AMD Radeon HD 8690M vs NVIDIA Quadro M5000M Comparison

NVIDIA Quadro M5000M and AMD Radeon HD 8690M are two mobile GPUs from different eras and market segments. The data shows a single head-to-head benchmark result: Geekbench OpenCL. In that test, the Quadro M5000M scores 22,920 points, while the Radeon HD 8690M manages 6,137 points. That is a decisive 273.5% advantage for the NVIDIA part. The Quadro M5000M also holds a higher average benchmark score of 6,481 compared to the Radeon’s 6,137, and it sits at the 37th percentile of all GPUs versus the 36th percentile for the AMD card. These figures establish a clear overall performance hierarchy, but the details of how and why the gap exists require a closer look at the architectural and specification differences.

Head-to-Head Benchmarks

The only directly comparable benchmark result in the data is Geekbench OpenCL. The NVIDIA Quadro M5000M achieves a score of 22,920, which is 273.5% higher than the AMD Radeon HD 8690M’s 6,137. This is not a marginal win; it is a landslide. In raw compute terms, the Quadro M5000M delivers 3.229 TFLOPS of FP32 throughput, while the Radeon HD 8690M provides 624.0 GFLOPS. That translates to a roughly five-fold difference in raw floating-point capability, which is consistent with the massive OpenCL score gap. The Quadro’s texture rate of 100.9 GTexel/s dwarfs the Radeon’s 19.50 GTexel/s, and the pixel rate of 67.26 GPixel/s versus 7.800 GPixel/s reinforces the same story. Every throughput metric in the data points in the same direction.

The Quadro M5000M also wins on memory bandwidth by a wide margin. It has 160.4 GB/s of bandwidth from an 8 GB GDDR5 frame buffer on a 256-bit bus. The Radeon HD 8690M has only 32.00 GB/s from 2 GB GDDR5 on a 64-bit bus. That is a 401.25% bandwidth advantage for the NVIDIA part. Memory bandwidth is often a bottleneck in compute workloads, so this gap alone would explain a substantial portion of the OpenCL delta. The Quadro’s memory clock is also higher at 1253 MHz (5 Gbps effective) versus the Radeon’s 1000 MHz (4 Gbps effective), further widening the effective throughput difference.

Neither card has any win in the other direction. The data shows winsA equal to 1 and winsB equal to 0. There is no benchmark result where the AMD card comes out ahead. Even in the Passmark suite, where the Quadro M5000M has scores for DirectX 9 (119), DirectX 10 (35), DirectX 11 (54), DirectX 12 (29), G2D (476), G3D (7062), and GPU compute (2756), the Radeon HD 8690M has no comparable entries in the data. The only shared test is Geekbench OpenCL, and the NVIDIA card wins it by a factor of 3.7. The average benchmark scores tell a similar tale: 6,481 for the Quadro versus 6,137 for the Radeon, a 5.6% advantage for NVIDIA in overall aggregate performance.

The Verdict

From the data alone, the NVIDIA Quadro M5000M is the superior product for any workload that relies on raw compute throughput, memory bandwidth, or pixel/texture processing. The 273.5% Geekbench OpenCL lead is the single most decisive piece of evidence. If a user is choosing between these two for tasks like GPU-accelerated rendering, scientific simulation, or any OpenCL-heavy application, the Quadro M5000M is the clear pick. Its 8 GB memory capacity is four times larger than the Radeon’s 2 GB, which matters for holding larger datasets in VRAM. The 256-bit bus versus the 64-bit bus also means the NVIDIA card will sustain high memory throughput under load, whereas the AMD card will saturate its narrow bus far more quickly.

