Intel Iris Xe MAX Graphics vs NVIDIA GeForce GTX 660 Ti Comparison

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and it favors the NVIDIA GeForce GTX 660 Ti. The GTX 660 Ti scores 15,113 points against the Intel Iris Xe MAX's 14,315 points, a 5.6% advantage. That is a meaningful but not overwhelming lead—enough to put the older discrete card ahead in raw compute workloads, but not enough to suggest a generational gap in performance.

Looking at the broader benchmark landscape, the GTX 660 Ti's average benchmark score across all tests is 15,063, placing it in the 57th percentile of all GPUs. Its nearest rivals include the AMD Radeon Pro 560X (15,082, just 0.1% ahead), the AMD Radeon RX 7600 (15,171, 0.7% ahead), the NVIDIA GeForce RTX 3050 OEM (15,199, 0.9% ahead), and the AMD Radeon 680M (15,270, 1.4% ahead). The GTX 660 Ti is essentially trading blows with these much newer parts, sitting within a 1.4% band of all four. That is remarkable for a card from the Kepler generation.

The Intel Iris Xe MAX, meanwhile, averages 14,315 points across its benchmarks, ranking in the 56th percentile. Its nearest rivals cluster even tighter: the AMD Radeon Vega 11 is 0.3% ahead (14,352), the NVIDIA GeForce GTX 1070 Ti is actually 0.3% behind (14,277), the NVIDIA GeForce GTX TITAN is 0.4% ahead (14,373), and the AMD Radeon RX Vega 11 is 0.5% ahead (14,385). The Iris Xe MAX sits squarely in the middle of a pack that includes a flagship from 2013 and a high-end card from 2017—an indication that Intel's first discrete GPU effort lands in a strange performance tier.

The head-to-head delta of 5.6% in OpenCL is the only direct comparison, and it mirrors the overall average score gap: the GTX 660 Ti's 15,063 average versus the Iris Xe MAX's 14,315 is a 5.2% difference. In practical terms, the GTX 660 Ti is the faster card in raw compute, but neither GPU is going to embarrass the other in a blind test.

The Verdict

The data paints a clear picture: the NVIDIA GeForce GTX 660 Ti is the stronger performer in the one metric where both are measured. A 5.6% OpenCL win is a win, and the GTX 660 Ti also holds a 748-point advantage in average benchmark score. If your priority is raw compute throughput, the choice is unambiguous.

However, context matters. The GTX 660 Ti is a 2012 discrete card with a 150 W TDP, dual-slot cooler, and two 6-pin power connectors. The Iris Xe MAX is a 2020 integrated-class part with a 25 W TDP, no power connectors, no display outputs, and no slot footprint. These are different products for different physical realities. The GTX 660 Ti needs a power supply rated at 450 W and a PCIe 3.0 x16 slot in a full-size chassis. The Iris Xe MAX runs off a 200 W PSU suggestion, uses PCIe 4.0 x8, and is classified as an IGP—meaning it lives on a motherboard or in a compact system where a discrete card simply will not fit.

So the verdict splits by use case. If you have a desktop with expansion slots and power to spare, the GTX 660 Ti is the better GPU by the numbers. If you are constrained by space, power budget, or the need for an integrated solution that does not require external connectors, the Iris Xe MAX is the only option that fits—and its performance is not far off, trailing by just 5.6% in OpenCL.

For a builder assembling a system today, the GTX 660 Ti's end-of-life status and 2012 release date mean it is a legacy part. The Iris Xe MAX is also end-of-life, but its 2020 release and 10 nm process give it a more modern foundation. Neither is a future-proof investment; both are historical artifacts in the benchmark database.

Where Each One Wins

The NVIDIA GeForce GTX 660 Ti wins in raw compute performance. Its 2.634 TFLOPS FP32 throughput outpaces the Iris Xe MAX's 2.534 TFLOPS, and its 109.8 GTexel/s texture rate is significantly higher than Intel's 79.20 GTexel/s. The GTX 660 Ti also has more shading units (1,344 vs. 768), more TMUs (112 vs. 48), and a much larger memory bandwidth (144.2 GB/s vs. 68.26 GB/s). In any workload that leans on texture fetches or raw shader throughput, the GTX 660 Ti should pull ahead.

The Intel Iris Xe MAX, despite losing the compute race, has its own advantages. It wins on pixel rate: 39.60 GPixel/s versus the GTX 660 Ti's 27.44 GPixel/s. That is a 44% advantage in fill-rate-bound scenarios, which could matter for certain older games or lightweight 2D workloads. The Iris Xe MAX also doubles the memory capacity (4 GB vs. 2 GB), which helps in scenarios where framebuffer usage exceeds 2 GB—though its lower bandwidth means those extra gigabytes come at a speed penalty.

The Iris Xe MAX also has a massive efficiency lead. Its 25 W TDP is one-sixth of the GTX 660 Ti's 150 W. For any system where heat or power draw is a constraint, the Intel part is the clear winner. The GTX 660 Ti requires a 450 W PSU and dual 6-pin connectors; the Iris Xe MAX needs only a 200 W PSU and no connectors at all. The Iris Xe MAX also supports PCIe 4.0 x8, while the GTX 660 Ti is stuck on PCIe 3.0 x16—a minor advantage for bandwidth-sensitive workloads if the rest of the system supports it.

