AMD Radeon Vega 8 Mobile vs NVIDIA GeForce GTX 675M Comparison

AMD Radeon Vega 8 Mobile and NVIDIA GeForce GTX 675M are both end-of-life mobile graphics solutions, but they represent two very different eras of GPU design. The data shows a clear, albeit narrow, victory for the AMD part in the single available benchmark, with the Vega 8 Mobile taking the Geekbench OpenCL test. However, the specifications reveal a fundamental architectural clash between a modern integrated processor and a legacy discrete mobile GPU, making the choice highly dependent on the specific use case.

Head-to-Head Benchmarks

The only direct benchmark comparison available is the Geekbench OpenCL test, and the results are closely contested. In this test, the AMD Radeon Vega 8 Mobile scores 7,435 points, while the NVIDIA GeForce GTX 675M scores 6,946 points. This gives AMD a 7% lead over NVIDIA in raw compute performance as measured by this workload. This is a significant result because it demonstrates that a modern IGP with 512 shading units can outperform a dedicated discrete GPU from 2012, which relies on a larger die but older architecture.

Digging into the nearest rivals provides further context for these scores. The AMD Radeon Vega 8 Mobile’s average benchmark score of 7,203 places it in the 39th percentile of all GPUs. Its closest rival is the NVIDIA GeForce GTX 750, which has an average score of 7,222, putting the Vega 8 Mobile just 0.3% behind it. This places the AMD IGP in the same performance bracket as a low-end discrete desktop card from NVIDIA, which is a notable achievement for an integrated part. The data also shows the Vega 8 Mobile is 0.4% ahead of the GTX 560 SE and 0.5% ahead of the Intel Iris Pro Graphics P580, confirming it sits comfortably in the mid-range of legacy hardware.

On the NVIDIA side, the GeForce GTX 675M’s average score of 6,946 also places it in the 39th percentile. Its closest rival is the AMD Radeon R5 M240, which scores 6,975, meaning the GTX 675M is 0.4% behind that part. It is also 1.1% behind the NVIDIA GeForce GTX 680M and 1.3% behind the NVIDIA T600. Interestingly, the GTX 675M is 1.7% ahead of the AMD FirePro M5100. While the percentile ranking is identical for both GPUs, the actual average score gap is 257 points in favor of the AMD part, reinforcing the 7% delta seen in the head-to-head test.

The single benchmark result is telling. It suggests that in pure compute throughput, the newer AMD architecture has an edge. The win for the Vega 8 Mobile is not a landslide, but it is a definitive one. The data indicates that the AMD Radeon Vega 8 Mobile is the faster GPU in this specific synthetic workload, overcoming its memory bandwidth limitations through sheer shader throughput and architectural efficiency.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon Vega 8 Mobile has a higher average benchmark score of 7,203, compared to the NVIDIA GeForce GTX 675M’s average score of 6,946.

Q: How do the two GPUs compare in the Geekbench OpenCL test?

A: The AMD Radeon Vega 8 Mobile wins the Geekbench OpenCL test with a score of 7,435, which is 7% higher than the NVIDIA GeForce GTX 675M’s score of 6,946.

Q: What is the process node difference between the two chips?

A: The AMD Radeon Vega 8 Mobile is manufactured on a 14 nm process at GlobalFoundries, while the NVIDIA GeForce GTX 675M uses a 40 nm process at TSMC.

Q: What type of memory does each GPU use?

A: The AMD Radeon Vega 8 Mobile uses System Shared memory, meaning its memory size, type, and bus width are all system-dependent. The NVIDIA GeForce GTX 675M uses 2 GB of dedicated GDDR5 memory on a 256-bit bus.

Q: What is the TDP of each GPU?

A: The AMD Radeon Vega 8 Mobile has a TDP of 25 W, while the NVIDIA GeForce GTX 675M has a TDP of 100 W.

Q: Which GPU supports Vulkan?

A: The AMD Radeon Vega 8 Mobile supports Vulkan 1.3, while the NVIDIA GeForce GTX 675M does not list Vulkan support in its specifications.

Architecture Differences

The architectural divide between these two GPUs is stark, reflecting a seven-year gap in design philosophy. The AMD Radeon Vega 8 Mobile is built on the GCN 5.0 architecture, using a chip codenamed Raven-M, and is part of the Vega IGP generation (Raven Ridge-M). This is a modern, power-efficient design fabricated on a 14 nm process at GlobalFoundries. The NVIDIA GeForce GTX 675M, by contrast, uses the aging Fermi 2.0 architecture, built around the GF114 chip and belonging to the GeForce 600M generation. It was manufactured on a 40 nm process at TSMC.

The transistor counts tell a compelling story of density. The AMD chip packs 4,940 million transistors into a 210 mm² die, resulting in a transistor density of 23.5 million transistors per square millimeter. The NVIDIA chip has 1,950 million transistors on a much larger 332 mm² die, yielding a density of only 5.9 million transistors per square millimeter. This massive density advantage is a direct result of the newer manufacturing process and modern design techniques, allowing AMD to fit more than twice the transistors onto a smaller physical footprint.

