NVIDIA GeForce GT 740 vs NVIDIA GeForce GTX 750 Ti Comparison

NVIDIA
GEFORCE

NVIDIA GeForce GT 740

CORE STATE GK107
VRAM 1024 MB
CLOCK SPEED —
TDP 64 W
BUS WIDTH 128 bit
ARCHITECTURE Kepler
nm
PROCESS 28 nm
LAUNCH DATE 2014
VS
NVIDIA
GEFORCE

GeForce GTX 750 Ti

CORE STATE GM107
VRAM 2 GB
CLOCK SPEED 1085 MHz
TDP 60 W
BUS WIDTH 128 bit
ARCHITECTURE Maxwell
nm
PROCESS 28 nm
LAUNCH DATE 2014

PERFORMANCE BENCHMARKS

geekbench_metal
2,363
5,182
geekbench_opencl
3,847
10,827
geekbench_vulkan
4,083
10,065
3dmark_3dmark_steel_nomad_dx12
N/A
117
passmark_directx_10
N/A
17
passmark_directx_11
N/A
26
passmark_directx_12
N/A
18
passmark_directx_9
N/A
60
passmark_g2d
N/A
528
passmark_g3d
N/A
3,903
passmark_gpu_compute
N/A
1,744

Analysis: NVIDIA GeForce GT 740 vs NVIDIA GeForce GTX 750 Ti

# Head-to-Head Benchmarks

The data paints a decisive picture: the NVIDIA GeForce GTX 750 Ti outperforms the NVIDIA GeForce GT 740 in every single benchmark where both cards were tested. The GTX 750 Ti wins all three head-to-head comparisons, and the margins are substantial.

In Geekbench Metal, the GTX 750 Ti scores 5,182 against the GT 740's 2,363. That represents a 54.4% deficit for the GT 740 — meaning the GTX 750 Ti delivers more than double the Metal performance. This is the closest of the three matchups, yet it is still a blowout by any reasonable measure.

The gap widens dramatically in Geekbench OpenCL. Here the GTX 750 Ti posts 10,827, while the GT 740 manages just 3,847. The delta reaches 64.5%, with the GTX 750 Ti outperforming its rival by nearly threefold. This is the largest margin in the entire head-to-head set, indicating that compute-oriented workloads are where the GTX 750 Ti's architectural advantages shine brightest.

Geekbench Vulkan tells a similar story. The GTX 750 Ti scores 10,065, compared to the GT 740's 4,083. The 59.4% delta places this result between the Metal and OpenCL gaps. Vulkan performance more than doubles from the GT 740 to the GTX 750 Ti, reinforcing the trend: across API-level benchmarks, the newer Maxwell-based card dominates.

Looking at the aggregate picture, the GT 740's average benchmark score across all its tests is 3,431, while the GTX 750 Ti's average is 2,953. This is a curious inversion — the GT 740 actually holds a higher average score despite losing every head-to-head test. The explanation lies in the benchmark mix. The GT 740's average draws from only Geekbench tests (Metal, OpenCL, Vulkan), all of which are compute-heavy. The GTX 750 Ti's average includes additional PassMark tests — DirectX 9, 10, 11, 12, G2D, G3D, and GPU Compute — several of which score very low (17 in DirectX 10, 18 in DirectX 12, 26 in DirectX 11) and drag down the mean. The PassMark G3D score of 3,903 and GPU Compute score of 1,744 are more representative of the GTX 750 Ti's real-world graphics capability, but they cannot offset the low DirectX synthetic results in the average calculation.

The percentile rankings reflect this oddity as well. The GT 740 sits at the 21st percentile among all GPUs, while the GTX 750 Ti sits at the 19th percentile. On the surface, this suggests the GT 740 is the better card — but the head-to-head results directly contradict that interpretation. The percentile field aggregates across all benchmarks in each card's respective suite, and the GTX 750 Ti's inclusion of legacy DirectX tests penalizes its standing. When the same test is run on both cards, the GTX 750 Ti wins every time.

