GPU Comparison

NVIDIA
GEFORCE

NVIDIA GeForce GT 640

CORE STATE GK107
VRAM 2 GB
CLOCK SPEED
TDP 65 W
BUS WIDTH 128 bit
ARCHITECTURE Kepler
nm
PROCESS 28 nm
LAUNCH DATE 2012
VS
NVIDIA
GEFORCE

Quadro 2000M

CORE STATE GF106
VRAM 2 GB
CLOCK SPEED
TDP 55 W
BUS WIDTH 128 bit
ARCHITECTURE Fermi
nm
PROCESS 40 nm
LAUNCH DATE 2011

PERFORMANCE BENCHMARKS

geekbench_metal
2,086
N/A
geekbench_opencl
3,736
3,434
geekbench_vulkan
3,809
N/A

Analysis: NVIDIA GeForce GT 640 vs NVIDIA Quadro 2000M

The Verdict

The data presents a clear split between two end-of-life NVIDIA parts aimed at different tasks. The Quadro 2000M, a Fermi-generation mobile workstation chip, and the GeForce GT 640, a Kepler-generation desktop card, are separated by more than just their form factors. Benchmark results show the GT 640 winning the only head-to-head test, the Geekbench OpenCL score, by a margin of 8.1%. The Quadro 2000M scores 3434, while the GT 640 scores 3736. This single data point suggests the GT 640 holds a raw compute advantage, but the story is more nuanced when considering their respective positions in the GPU landscape.

Looking at the percentile rankings, the two are nearly inseparable. The Quadro 2000M sits at the 21st percentile of all GPUs, while the GT 640 rests at the 20th percentile. This means both are firmly in the lower tier of performance, with the GT 640 being marginally lower-ranked despite its higher benchmark score. This discrepancy highlights that the GT 640’s average benchmark score of 3210 is dragged down by its additional tests, Metal and Vulkan, which the Quadro lacks entirely. The Quadro’s average benchmark score is 3434, which is exactly its OpenCL score, while the GT 640’s average is 3210, pulled below its OpenCL result by the lower Metal and Vulkan numbers.

For buyers, the choice hinges on platform and software needs. The Quadro 2000M is a mobile MXM module with portable-device-dependent display outputs, making it a laptop upgrade part. The GT 640 is a single-slot desktop card with PCIe 3.0 x16 interface and fixed display outputs (DVI, HDMI 1.4a, DisplayPort 1.2). The data shows the GT 640 is the faster compute part in OpenCL, but the Quadro’s workstation lineage suggests it was designed for certified professional applications. However, the fact pack contains no professional application benchmarks to confirm this. Strictly from the data, the GT 640 is the better choice for raw OpenCL performance on a desktop, while the Quadro 2000M is the only option for MXM-equipped laptops.

Architecture Differences

The architectural gap between these two GPUs is significant. The Quadro 2000M uses the GF106 chip, built on Fermi architecture, which is fabricated on a 40 nm process at TSMC. The GT 640 uses the GK107 chip, based on Kepler architecture, manufactured on a 28 nm process, also at TSMC. This process shrink from 40 nm to 28 nm is a fundamental difference. The GT 640 packs more transistors into a smaller die: 1,270 million transistors on a 118 mm² die, versus 1,170 million transistors on a 238 mm² die for the Quadro. This yields a transistor density of 10.8 million per square millimeter for the GT 640, more than double the Quadro’s 4.9 million per square millimeter.

The shading unit count tells a similar story. The GT 640 has 384 shading units, exactly double the Quadro’s 192. Both have 32 texture mapping units and 16 render output units, so the texture and pixel throughput differences come from clock speeds rather than unit counts. The GT 640’s pixel rate is 7.216 GPixel/s versus the Quadro’s 4.400 GPixel/s, and its texture rate is 28.86 GTexel/s versus 17.60 GTexel/s. The FP32 compute is 692.7 GFLOPS for the GT 640 versus 422.4 GFLOPS for the Quadro, a 64% advantage for the newer card.