In API support, the Iris Xe MAX is more modern: DirectX 12 (12_1) versus the GTX 660 Ti's DirectX 12 (11_0), and Vulkan 1.4 versus Vulkan 1.2.175. Both support OpenGL 4.6. For newer titles that leverage DX12_1 features, the Intel part has a compatibility edge.

FAQ

Q: Which GPU is faster in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 660 Ti scores 15,113 versus the Intel Iris Xe MAX's 14,315, a 5.6% advantage for the NVIDIA card.

Q: Does the Intel Iris Xe MAX have more memory?

A: Yes, it has 4 GB of LPDDR4X memory compared to the GTX 660 Ti's 2 GB of GDDR5. However, the GTX 660 Ti's memory bandwidth is more than double at 144.2 GB/s versus 68.26 GB/s.

Q: Which GPU has better API support?

A: The Intel Iris Xe MAX supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 660 Ti supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Q: What is the power draw difference?

A: The GTX 660 Ti has a 150 W TDP and requires a 450 W PSU with two 6-pin connectors. The Iris Xe MAX has a 25 W TDP and only needs a 200 W PSU, with no power connectors.

Q: Can the Intel Iris Xe MAX be used as a standalone graphics card?

A: No, it has no display outputs and is classified as an IGP (integrated graphics processor). The GTX 660 Ti has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2 outputs.

Q: How do these GPUs compare to their nearest rivals?

A: The GTX 660 Ti's average score of 15,063 is within 1.4% of the AMD Radeon 680M (15,270) and within 0.9% of the RTX 3050 OEM (15,199). The Iris Xe MAX's average of 14,315 is within 0.5% of the AMD Radeon RX Vega 11 (14,385) and sits right next to the GTX 1070 Ti (14,277, 0.3% behind).

Architecture Differences

The two GPUs come from entirely different architectural lineages. The NVIDIA GeForce GTX 660 Ti uses the GK104 chip built on Kepler architecture, manufactured on a 28 nm process at TSMC. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0M per mm². The chip has 1,344 shading units, 112 texture mapping units, and 24 ROPs.

The Intel Iris Xe MAX uses the DG1 chip on Intel's Generation 12.1 Xe Graphics architecture. It is built on a 10 nm process at Intel's own fabs, with a die size of just 95 mm². Transistor count and density are not listed for this chip. The Iris Xe MAX has 768 shading units, 48 TMUs, and 24 ROPs—the same ROP count as the GTX 660 Ti, but half the TMUs and nearly half the shaders.

The clock behavior differs significantly. The GTX 660 Ti has a 915 MHz base clock and 980 MHz boost. The Iris Xe MAX starts at a low 300 MHz base but boosts to 1650 MHz—a much wider frequency range. Memory clocks also differ: the GTX 660 Ti runs at 1502 MHz (6 Gbps effective) while the Iris Xe MAX runs at 2133 MHz (4.3 Gbps effective).

Compute capabilities diverge on FP16. The Iris Xe MAX explicitly supports FP16 at 5.069 TFLOPS (2:1 ratio), while the GTX 660 Ti lists no FP16 figure. FP32 is close: 2.634 TFLOPS for NVIDIA versus 2.534 TFLOPS for Intel.

The Iris Xe MAX is a Generation 12.1 part, succeeding Intel's "Graphics" line and preceding "Alchemist." The GTX 660 Ti follows the GeForce 500 series and precedes the GeForce 700 series. Both are end-of-life products.

Specification Differences

The most striking difference is form factor and power. The GTX 660 Ti is a dual-slot card measuring 241 mm (9.5 inches) in length, requiring two 6-pin power connectors and a 450 W PSU. The Iris Xe MAX is classified as an IGP with no slot width, no length dimension, no power connectors, and only a 200 W PSU suggestion. The GTX 660 Ti's TDP is 150 W; the Iris Xe MAX's is 25 W.

Memory configurations could not be more different. The GTX 660 Ti uses 2 GB of GDDR5 on a 192-bit bus with 144.2 GB/s bandwidth. The Iris Xe MAX uses 4 GB of LPDDR4X on a 128-bit bus with 68.26 GB/s bandwidth. The GTX 660 Ti has more than double the bandwidth but half the capacity.

The bus interface also differs: PCIe 3.0 x16 for NVIDIA versus PCIe 4.0 x8 for Intel. Display outputs are exclusive to the GTX 660 Ti (2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2); the Iris Xe MAX has none. The GTX 660 Ti's launch MSRP was 299 USD; no launch MSRP is listed for the Iris Xe MAX.

Release dates are eight years apart: August 2012 for the GTX 660 Ti, October 2020 for the Iris Xe MAX. The GTX 660 Ti's process node is 28 nm; the Iris Xe MAX's is 10 nm. The NVIDIA chip is produced by TSMC, the Intel chip by Intel itself. Transistor counts and die sizes differ dramatically: 3,540 million transistors on 294 mm² versus an undisclosed count on 95 mm².