AMD Radeon HD 8790M vs NVIDIA Quadro M5000M Comparison

NVIDIA Quadro M5000M and AMD Radeon HD 8790M come from different eras of mobile graphics, and the benchmark data shows a clear generational gap. The Quadro M5000M is a high-end professional mobile GPU built on Maxwell 2.0, while the Radeon HD 8790M is a mid-range part from AMD’s GCN 1.0 generation. Both are end-of-life MXM modules, but their performance profiles and feature sets diverge sharply in every measurable way.

Head-to-Head Benchmarks

The only two benchmarks that directly compare both GPUs are Geekbench OpenCL and Geekbench Vulkan, and the NVIDIA Quadro M5000M wins both by a massive margin. In Geekbench OpenCL, the Quadro M5000M scores 22920 against the Radeon HD 8790M’s 5017, a delta of 356.8%. That is not a marginal lead; it is a fourfold-plus difference in raw compute throughput. The Vulkan result is similarly lopsided: the Quadro M5000M posts 24875 versus 6365 for the Radeon, a 290.8% advantage. These are the only head-to-head tests recorded, and the data shows the Quadro M5000M winning both, with winsA equal to 2 and winsB equal to 0.

Looking at the broader benchmark suite, the Quadro M5000M also has more recorded data points. It scores 7062 in Passmark G3D, 2756 in Passmark GPU Compute, 476 in Passmark G2D, 119 in Passmark DirectX 9, 54 in Passmark DirectX 11, 35 in Passmark DirectX 10, and 29 in Passmark DirectX 12. Its average benchmark score is 6481, placing it at the 37th percentile of all GPUs. The Radeon HD 8790M, by contrast, only has the two Geekbench results recorded, and its average benchmark score is 5691, which puts it at the 33rd percentile. The difference in average score, 6481 versus 5691, is roughly 13.9% in favor of the NVIDIA part, though the Radeon’s average is based on far fewer tests.

The Radeon HD 8790M’s nearest rivals in the database include the Intel Iris Pro Graphics P6300 (5712, delta -0.4%), NVIDIA GeForce GTX 670MX (5721, delta -0.5%), and NVIDIA GeForce GTX 550 Ti (5731, delta -0.7%). Interestingly, the NVIDIA Quadro M500M appears as a rival with an average score of 5604, meaning the Radeon HD 8790M is 1.6% ahead of that particular Quadro M-class chip. For the Quadro M5000M, its nearest rivals are AMD Radeon Vega 10 Mobile (6476, delta 0.1%), NVIDIA GeForce GT 555M (6493, delta -0.2%), NVIDIA GeForce GTX 670M (6513, delta -0.5%), and Intel UHD Graphics P750 (6554, delta -1.1%). These deltas are all within roughly 1.1%, indicating that the Quadro M5000M sits in a tight performance cluster, whereas the Radeon HD 8790M is grouped with noticeably weaker parts.

Where Each One Wins

The NVIDIA Quadro M5000M wins everywhere that data exists. In compute workloads, its Geekbench OpenCL score of 22920 is over four times higher than the Radeon’s 5017, which points to advantages in general-purpose GPU tasks like rendering, simulation, and machine learning inference. The Vulkan score of 24875 versus 6365 suggests the Quadro M5000M also handles modern graphics APIs far better, which matters for games and applications that use Vulkan for lower overhead and better multi-threaded performance.

The Radeon HD 8790M has no recorded wins in any benchmark. That said, it does not necessarily fail in all scenarios. Its Passmark-style metrics are simply not present in the database, so a direct comparison in DirectX 9, 10, 11, or 12 is impossible. The only data available shows the Quadro M5000M ahead in both OpenCL and Vulkan. For legacy DirectX titles, the Quadro M5000M’s own Passmark scores (119 in DX9, 54 in DX11, 35 in DX10, 29 in DX12) give some reference, but without the Radeon’s numbers, no head-to-head verdict can be made there.

For use-case guidance, the Quadro M5000M is the clear choice for compute-heavy mobile workstations: 3D modeling, scientific computing, and any workload that leverages OpenCL. The Radeon HD 8790M, with only 691.2 GFLOPS of FP32 performance versus the Quadro’s 3.229 TFLOPS, is suited for lighter duties, but the benchmark data does not support any scenario where it beats the Quadro M5000M. If power draw is a concern, the Radeon may be easier to cool, but no TDP is recorded for it, so that remains qualitative.

Architecture Differences

The two GPUs come from different architectural generations and different design philosophies. The NVIDIA Quadro M5000M uses the GM204 chip, built on the Maxwell 2.0 architecture, manufactured by TSMC on a 28 nm process. It packs 5,200 million transistors into a 398 mm² die, yielding a transistor density of 13.1M per mm². The Radeon HD 8790M uses AMD’s Mars chip, based on GCN 1.0, also on TSMC’s 28 nm node, but with only 950 million transistors on a 77 mm² die, giving a density of 12.3M per mm². The Quadro M5000M is the larger, more complex chip by a wide margin.

