⋮    ⋮  

AMD 7nm “Super Secret” Navi GPU Spotted In Driver, 2H 2018 Launch Expected

Author Photo
Dec 18, 2017
191Shares
Submit

AMD’s upcoming next generation 7nm based graphics architecture codenamed “Navi” has reportedly been spotted in Linux driver code. The all new GPU architecture is officially slated to debut next year, with all whispers indicating a debut in the latter half of the year.

Navi was peculiarly hidden behind a “Super Secret” ID. No, really, that’s what AMD’s driver whiz’s have listed it under. The code reads:

amd-radeon-software-adrenalin-edition-2Related AMD Radeon Software Adrenalin 2019 Edition’s 25+ New Features Leaked – Out Dec 13

“new_chip.gfx10.mmSUPER_SECRET.enable[0:0]”

The Vega architecture is GFX9. Navi is GFX10, which is what gave this new “Super Secret” chip away.

AMD Preps “Super Secret” Navi Chip


Traditionally, the Linux driver code would be updated using the names of upcoming graphics architectures and that’s usually how the hardware leak scene becomes aware of impending new GPUs.  This time however the RTG software team’s use of “Super Secret” appears to be somewhat of a humorous attempt to keep the code name of this particular chip under wraps.

That’s right, we’re not talking about the entire Navi architecture, this code is referring to one specific Navi chip. Seeing how the Linux code has to be updated months in advance of Navi’s arrival, they would’ve had no choice but to add it at some point. Had the specific chip code name been included we would’ve had ways to track it across a bunch of online databases it would appear in the future. However, I guess we’ll take what we can get, and that’s all we are going to get right now.

amd-radeon-feature-wccftechRelated Rumor: AMD Navi RX 3080 to Launch Mid 2019 & Take On RTX 2080

One Scalable Navi GPU To Rule Them All – The Legend Of The Wonder Glue

There’s been talk in the industry that Navi could very well be AMD’s first attempt at “gluing” several small, high yielding 7nm GPU dies to make bigger and more powerful graphics chips. This would work similarly to what we’ve already seen the company do using Infinity Fabric to link 4 Zeppelin dies to create 32 core EPYC processors and 2 Zeppelin dies to create the ever popular Threadripper chips.

Threadripper MCM, Image credit : Der8auer

Leveraging the same Infinity Fabric advantage on the GPU side would allow the company to launch Navi sooner, improve yields, clock speeds and reduce its overall cost structure. We’ve seen some evidence that NVIDIA is also pursuing similar multi-chip-module technologies and are currently conducting feasibility studies. Although this would take several generations to materialize from a proof of concept to an actual product. AMD, on the other hand, has already launched products using the technology.

The next step would be to transfer that expertise to the GPU side of the company, something that we believe the company has started to work on with Vega, which is the first GPU architecture to support the “wonder glue” technology, Infinity Fabric. We’ve already explored how all of this would work in a report published by our own Usman Pirzada back in October that I’d highly recommend you check out.

Navi – Scalability & Next Generation Memory ( GDDR6 )


This would explain AMD’s use of the term “scalability” as a brand new feature that’s due with Navi. The fact that Infinity Fabric would enable it to utilize a single Navi GPU die that effectively scales and can be used to cover all markets top to bottom.

On the memory side, “nextgen” could refer to either HBM3 or GDDR6. Whether some Navi parts will make use of HBM3 while others GDDR6 is subject to speculation at this point. What’s not up for conjecture however is the fact that we know, thanks to one credible leak, that the company has been working on a GDDR6 memory interface that we have very little doubt will make it into Navi.

Navi is said to land in the second half of 2018, with some talk about a late summer debut. Although with 7nm being such a wild card, nothing is really set in stone. NVIDIA’s upcoming 12nm based Volta based GeForce lineup is setting up to land in the first half of next year. We also have second generation Ryzen Pinnacle Ridge parts coming out early next year and Intel’s impending 8 core 9700K in the summer. All in all, 2018 is definitely shaping up to be one enormously exciting year for hardware.

AMD Radeon Graphics Architectures

WCCFTech Year Consumer
Product
GPU Process Transistors In Billions Performance Memory Bandwidth
Southern Islands 2012 HD 7970 Ghz Tahiti 28nm 4.3 4.1 TFLOPS 3GB GDDR5 264GB/s
Volcanic Islands 2013 R9 290X Hawaii 28nm 6.2 5.6 TFLOPS 4GB GDDR5 320GB/s
Caribbean Islands 2015 R9 Fury X Fiji 28nm 8.9 8.6 TFLOPS 4GB HBM1 512GB/s
POLARIS 2016 RX 480 Polaris 10 14nm 5.7 5.2 TFLOPS 8GB GDDR5 256 GB/s
VEGA 2017 RX Vega 64 Vega 10 14nm 12.5 13 TFLOPS 8GB HBM2 480GB/s
VEGA Refresh
VEGA 2
2018

2018
TBA

TBA
TBA

Vega 20
12nm

7nm
12.5

TBA
TBA

TBA
TBA

32GB HBM2
TBA

1 TB/s
NAVI 2019 TBA Navi 10 7nm TBA TBA Nextgen Memory TBA
Wccftech GeForce 10 Series Tesla V100 GeForce 20/11 Series AMD 400 Series AMD 500 Series AMD 600 Series
Architecture Pascal Volta Turing Polaris Polaris / Vega Navi
Process Node TSMC 16nm FF TSMC 12nm FF TSMC 12nm FF Samsung 14nm LPP Samsung 14nm LPP TSMC 7nm FinFET
Memory GDDR5/X HBM2 GDDR6 GDDR5 GDDR5 / HBM2 GDDR6/HBM2
Year 2016 2017 2018 2016 2017 2019
GPUs GP102, GP104, GP106, GP107 GV110 TU102, TU104, TU106, TU107 Polaris 10, Polaris 11 Polaris 20, Polaris 21, Vega 10, Vega 11 Navi 10, Navi 11
(Navi Chiplet?)

AMD Ryzen CPU Roadmap

WCCFTech AMD Summit Ridge AMD Pinnacle Ridge
Product Family Ryzen 1000 Series Ryzen 2000 Series
Process Node 14nm 12nm
CPU Cores Up to 8 Up to 8
GPU Architecture N/A N/A
TDP 65W-95W 65W-95W
Socket AM4 AM4
Memory Support DDR4 DDR4
Launch Q1 2017 Q1 2018

Intel's 9th Generation Unlocked SKUs Specs:

Processor Name Intel Core i5-8600K Intel Core i5-9600K Intel Core i7-8700K Intel Core i7-9700K
Core Family Coffee Lake 8th Gen Coffee Lake Refresh 9th Gen? Coffee Lake 8th Gen Coffee Lake Refresh 9th Gen?
Process 14nm++ TBA 14nm++ TBA
Cores / Threads 6/6 6/12 6/12 8/16
Clock Speeds (Base) 3.6 GHz TBA 3.7 GHz TBA
Clock Speeds (Boost) 4.3 GHz TBA 4.7 GHz TBA
TDP 95W 92W 95W TBA
L3 Cache 9 MB 12 MB 12 MB 16 MB
Price $257 US TBA $359 US TBA
Launch 2017 2018 2017 2018

 

Source: Via ComputerBase.de

Submit