GPU Comparison

# NVIDIA GeForce GTX 970M vs NVIDIA GeForce 930M

The NVIDIA GeForce GTX 970M and NVIDIA GeForce 930M occupy opposite ends of the mobile GPU spectrum within the same GeForce 900M generation. The GTX 970M is built on the GM204 chip with Maxwell 2.0 architecture, while the 930M uses the smaller GM108S chip with first-generation Maxwell. Their average benchmark scores, 4628 for the GTX 970M versus 4388 for the 930M, show only a 5.5% gap on paper, but the head-to-head results reveal a far more dramatic story. The GTX 970M sits at the 27th percentile of all GPUs, while the 930M lands at the 26th percentile, placing both firmly in the lower-middle tier of the overall performance distribution. Yet their architectural differences suggest these two chips were designed for entirely different workloads and user expectations.

FAQ

Q: How much faster is the GTX 970M than the 930M in OpenCL performance?

A: The GTX 970M scores 18,946 in Geekbench OpenCL, which is 275.5% higher than the 930M's 5,046. This means the GTX 970M delivers roughly 3.75 times the compute throughput in this workload.

Q: What is the difference in Vulkan benchmark scores between the two GPUs?

A: In Geekbench Vulkan, the GTX 970M achieves 18,292 points while the 930M scores 3,729. The GTX 970M leads by 390.5%, making Vulkan the largest performance gap between the two in any head-to-head test.

Q: Which GPU has more shading units and how does that affect performance?

A: The GTX 970M features 1,280 shading units compared to 384 on the 930M, a 3.3x difference. This directly translates into the GTX 970M's 2.657 TFLOPS FP32 throughput versus the 930M's 421.6 GFLOPS, explaining the massive compute benchmark advantages.

Q: Do both GPUs support the same DirectX version?

A: No. The GTX 970M supports DirectX 12 (12_1), while the 930M is limited to DirectX 12 (11_0). The GTX 970M's Maxwell 2.0 architecture includes feature level 12_1 capabilities that the older Maxwell implementation on the 930M lacks.

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

A: The GTX 970M offers 120.3 GB/s bandwidth from 6 GB of GDDR5 memory on a 192-bit bus, while the 930M provides 12.80 GB/s from 2 GB of DDR3 on a 64-bit bus. That is a 9.4x difference in memory bandwidth.

Q: How do their transistor counts and die sizes compare?

A: The GTX 970M packs 5,200 million transistors on a 398 mm² die, while the 930M has 1,020 million transistors on a 77 mm² die. Both use the same 28 nm TSMC process, resulting in nearly identical transistor densities of 13.1M/mm² and 13.2M/mm² respectively.

Architecture Differences

The architectural gap between these two GPUs is fundamental, not incremental. The GTX 970M uses the GM204 chip with Maxwell 2.0 architecture, which introduces significant improvements over the original Maxwell design found in the 930M's GM108S chip. Both are fabricated on TSMC's 28 nm process, but that is where the similarities end.

The GTX 970M's GM204 die contains 5,200 million transistors spread across 398 mm², while the 930M's GM108S packs just 1,020 million transistors into 77 mm². This 5.1x transistor advantage and 5.2x die area advantage give the GTX 970M vastly more compute resources. The shading unit count tells the story: 1,280 versus 384, a 3.3x difference. Texture mapping units number 80 versus 24, and ROPs stand at 48 versus 8, a 6x difference in pixel processing capability.

Clock speeds further amplify the gap. The GTX 970M runs at a 924 MHz base clock with a 1038 MHz boost, while the 930M operates at a flat 549 MHz with no boost capability. This means the GTX 970M benefits from dynamic clock scaling under load, while the 930M runs at a constant frequency. The GTX 970M's memory clock of 1253 MHz (5 Gbps effective) on GDDR5 vastly outperforms the 930M's 800 MHz (1600 Mbps effective) on DDR3.

The GTX 970M's Maxwell 2.0 architecture supports DirectX 12 (12_1) feature level, while the 930M's original Maxwell only reaches DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4, but the GTX 970M's newer architecture includes additional hardware features for modern rendering techniques. The GTX 970M uses an MXM-B (3.0) bus interface with an MXM Module slot width, while the 930M uses PCIe 3.0 x8 and is integrated as an IGP. The GTX 970M also has a listed TDP of 33 W for the 930M, though the GTX 970M's TDP is not specified in the data.

Head-to-Head Benchmarks

The head-to-head benchmark data shows a complete dominance by the GTX 970M, winning both tests with staggering margins. In Geekbench OpenCL, the GTX 970M scores 18,946 against the 930M's 5,046, a delta of 275.5%. This is not a marginal victory but a generational leap in compute performance. The OpenCL test exercises general-purpose GPU compute, and the GTX 970M's 2.657 TFLOPS FP32 throughput simply overwhelms the 930M's 421.6 GFLOPS.

The Geekbench Vulkan benchmark tells an even more one-sided story. The GTX 970M achieves 18,292 points, while the 930M manages only 3,729, a 390.5% delta in favor of the GTX 970M. Vulkan is a low-overhead graphics API that benefits from raw hardware throughput, and the GTX 970M's combination of 1,280 shading units, 80 TMUs, and 48 ROPs provides massive parallel processing capability. The 930M's 384 shading units, 24 TMUs, and 8 ROPs are simply outclassed.

