AMD Radeon HD 8750M vs Intel UHD Graphics P750 Comparison

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The Intel UHD Graphics P750 scores 6554, while the AMD Radeon HD 8750M scores 5970. The Intel part leads by 9.8% in this benchmark.

Q: How does the Intel UHD Graphics P750 compare to its nearest rivals?

A: The P750 sits within a tight cluster: it is 0.4% behind the AMD Radeon HD 7730M (6581), 0.9% behind the AMD Radeon R7 M460 (6612), and 0.6% ahead of the NVIDIA GeForce GTX 670M (6513). Its 38th percentile ranking among all GPUs places it in the lower-middle tier of the database.

Q: What are the architectural foundations of each GPU?

A: The Intel UHD Graphics P750 is built on Generation 12.1 architecture using Rocket Lake silicon on Intel's 14 nm+++ process. The AMD Radeon HD 8750M uses GCN 1.0 architecture on the Mars chip, manufactured by TSMC on a 28 nm node with 950 million transistors on a 77 mm² die.

Q: What memory configurations do these GPUs use?

A: The Intel UHD Graphics P750 uses System Shared memory with a System Shared bus width and System Dependent bandwidth. The AMD Radeon HD 8750M has 1024 MB of dedicated DDR3 memory on a 128-bit bus, delivering 28.80 GB/s of bandwidth.

Q: Which API levels are supported by each GPU?

A: The Intel part supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The AMD part supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The Intel GPU has the higher DirectX feature level and the newer Vulkan version.

Q: What is the production status of these GPUs?

A: Both are end-of-life products. The AMD Radeon HD 8750M was released on February 25, 2013, and its predecessor is London, while its successor is Gem System.

Architecture Differences

The Intel UHD Graphics P750 and AMD Radeon HD 8750M represent fundamentally different design philosophies. Intel's part comes from the Generation 12.1 architecture, implemented on the Rocket Lake chip using Intel's 14 nm+++ process. This is an integrated graphics processor (IGP) that connects via Ring Bus and draws its memory from the system pool. The AMD Radeon HD 8750M, by contrast, is a discrete mobile GPU based on GCN 1.0, using the Mars chip fabricated by TSMC on a 28 nm process. The AMD chip contains 950 million transistors within a 77 mm² die, giving a transistor density of 12.3 million per mm².

The compute resource allocation differs sharply between the two. The Intel GPU fields 256 shading units, 64 texture mapping units, and 32 ROPs. The AMD GPU has 384 shading units but only 24 TMUs and 8 ROPs. This means AMD packs 50% more shader processors, while Intel provides 2.7x the texture units and 4x the ROP count. The result is a different balance of throughput: Intel's architecture emphasizes pixel and texture work, while AMD's design concentrates on raw shader execution.

Clock behavior also diverges. The Intel part runs at a 350 MHz base clock and boosts to 1300 MHz. The AMD part runs at a 775 MHz base and boosts to 825 MHz, with memory clocked at 900 MHz for 1800 Mbps effective. The Intel GPU's boost clock is 475 MHz higher than AMD's boost clock, a 57.6% higher figure. The AMD GPU's memory runs at 28.80 GB/s of dedicated bandwidth on a 128-bit bus. The Intel GPU's memory bandwidth is System Dependent.

The pixel rate and texture rate figures underscore the architectural difference. Intel achieves 41.60 GPixel/s and 83.20 GTexel/s, while AMD achieves 6.600 GPixel/s and 19.80 GTexel/s. Intel's pixel rate is 6.3x higher, and the texture rate is 4.2x higher. The FP32 throughput is close: Intel delivers 665.6 GFLOPS, AMD delivers 633.6 GFLOPS, a 5.1% Intel advantage. Intel also lists FP16 at 1,331.2 GFLOPS with a 2:1 ratio; AMD lists no FP16 figure.

The power profile differs as well. The Intel part has a TDP of 15 W and is an IGP, meaning no slot width applies. The AMD Radeon HD 8750M has no TDP recorded in the database and no slot width specified, though it is a discrete mobile part. The Intel part's display outputs are Motherboard Dependent, while the AMD part has no display output information recorded.

Head-to-Head Benchmarks

The database contains a single head-to-head comparison between these two GPUs: the Geekbench OpenCL test. In this test, the Intel UHD Graphics P750 scores 6554 against the AMD Radeon HD 8750M's 5970. The Intel GPU wins by 9.8%, a decisive margin given that both parts are end-of-life mobile or integrated solutions. The Intel part wins 1 test in the head-to-head set, while the AMD part wins 0.

The 9.8% gap corresponds to a difference of 584 points on the Geekbench OpenCL scale. To contextualize this within the broader database, the AMD Radeon HD 8750M's nearest rivals include the NVIDIA Quadro K4000 at 5982, the NVIDIA Quadro K4000M at 5986, the NVIDIA Quadro K620M at 5957, and the AMD Radeon HD 8730M at 5955. The AMD part's 5970 score places it 0.2% behind the Quadro K4000, 0.3% ahead of the K620M, 0.2% behind the K4000M, and 0.3% ahead of the HD 8730M. The AMD part ranks at the 34th percentile of all GPUs, below the Intel part's 38th percentile.

The Intel part's nearest rivals show a similarly tight grouping around the 6500 mark. The AMD Radeon HD 7730M at 6581 is 0.4% ahead of Intel's 6554 score, the NVIDIA GeForce GTX 670M at 6513 is 0.6% behind Intel's 6554, the AMD Radeon R7 M460 at 6612 is 0.9% ahead of Intel's 6554, and the NVIDIA GeForce GT 555M at 6493 is 0.9% behind Intel's 6554. Intel's 6554 score fits within 0.9% of three rivals in either direction, indicating that the P750 is a mid-pack performer relative to GPUs from the same era.

