Windows 10 Combined VRAM Performance Update

[AKK_K]

WINDOWS 10 UMP
UWP (Universal Windows Platform) allows developers to create experiences that are easily deployed across all Windows 10 devices,
from PCs to tablets to phones to Xbox One. When it comes to a UWP game on Windows 10 PCs, we?re early in our journey.
We?re listening to the feedback from the community ? multiple GPUs, SLI, crossfire, v-sync, etc.
We?re embracing the feedback and working to ensure gamers on Windows 10 have a great experience.
We?ll have more to discuss in the coming months.? ? a Microsoft spokesperson

http://www.pcper.com/news/Graphics-Cards/AMD-Add-DirectFlip-DX12-Indicates-FreeSync-Incompatible-UWP-Games

http://www.pcper.com/news/Graphics-Cards/Microsoft-Comments-UWP-Games-and-Full-Screen-Capability

[AKK_K]

PresentMon: Frame Time Performance Data for DX12 / UWP Games

http://www.pcper.com/reviews/Graphics-Cards/PresentMon-Frame-Time-Performance-Data-DX12-UWP-Games

So let's look at some data provided by PresentMon and how it compares to other tools in the market.

Intel Core i7-5960X + X99
NVIDIA GTX 980 Ti (364.00)
AMD Fury X (16.3)

gears-980tivsfuryx_0.jpg

I made the red line representing the Fury X semi-transparent to help with visualizing the results,
but otherwise we are looking at PresentMon data as provided and described above.
The green team is consistently running at lower frame times (higher frame rates) with the GTX 980 Ti
and it shows significantly fewer spikes in frame times than the Fury X.
It should be noted that the benchmark and game play of Gears feel SIGNIFICANTLY better
with the recently released 16.3 driver from AMD than with 16.2 even though the overall advantage still lies with NVIDIA.

[GreenHerb]

รอเกมมาส่ง T_T

[AKK_K]

ปู่มาแล้วหลานเอ้ย

[AKK_K]

Intel Skylake GT4e GPU chip pictured

Intel GPU ที่แรงกว่า Iris Pro 50%


10.03.2016: Congatec, a supplier of industrial computer modules, has released a picture of its Conga-TS170 module which uses Intel's 6th generation Core Skylake CPU with Intel Gen9 HD Graphics (GT4e) chip allowing us to see the large GPU chip for the first time.

The Congatec Conga-TS170 module is based on Intel's Core i7-6820EQ chip, which is a mobile Skylake-H quad-core CPU with support for Intel Hyper-Threading, packs 8MB of L3 cache and works at 2.8GHz base and 3.5GHz Turbo clock.

This CPU is interesting as it comes with Intel's flagship Gen9 HD Graphics (GT4e) GPU with 72 Execution Units, 128MB of eDRAM and maximum GPU clock of 1150MHz.

The Conga-TS170 module is aimed at graphics intensive applications and thanks to the pictures of the module we are able to check out the GT4e GPU chip for the first time. What we can see from the picture of the module, the GT4e chip is rather large and very long.

The die appears to be somewhere in the range of 210-220mm2 in size and includes 77mm2 eDRAM chip. According to details, over 65 percent of the die is reserved for the IGP and this chip should provide plenty of graphics power.

Hopefully, we will soon see some benchmarks as well which will confirm Intel's earlier statements that the GT4e is up to 50 percent faster in 3D Gaming, compared to the Skylake Iris Pro chip.

http://www.ocaholic.ch/modules/news/article.php?storyid=14358

http://www.legitreviews.com/intel-skylake-gt4e-core-photo-shows-long-chip_179708

[GreenHerb]

สนแต่ The Division อ่ะตอนนี้

[AKK_K]

Nvidia Async

The 2nd gen Maxwell can support async compute, but it don't support multi-engine concurrency.
This is because the compute and the graphics engine cannot be activated at the same time.
All nVidia GPUs have one execution slot reserved for compute shaders in the graphics engine.
And this is the engine what nVidia must use for D3D12, because Microsoft require hardware support for resource barriers.
The actual hardwares from nVidia don't support this in the independent compute engines.
The second problem is that the SMX can only execute one type of WARP at the same time,
and a full cache and scheduler flush required to switch between modes.
This switch is very-very slow, so mixed compute and graphics instructions will just slow down the hardware.