These benchmark results align with the theoretical specifications. The GTX 970M's pixel rate of 49.82 GPixel/s compares to the 930M's 4.392 GPixel/s, an 11.3x advantage. Texture rate differs similarly: 83.04 GTexel/s versus 13.18 GTexel/s, a 6.3x gap. The memory bandwidth difference of 120.3 GB/s versus 12.80 GB/s is perhaps the most striking disparity, as it affects every memory-bound workload. The GTX 970M also has 6 GB of GDDR5 memory versus 2 GB of DDR3, meaning it can hold larger textures and datasets without spilling to system memory.

Specification Differences

The specification table reveals systematic differences across every major component. The GTX 970M uses the GM204 chip with Maxwell 2.0 architecture, while the 930M uses GM108S with original Maxwell. Process nodes are identical at 28 nm from TSMC, but transistor counts diverge sharply: 5,200 million versus 1,020 million. Die sizes follow suit at 398 mm² versus 77 mm², with nearly identical transistor densities of 13.1M/mm² and 13.2M/mm².

Clock speeds show the GTX 970M running at 924 MHz base and 1038 MHz boost, while the 930M is locked at 549 MHz with no boost. Memory configurations differ completely: the GTX 970M has 6 GB GDDR5 on a 192-bit bus with 120.3 GB/s bandwidth, while the 930M has 2 GB DDR3 on a 64-bit bus with 12.80 GB/s. Memory clocks are 1253 MHz (5 Gbps effective) versus 800 MHz (1600 Mbps effective).

Compute resources scale dramatically: shading units 1280 versus 384, TMUs 80 versus 24, and ROPs 48 versus 8. The GTX 970M achieves 49.82 GPixel/s pixel rate and 83.04 GTexel/s texture rate, compared to the 930M's 4.392 GPixel/s and 13.18 GTexel/s. FP32 throughput stands at 2.657 TFLOPS versus 421.6 GFLOPS. The GTX 970M's TDP is not listed, while the 930M has a 33 W TDP. The GTX 970M uses an MXM-B (3.0) interface with MXM Module slot width, while the 930M uses PCIe 3.0 x8 as an IGP. Both have no power connectors and portable-device-dependent display outputs.

Where Each One Wins

The GTX 970M wins in every measurable benchmark category. In the two head-to-head tests, Geekbench OpenCL and Geekbench Vulkan, it leads by 275.5% and 390.5% respectively. Its architectural advantages in shading units, TMUs, ROPs, and memory bandwidth make it superior for any graphics-intensive application, from gaming to content creation. The 6 GB GDDR5 frame buffer versus 2 GB DDR3 means the GTX 970M can handle higher resolutions, larger textures, and more complex scenes without memory-related stuttering.

The 930M's advantages are more subtle and relate to integration rather than performance. Its 33 W TDP suggests lower power consumption, making it suitable for thinner, lighter laptops where battery life and thermal management take priority. The IGP form factor with PCIe 3.0 x8 interface indicates it is designed for basic multimedia tasks, web browsing, and light productivity rather than gaming or compute workloads. Its 77 mm² die size and 1,020 million transistors make it a low-cost, low-complexity solution for entry-level laptops.

The 930M does compete favorably with its nearest rivals, sitting within 2.2% of the AMD FirePro W2100 and 0.7% above the Intel Iris Pro Graphics 5200. The GTX 970M, meanwhile, trades blows with the NVIDIA Quadro M3000M (0.2% delta) and sits just 0.6% below the AMD Radeon R5 M320, showing it is competitive within its mobile GPU tier.

The Verdict

The data presents a clear hierarchy. The GTX 970M is the overwhelmingly more capable GPU, winning both head-to-head benchmarks by margins of 275.5% and 390.5%. Its 1,280 shading units, 48 ROPs, and 120.3 GB/s memory bandwidth place it in a different performance class entirely. Users who need to run modern games, GPU-accelerated applications, or any compute-heavy workload should choose the GTX 970M without hesitation.

The 930M serves a fundamentally different purpose. With a 33 W TDP, 384 shading units, and 12.80 GB/s memory bandwidth, it is built for efficiency and basic functionality rather than performance. It suits users who primarily need GPU acceleration for everyday tasks like video playback, casual photo editing, or older games. Its 2 GB DDR3 memory and 64-bit bus are adequate for these workloads but will struggle with modern titles or professional applications.

The percentile rankings, 27th for the GTX 970M and 26th for the 930M, might suggest near-parity, but the head-to-head data contradicts this. The GTX 970M's average benchmark score of 4628 versus the 930M's 4388 understates the real-world gap because the averages include different benchmark suites. When directly compared on identical tests, the GTX 970M delivers 3.75x to 4.9x the performance. The 930M's competitive standing against integrated graphics like the Intel Iris Pro 5200 confirms its position as an entry-level solution, while the GTX 970M's rivalry with the Quadro M3000M demonstrates its workstation-capable performance. Choose the GTX 970M for performance; choose the 930M only if power efficiency and minimal thermal footprint are the absolute priorities.