The benchmark data suggests that the Intel UHD Graphics P750 punches above its integrated classification. Despite being an IGP with system-shared memory, its 6554 Geekbench OpenCL score exceeds the discrete AMD Radeon HD 8750M's 5970 by 9.8%, and it sits closer to the discrete AMD Radeon HD 7730M within 0.4%. The Intel part's 665.6 GFLOPS FP32 compute output is 32.0 GFLOPS ahead of the AMD part's 633.6 GFLOPS, reinforcing the OpenCL performance lead.

Specification Differences

The two GPUs differ across nearly every measurable specification category. The Intel UHD Graphics P750 has a base clock of 350 MHz, boost clock of 1300 MHz, and memory clock of System Shared. The AMD Radeon HD 8750M has a base clock of 775 MHz, boost clock of 825 MHz, and memory clock of 900 MHz at 1800 Mbps effective. The Intel part's boost clock is 1300 MHz, which is 475 MHz above the AMD part's boost of 825 MHz.

Memory configuration differs completely. Intel uses System Shared memory with a System Shared bus width and System Dependent bandwidth. AMD uses 1024 MB of DDR3 on a 128-bit bus with 28.80 GB/s bandwidth. The shading unit count is 256 for Intel versus 384 for AMD. Texture mapping units are 64 for Intel versus 24 for AMD. ROPs are 32 for Intel versus 8 for AMD.

Pixel rate is 41.60 GPixel/s for Intel versus 6.600 GPixel/s for AMD. Texture rate is 83.20 GTexel/s for Intel versus 19.80 GTexel/s for AMD. FP32 compute is 665.6 GFLOPS for Intel versus 633.6 GFLOPS for AMD. Intel lists FP16 at 1,331.2 GFLOPS (2:1); AMD lists no FP16 value.

The process node is 14 nm+++ for Intel versus 28 nm for AMD. The foundry is Intel for the P750 versus TSMC for the Radeon. The transistor count is only listed for AMD at 950 million; Intel's is not recorded. The die size is only listed for AMD at 77 mm²; Intel's is not recorded. Transistor density is only listed for AMD at 12.3M per mm².

The bus interface differs: Intel uses Ring Bus, AMD uses PCIe 3.0 x8. The TDP is 15 W for Intel; AMD has no TDP recorded. The slot width is IGP for Intel; AMD has no slot width recorded. Display outputs are Motherboard Dependent for Intel; AMD has no display output data.

The API support differs: Intel supports DirectX 12 (12_1) and Vulkan 1.4, while AMD supports DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6. The production status is end-of-life for both. The release date is only recorded for AMD: February 25, 2013. The predecessor is London for AMD, and the successor is Gem System; Intel lists no predecessor or successor.

The power connectors, suggested PSU, dimensions, and launch MSRP are either absent or null for both parts. The transistor density figure of 12.3M per mm² for AMD stands alone, as Intel's process details are not quantified in the same way.

The Verdict

The benchmark data points to the Intel UHD Graphics P750 as the stronger GPU in raw OpenCL compute. Its 6554 Geekbench OpenCL score beats the AMD Radeon HD 8750M's 5970 by 9.8%, a clear win in the only head-to-head test recorded. The Intel part also holds the higher percentile rank at 38 versus AMD's 34, and it delivers 665.6 GFLOPS of FP32 compute against AMD's 633.6 GFLOPS, a 5.1% advantage.

The decision between these two GPUs depends on the use case. For users who prioritize texture and pixel throughput, the Intel part offers 83.20 GTexel/s and 41.60 GPixel/s, figures that dwarf the AMD part's 19.80 GTexel/s and 6.600 GPixel/s. The Intel GPU's 4x ROP advantage and 2.7x TMU advantage make it the pick for fill-rate-bound workloads. The AMD part counters with more shading units, 384 versus 256, which could benefit shader-heavy tasks in theory, but the recorded OpenCL benchmark does not reflect that advantage. The AMD part's 5970 score is lower than every one of its nearest rivals: the NVIDIA Quadro K4000 at 5982, the Quadro K4000M at 5986, the Quadro K620M at 5957, and the Radeon HD 8730M at 5955. The AMD GPU is the weaker performer in this comparison, sitting at the 34th percentile.

The Intel part's nearest rivals all sit within 0.9% of its score, with the AMD Radeon HD 7730M at 6581 and the NVIDIA GeForce GTX 670M at 6513. The P750's 6554 score places it in a competitive band, slightly below the HD 7730M but ahead of the GTX 670M. The Intel GPU's integrated nature and 15 W TDP make it suited for systems requiring modest power draw, while the AMD Radeon HD 8750M has no TDP recorded in the database, making power comparison impossible from the data.

For a user choosing between these two, the Intel UHD Graphics P750 is the recommended pick based on the recorded scores. It wins the head-to-head benchmark, has a higher percentile ranking, and offers higher pixel and texture rates. The AMD Radeon HD 8750M, with its larger shader count and dedicated 1024 MB memory buffer, might be preferable in a system where dedicated VRAM matters, but the benchmark data does not show that translating into superior OpenCL performance. The AMD part's 28.80 GB/s bandwidth and 128-bit bus are fixed resources, whereas the Intel part's memory bandwidth is System Dependent and could vary with the host platform.

The verdict from the database is straightforward: the Intel UHD Graphics P750 leads in every recorded performance metric, and the AMD Radeon HD 8750M trails by 9.8% in OpenCL. The Intel part also trails 4 points in percentile rank as well, 38 versus 34. The data supports picking the Intel part for compute workloads and for fill-rate-sensitive applications. The AMD part's only strengths are its higher shader count and dedicated memory capacity of 1024 MB, neither of which appears in the recorded benchmark results as a winning differentiator.