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

Radeon Navi GPUs – AMD’s Brand New GPU Architecture, Aimed at Mainstream and High-Performance Gamers

Coming to Navi, AMD’s next biggest GPU architecture release in 2 years, the company would be following a similar strategy of launching it close to the mainstream CPU family.
This would give AMD an opportunity to highlight their upcoming mainstream Radeon family to potential buyers. The 7th July release date for Navi is also mentioned but considering there are rumors of a delay and the fact that AMD is under pressure from
TSMC’s overall manufacturing capabilities, we may get an announcement rather than a release on the aforementioned date.

With AMD Radeon VII released, AMD would put their full efforts towards Navi rather than the Vega architecture which has now existed as AMD’s high-end offering since 2017.
AMD’s Radeon Vega GPUs were designed with HPC in mind but they also served the gaming market.
There are several prosumer products based around Vega, you can even call the Radeon VII a prosumer card but that might change with Navi.



With Radeon Navi GPUs, AMD is said to be making gaming a first priority. You see, other than mainstream and high-performance desktop/mobility cards,
Navi has long been rumored to power the next-generation consoles from Sony and Microsoft.
AMD talked about partnerships on future platforms with Microsoft’s head of Gaming, Phil Spencer, at CES 2019, further confirming what we were being told.


“As we look forward to future platforms that we’re building and work that we’re doing, the partnership and the innovations that we’ve seen in the past have led to what we’ve been able to do today,
I think they’re going to be critically important to our future endeavors. I’m really looking forward to showing those to people more in the future.”

– Phil Spencer, Head of Gaming and Xbox at Microsoft

AMD Navi GPUs have been talked as the next big thing coming out from RTG and is likely the last GPU generation to feature GCN architecture as its DNA. With Navi, AMD is going to focus on brand new design changes to the architecture that would help boost its efficiency and performance.
For comparison, Radeon VII utilizes the 7nm architecture but it is essentially a die shrunk of an existing architecture.
The Navi GPUs will be designed around 7nm from the start and that would definitely help AMD take full advantage of the 7nm process node.

Previously in interviews, AMD’s CEO, Lisa Su, and AMD’s CTO, Mark Papermaster, both had confirmed a refresh of sorts across their Radeon program.

“We are really excited to start on the high end with our 7nm Radeon VII […] and you will see the announcements over the course of the year as we refresh across our Radeon program.”
“What we do over the course of the year is what we do every year. We’ll round out the whole roadmap,”


— Mark Papermaster (via The Street)


“We believe, we will be very competitive overall and that includes the high-end of the GPU market. Obviously there are new products out there from our competition.
We will have our set of new products as well and we will be right there in the mix”

“As it relates to ray tracing in particular I think it’s an important technology, but as with all important technologies it takes time to really have the ecosystem adopt [it].
And we’re working very closely with the ecosystem on both hardware and software solutions and expect that ray tracing will be an important element especially as it gets more into the mainstream, frankly, of the market.”


— AMD CEO Lisa Su – Nov 27, 2018 – 22nd Annual Credit Suisse Technology, Media & Telecom Conference

 

AMD states that all of its DX12 GPUs support real-time ray tracing via Microsoft’s DXR fallback layer

“Real-time ray tracing” are the latest and hottest buzzwords on the PC.
And while NVIDIA users have been enjoying these effects, AMD claimed that all of its current DX12 graphics cards support ray tracing via Microsoft’s DXR fallback layer.

According to AMD themselves, their current stack does support the fallback layer.
https://twitter.com/coreteks/status/1098174427609612288/photo/1
— coreteks (@coreteks) February 20, 2019

Now as you may know, only the RTX series is currently capable of running the real-time ray tracing effects in the latest games and that’s because AMD has not added support for the fallback layer on its drivers.
From what we know so far, the only GPU that supports Microsoft’s fallback layer is the NVIDIA Titan V, something that explains why this particular GPU was able to run Battlefield 5 (something we’ve covered in the past).

Performance, however, appears to be underwhelming via this “emulation/software” method.
This could explain why AMD has not enabled the real-time ray tracing fallback layer on its drivers as its GPUs currently lack hardware components that could accelerate the ray tracing calculations.
In short, and while theoretically AMD’s DX12 GPUs can support real-time ray tracing.

AMD is free to provide DXR support through their D3D12 drivers though. Any D3D12 GPU is capable of running this DXR code, since it is just an extension of DirectCompute.