In the nearest-rival comparisons, the GT 740's closest competitor is the NVIDIA Quadro 2000M at 3,434 average score, a mere 0.1% difference. The Intel HD Graphics P4600 trails by 1.2%, the GeForce 920MX sits 2.7% ahead, and the Intel HD Graphics 530 lags by 3%. For the GTX 750 Ti, the nearest rival is the NVIDIA GeForce GTX 860M at 2,967, a 0.5% difference, with the GeForce 820A 1% behind and the Quadro P600 1% ahead. Notably, the NVIDIA GeForce RTX 4060 Ti 8 GB appears as a rival with a 1.4% delta — a modern card that scores within a hair of the GTX 750 Ti's average, which says more about the benchmark suite's weighting than about real-world parity.

# Where Each One Wins

The GT 740's case is narrow but real. In the only benchmarks where both cards appear — Geekbench Metal, OpenCL, and Vulkan — the GT 740 loses all three. There is no test in the data where the GT 740 comes out ahead. Its wins, if they can be called that, are entirely contextual. The GT 740's average benchmark score of 3,431 is higher than the GTX 750 Ti's 2,953, but that is an artifact of test selection, not performance superiority.

For the GT 740, the practical takeaway is limited to scenarios where the card's lower compute ceiling is acceptable. Its 384 shading units and 32 texture mapping units are less than half of the GTX 750 Ti's 640 shading units and 40 TMUs, and its FP32 throughput of 762.6 GFLOPS is barely above half of the GTX 750 Ti's 1,388.8 GFLOPS. The GT 740's pixel rate of 7.944 GPixel/s is less than half of the GTX 750 Ti's 17.36 GPixel/s. These are not speculative differences — they are measured specifications that translate directly to the benchmark deltas.

The GTX 750 Ti wins every category that matters. It holds a 54.4% advantage in Metal, a 64.5% advantage in OpenCL, and a 59.4% advantage in Vulkan. Its memory subsystem is also superior: 2 GB of GDDR5 versus 1 GB, with bandwidth of 86.40 GB/s versus 80.19 GB/s. The GTX 750 Ti's base clock of 1,020 MHz and boost clock of 1,085 MHz are explicit in the data, while the GT 740's clocks are not listed at all — a sign that the GT 740 operates at lower frequencies, consistent with its reduced power envelope and older Kepler design.

The GTX 750 Ti also wins on efficiency. Its TDP is 60 watts, lower than the GT 740's 64 watts, despite delivering dramatically higher performance. The GTX 750 Ti requires no power connectors, while the GT 740 needs a single 6-pin connector. Both cards suggest a 250-watt PSU, but the GTX 750 Ti achieves that with headroom to spare. The GTX 750 Ti's transistor density of 12.6 million per square millimeter on a 148 mm² die — versus the GT 740's 10.8 million per mm² on a 118 mm² die — indicates a more modern design that extracts more performance from each transistor.

# The Verdict

The data is unambiguous. The NVIDIA GeForce GTX 750 Ti is the superior card in every head-to-head benchmark, with deltas ranging from 54.4% to 64.5%. There are zero wins for the GT 740 in the direct comparisons. If the decision is between these two specific cards, the GTX 750 Ti is the only rational choice for any workload that involves Metal, OpenCL, or Vulkan APIs.

The GT 740's only statistical advantage — its higher average benchmark score of 3,431 versus 2,953 — is a measurement artifact. The GTX 750 Ti's average is dragged down by low-scoring PassMark DirectX tests (17 in DX10, 18 in DX12, 26 in DX11) that have no counterpart in the GT 740's benchmark suite. When the same test runs on both cards, the GTX 750 Ti wins outright. The GT 740's 21st percentile ranking versus the GTX 750 Ti's 19th percentile is likewise misleading; the percentile field reflects the full benchmark set for each card, not a like-for-like comparison.

Who should pick the GT 740? Strictly from the data, only someone who cannot use the GTX 750 Ti at all. The GT 740 is a single-slot card with a 64-watt TDP, which is lower than the GTX 750 Ti's 60-watt TDP — wait, that is not lower; the GTX 750 Ti is actually more power-efficient. The GT 740 does have the same 145 mm length and same display outputs (2x DVI, 1x mini-HDMI 1.4a) as the GTX 750 Ti. But there is no performance scenario in the data where the GT 740 comes out ahead.