The AMD Radeon HD 8690M is not without a role, however. Its average benchmark score of 6,137 places it just 0.3% behind the Intel Iris Pro Graphics 6200 and 1% behind the NVIDIA RTX A400 and GeForce MX230, according to nearest rival data. The Quadro M5000M, by contrast, is within 0.1% of the AMD Radeon Vega 10 Mobile and 0.2% of the GeForce GT 555M. This suggests the Quadro is competing in a slightly higher performance tier, even if its percentile ranking is only one point higher. For users who need a mobile GPU for basic display output, light 2D work, or legacy applications that do not stress compute, the Radeon HD 8690M is a functional option. It also consumes no stated TDP in the data, whereas the Quadro is rated at 100 W, which may be relevant for thermal-constrained laptops.

The verdict is straightforward: the Quadro M5000M wins on every measurable benchmark and specification that matters for performance. The Radeon HD 8690M only makes sense if the workload is trivial and the lower power envelope is a priority, but the data does not quantify that power difference. The Quadro’s 100 W TDP is the only power figure available, and the Radeon has no TDP listed. For anyone choosing based on benchmark results, the NVIDIA card is the only defensible choice.

Architecture Differences

The two GPUs come from fundamentally different architectures. The NVIDIA Quadro M5000M uses the Maxwell 2.0 architecture on the GM204 chip, fabricated on a 28 nm process at TSMC. The AMD Radeon HD 8690M uses the GCN 1.0 architecture on the Sun chip, also fabricated on a 28 nm process at TSMC. Both are end-of-life products, but their design philosophies differ sharply.

The GM204 chip is a large die at 398 mm² with 5,200 million transistors. That translates to a transistor density of 13.1 million transistors per square millimeter. The Sun chip is much smaller at 56 mm² with 690 million transistors, giving a density of 12.3 million per square millimeter. The Quadro’s die is over seven times larger in area and packs over seven times the transistor count. This scale difference is the root cause of the performance gap. More transistors mean more shading units, texture units, and ROPs, which directly feed the benchmark scores.

The Maxwell 2.0 architecture emphasizes high per-clock efficiency and is known for its strong compute throughput relative to power. The Quadro M5000M has 1,536 shading units, 96 TMUs, and 64 ROPs. The GCN 1.0 architecture in the Radeon HD 8690M is an earlier design that was more compute-oriented but less efficient per transistor. It has 320 shading units, 20 TMUs, and only 8 ROPs. The ROP count difference is particularly stark: 64 versus 8, which explains the pixel rate gap of 67.26 GPixel/s versus 7.800 GPixel/s. The texture rate gap of 100.9 GTexel/s versus 19.50 GTexel/s is directly tied to the 96 versus 20 TMU count.

Both architectures support DirectX 12, but the Quadro M5000M reaches DirectX 12 (12_1), while the Radeon HD 8690M is limited to DirectX 12 (11_1). The Quadro also supports Vulkan 1.4, while the Radeon supports Vulkan 1.2.170. Both support OpenGL 4.6. The Quadro’s newer feature set means it can handle more recent rendering techniques and compute APIs. The Radeon’s older GCN 1.0 design predates several optimizations in later GCN revisions, and its 64-bit memory bus is a severe architectural bottleneck that no amount of clock speed can overcome.

Specification Differences

The specification sheet reveals a long list of differences, and nearly every one favors the NVIDIA Quadro M5000M. The process node is identical at 28 nm, and both are made by TSMC, but everything else diverges.

Clock speeds: The Quadro M5000M has a base clock of 962 MHz and a boost clock of 1051 MHz. The Radeon HD 8690M has a base clock of 925 MHz and a boost clock of 975 MHz. The NVIDIA card runs slightly faster, but the real difference is in the parallel hardware, not the clocks.

Memory: The Quadro has 8 GB of GDDR5 on a 256-bit bus, yielding 160.4 GB/s of bandwidth. The Radeon has 2 GB of GDDR5 on a 64-bit bus, yielding 32.00 GB/s. The Quadro’s memory clock is 1253 MHz (5 Gbps effective) versus the Radeon’s 1000 MHz (4 Gbps effective).

