NVIDIA P104-100 vs NVIDIA T600 Mobile Comparison

NVIDIA’s T600 Mobile and P104-100 occupy the same overall performance percentile in the database, but the data shows they are fundamentally different tools. The T600 Mobile is a low-power Turing-based mobile solution, while the P104-100 is a Pascal-era mining card with no display outputs. Their average benchmark scores are nearly identical — 32,849 for the T600 Mobile versus 32,747 for the P104-100, a delta of just 0.3% — yet the head-to-head results reveal a decisive split. This analysis breaks down where each GPU wins, what the numbers mean, and how their architectures dictate their roles.

Where Each One Wins

The P104-100 is the clear performance winner in every shared benchmark test. In the two available head-to-head comparisons, the P104-100 beats the T600 Mobile by 31.3% in Geekbench OpenCL and by 33.1% in Geekbench Vulkan. That is not a marginal gap; it is a commanding lead in raw compute throughput. The P104-100’s advantage stems from its much larger GPU: 1,920 shading units, 120 texture mapping units, and 64 ROPs, compared to the T600 Mobile’s 896 shading units, 56 TMUs, and 32 ROPs. It also delivers more than double the FP32 throughput — 6.655 TFLOPS versus 2.527 TFLOPS — and more than double the pixel rate (110.9 GPixel/s versus 45.12 GPixel/s).

The T600 Mobile, by contrast, wins on efficiency and integration. Its thermal design power is listed at 40 W, whereas the P104-100 has no TDP listed but requires a 200 W suggested PSU and a single 8-pin power connector. The T600 Mobile is an IGP (integrated graphics processor) with no power connectors, making it suitable for compact, portable systems. The P104-100 is a dual-slot card measuring 267 mm (10.5 inches) with no display outputs, which disqualifies it from any conventional desktop or workstation use case that requires a monitor. The T600 Mobile is end-of-life but was designed for mobile workstations; the P104-100 was explicitly a mining GPU, as its generation label indicates.

In terms of benchmark percentiles, both cards sit at the 76th percentile among all GPUs, meaning they are statistically peers in aggregate performance. The nearest rivals for the T600 Mobile include the P104-100 (delta 0.3%), AMD Radeon RX 7800 XT (0.7%), AMD FirePro S9300 X2 (0.9%), and NVIDIA T550 Mobile (-0.9%). The P104-100’s nearest rivals are the T600 Mobile (-0.3%), RX 7800 XT (0.4%), FirePro S9300 X2 (0.6%), and AMD FirePro S10000 (1.1%). So while the P104-100 dominates the specific head-to-head tests, the overall average scores put them in the same league.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA T600 Mobile edges out the P104-100, with an average score of 32,849 versus 32,747. That is a 0.3% difference in favor of the T600 Mobile.

Q: How much faster is the P104-100 in Geekbench OpenCL?

A: The P104-100 scores 51,663 in Geekbench OpenCL, while the T600 Mobile scores 35,486. This gives the P104-100 a 31.3% lead.

Q: Does the T600 Mobile support display outputs?

A: Yes, the T600 Mobile’s display outputs are listed as "Portable Device Dependent," meaning they vary by the host laptop or mobile device. The P104-100, in contrast, has no outputs at all.

Q: What are the memory specifications for each card?

A: The T600 Mobile has 4 GB of GDDR6 on a 128-bit bus, yielding 192.0 GB/s bandwidth. The P104-100 also has 4 GB, but it is GDDR5X on a 256-bit bus, yielding 320.3 GB/s bandwidth.

Q: Which card has a higher boost clock?

A: The P104-100 boosts to 1733 MHz, while the T600 Mobile boosts to 1410 MHz. The P104-100’s base clock is also higher at 1607 MHz versus 780 MHz.

Q: Are both cards still in production?

A: No. Both are listed as end-of-life. The T600 Mobile was released in April 2021, and the P104-100 was released in December 2017.

Head-to-Head Benchmarks

The two GPUs were tested in exactly two common benchmarks: Geekbench OpenCL and Geekbench Vulkan. In both, the P104-100 wins decisively.

Starting with Geekbench OpenCL, the P104-100 posts a score of 51,663 against the T600 Mobile’s 35,486. That is a delta of -31.3% from the perspective of the T600 Mobile, meaning the T600 Mobile scores roughly two-thirds of the P104-100’s result. The margin is consistent with the hardware disparity: the P104-100 has 2.14 times the shading units, 2.14 times the TMUs, and 2.63 times the FP32 throughput. The memory bandwidth difference is also stark — 320.3 GB/s versus 192.0 GB/s — which directly benefits OpenCL workloads that are memory-bound.

