018 - link
If you think through-package interconnects compare to through-silicon interconnects, then I have some HBM on DIMMs to sell you.Spunjji - Thursday, December 13, 2018 - link
Noice. :DIII-V - Wednesday, December 12, 2018 - link
I love how everyone thinks AMD is the pioneer with chiplets. They're not. That would be Marvell.And Intel themselves has been hinting that it's a good way to go, looking at their EMIB solution.
But AMD fan boys are a special breed of stupid...
sgeocla - Wednesday, December 12, 2018 - link
The electric car was pioneered more than a hundred years.It's one thing to pioneer something and a whole different thing to actually develop it into something that is affordable to millions and drags the whole industry forward.
If you think pioneering is all there is to it I have hundreds of grapehene battery designs you should invest you narrow-minded-driven life savings into.
evernessince - Wednesday, December 12, 2018 - link
You have some issues buddy. How about not being toxic next time.