Core counts differ substantially. The Quadro M5000M has 1536 shading units, 96 texture mapping units, and 64 ROPs. The Radeon HD 8790M has 384 shading units, 24 TMUs, and just 8 ROPs. That ROP count is particularly low, which explains the Radeon’s pixel rate of 7.200 GPixel/s versus the Quadro’s 67.26 GPixel/s. Texture rate is similarly lopsided: 100.9 GTexel/s for the Quadro M5000M versus 21.60 GTexel/s for the Radeon. FP32 compute is 3.229 TFLOPS for the NVIDIA part, compared to 691.2 GFLOPS for the AMD part, a factor of 4.7.

Memory configurations are also very different. The Quadro M5000M has 8 GB of GDDR5 on a 256-bit bus, with memory clocked at 1253 MHz (5 Gbps effective) and bandwidth of 160.4 GB/s. The Radeon HD 8790M has 2 GB of GDDR5 on a 128-bit bus, at 1000 MHz (4 Gbps effective), yielding 64.00 GB/s. The Quadro M5000M has four times the memory capacity and 2.5 times the bandwidth. This matters for large datasets and high-resolution textures.

API support differs as well. The Quadro M5000M supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The Radeon HD 8790M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Both are end-of-life, but the NVIDIA part has a newer feature level in both DirectX and Vulkan. The Quadro M5000M also has a higher base clock of 962 MHz and boost of 1051 MHz, versus the Radeon’s 850 MHz base and 900 MHz boost. The Radeon’s memory clock is lower as well, at 1000 MHz versus 1253 MHz.

FAQ

Q: Which GPU has higher compute performance in OpenCL?

A: The NVIDIA Quadro M5000M scores 22920 in Geekbench OpenCL, which is 356.8% higher than the AMD Radeon HD 8790M’s 5017.

Q: Is the Radeon HD 8790M competitive in Vulkan?

A: No. The Radeon HD 8790M scores 6365 in Geekbench Vulkan, while the Quadro M5000M scores 24875, a 290.8% difference in favor of NVIDIA.

Q: How does memory bandwidth compare between the two?

A: The Quadro M5000M has 160.4 GB/s of bandwidth from 8 GB of GDDR5 on a 256-bit bus. The Radeon HD 8790M has 64.00 GB/s from 2 GB of GDDR5 on a 128-bit bus.

Q: What are the transistor counts and die sizes?

A: The Quadro M5000M has 5,200 million transistors on a 398 mm² die. The Radeon HD 8790M has 950 million transistors on a 77 mm² die.

Q: Which GPU supports a newer Vulkan version?

A: The Quadro M5000M supports Vulkan 1.4, while the Radeon HD 8790M supports Vulkan 1.2.170.

Q: What is the average benchmark score for each GPU?

A: The Quadro M5000M has an average benchmark score of 6481, and the Radeon HD 8790M has an average of 5691. The Quadro M5000M also sits at the 37th percentile of all GPUs, versus the 33rd percentile for the Radeon.

Specification Differences

The two GPUs differ in nearly every specification. The Quadro M5000M uses the GM204 chip with Maxwell 2.0 architecture, while the Radeon HD 8790M uses the Mars chip with GCN 1.0. Both are 28 nm and made by TSMC, but the Quadro has 5,200 million transistors versus 950 million, and a die size of 398 mm² versus 77 mm². Transistor density is 13.1M per mm² for NVIDIA and 12.3M per mm² for AMD.

Clock speeds are higher on the Quadro M5000M: base 962 MHz and boost 1051 MHz, compared to 850 MHz base and 900 MHz boost for the Radeon. Memory is also faster on the NVIDIA part: 1253 MHz (5 Gbps effective) versus 1000 MHz (4 Gbps effective). Memory capacity is 8 GB versus 2 GB, bus width is 256-bit versus 128-bit, and bandwidth is 160.4 GB/s versus 64.00 GB/s.

Shading units are 1536 versus 384, TMUs are 96 versus 24, and ROPs are 64 versus 8. Pixel rate is 67.26 GPixel/s versus 7.200 GPixel/s, texture rate is 100.9 GTexel/s versus 21.60 GTexel/s, and FP32 is 3.229 TFLOPS versus 691.2 GFLOPS. The Quadro M5000M has a TDP of 100 W, while the Radeon HD 8790M has no TDP recorded. Bus interfaces differ too: MXM-B (3.0) for NVIDIA versus MXM-A (3.0) for AMD. DirectX support is 12 (12_1) for NVIDIA versus 12 (11_1) for AMD, and Vulkan support is 1.4 versus 1.2.170. Release dates are 2015-08-17 for the Quadro M5000M and 2013-03-31 for the Radeon HD 8790M.

The Verdict

The data is unambiguous. The NVIDIA Quadro M5000M outperforms the AMD Radeon HD 8790M in every recorded benchmark, with deltas of 356.8% in OpenCL and 290.8% in Vulkan. It has more memory, more bandwidth, more shaders, higher clock speeds, and a newer architecture. The Quadro M5000M is the pick for anyone needing a mobile workstation GPU for compute-intensive tasks, modern API support, or large memory buffers. The Radeon HD 8790M, with its smaller die and lower compute throughput, is only suitable for lighter workloads, and even then, the benchmark data does not show it winning anywhere. For a builder choosing between these two, the Quadro M5000M is the superior part across the board. The Radeon’s only consolation is that it came earlier and uses less silicon, but that does not translate into any performance advantage in the database.