AMD Radeon Vega 8 Mobile

AMD Radeon Vega 8 Mobile is an integrated GPU from AMD, using the Raven-M chip and built on the GCN 5.0 architecture. The die is manufactured by GlobalFoundries on a 14 nm process and contains 4,940 million transistors in a 210 mm² area. It belongs to the Vega IGP (Raven Ridge-M) generation, with 512 shading units, 32 TMUs, and 8 ROPs, running at a 300 MHz base clock and a 1101 MHz boost clock. The product is end-of-life; its release date is January 7, 2019, and its predecessor and successor are GCN 3.0 IGP and Navi II IGP.

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

The memory size, memory type, and bus width are all listed as System Shared. There is no fixed VRAM allocation and no fixed memory bus width. Memory bandwidth is System Dependent, so the actual throughput is set by the host platform’s memory subsystem rather than by the GPU alone. The memory clock is also System Shared.

For higher resolutions, this shared design means the GPU and the host CPU draw from the same memory channels. A system with a faster memory configuration will deliver more bandwidth to the GPU, while a constrained memory setup will hold the GPU back. This makes high-resolution performance highly platform-dependent: the same Vega 8 Mobile can behave differently depending on the system memory it is paired with.

The bus interface is IGP, and the display outputs are Portable Device Dependent. That reinforces the integrated nature of this GPU, since it does not use a discrete memory interface. The fixed graphics pipeline rates are 8.808 GPixel/s for pixel fill and 35.23 GTexel/s for texture fill, but these rates do not remove the dependency on shared system memory for real-world workloads.

How It Compares

The nearest rival data places the Vega 8 Mobile in a tight cluster of comparable GPUs.

Against the NVIDIA GeForce GTX 750, the two GPUs are tied. The Vega 8 Mobile has an average benchmark score of 7203, exactly matching the GTX 750’s average score of 7203, and the deltaPct is 0.

Against the NVIDIA GeForce GTX 680M, the gap is negligible. The GTX 680M has an average score of 7193 and the deltaPct is 0.1, putting the Vega 8 Mobile marginally ahead.

Against the NVIDIA Quadro 5000, the Vega 8 Mobile falls slightly behind. The Quadro 5000 has an average score of 7315, and the deltaPct of -1.5 shows that the Quadro 5000 is ahead by about 1.5%.

Against the NVIDIA T600, the Vega 8 Mobile has the edge. The T600 has an average score of 7068, and the deltaPct of 1.9 puts the Vega 8 Mobile ahead by about 1.9%.

Benchmark Performance

The Vega 8 Mobile scores 7435 in Geekbench OpenCL and 6970 in Geekbench Vulkan. Its average benchmark score is 7203. That average places it at the 39th percentile of all GPUs in the database.

The deltaPct values show that the nearest rivals are all within a few percent of each other. The 0% difference against the GTX 750 means the two GPUs are effectively interchangeable in this benchmark set. The 0.1% difference against the GTX 680M is a negligible margin. The 1.5% gap to the Quadro 5000 is small but visible in the data. The 1.9% lead over the T600 gives the Vega 8 Mobile a slightly better standing than that rival.

Looking at the two listed tests, the OpenCL result is higher than the average benchmark score, while the Vulkan result is lower than the average benchmark score. This indicates that API selection can change where the GPU lands relative to its own average. Rival average scores span from 7068 for the T600 to 7315 for the Quadro 5000, so the Vega 8 Mobile sits near the middle of its immediate competitive set.

FAQ

Q: Does this GPU have dedicated ray tracing cores?

A: The data does not list an RT core count. The rtCores field is null, so hardware ray tracing capacity is not quantified in this listing.

Q: Does it have tensor cores?

A: No tensor core count is provided. The tensorCores field is null, so AI acceleration hardware is not specified in the data.

Q: What graphics APIs are supported?

A: The Vega 8 Mobile supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3.

Q: How much dedicated VRAM does it have?

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

Q: What are the clock speeds?

A: The base clock is 300 MHz and the boost clock is 1101 MHz. The memory clock is listed as System Shared.

Q: What is the power connector requirement?

A: The power connectors field is None. The slot width is IGP, so the GPU is not designed to take discrete power connectors.

Ray Tracing and Feature Set

The Vega 8 Mobile is built on GCN 5.0 and uses a 14 nm process from GlobalFoundries, with the Raven-M chip. Its core configuration consists of 512 shading units, 32 TMUs, and 8 ROPs. These resources define its conventional rasterization pipeline.

The list includes no RT core count and no tensor core count. This means the data set does not attribute dedicated ray tracing or tensor processing hardware to this GPU. The feature set is therefore centered on standard shader and texturing work rather than dedicated hardware-accelerated ray tracing.

Compute throughput is listed as 1,127.4 GFLOPS for FP32 and 2.255 TFLOPS for FP16, with the FP16 result specified at a 2:1 ratio. That gives the GPU additional throughput on FP16 operations relative to FP32. API support includes DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3, covering a broad range of modern graphics workloads. The bus interface remains IGP, and the display outputs are Portable Device Dependent, so feature delivery depends on the host platform.

Power and Cooling

The TDP is 25 W, making this a low-power integrated GPU. The slot width is IGP, so it is not a discrete expansion card. Power connectors are listed as None, meaning there is no separate GPU power cable requirement in the data.

The suggested PSU field is not populated, so no power supply recommendation is provided in the data. Because this is an IGP, power delivery and cooling are handled by the host device rather than by a discrete board with its own connectors. The dimensions are not specified, which also points to a product that is integrated into a platform instead of shipped as an add-in card. The end-of-life status and portable-device-dependent display outputs further indicate that this GPU is tied to the life cycle of the systems it shipped in.