New Upcoming ATI/AMD GPU's Thread: Leaks, Hopes & Aftermarket GPU's

Some new Net Lags comparison:

Spoiler

More Here:
->

 

Care to explain?

 

I like it (so i share), today i found it on Web

-> https://www.humblebundle.com/store/oddworld-abes-oddysee

Here have a cokie.

 

Almost here
I wonder if it have SM 6.0



-> Run Store then Search for FM7 Demo

 

Already got that...thought it was something Radeon related.

 

FM7
Fury on my Default Easy Gamer 1005/550 1169v
Almost All settings Maxed at 1440p 4xMSAA | 60FPS no-Vsync (Ave. 61, CAP in ReliVe to 62)

So smooth, and GFX are stunning -> Best looking Racer ATM, waiting for GTR3

Some screens:
===

Spoiler

 

Asus rog strix Radeon rx Vega 64:

-> http://greentechreviews.ru/2017/09/18/obzor-videokarty-asus-rog-strix-radeon-rx-vega-64

 

AMD Is Building A Self-Driving Car AI Chip For Tesla

AMD has reportedly won another major semicustom chip design contract, this time around from Tesla. Both companies are said to be working together on a new AI accelerator for self-driving cars, reports CNBC.

Over 50 Tesla engineers are reportedly involved in the project which is being led by none other than legendary ex-AMD microprocessor and system architect Jim Keller. Keller’s resume includes desgining Apple’s A4 and A5 chips as well as developing AMD’s most successful processors, including the original Athlon64 in the early 2000s and the company’s latest Zen microarchitecture.

Tesla has reportedly already received samples of the new semi-custom chip and is currently testing it. This is Tesla’s first attempt at making its own purpose-built AI chip and is said to be leveraging AMD IP mixed with its own. The company currently relies on GPUs from AMD’s rival, NVIDIA, for its self-driving cars.

The new AI chip is believed to be the first major step towards Tesla’s goal of delivering totally autonomous vehicles to consumers by 2019. Last month we reported that AMD’s next generation Navi graphics architecture will feature AI specific hardware. Navi is expected to launch some time in 2019, which coincides with Elon Musk’s promise to deliver completely autonomous vehicles by 2019.

UPD.

• Jim Keller (legendary microprocessor architect and formerly of Apple and AMD) joined Tesla Autopilot as the VP of Autopilot Hardware Engineering.
• Peter Bannon (formerly of Apple) followed Jim to Tesla Autopilot.
• David Glasco (formerly worked at Intel, IBM and NVIDIA) was Senior Director of Server SOC Architecture at AMD before joining Tesla.
• Thaddeus Fortenberry (formerly Cloud Server Architect at AMD) also joined Tesla Autopilot in the coming weeks.
• Debjit Das Sarma (former AMD fellow and CPU Lead Architect) also joined Tesla soon after.
• Keith Witek (formerly AMD Corporate Vice President of Strategy and Corporate Development) joined Tesla as the Director of Autopilot Enablement and Associate General Counsel.
• Junli Gu (Formerly Member Technical Staff, Machine Learning at AMD) joined Tesla as the Tech Lead for Machine Learning, Autopilot

-> http://wccftech.com/amd-making-semicustom-self-driving-car-ai-chip-tesla

 

Make PC Gaming Great again! -> DX12/Vulcan is the Future!

 

Is DeusEx MkD a new Crysis in terms of Taxing GPUs and Optimisations?
So many new GFX candys

 

I did like some of the tech the game used but it's also a pretty demanding on VRAM but as long as you have some 6+ GB available on the GPU the game can look really impressive. (With little stuttering.)
Improved PureHair/TressFX 3.x, high-resolution SSR, CHS shadow support although it also cuts down the shadow draw distance unless that was fixed, mix of tessellation for some surfaces like characters and parallax for others such as cobblestone and things like depth of field, motion blur, chromatic aberration, subsurface scatter, light-rays and of course the material system. (And tons of other shaders if one were to go into it more in-depth.)

Quite a step up in terms of tech over Human Revolution at least.
(DX12 is quite unstable though but there is some performance gains if it doesn't crash at least on AMD GPU's and depending on CPU load.)

I think the engine might be using multi-threaded rendering too but that's showing up in more and more games lately for both DX11 and DX12 when used.