 

AMD Drops The Price Bomb on Radeon RX Vega 56 Graphics Card, Now Available For $279 US

The Radeon Vega 56 also comes with AMD’s Raise the Game Fully Loaded Bundle which includes 3 AAA title.
These titles are Resident Evil 2 Remake, Devil May Cry 5 and the Division 2 which is $180 US worth of value.



AMD is offering a huge price cut on their Radeon RX Vega 56 graphics card by offering it at a price point of $279 US.
The graphics card launched last year and was a really popular graphics card that offered performance close to the flagship Radeon Vega 64 once it was overclocked.

Flash it & UV -> The Greatness awaits

 

nice!

 

AMD Radeon RX 590 and Radeon RX 580 Rumored To Get Price Cuts In The Coming Weeks – Prices Down To $229 US and $199 US, Respectively

The rumor states that the cards to be hit with the new price cuts would be the AMD Polaris based mainstream cards.
These include the Radeon RX 590 and the Radeon RX 580. The Radeon RX 580 was based on the Polaris 20 GPU architecture while the Radeon RX 590 is based on the Polaris 30 GPU architecture.
It was also recently rumored that AMD might be introducing their Radeon Navi GPUs in July aiming the mainstream segment so it’s highly likely that they are clearing inventory of Polaris cards before the arrival of next-gen parts.

Rumor: MSRP of RX580 8GB at 199$ and of RX590 at 229$ in next weeks.
— Bits And Chips – Eng (@BitsAndChipsEng) February 22, 2019

Both cards have the same GPU configuration of 2304 stream processors, 144 TMUs and 32 ROPs. Both cards even feature the same memory configuration of 8 GB GDDR5 VRAM, running across a 256-bit wide bus interface at a clock speed of 8000 MHz.
The only differences between the cards are their manufacturing process and clock speeds.
While the RX 580 is more close to the RX 480 with its 14nm process node and up to 1340 MHz clocks, the RX 590 utilizes the 12nm process to deliver clock speeds of up to 1545 MHz.

 

A lot of useful information (and mandatory pr), but thanks for posting; what a time to be alive, but the waiting game is killing me!

 

Thats the only thing AMD can do to fight nvidia, reducing prices. Navi will probably do what Polaris did two years ago, reclaim the midrange to high end for AMD again.

 

^^ Yo, im waiting till Arturus (next Big Chip)
I have a lot of headroom on my Vega, so VII is not mandatory update for me

1760-1800MHz is doable on my GPU.
Already ppl are posting (on OCnet) 29k GPU scores with LC.

 

Most of us are not interensted in a next big chip but upcoming great performing midrange gpus. If Navi can pull 2070 performance at 300$ it will be a killer card.

First they need to replace Polaris then Vega.

 

Yup, it's true.
Mid range is Main 'Vein' for industry.
Priced right with right performance can benefit all of Us (e.g. Beautiful & well optimised games).

 

Wouldn't that be great... If only AMD had a 2080 performing card for $200 cheaper than the 2080. Why does anyone even think that AMD is going to undercut Nvidia?

 

Seeing the market data showing the slow adoptance of RTX, and the reaction to the VII pricing, I think AMD will see that the market really isn't ready to bear those kind of pricings if they expect to move volume.

 

Finally some new Tech Demo by ATI/AMD

 

AMD's patent about variable rate super-sampling and sub-sampling:
http://www.freepatentsonline.com/20190066371.pdf
Yes, this thing has potential to not only decrease IQ for performance boost, but increase details.

 

**SPONSOR** is Power Color
Valid until March 3rd
Radeon RX 580 ($169.99) +2 games : -> https://www.youtube.com/redirect?q=https://www.newegg.com/Product/Product.aspx?Item=N82E16814131720&Description=rx%20580&cm_re=rx_580-_-14-131-720-_-Product&redir_token=yyUSKaq8cRV4GTud2fyadDMQFYh8MTU1MTU1MTg4NEAxNTUxNDY1NDg0&v=3dgZuUKgzZQ&event=video_description

 

• FreeSync 2
  
• Peak Brightness (Typical) = 600cd/m2 cd/㎡

So it's new HDR 600 QDot 1440p 144Hz !

-> https://www.samsung.com/levant/monitors/c27hg70/LC27HG70QQMXZN/

Not avaible in America or EU yet

 

AMD Navi GPUs might use Variable Rate Shading tech

A patent filing by AMD has recently surfaced, indicating the chip designer's intent to use Variable Rate Shading (VRS) technology to speed up output from future GPUs.
The patent was originally filed back in 2017 and explores VRS being used by next-gen GPUs. At that time, the furthest ahead the AMD GPU roadmap went is to the Navi GPU, offering "scalability and next-gen memory".
Note that Nvidia has already implemented VRS to work with its Turing GPUs, and you can read more about that on page 6 of the HEXUS Editor's examination of the Turing architecture.



