AMD Instinct MI355X vs Intel UHD Graphics 770 Mobile Comparison

Head-to-Head Benchmarks

The recorded data contains no direct benchmark comparisons between the AMD Instinct MI355X and the Intel UHD Graphics 770 Mobile. Both entries show zero benchmark scores, zero average scores, and zero recorded wins in head-to-head testing. The percentile rankings for both GPUs sit at 50, which places them at the midpoint of all GPUs tracked in the database, though this figure carries limited weight given the absence of actual performance measurements.

What can be established from the specification data is the scale of the disparity. The AMD Instinct MI355X delivers 78.64 TFLOPS of FP32 compute, while the Intel UHD Graphics 770 Mobile delivers 819.2 GFLOPS. That is a 96-fold difference in raw single-precision throughput. The texture rate tells a similar story: 2,457.6 GTexel/s for the AMD part versus 25.60 GTexel/s for the Intel integrated solution. The pixel rate comparison is one of the few areas where the Intel GPU does not show zero, posting 12.80 GPixel/s against the AMD part's 0 MPixel/s, which reflects the MI355X's lack of traditional display output capability rather than any performance weakness.

The memory subsystem gap is even more pronounced. The MI355X uses 288 GB of HBM3e memory across an 8192-bit bus, producing 8.19 TB/s of bandwidth. The Intel part uses system shared memory with a bus width and bandwidth listed as "System Shared" and "System Dependent," respectively. No fixed bandwidth figure exists for comparison because the performance is tied to the host system's memory configuration.

Clock behavior differs substantially. The AMD GPU base clock is 1000 MHz with a boost of 2400 MHz. The Intel GPU runs a 300 MHz base clock with a 1600 MHz boost. The AMD memory clock is listed as 2000 MHz with 8 Gbps effective data rate, while the Intel memory clock is simply "System Shared."

FAQ

Q: Which GPU has more shading units?

A: The AMD Instinct MI355X has 16,384 shading units. The Intel UHD Graphics 770 Mobile has 256 shading units. The AMD part also has 1,024 texture mapping units, while the Intel part has 16.

Q: What is the thermal design power difference?

A: The AMD Instinct MI355X is rated at 1400 W TDP. The Intel UHD Graphics 770 Mobile is rated at 15 W TDP. The AMD part requires a suggested power supply of 1800 W, while the Intel part has no separate power supply requirement listed.

Q: Are these GPUs from the same manufacturing process?

A: No. The AMD Instinct MI355X uses a 3 nm process at TSMC and contains 185,000 million transistors on a 2380 mm² die. The Intel UHD Graphics 770 Mobile uses a 10 nm process at Intel, with no transistor count or die size recorded.

Q: What memory types do these GPUs use?

A: The AMD Instinct MI355X uses HBM3e memory with 288 GB capacity and an 8192-bit bus. The Intel UHD Graphics 770 Mobile uses system shared memory, where the type, size, bus width, and bandwidth are all tied to the host system.

Q: Do both GPUs support the same graphics APIs?

A: No. The Intel UHD Graphics 770 Mobile supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The AMD Instinct MI355X lists N/A for DirectX, OpenGL, and Vulkan, indicating it is not designed for conventional graphics API workloads.

Q: When were these products released?

A: The AMD Instinct MI355X has a release date of June 11, 2025. The Intel UHD Graphics 770 Mobile has a release date of January 3, 2023.

Architecture Differences

The AMD Instinct MI355X is built on the CDNA 4.0 architecture, which is AMD's compute-optimized design lineage aimed at data center acceleration. The chip is designated MI350 256CU, indicating a large compute unit count. It belongs to the Instinct (MIx) generation and uses a 3 nm process node at TSMC. The transistor count reaches 185,000 million, packed into a 2380 mm² die, yielding a transistor density of 77.7 million transistors per square millimeter.

The Intel UHD Graphics 770 Mobile uses the Generation 12.2 architecture, which is Intel's integrated graphics design. The chip is based on Raptor Lake and belongs to the HD Graphics-M (Raptor Lake) generation. It uses a 10 nm process at Intel. No transistor count, die size, or density figures are recorded for this part, which reflects its nature as an integrated GPU sharing the processor die.

The AMD part has no display outputs, while the Intel part lists "Portable Device Dependent" display outputs. This confirms that the MI355X is an accelerator without graphics output functionality, whereas the Intel integrated GPU exists primarily to drive displays on mobile systems.

The AMD part uses a PCIe 5.0 x16 bus interface and is formatted as an OAM Module. The Intel part uses a Ring Bus interface and is classified as an IGP. The AMD GPU has no power connectors listed and requires a suggested PSU of 1800 W. The Intel GPU has no power connector information and no suggested PSU, consistent with an integrated solution drawing power from the host processor's power delivery.

FP16 compute ratios differ. The AMD part lists FP16 at 78.64 TFLOPS with a 1:1 ratio to FP32, meaning it does not gain a throughput advantage in reduced precision. The Intel part lists FP16 at 1.638 TFLOPS with a 2:1 ratio, meaning its half-precision throughput is double its FP32 rate.