The number one reason why nVidia did this to the 2nd gen Maxwell is the efficiency for the mobile/ultramobile market.
This design has some serious limits with D3D12 and Vulkan,
but in the other hand the power consumption is better than what AMD can provide with that robust scheduling in their GCN designs.

DX12 Multi engine capabilties of recent AMD and Nvidia hardware

What are engines and queues?

เรื่องมันยาว

http://ext3h.makegames.de/DX12_Compute.html

[AKK_K]

voxel (n) ปริมาตรที่เล็กที่สุดที่รวมตัวกันเป็นภาพสามมิติ ย่อมาจาก volume pixel

Voxel From Wikipedia, the free encyclopedia

A voxel represents a value on a regular grid in three-dimensional space.
As with pixels in a bitmap, voxels themselves do not typically have their position (their coordinates) explicitly encoded along with their values.
Instead, the position of a voxel is inferred based upon its position relative to other voxels
(i.e., its position in the data structure that makes up a single volumetric image).
In contrast to pixels and voxels, points and polygons are often explicitly represented by the coordinates of their vertices.
A direct consequence of this difference is that polygons are able to efficiently represent simple 3D structures with lots of empty or homogeneously filled space,
while voxels are good at representing regularly sampled spaces that are non-homogeneously filled.
Voxels are frequently used in the visualization and analysis of medical and scientific data.
Some volumetric displays use voxels to describe their resolution. For example, a display might be able to show 512?512?512 voxels.

The word voxel is based on a contraction of vox ("volume") and el (for "element");[1] similar formations with el for "element" include the words pixel and texel.

[AKK_K]

The Basics of GPU Voxelization
What Can You Do With Voxelization?
What is GPU Voxelization?

/แต่บางกรณีมันทำให้เกิดความผิดพลาด
Oh No! Cracks During Voxelization!
What? More Holes!

Extra Voxels Generated By Expanded Primitive Edges For Conservative Rasterization

. . .

https://developer.nvidia.com/content/basics-gpu-voxelization

[AKK_K]

Voxel Global Illumination (VXGI) is a stunning advancement,
delivering incredibly realistic lighting, shading and reflections to next-generation games and game engines.

WHAT VXGI REALLY IS
A software library that computes approximate indirect illumination
- Works on any DX11 GPU, faster on Maxwell--
- Has to be integrated into rendering engines
- UE4 integration available
- One bounce of indirect illumination

An algorithm inspired by SVOGI
-Voxel cone tracing
-Clip-map instead of an octree
-Handles large and dynamic scenes well, no preprocessing

http://on-demand.gputechconf.com/gtc/2015/presentation/S5670-Alexey-Panteleev.pdf

[AKK_K]

NVIDIA VXAO (Voxel Ambient Occlusion)

Nvidia calls VXAO ? the highest quality Ambient Occlusion solution available in PC games today.?

[AncienTi]

http://tombraider.tumblr.com/post/140859222830/dev-blog-bringing-directx-12-to-rise-of-the-tomb

Rise of Tomb Rider ลง DX 12 ให้มาแล้วจ้า

[AKK_K]

Rise of the Tomb Raider patch adds DX12 and Nvidia VXAO (Voxel Ambient Occlusion)

by Jeff Kampman ? 10:51 AM on March 11, 2016
Rise of the Tomb Raider is getting its fifth patch today, and it's a big one.
Developer Nixxes Software says this patch will let the game take advantage of the DirectX 12 API
and Nvidia's voxel ambient occlusion (or VXAO) technology.

The arrival of Nvidia voxel ambient occlusion in Rise of the Tomb Raider is a bit more mysterious.
Nvidia hasn't documented this feature in its online reference materials yet,
but it appears to be a subset of the company's work on voxel global illumination (or VXGI).
Voxel global illumination can produce highly realistic lighting effects in a scene,
so we'd expect a better-looking ambient occlusion effect in this case.
Nixxes says the feature will only work on Nvidia Maxwell graphics cards, and only then in the game's DX11 mode.

VXAO ทำงานที่ DX11

http://techreport.com/news/29841/rise-of-the-tomb-raider-patch-adds-dx12-and-nvidia-vxao

[AKK_K]

380X การ์ดระดับกลางที่น่าจะดีที่สุดสำหรับ The Division
แรงกว่า GTX960 30-40%
02-Ingame-vs-Walk_w_600.jpg

[AKK_K]

และ gamework

12-Gameworks-High-End_w_600.jpg