AMD Radeon Vega 11

The AMD Radeon Vega 11 is an integrated graphics processor based on GCN 5.0, manufactured by GlobalFoundries on a 12 nm process. The Picasso chip contains 4,940 million transistors on a 210 mm² die, for a transistor density of 23.5 million per mm². It is an end-of-life IGP released on September 29, 2019, with a 15 W TDP, no power connectors, and no suggested PSU in the data. Its average benchmark score is 14,352, and it sits at the 55th percentile of all GPUs. The four closest listed rivals are separated from it by no more than the 0.5% delta recorded in the data.

Power and Cooling — TDP, PSU recommendation, connector requirements

At 15 W, the data places this part at a deliberately low power level. The slot width is listed as IGP, and the bus interface is also IGP, confirming that this is not a separate expansion card. Power connectors are recorded as None, and the suggested PSU field is null, so the data does not define a separate power supply requirement. The base clock is 300 MHz and the boost clock is 1400 MHz, with the memory clock marked System Shared. There is no separate game clock in the data, only the base and boost clocks.

Because the dimensions fields are all null, the listing contains no card length, height, or width. That is consistent with an integrated part rather than a discrete board. The 15 W TDP and IGP slot width mean the thermal and power delivery context is the host processor, not a standalone graphics card. The practical takeaway from the connector field is direct: no auxiliary power connector is needed.

Ray Tracing and Feature Set — RT/tensor cores, API support from facts

The generation field is Vega IGP (Picasso), and the architecture is GCN 5.0. The rtCores and tensorCores fields are both null, so the database does not assign any dedicated ray tracing cores or tensor cores to this part. API support is DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. The feature set also includes 704 shading units, 44 texture mapping units, and 8 ROPs.

Compute throughput is listed at 1.971 TFLOPS for FP32 and 3.942 TFLOPS for FP16, with the FP16 path running at a 2:1 ratio. Pixel throughput is 11.20 GPixel/s, and texture throughput is 61.60 GTexel/s. Display outputs are motherboard dependent, which means the actual video outputs come from the host board rather than from the GPU alone. The API list gives the Vega 11 modern driver-facing support across DirectX, OpenGL, and Vulkan, even though dedicated ray tracing hardware is not part of the specification.

Benchmark Performance — analyze scores vs rivals with exact % deltas

Benchmark results are available for three GeekBench workloads. Metal scores 17,392, OpenCL scores 13,392, and Vulkan scores 12,273. The average benchmark score is 14,352, and the percentile versus all GPUs is 55. Metal is the highest score among the three, Vulkan is the lowest, and OpenCL sits between them. That spread shows API-dependent behavior: the same hardware can rank differently depending on which compute or graphics interface is used.

Against the nearest rivals, the margins are very narrow. Intel Iris Xe MAX Graphics averages 14,315, with a deltaPct of 0.3. AMD Radeon RX Vega 11 averages 14,314, also with a deltaPct of 0.3. NVIDIA GeForce GTX 1660 SUPER averages 14,286, with a deltaPct of 0.5. AMD Radeon RX 5500 XT averages 14,389, with a deltaPct of -0.3. Read directly, the Vega 11 is 0.3% ahead of the Intel part, 0.3% ahead of the RX Vega 11 rival, 0.5% ahead of the GTX 1660 SUPER, and 0.3% behind the RX 5500 XT.

The negative delta for the RX 5500 XT makes it the only listed rival that leads the Vega 11. Even that lead is small. The data describes a tightly packed cluster of results rather than a clear performance hierarchy. The Vega 11’s 55th percentile position sits just above the database midpoint, reinforcing the idea that this is a middle-tier result in aggregate terms.

Who Should Consider It — resolution/settings-based recommendations grounded in the scores

Because this is an IGP with a 15 W TDP, no power connectors, and no discrete PSU specification, it belongs in systems that do not use a separate graphics card. The display outputs are motherboard dependent, so connectivity depends on the host board. The production status is end-of-life, and its predecessor is GCN 3.0 IGP while its successor is listed as Vega II IGP, which frames the Vega 11 as one step in a line of integrated GPUs rather than a forward-looking component.

The 55th percentile places the average score above half of the GPUs in the database. At the same time, all four nearest rivals are within 0.5%, so the Vega 11 does not stand apart from comparable hardware. Users with moderate graphical demands can use the aggregate score as a baseline for expectation setting. The data contains no per-game resolution or settings results, so specific high-resolution settings cannot be quoted from this listing. What the facts do support is a modest performance tier, with the main caveat being the memory subsystem described below.

Memory Subsystem — VRAM size/type, bus width, bandwidth and what it means for high resolutions

The memory size, type, and bus width are all listed as System Shared. The memory clock is also System Shared, and bandwidth is System Dependent. There is no dedicated VRAM capacity, no fixed bus width, and no fixed bandwidth number in the fact pack. This means the GPU relies on host system memory rather than on a private memory pool.

The fixed throughput limits are 11.20 GPixel/s and 61.60 GTexel/s, driven by 8 ROPs and 44 TMUs. For high resolutions, the absence of a fixed memory bandwidth is important. Because bandwidth is System Dependent, two systems using the same Vega 11 could behave differently depending on the host memory configuration. The data does not provide a single memory bandwidth figure to use for resolution scaling, so high-resolution expectations have to be platform dependent rather than strictly GPU dependent.

FAQ

Q: What architecture does the AMD Radeon Vega 11 use?

A: It uses GCN 5.0, with the Picasso chip built on GlobalFoundries' 12 nm process.

Q: Does it have dedicated ray tracing or tensor cores?

A: No. The data lists rtCores as null and tensorCores as null, so no dedicated ray tracing or tensor hardware is specified.

Q: What is the TDP and power connector requirement?

A: TDP is 15 W. Power connectors are None, and no suggested PSU is listed.

Q: How much VRAM does it have?

A: It has no dedicated VRAM. Memory size, type, and bus width are all System Shared, and bandwidth is System Dependent.

Q: What are the supported APIs?

A: DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3.

Q: What is its current production status?

A: It is end-of-life. The predecessor is GCN 3.0 IGP and the successor is Vega II IGP.

How It Compares

Intel Iris Xe MAX Graphics: The Intel part averages 14,315, and the deltaPct is 0.3. The Vega 11 is 0.3% ahead in the aggregate data, so the two effectively occupy the same performance position.

AMD Radeon RX 5500 XT: With an average score of 14,389 and a deltaPct of -0.3, the RX 5500 XT is the only listed rival ahead of the Vega 11. The lead is small, but the ordering puts this rival at the top of the group.

AMD Radeon RX Vega 11: The similarly named rival averages 14,314, with a deltaPct of 0.3. The Vega 11 is 0.3% ahead, meaning the close names are matched by close average scores.

NVIDIA GeForce GTX 1660 SUPER: The NVIDIA part averages 14,286, with a deltaPct of 0.5. The Vega 11 is 0.5% ahead, which is the largest positive delta in the rival list. Even so, the gap remains small.