GPU microarchitecture by AMD
RDNA 3 is a GPU microarchitecture designed by AMD, released with the Radeon RX 7000 series on December 13, 2022.
Background
On June 9, 2022, AMD held their Financial Analyst Day where they presented a client GPU roadmap which contained mention of RDNA 3 coming in 2022 and RDNA 4 coming in 2024.[1] AMD announced to investors their intention to achieve a performance-per-watt uplift of over 50% with RDNA 3 and that the upcoming architecture would be built using chiplet packaging on a 5 nm process.[2]
A sneak preview for RDNA 3 was included towards the end of AMD's Ryzen 7000 unveiling event on August 29, 2022. The preview included RDNA 3 running gameplay of Lies of P, AMD CEO Lisa Su confirming that a chiplet design would be used, and a partial look at AMD's reference design for an RDNA 3 GPU.[3]
Full details for the RDNA 3 architecture were unveiled on November 3, 2022 at an event in Las Vegas.[4]
Architectural details
Chiplet packaging
For the first time ever in a consumer GPU, RDNA 3 utilizes modular chiplets. AMD previously had great success with its use of chiplets in its Ryzen desktop and Epyc server processors.[5]
The decision to move to a chiplet-based GPU microarchitecture was led by AMD Senior Vice President Sam Naffziger who had also lead the chiplet initiative with Ryzen and Epyc.[6] The development of RDNA 3's chiplet architecture began towards the end of 2017 with Naffziger leading the AMD graphics team in the effort.[7]
Memory Cache Dies (MCDs)
With a respective 2.05 billion transistors, each Memory Cache Die (MCD) contains large blocks of L3 cache and two physical 32-bit GDDR6 memory interfaces for a combined 64-bit interface per MCD.[8] The Radeon RX 7900 XTX has a 384-bit memory bus through the use of six MCDs while the RX 7900 XT has a 320-bit bus due to its five MCDs.
Chiplet interconnects
The chiplet interconnects have a bandwidth of 5.3 TB/s.[9]
Process node
According to Naffziger, cache and SRAM do not scale as linearly as logic does on advanced nodes like N5 in terms of density and power consumption so they can instead be fabricated on the cheaper, more mature N6 node. The use of smaller chiplet dies rather than one large monolithic die is beneficial for maximizing wafer yields as more dies can be fitted onto a single wafer.[7]
Compute Units
RDNA 3 icludes improved dual-issue shader ALUs with the ability to execute two instructions per cycle. It can contain up to 96 graphics Compute Units that can provide up to 61 TFLOPS of compute.[10]
RDNA 3 has dedicated AI acceleration with Wave MMA (matrix multiply-accumulate) instructions,[11] which can improve AI-based performance by 2.7x and also benefits ray tracing instructions, similar to Nvidia's Tensor cores.[10]
Ray tracing
Each RDNA 3 Compute Unit contains one ray tracing accelerator. The overall number of ray tracing accelerators is increased due to the higher number of Compute Units, though the number of ray tracing accelerators per Compute Unit has not increased over RDNA 2.
Clock speeds
RDNA 3 was designed to support high clock speeds. On RDNA 3, clock speeds have been decoupled with the front end operating at a 2.5 GHz frequency while the shaders operate at 2.3 GHz. The shaders operating at a lower clock speed gives up to 25% power savings according to AMD and RDNA 3's shader clock speed is still 15% faster than RDNA 2.[12]
Cache and memory subsystem
RDNA 3 GPUs use GDDR6 memory rather than faster GDDR6X due to the latter's increased power consumption.
16 MB Infinity Cache is included on each MCD. Theoretically, additional L3 cache could be added to the MCDs via AMD's 3D V-Cache die stacking technology as the MCDs contain unused TSV connection points.[13][14]
Power efficiency
AMD claims that RDNA 3 achieves a 54% increase in performance-per-watt which is in line with their previous claims of 50% performance-per-watt increases for both RDNA and RDNA 2.
Media engine
RDNA 3 is the first RDNA architecture to have a dedicated media engine. It is built into the GCD and is based on VCN 4.0 encoding and decoding core.[15] AMD's AMF AV1 encoder is comparable in quality to Nvidia's NVENC AV1 encoder but can handle a higher number of simultaneous encoding streams compared to the limit of 3 on the GeForce RTX 40 series.[16]
Supported encoding frame rates (FPS) per resolution and video coding format[17]
| Resolution
|
H.264
|
H.265
|
AV1
|
| 1080p
|
|
|
|
| 1440p
|
|
|
|
| 4K
|
180
|
180
|
240
|
| 8K
|
48
|
48
|
60
|
Display engine
RDNA 3 GPUs feature a new display engine called the "Radiance Display Engine". AMD touted its support for DisplayPort 2.1 UHBR 13.5, delivering up to 54 Gbit/s bandwidth for high refresh rates at 4K and 8K resolutions.[18] DisplayPort 2.1 can support 4K at 480 Hz and 8K at 165 Hz with Display Stream Compression (DSC). The previous DisplayPort 1.4 standard with DSC was limited to 4K at 240 Hz and 8K at 60 Hz.