In Geekbench Vulkan, the gap widens slightly. The P104-100 scores 45,165, and the T600 Mobile scores 30,211, yielding a delta of -33.1%. This suggests the P104-100’s advantage is not just in compute but also in graphics-related tasks, likely due to its higher pixel rate (110.9 GPixel/s versus 45.12 GPixel/s) and texture rate (208.0 GTexel/s versus 78.96 GTexel/s). The P104-100’s 256-bit memory bus provides more than 66% additional bandwidth, which helps feed its larger ROP count.

Despite these sweeping wins, the aggregate benchmark averages tell a different story. The T600 Mobile’s average score of 32,849 is actually higher than the P104-100’s 32,747. This is possible because the average includes additional benchmarks beyond the two shared tests, and the T600 Mobile’s 76th percentile ranking is identical to the P104-100’s. The takeaway: in direct comparison, the P104-100 is substantially faster, but the broader dataset shows both cards sitting at the same performance tier.

Specification Differences

The most obvious differences are in memory and compute resources. The T600 Mobile uses 4 GB of GDDR6 on a 128-bit bus with 192.0 GB/s bandwidth. The P104-100 uses 4 GB of GDDR5X on a 256-bit bus with 320.3 GB/s bandwidth. The P104-100’s bus width is double, and its memory clock is lower (1251 MHz versus 1500 MHz) but the effective data rate of 10 Gbps versus 12 Gbps is offset by the wider bus.

Shading units: 896 for the T600 Mobile, 1,920 for the P104-100. TMUs: 56 versus 120. ROPs: 32 versus 64. These are not incremental differences; the P104-100 has more than twice the core resources. The FP32 compute is 2.527 TFLOPS for the T600 Mobile and 6.655 TFLOPS for the P104-100. Pixel rate is 45.12 GPixel/s versus 110.9 GPixel/s, and texture rate is 78.96 GTexel/s versus 208.0 GTexel/s.

Clocks differ significantly as well. The T600 Mobile runs at a 780 MHz base and 1410 MHz boost. The P104-100 runs at 1607 MHz base and 1733 MHz boost — over twice the base clock. This is partly due to the different architectures and power envelopes.

Physical specifications diverge sharply. The T600 Mobile is an IGP with no slot width, no power connectors, and a 40 W TDP. The P104-100 is a dual-slot card, 267 mm long, with a single 8-pin power connector and a suggested PSU of 200 W. The bus interface also differs: the T600 Mobile uses PCIe 3.0 x16, while the P104-100 uses PCIe 1.0 x4 — a major limitation for the P104-100 in any host system.

Display outputs: the T600 Mobile is "Portable Device Dependent," meaning it relies on the laptop’s outputs. The P104-100 has no outputs at all, which is expected for a mining card. The T600 Mobile’s generation is listed as "Quadro Turing-M (Tx000)" and its predecessor is "Quadro Pascal-M" with a successor of "Ampere-MW." The P104-100’s generation is simply "Mining GPUs" with no predecessor or successor listed.

Architecture Differences

The two GPUs come from different NVIDIA architectures. The T600 Mobile is built on Turing (chip TU117), fabricated on a 12 nm process at TSMC. The P104-100 is built on Pascal (chip GP104), fabricated on a 16 nm process, also at TSMC. Turing is the newer architecture, and it is reflected in the T600 Mobile’s feature set.

Transistor counts differ substantially. The T600 Mobile has 4,700 million transistors on a 200 mm² die, yielding a transistor density of 23.5M per mm². The P104-100 has 7,200 million transistors on a 314 mm² die, with a density of 22.9M per mm². The P104-100’s die is 57% larger, and its transistor count is 53% higher, but the densities are nearly identical — a sign that the 12 nm process on the T600 Mobile offers only marginal density improvement over 16 nm.

Memory type is another architectural split: GDDR6 on the T600 Mobile versus GDDR5X on the P104-100. The T600 Mobile’s memory clock is 1500 MHz with 12 Gbps effective, while the P104-100’s is 1251 MHz with 10 Gbps effective. However, the P104-100’s 256-bit bus overcomes the lower per-pin speed.

FP16 performance is a notable architectural divergence. The T600 Mobile achieves 5.053 TFLOPS FP16 with a 2:1 ratio relative to FP32. The P104-100 achieves only 104.0 GFLOPS FP16, with a 1:64 ratio — meaning its FP16 throughput is negligible. This is a hallmark of Pascal’s design, which did not prioritize half-precision compute. Turing, by contrast, offers full-rate FP16, which matters for inference workloads.

Neither card has ray tracing cores or tensor cores, so those features are absent from both. The API support is identical: DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The T600 Mobile’s release date is April 2021, while the P104-100’s is December 2017 — a gap of over three years that explains the architectural generational difference. The T600 Mobile also has a production status of end-of-life, as does the P104-100, making both legacy products. The T600 Mobile’s 40 W TDP versus the P104-100’s 200 W suggested PSU requirement illustrates the design philosophy: Turing efficiency versus Pascal brute force.