Compute Unit

RDNA 2 contains a significant increase in the number of Compute Units (CUs) with a maximum of 80, a doubling from the maximum of 40 in the Radeon RX 5700 XT.^[1] Each Compute Unit contains 64 shader cores.^[7] CUs are organized into groups of two named Work Group Processors with 32 KB of shared L0 cache per WGP. Each CU contains two sets of an SIMD32 vector unit, an SISD scalar unit, textures units, and a stack of various caches.^[8] New low precision data types like INT4 and INT8 are new supported data types for RDNA 2 CUs.^[8]

The RDNA 2 graphics pipeline has been reconfigured and reordered for greater performance-per-watt and more efficient rendering by moving the caches closer to the shader engines. A new mesh shaders model allows shader rendering to be done in parallel using smaller batches of primitives called "meshlets". As a result, the mesh shaders feature enables greater control of the GPU geometry pipeline.^[9]

Ray tracing

Real-time hardware accelerated ray tracing is a new feature for RDNA 2 which is handled by a dedicated ray accelerator inside each CU.^[10] Ray tracing on RDNA 2 relies on the more open DirectX Raytracing protocol rather than the Nvidia RTX protocol.^[11]

In February 2023, it was reported that driver updates had boosted ray tracing performance by up to 40% using DirectX Raytracing.^[12]

Clock speeds

With RDNA 2 using the same 7 nm node as RDNA, AMD claims that RDNA 2 achieves a 30% frequency increase over its predecessor while using the same power.^[13]

Cache and memory subsystem

In addition to the traditional L1 and L2 caches that GPUs possess, RDNA 2 adds a new global L3 cache that AMD calls "Infinity Cache".^[14] This was done to avoid the use of a wider memory bus while still being able to maintain the same data bandwidth. Product technology architect Sam Naffziger said that, without Infinity Cache, "We were looking at the daunting prospect of having to put a 512-bit interface and all the power, area and expense associated with that".^[15] Using a wider memory bus requires more power which is in conflict with AMD's increased performance-per-watt goals for RDNA 2. AMD engineers ran tests comparing RDNA 2 silicon featuring a large on-die cache and with wider memory buses. They discovered that having such a cache would aid in the re-use of temporal and spatial data when the GPU is rendering a complex image. It is beneficial for the GPU's compute units to have fast access to a physically close cache rather than searching for data in video memory. AMD claims that RDNA 2's 128 MB of on-die Infinity Cache "dramatically reduces latency and power consumption".^[16] The GPU having access to a large L2 or L3 cache allows it to more quickly access necessary data compared to accessing VRAM or system RAM. The Infinity Cache is made up of two sets of 64 MB cache that can run on its own clock rate independent from the GPU cores. The Infinity Cache has a peak internal transfer bandwidth of 1986.6 GB/s and results in less reliance being placed on the GPU's GDDR6 memory controllers.^[8] Each Shader Engine now has two sets of L1 caches. The large cache of RDNA 2 GPUs give them a higher overall memory bandwidth compared to Nvidia's GeForce RTX 30 series GPUs.

Power efficiency

AMD claims that RDNA 2 achieves up to a 54% increase in performance-per-watt over the first RDNA microarchitecture.^[16] 21% of that 54% improvement is attributed to performance-per-clock enhancements, in part due to the addition of Infinity Cache.^[17]

Media engine

RDNA 2 uses the VCN 3.0, VCN 3.1, and VCN 3.1.2 video decoding blocks in its media engine.^[18][19] It adds support for AV1 decoding at up to 8K resolution, though AV1 hardware encoding support would not come until RDNA 3 in 2022.^[20][21] However, the low-end Navi 24 die and iGPUs based on RDNA 2.0 do not contain any media encoders and cannot decode AV1 as a result.^[18]

Desktop

Mobile

More information Model (Code name), Release date ...

Model (Code name)	Release date	Architecture & fab	Transistors & die size	Core		Fillrate[lower-alpha 1][lower-alpha 2][lower-alpha 3]		Processing power[lower-alpha 1][lower-alpha 4] (GFLOPS)			Infinity Cache	Memory				HW Decoder			HW Encoder			TDP	Bus interface
				Config[lower-alpha 5]	Clock[lower-alpha 1] (MHz)	Texture (GT/s)	Pixel (GP/s)	Half	Single	Double		Size	Bandwidth (GB/s)	Bus type & width	Clock (MT/s)	AV1	H265	4K H264	AV1	H265	4K H264
Radeon RX 6300M (Navi 24)[lower-alpha 6][64]	Jan 4, 2022	RDNA 2 TSMC N6	5.4×109 107 mm2	768:64:32:12 12 CU	1512	97.76	48.38	6,270	3,130	195.6	8 MB	2 GB	64	GDDR6 32-bit	16000							25 W	PCIe 4.0 ×4
Radeon RX 6450M (Navi 24)[65]	Jan 4, 2023				2220	118.10	71.04	7,600	3,780	236.3	16 MB	4 GB	128	GDDR6 64-bit								50 W
Radeon RX 6550S (Navi 24)[66]				1024:64:32:16 16 CU	2170	154.20	69.44	9,900	4,900	306.3
Radeon RX 6500M (Navi 24)[lower-alpha 6][67]	Jan 4, 2022				2191	155.7	70.11	9,970	4,980	311.2
Radeon RX 6550M (Navi 24)[68]	Jan 4, 2023				2560	182.10	81.92	11,600	5,800	362.5			144		18000							80 W
Radeon RX 6600S (Navi 23)[69]	Jan 4, 2022	RDNA 2 TSMC N7	11.06×109 237 mm2	1792:128:64:28 28 CU	1881	244.2	120.3	15,630	7,810	448.1	32 MB		224	GDDR6 128-bit	14000								PCIe 4.0 ×8
Radeon RX 6700S (Navi 23)[70]					1890	247.5	120.9	15,840	7,920	495.0		8 GB
Radeon RX 6600M (Navi 23)[71]	May 31, 2021				2177	274.2	139.3	17,550	7,800	487.5												100 W
Radeon RX 6650M (Navi 23)[72]	Jan 4, 2022				2222	276.6	139.3	17,700	8,850	553.1			256		16000							120 W
Radeon RX 6800S (Navi 23)[73]				2048:128:64:32 32 CU	1975	288.0	134.4	18,430	9,220	576.5												100 W
Radeon RX 6650M XT (Navi 23)[74]					2162	311.5	142.2	19,940	9,970	623.1												120 W
Radeon RX 6700M (Navi 22)[75]	May 31, 2021		17.2×109 335 mm2	2304:144:64:36 36 CU	2300	331.4	147.2	21,209	10,605	662.1	80 MB	10 GB	320	GDDR6 160-bit								135 W	PCIe 4.0 ×16
Radeon RX 6800M (Navi 22)[76]				2560:160:64:40 40 CU	2300	368.0	147.2	23,550	11,780	736.2	96 MB	12 GB	384	GDDR6 192-bit								145+ W
Radeon RX 6850M XT (Navi 22)[77]	Jan 4, 2022				2580	415.6	157.6	26,430	13,209	825.6			432		18000							165 W

Close

Workstation

Integrated graphics processing units (iGPUs)

Consoles