The Verdict

The data indicates that these two GPUs serve entirely different purposes and should not be viewed as competing products. The AMD Instinct MI355X is a data center accelerator with 288 GB of HBM3e memory, 8.19 TB/s of memory bandwidth, and 78.64 TFLOPS of FP32 compute. It requires a 1400 W TDP and an 1800 W suggested power supply. It has no display outputs and no supported graphics APIs. This is a compute engine for server workloads.

The Intel UHD Graphics 770 Mobile is an integrated GPU designed for mobile systems. It delivers 819.2 GFLOPS of FP32 compute, uses system shared memory, and carries a 15 W TDP. It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, and its display outputs depend on the portable device it is integrated into. This is a display and light-compute solution for laptops.

A buyer selecting between these parts would be choosing based on use case, not performance level. The MI355X's massive memory capacity and bandwidth, combined with its compute throughput, suit large-scale data center tasks. The Intel GPU's low power draw and API support suit mobile computing and display output. The data records no benchmark overlap to suggest any scenario where these two would be evaluated side by side.

The production status differs as well. The Intel part is listed as Active, with a successor in Arc Graphics-M. The AMD part has no production status recorded, but its predecessor is listed as Radeon Instinct, and its release date of June 2025 places it as a recent data center product.

Specification Differences

The following fields differ between the two parts:

• Process Node: AMD uses 3 nm at TSMC. Intel uses 10 nm at Intel.
• Transistors: AMD has 185,000 million. Intel has no recorded value.
• Die Size: AMD has 2380 mm². Intel has no recorded value.
• Transistor Density: AMD has 77.7M per mm². Intel has no recorded value.
• Base Clock: AMD runs 1000 MHz. Intel runs 300 MHz.
• Boost Clock: AMD reaches 2400 MHz. Intel reaches 1600 MHz.
• Memory Clock: AMD uses 2000 MHz with 8 Gbps effective. Intel uses system shared memory.
• Memory Size: AMD has 288 GB. Intel has system shared memory.
• Memory Type: AMD uses HBM3e. Intel uses system shared memory.
• Memory Bus Width: AMD uses 8192 bit. Intel uses system shared memory.
• Memory Bandwidth: AMD has 8.19 TB/s. Intel has system dependent bandwidth.
• Shading Units: AMD has 16,384. Intel has 256.
• TMUs: AMD has 1,024. Intel has 16.
• ROPs: AMD has 0. Intel has 8.
• Pixel Rate: AMD is 0 MPixel/s. Intel is 12.80 GPixel/s.
• Texture Rate: AMD is 2,457.6 GTexel/s. Intel is 25.60 GTexel/s.
• FP32: AMD is 78.64 TFLOPS. Intel is 819.2 GFLOPS.
• FP16: AMD is 78.64 TFLOPS (1:1). Intel is 1.638 TFLOPS (2:1).
• TDP: AMD is 1400 W. Intel is 15 W.
• Slot Width: AMD is OAM Module. Intel is IGP.
• Power Connectors: AMD lists None. Intel lists no value.
• Suggested PSU: AMD lists 1800 W. Intel lists no value.
• Bus Interface: AMD uses PCIe 5.0 x16. Intel uses Ring Bus.
• Display Outputs: AMD has no outputs. Intel is portable device dependent.
• DirectX: AMD is N/A. Intel is 12 (12_1).
• OpenGL: AMD is N/A. Intel is 4.6.
• Vulkan: AMD is N/A. Intel is 1.4.
• Dimensions: AMD is 102 mm by 165 mm. Intel has no recorded dimensions.
• Release Date: AMD is June 11, 2025. Intel is January 3, 2023.
• Predecessor: AMD lists Radeon Instinct. Intel lists no value.
• Successor: AMD lists no value. Intel lists Arc Graphics-M.
• Production Status: AMD lists no value. Intel lists Active.

Where Each One Wins

The AMD Instinct MI355X wins decisively in every compute-oriented category recorded. Its FP32 throughput of 78.64 TFLOPS exceeds the Intel part's 819.2 GFLOPS by nearly two orders of magnitude. Its texture rate of 2,457.6 GTexel/s is roughly 96 times higher. Its memory bandwidth of 8.19 TB/s, memory capacity of 288 GB, and 8192-bit bus have no equivalent on the Intel side, which relies entirely on host system memory. The 16,384 shading units and 1,024 TMUs provide the execution resources for large-scale parallel workloads.

The Intel UHD Graphics 770 Mobile wins in the categories that matter for integrated graphics. It has 8 ROPs versus the AMD part's 0, giving it a pixel rate of 12.80 GPixel/s while the AMD part produces 0 MPixel/s. It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, all of which are listed as N/A on the AMD part. Its display outputs are functional, whereas the AMD part has no outputs. Its 15 W TDP makes it suitable for mobile systems, while the AMD part demands 1400 W. The Intel part also has a listed active production status with a successor in Arc Graphics-M, while the AMD part has no production status recorded.

The use case split is therefore clear. The MI355X is for data center compute, machine learning training, and high-performance acceleration where memory capacity and bandwidth dominate. The UHD Graphics 770 Mobile is for laptop display output, video playback, and light graphics work where power efficiency and API compatibility matter. Neither part can substitute for the other in its intended role.