Who should pick the GTX 750 Ti? Anyone who wants more than double the Metal performance, nearly triple the OpenCL performance, or more than double the Vulkan performance. The GTX 750 Ti also offers twice the memory capacity (2 GB versus 1 GB), higher memory bandwidth, more shading units, more TMUs, higher pixel rate, higher texture rate, and nearly double the FP32 throughput. It does all of this while drawing less power and requiring no external power connector.

The GTX 750 Ti's launch MSRP was 149 USD. The GT 740's launch MSRP was 89 USD. The GTX 750 Ti costs 67% more at launch, but delivers 118% more performance in the closest benchmark and 181% more in the widest gap. The data shows no scenario where the GT 740's lower price translates into competitive performance.

# FAQ

Q: Which card wins in Geekbench Metal?

A: The NVIDIA GeForce GTX 750 Ti wins decisively, scoring 5,182 against the GT 740's 2,363, a 54.4% advantage.

Q: What is the largest performance gap between the two cards?

A: The largest gap is in Geekbench OpenCL, where the GTX 750 Ti scores 10,827 versus the GT 740's 3,847, a 64.5% difference.

Q: Does the GT 740 win any head-to-head benchmark?

A: No. The GT 740 loses all three head-to-head tests: Geekbench Metal, OpenCL, and Vulkan. Its wins count is zero.

Q: How do the average benchmark scores compare?

A: The GT 740 has a higher average benchmark score (3,431) than the GTX 750 Ti (2,953), but this is due to the GTX 750 Ti's inclusion of low-scoring PassMark DirectX tests (17 in DX10, 18 in DX12, 26 in DX11) that are absent from the GT 740's suite.

Q: Which card has more memory?

A: The GTX 750 Ti has 2 GB of GDDR5 memory, while the GT 740 has 1 GB. The GTX 750 Ti also has higher bandwidth at 86.40 GB/s versus 80.19 GB/s.

Q: What are the power requirements for each card?

A: The GTX 750 Ti has a 60-watt TDP and requires no power connectors, while the GT 740 has a 64-watt TDP and requires a single 6-pin connector. Both list a 250-watt suggested PSU.

# Architecture Differences

The two cards represent different architectural generations from NVIDIA, both built on the same 28 nm process at TSMC. The GT 740 uses the GK107 chip based on the Kepler architecture, while the GTX 750 Ti uses the GM107 chip based on the Maxwell architecture. Both belong to the GeForce 700 generation and share the same predecessor (GeForce 600) and successor (GeForce 900).

The GK107 die measures 118 mm² and contains 1,270 million transistors, yielding a transistor density of 10.8 million per mm². The GM107 die is larger at 148 mm² and packs 1,870 million transistors, achieving a higher density of 12.6 million per mm². This 600-million-transistor increase and denser layout reflect Maxwell's more efficient execution architecture.

Core configuration differs substantially. The GT 740 has 384 shading units, 32 texture mapping units, and 16 ROPs. The GTX 750 Ti has 640 shading units, 40 TMUs, and 16 ROPs. The shading unit count is 66.7% higher on the GTX 750 Ti, and TMUs are 25% higher. ROPs are identical at 16, which means the GTX 750 Ti's higher pixel rate of 17.36 GPixel/s comes from its higher clock speeds rather than additional ROP hardware.

Clock speeds tell a partial story. The GTX 750 Ti has an explicit base clock of 1,020 MHz and a boost clock of 1,085 MHz. The GT 740's base and boost clocks are not listed in the data, but its memory clock is 1,253 MHz (5 Gbps effective) versus the GTX 750 Ti's 1,350 MHz (5.4 Gbps effective). The GTX 750 Ti's memory bandwidth of 86.40 GB/s exceeds the GT 740's 80.19 GB/s despite both using a 128-bit bus — the higher memory clock does the work.