As for VRAM consumption and RAM usage in general there's a possible memory leak so it builds up over time and even for 1920x1080 the ultra texture quality can see VRAM usage hover around 6 GB especially in the two Prague hub zones of the game where there's more aggressive texture streaming employed. (The winding streets also allow for some nice culling though and LOD scaling though it's still very demanding on the GPU.) Exploration and detail is really nice too with tons of things to find and numerous layers ranging from above ground open windows leading into small closed of rooms and then the sewer section. Loading times are pretty poor though so traveling between Prague 1 and Prague 2 was a slow process and a lot of the 12 or so side quests throughout the game have you going back and forth. (Tons of content that also doesn't seem to lead anywhere hinting at unimplemented content and just things in general fleshing out some of the game world a bit since Jensen can't solve everything.)

Bank DLC was fun though I wasn't a huge fan of the prison one, would be nice to see Square / Eidos returning to and wrapping up the story but it's on ice for the moment and some Avengers game is being developed currently as I understood it.
Hitman is pretty neat too even if it doesn't have quite the same tech, devs being able to break off from Square was a nice move too allowing them to develop the game as they want though I don't know if much more is planned for it after the last patch doing some lighter work and smaller tweaks here and there. (Removed the DRM too, well the non-Steam one at least ha ha.)

 

Need for Speed Payback PC Requirements:

Minimum Requirements for 720p30 at low settings

• OS: 64-bit Windows 7 or later
• CPU: Intel i3 6300 @ 3.8GHz or AMD FX 8150 @ 3.6GHz with 4 hardware threads
• RAM: 6GB
• DISC DRIVE: DVD ROM drive required for installation only
• HARD DRIVE: 30GB
• VIDEO: NVIDIA GeForce® GTX 750 Ti or AMD Radeon™ HD 7850 or equivalent DX11 compatible GPU with 2GB of memory
• DirectX: 11 Compatible video card or equivalent
• INPUT: Dual analog controller
• ONLINE CONNECTION REQUIREMENTS: 192 KBPS or faster Internet connection

Recommended Requirements for 1080p60 at high settings

• OS: 64-bit Windows 10 or later
• CPU: Intel i5 4690K @ 3.5GHz or AMD FX 8350 @ 4.0GHz with 4 hardware threads
• RAM: 8GB
• DISC DRIVE: DVD ROM drive required for installation only
• HARD DRIVE: 30GB
• VIDEO: AMD Radeon™ RX 480 4GB, NVIDIA GeForce® GTX 1060 6GB or equivalent DX11 compatible GPU with 4GB of memory
• DirectX: 11 Compatible video card or equivalent
• INPUT: Dual analog controller
• ONLINE CONNECTION REQUIREMENTS: 512 KBPS or faster Internet connection

 

AMD Preparing To Launch 13 Vega-11 Based Graphics Cards, GPU Passes Certification

AMD is readying several new graphics cards based on its Vega 11 GPU which has just passed its final manufacturing certification. The graphics cards will be based on AMD’s yet unreleased Vega 11 XT and Vega 11 Pro GPUs.

The new graphics cards are expected to replace AMD’s Polaris 10/20 GPUs found in the RX 480/70 and 580/70 graphics cards on the desktop as well as in laptops. The company is also preparing a variety of new Radeon Pro and Radeon Instinct accelerators based on the new GPU

AMD Preparing To Launch 13 New Graphics Cards Based On Vega 11 GPU
A few days ago we reported that AMD’s Vega 11 GPU is entering production and that company had placed wafer orders some time ago with Globalfoundries, its manufacturing partner to produce the GPU dies. Yesterday, 13 different graphics card variants based on the new Vega 11 GPU passed their RRA certification in South Korea. Which means that Vega 11 based graphics cards are now ready to enter the market.



The RRA certification for Vega 10 appeared just a month ahead of AMD’s Radeon Vega Frontier Edition announcement, so we’re definitely nearing Vega 11’s debut. There were only six graphics board variants based on the Vega 10 GPU mentioned in its RRA ceritifaction, Vega 11 has 13.

Unlike Vega 10 which was exclusive to the desktop, Vega 11 will be the first ever HBM based graphics card to come to notebooks. So a number of those 13 certified boards are going to be mobile variants. If things go according to plan for AMD, Vega 11 should be ready to go into notebooks in time for the holiday season alongside its Raven Ridge APUs.

According to information that surfaced a couple of months ago, we know that two of the 13 graphics cards will be RX Vega boards and two more will be Radeon Pro boards. Finally, an unknown number of the 13 variants are Radeon Instinct accelerators.

AMD Radeon RX Vega 32 & RX Vega 28 (RUMOR)
Whispers around the industry are that the two RX Vega boards based on Vega 11 will be RX Vega 32 and RX Vega 28. Vega 11 XT is rumored to have 2048 GCN stream processors, a 1024-bit memory interface and 4GB of HBM2, while Vega 11 Pro is rumored to have 1792 stream processors and the same memory interface and capacity. The cards are expected to go up against NVIDIA’s GTX 1060. As with all rumors take this information with a grain of salt. We should be able to learn more fairly soon, so stay tuned.