A big clue to the attraction of VRS lies within its name. Its applicability is most relevant to VR headset rendering where the centre of the displays can offer the sharpest detail with lower fidelity required towards the edges and corners of the HMD displays.
As the HEXUS article discussed, there are various other ways to apply VRS, such as Content- and Motion-Adaptive Shading so it can be more useful to use in gaming using a traditional monitor, for example.



It certainly isn't set in stone that VRS is coming to AMD Navi GPUs, and AMD hasn't publically indicated it will.
However, it would be a logical move to 'keep up with the Joneses' and make more efficient use of AMD GPUs in PCs and consoles.
Furthermore, VRS is particularly useful in VR display rendering in HMDs which are expected to continue to progress and grow in popularity for PC and console users.

THX to HEXUS

Here supporting related article

-> https://kotaku.com/horizon-zero-dawn-uses-all-sorts-of-clever-tricks-to-lo-1794385026

 

In simple terms, VRS will apply full details on sections of the frames where most of the attention is focused while reducing the rendering load on other parts, thus increasing overall performance.
This method specifically helps when running games at higher resolutions. Following are some of the benefits of VRS:

Benefits of Apply Variable Rate Shading

• Variable Rate Shading is a new rendering technique enabled by Turing GPUs. It increases rendering performance by applying full GPU shading horsepower to detailed areas of the scene, and less GPU horsepower to less detailed areas.
• Variable Rate Shading works by varying the number of pixels that can be processed by a single pixel shader operation. Single pixel shading operations can now be applied to a block of pixels, allowing applications to effectively vary the shading rate in different areas of the screen.
• Variable Rate Shading can be used to render more efficiently in VR by rendering to a surface that more closely approximates the lens corrected image that is output to the headset display. This avoids rendering many pixels that would be discarded before the image is output to the VR headset.
• Coupled with eye tracking, Variable Rate Shading can be applied to maximize quality in the foveated area where the eye is directly looking and reduce the shading rate in the periphery. This method can improve rendering performance without noticeably impacting image quality.

 

AMD Stands for Open SMAA and TAA Solutions

A report via PCGamesN places AMD's stance on NVIDIA's DLSS as a rather decided one: the company stands for further development of SMAA (Enhanced Subpixel Morphological Antialiasing) and TAA (Temporal Antialising) solutions on current, open frameworks,
which, according to AMD's director of marketing,
Sasa Marinkovic,
"(...) are going to be widely implemented in today's games, and that run exceptionally well on Radeon VII", instead of investing in yet another proprietary solution.
While AMD pointed out that DLSS' market penetration was a low one, that's not the main issue of contention.
In fact, AMD decides to go head-on against NVIDIA's own technical presentations, comparing DLSS' image quality and performance benefits against a native-resolution, TAA-enhanced image - they say that SMAA and TAA can work equally as well without
"the image artefacts caused by the upscaling and harsh sharpening of DLSS."

Of course, AMD may only be speaking from the point of view of a competitor that has no competing solution.
However, company representatives said that they could, in theory, develop something along the lines of DLSS via a GPGPU framework - a task for which AMD's architectures are usually extremely well-suited.
But AMD seems to take the eyes of its DLSS-defusing moves, however, as AMD's Nish Neelalojanan, a Gaming division exec, talks about potential DLSS-like implementations across

"Some of the other broader available frameworks, like WindowsML and DirectML", and that these are "something we [AMD] are actively looking at optimizing…
At some of the previous shows we've shown some of the upscaling, some of the filters available with WindowsML, running really well with some of our Radeon cards."
So whether it's an actual image-quality philosophy, or just a competing technology's TTM (time to market) one, only AMD knows.

 

Improvements upon SMAA would be interesting, a return for SMAA 2x and 4x would be nice to see though MSAA is kinda disappearing in favor of TAA though temporal AA using data from previous frames also causes a bit of blur even if image stability is good and aliasing particularly of objects in motion is greatly lessened.

Hmm wonder if super sampling temporal anti-aliasing could be more developed to maybe counter it a bit though at a larger performance hit albeit less than full render super-sampling and having the entire image going through DSR/VSR or similar will be interesting to hear more about this and what else AMD (And NVIDIA.) are planning for 2019 and beyond.