Compute throughput is where Maxwell's efficiency shows. The GT 740 delivers 762.6 GFLOPS of FP32 performance, while the GTX 750 Ti delivers 1,388.8 GFLOPS — an 82% advantage. Texture rate follows a similar pattern: 31.78 GTexel/s for the GT 740 versus 43.40 GTexel/s for the GTX 750 Ti, a 36.6% gap. Pixel rate is more lopsided, with the GTX 750 Ti's 17.36 GPixel/s more than doubling the GT 740's 7.944 GPixel/s.

API support shows a notable divergence. Both cards support DirectX 12 (11_0) and OpenGL 4.6, but the Vulkan versions differ: the GT 740 supports Vulkan 1.2.175, while the GTX 750 Ti supports Vulkan 1.4. This newer Vulkan implementation on the Maxwell card likely contributes to its Vulkan benchmark advantage, though the compute hardware gap is the primary factor.

Memory capacity differs as well: 1 GB on the GT 740 versus 2 GB on the GTX 750 Ti, both GDDR5 on a 128-bit bus. The 2 GB capacity gives the GTX 750 Ti materially more headroom for modern workloads.

Power characteristics favor the Maxwell card. Despite delivering roughly double the performance in several metrics, the GTX 750 Ti has a TDP of 60 watts versus the GT 740's 64 watts. The GTX 750 Ti requires no power connectors, while the GT 740 needs a single 6-pin. Both are single-slot cards with identical dimensions (145 mm / 5.7 inches length) and identical display outputs (2x DVI, 1x mini-HDMI 1.4a). Both use a PCIe 3.0 x16 interface, and both are end-of-life products.

The release dates differ by about three months: the GTX 750 Ti launched on 2014-02-17, and the GT 740 launched on 2014-05-28. The GTX 750 Ti's earlier release with a more advanced architecture set the performance baseline that the GT 740 never approached.

DETAILED SPECIFICATIONS

SPECIFICATION
GT 740
GTX 750 Ti
Core Specs
Shading Units
384
640 +66.7%
Shaders
384
640 +66.7%
TMUs
32
40 +25.0%
ROPs
16
16 0.0%
Clocks
Base Clock
—
1020 MHz
Boost Clock
—
1085 MHz
GPU Clock
993 MHz
—
Memory Clock
1253 MHz 5 Gbps effective
1350 MHz 5.4 Gbps effective
Memory
Memory Size
1024 MB
2 GB
VRAM (MB)
1,024
2,048 +100.0%
Memory Type
GDDR5
GDDR5
Memory Bus
128 bit
128 bit
Bandwidth
80.19 GB/s
86.40 GB/s
Cache
L1 Cache
16 KB (per SMX)
64 KB (per SMM)
L2 Cache
256 KB
2 MB
Performance
Pixel Rate
7.944 GPixel/s
17.36 GPixel/s
Texture Rate
31.78 GTexel/s
43.40 GTexel/s
FP32 (TFLOPS)
762.6 GFLOPS
1,388.8 GFLOPS
FP64 (TFLOPS)
31.78 GFLOPS (1:24)
43.40 GFLOPS (1:32)
Power
TDP
64 W
60 W
TDP (W)
64
60 -6.3%
Suggested PSU
250 W
250 W
Power Connectors
1x 6-pin
None
Architecture
Architecture
Kepler
Maxwell
GPU Name
GK107
GM107
Generation
GeForce 700
GeForce 700
Process Size
28 nm
28 nm
Transistors
1,270 million
1,870 million
Die Size
118 mm²
148 mm²
Foundry
TSMC
TSMC
Density
10.8M / mm²
12.6M / mm²
API Support
DirectX
12 (11_0)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.2.175
1.4
OpenCL
3.0
3.0
CUDA
3.0
5.0
Shader Model
6.5 (5.1)
6.7 (5.1)
Physical
Slot Width
Single-slot
Single-slot
Length
145 mm 5.7 inches
145 mm 5.7 inches
Outputs
2x DVI1x mini-HDMI 1.4a
2x DVI1x mini-HDMI 1.4a
Bus Interface
PCIe 3.0 x16
PCIe 3.0 x16
Other
Launch Price
89 USD
149 USD
Production
End-of-life
End-of-life
Predecessor
GeForce 600
GeForce 600
Successor
GeForce 900
GeForce 900
View GeForce GT 740 Details View GeForce GTX 750 Ti Details