Compute resources: The Quadro has 1,536 shading units, 96 TMUs, and 64 ROPs. The Radeon has 320 shading units, 20 TMUs, and 8 ROPs. FP32 throughput is 3.229 TFLOPS for the Quadro and 624.0 GFLOPS for the Radeon.

Power and interface: The Quadro is rated at 100 W TDP and uses an MXM-B (3.0) bus interface with a slot width of MXM Module. The Radeon has no TDP listed and uses a PCIe 3.0 x8 interface. The Quadro has no power connectors listed, which is typical for MXM modules that draw power from the slot. The Radeon has no power connector data either.

Release dates: The Quadro M5000M was released on 2015-08-17, while the Radeon HD 8690M was released on 2013-02-28. The two and a half year gap explains the architectural improvements in the NVIDIA part. The Quadro’s predecessor is Quadro Kepler-M, and its successor is Quadro Pascal-M. The Radeon’s predecessor is London, and its successor is Gem System. Both are production status end-of-life.

FAQ

Q: How much faster is the NVIDIA Quadro M5000M in Geekbench OpenCL?

A: The Quadro M5000M scores 22,920, which is 273.5% higher than the Radeon HD 8690M’s 6,137.

Q: What is the memory bandwidth difference between the two GPUs?

A: The Quadro M5000M has 160.4 GB/s of bandwidth from an 8 GB GDDR5 frame buffer on a 256-bit bus. The Radeon HD 8690M has 32.00 GB/s from 2 GB GDDR5 on a 64-bit bus.

Q: Which GPU has more shading units?

A: The NVIDIA Quadro M5000M has 1,536 shading units, while the AMD Radeon HD 8690M has 320.

Q: Do both GPUs support DirectX 12?

A: Yes, but the Quadro M5000M supports DirectX 12 (12_1), while the Radeon HD 8690M supports DirectX 12 (11_1).

Q: What is the FP32 compute throughput of each card?

A: The Quadro M5000M delivers 3.229 TFLOPS, and the Radeon HD 8690M delivers 624.0 GFLOPS.

Q: Which GPU has a higher pixel rate?

A: The Quadro M5000M has a pixel rate of 67.26 GPixel/s, compared to the Radeon HD 8690M’s 7.800 GPixel/s.

Where Each One Wins

The NVIDIA Quadro M5000M wins in every category that the data covers. For compute-heavy workloads such as OpenCL acceleration, machine learning inference, or any GPGPU task, the 273.5% lead in Geekbench OpenCL makes it the only viable choice. The 8 GB memory capacity is four times larger, allowing larger datasets to reside in VRAM without spilling to system memory. The 256-bit bus and 160.4 GB/s bandwidth ensure that memory-bound kernels are not starved. The 64 ROPs deliver 67.26 GPixel/s, which is essential for high-resolution rendering or heavy post-processing effects. The Quadro’s 100 W TDP is the only power figure in the data, so it is the reference point for thermal design, but no rival power draw is listed for comparison.

The AMD Radeon HD 8690M has no benchmark wins in the data. Its only role is as a low-end legacy mobile part. With 320 shading units and 8 ROPs, it can handle basic 2D desktop compositing, video playback, and very light 3D workloads from an era before its release. Its 2 GB memory is sufficient for older games at low resolution and settings, but the 64-bit bus limits any memory-intensive task. The Radeon’s nearest rivals include the Intel Iris Pro Graphics 6200 and NVIDIA GeForce MX230, both of which score within 1% of its average benchmark score, indicating it sits at the very bottom of the mobile GPU performance spectrum. The Quadro M5000M, by contrast, is within 0.5% of the GeForce GTX 670M and just 1.1% behind the Intel UHD Graphics P750, placing it in a mid-range tier that is far above the Radeon.

In practical terms, the Quadro M5000M is for professionals or enthusiasts who need mobile GPU compute without compromise. The Radeon HD 8690M is for basic display output in a laptop that will never be asked to do heavy graphics work. The data is unambiguous: one GPU is a compute workhorse, and the other is a legacy entry-level part with no measurable advantage in any tested benchmark.