 

Oh, wow. So they really did go all out on HBM.

Can't wait to see the HBM Raven ridge models.... (here's hoping).

 

Take my money and give me my Vega laptop already.

 

AMD Bids Farewell To CrossFire (Name) After 12 Years, Retiring Brand In Favor Of mGPU

AMD revealed today that it no longer refers to multi-GPU support by its long-standing Radeon brand known as CrossFire. This revelation came via PCWorld’s Brad Chacos, who was curious as to why Crossfire wasn’t mentioned in the company’s latest Radeon Software 17.9.2 drivers. He proceeded to email AMD to inquire about this apparent omission.

To his surprise AMD’s answer was that it’s transitioning away from the CrossFire brand, referring to it as a DirectX 11 era technology. “CrossFire isn’t mentioned because it technically refers to DX11 applications,” an AMD PR person told PCWorld. “In DirectX 12, we reference multi-GPU as applications must support mGPU, whereas AMD has to create the profiles for DX11. We’ve accordingly moved away from using the CrossFire tag for multi-GPU gaming.”

AMD further elaborated by saying that this is simply a branding decision not a technical one. The company is by no means ending multi-GPU support for its graphics cards, in fact its latest drivers enable multi-GPU support for its recently released Radeon RX Vega graphics cards with promises of 80%+ performance scaling.

Spoiler

Multi GPU In The New DirectX 12 Era
Gone are the days of old when graphics card vendors had to hand-craft and optimize CrossFire & SLI profiles for multi-GPU to work at all in games. DirectX 12, unlike all of its predecessors, delegates this responsibility entirely to the game developer via a feature called Explicit Multi-Adapter. The API enables low-level access to the hardware, allowing developers to access each GPU directly and independently.



There are two types of Explicit Multi-Adapter modes in DirectX12 “Linked” and “Unlinked”. Linked mode is only available when there are multiple identical GPUs in the system, similar to what CrossFire and SLI setups look like today. It allows for the computing/graphics resources and the memory pools of multiple GPUs to be combined into one larger addressable unit. This mode works by delegating similarly tasking workloads to each GPU, essentially cutting the overall work in half to complete it in a shorter amount of time.

One GPU could render one portion of the frame while the second could render another portion of the same frame for example. This is called Split Frame Rendering, SFR for short. This technique negates the limitations inherent in the traditional alternate frame rendering – AFR – technique used today in DX11 and older games, which involves mirroring the memory pools and introduces the need for frame pacing.



Unliked mode is the second multi-GPU option under DX12. This mode is designed to take advantage of GPUs with different capabilities and even different vendors at once in the system. Yes, it allows developers to force NVIDIA and AMD GPUs to work together. The mode enables this through a thin abstraction layer allowing the different GPUs to swap data back and forth while giving developers total control. This mode enables developers to make use of discrete GPUs alongisde integrated GPUs or other discrete GPUs. With each GPU being treated as a completely independent graphics engine that the developers can choose to utilize for whatever they want.

This brings is to the concepts of CrossFire and SLI, which were introduced over a decade ago. As you can clearly tell the new new DirectX 12 multi-GPU capabilities have very little to do with SLI and CrossFire, which are essentially proprietary algorithms developed by each vendor to enable alternate frame rendering via the driver layer. Hence calling DirectX 12 multi-GPU support SLI or CrossFire is simply not accurate.

DX12 in and of itself does not rely on the proprietary SLI and CrossFire technologies that NVIDIA and AMD have developed. DX12 multi-GPU support stands on its own two feet. All that’s required of the vendor is to make sure that their hardware and software is compliant with the multi-GPU features built into DirectX 12. This is why AMD isn’t calling multi-GPU support in its latest driver release CrossFire, because it simply is not. Not anymore.

 

-> http://www.tomshardware.com/reviews/acer-xr382cqk-curved-freesync-monitor,5067.html

 

I'm still very skeptical of mGPU. It remains to be seen if developers will be interested at all in supporting it. I would really love to see it happen, But i'm just not confident.

 

AMD/ATI
CASE STUDIES
From Tesla, Apple and Dr Pepper Snapple Group to Sony, MS and HP, the world's top companies depend on AMD. Take a closer look at the stories behind the products and discover how your home security system, your flight, your gaming console, your soda, and a stroll through Times Square could all have one thing in common - AMD.

-> http://www.amd.com/en-us/solutions/amd-everywhere-case-studies