The processing core configurations also differ significantly. The AMD Radeon Vega 8 Mobile features 512 shading units, 32 texture mapping units (TMUs), and 8 render output units (ROPs). The NVIDIA GeForce GTX 675M, in contrast, has 384 shading units, 64 TMUs, and 32 ROPs. While the AMD part has more shaders, the NVIDIA part has double the TMUs and four times the ROPs. This suggests the NVIDIA GPU was designed to handle texture-heavy and rasterization-heavy workloads more efficiently, while the AMD GPU focuses more on raw shader compute.

The memory architecture is another fundamental difference. The AMD Radeon Vega 8 Mobile relies on System Shared memory, making its bandwidth and capacity entirely dependent on the host system's RAM. The NVIDIA GeForce GTX 675M, however, has a dedicated 2 GB pool of GDDR5 memory on a 256-bit bus, providing a fixed bandwidth of 96.00 GB/s. This is a critical distinction for gaming, as dedicated VRAM offers predictable performance, whereas the AMD IGP's performance can vary based on system configuration.

Specification Differences

The specification sheets for these two GPUs highlight the key differences in their design targets. The most obvious is the process node: 14 nm for AMD versus 40 nm for NVIDIA. This leads to a massive difference in transistor density, with the AMD chip having 23.5 million transistors per square millimeter versus NVIDIA's 5.9 million.

The clocks are also specified differently. The AMD Radeon Vega 8 Mobile has a base clock of 300 MHz and a boost clock of 1,101 MHz. The NVIDIA GeForce GTX 675M has no listed base or boost clocks, only a memory clock of 750 MHz, which translates to 3 Gbps effective. The memory configuration is a major differentiator: AMD uses System Shared memory, while NVIDIA uses 2 GB of GDDR5 with a 256-bit bus and 96.00 GB/s bandwidth.

The compute unit counts differ as well. AMD has 512 shading units, 32 TMUs, and 8 ROPs. NVIDIA has 384 shading units, 64 TMUs, and 32 ROPs. The pixel rate and texture rate reflect these differences: AMD achieves 8.808 GPixel/s and 35.23 GTexel/s, while NVIDIA achieves 9.920 GPixel/s and 39.68 GTexel/s. The FP32 performance is higher on the AMD part at 1,127.4 GFLOPS versus 952.3 GFLOPS for NVIDIA. AMD also lists FP16 performance of 2.255 TFLOPS, while NVIDIA does not list any FP16 capability.

The power and physical specifications are also notably different. The AMD Radeon Vega 8 Mobile has a TDP of just 25 W and is an IGP with no power connectors, fitting into a "Portable Device Dependent" slot. The NVIDIA GeForce GTX 675M has a much higher TDP of 100 W, is an MXM Module using an MXM-B (3.0) interface, and also has no power connectors. The API support differs, with AMD supporting DirectX 12 (12_1) and Vulkan 1.3, while NVIDIA supports DirectX 12 (11_0) and has no Vulkan support listed.

Where Each One Wins

The AMD Radeon Vega 8 Mobile wins in the compute benchmark and in architectural efficiency. Its 7% lead in Geekbench OpenCL, coupled with its higher FP32 throughput of 1,127.4 GFLOPS, makes it the stronger choice for compute-heavy applications like OpenCL-accelerated workloads. Its modern GCN 5.0 architecture, 14 nm process, and support for Vulkan 1.3 make it a more future-proof platform for modern APIs. The lower TDP of 25 W also makes it the clear winner for power-constrained environments, such as thin-and-light laptops where battery life and thermal headroom are at a premium. Its high transistor density of 23.5M / mm² shows that it is a marvel of modern engineering, delivering competitive performance in a compact IGP form factor.

The NVIDIA GeForce GTX 675M wins in specific hardware capabilities that favor traditional gaming workloads. Despite having fewer shading units, it has double the TMUs (64 vs 32) and four times the ROPs (32 vs 8). This gives it a higher pixel rate of 9.920 GPixel/s and a higher texture rate of 39.68 GTexel/s compared to the AMD part. This suggests that in fill-rate-limited scenarios, the NVIDIA GPU could pull ahead. Its dedicated 2 GB of GDDR5 memory on a 256-bit bus with 96.00 GB/s of bandwidth is a significant advantage over system-shared memory, ensuring that textures and frame buffers are always local and fast. This makes it the more reliable choice for gaming at higher resolutions and detail settings where VRAM capacity and bandwidth are critical.

The Verdict

The benchmark data is clear: the AMD Radeon Vega 8 Mobile is the more powerful GPU in synthetic compute tests, beating the NVIDIA GeForce GTX 675M by 7% in Geekbench OpenCL. It achieves this with a fraction of the power draw, using only 25 W versus 100 W. For any user prioritizing raw compute performance, modern API support, or power efficiency, the AMD Radeon Vega 8 Mobile is the definitive choice.

However, the NVIDIA GeForce GTX 675M still holds a niche for users who require the characteristics of a dedicated discrete GPU. Its 2 GB of dedicated GDDR5 memory and 256-bit memory bus provide a level of performance consistency that an IGP cannot match, as the AMD part’s performance is "System Dependent". Its superior fill rates, stemming from more TMUs and ROPs, could offer advantages in specific gaming scenarios that are not captured by the single compute benchmark. Ultimately, the choice depends on whether the user values modern compute efficiency or legacy discrete memory bandwidth. For most modern workloads, the AMD Radeon Vega 8 Mobile is the better product. For older gaming applications that rely on dedicated VRAM, the GTX 675M may still be a viable option.