Memory configurations are nearly identical. Both use 2 GB of DDR3 with a 128-bit bus width. The bandwidth is essentially the same: 28.80 GB/s for the Quadro and 28.51 GB/s for the GT 640. The memory clocks are close, with the Quadro at 1800 Mbps effective and the GT 640 at 1782 Mbps effective. Power consumption differs modestly, with the Quadro at 55 W and the GT 640 at 65 W. The GT 640 also has a suggested PSU rating of 250 W, while the Quadro has none listed. API support shows another divergence: both support DirectX 12 (11_0) and OpenGL 4.6, but the GT 640 adds Vulkan 1.2.175 support, while the Quadro has no Vulkan API listed.

Where Each One Wins

The GT 640 wins the only direct benchmark comparison, the Geekbench OpenCL test, with a score of 3736 versus 3434 for the Quadro 2000M. That is an 8.1% lead. This suggests the GT 640 is the stronger choice for OpenCL compute workloads. The GT 640 also offers a broader API surface with Vulkan support, which could be relevant for modern cross-platform applications. Its desktop form factor with dedicated display outputs (DVI, HDMI 1.4a, DisplayPort 1.2) makes it easy to connect to a variety of monitors.

The Quadro 2000M, despite losing the OpenCL test, holds its own in specific contexts. Its MXM module form factor means it is designed for portable systems, and the display output is described as "Portable Device Dependent," indicating it relies on the laptop’s own panel and connectors. The Quadro’s Fermi architecture, while older, was a workstation-focused design. Its nearest rivals include the NVIDIA GeForce GT 740 (average score 3431, delta 0.1%), Intel HD Graphics P4600 (3389, delta 1.3%), and Intel HD Graphics 530 (3332, delta 3.1%). This places the Quadro just ahead of those integrated and low-end parts. The GT 640’s nearest rivals include the Intel Arc Pro B60 (3182, delta 0.9%), NVIDIA Quadro P1000 (3163, delta 1.5%), and NVIDIA GeForce 920M (3287, delta negative 2.3%). So the GT 640 is positioned relative to slightly different competition.

For gaming or general desktop use, the GT 640’s higher raw scores and Vulkan support make it the more capable part. For a laptop requiring a specific MXM upgrade, the Quadro 2000M is the only viable option between the two. The data does not support a clear workstation-performance advantage for the Quadro, as no professional benchmarks are present. The GT 640’s additional Metal and Vulkan benchmark scores (2086 and 3809 respectively) show it can handle those APIs, while the Quadro has no such data.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The NVIDIA GeForce GT 640 scores 3736 in Geekbench OpenCL, while the NVIDIA Quadro 2000M scores 3434. The GT 640 leads by 8.1%.

Q: What is the difference in shading unit counts?

A: The GeForce GT 640 has 384 shading units, which is exactly double the Quadro 2000M’s 192 shading units.

Q: Do both cards support the same APIs?

A: Both support DirectX 12 (11_0) and OpenGL 4.6. However, the GeForce GT 640 also supports Vulkan 1.2.175, while the Quadro 2000M has no Vulkan API listed.

Q: How do their memory configurations compare?

A: Both have 2 GB of DDR3 memory with a 128-bit bus width. The Quadro 2000M has a bandwidth of 28.80 GB/s, and the GT 640 has 28.51 GB/s, making them nearly identical.

Q: What are the form factor differences?

A: The Quadro 2000M is an MXM Module with a bus interface of MXM-A (3.0) and portable-device-dependent display outputs. The GT 640 is a single-slot desktop card with a PCIe 3.0 x16 interface and fixed outputs: 1x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2.

Q: How do the two compare in terms of overall benchmark averages?

A: The Quadro 2000M has an average benchmark score of 3434, while the GT 640’s average is 3210. The GT 640’s average is lower because it includes Metal and Vulkan scores (2086 and 3809) alongside its higher OpenCL score.

Head-to-Head Benchmarks

The only direct head-to-head benchmark available is Geekbench OpenCL. In this test, the GeForce GT 640 scores 3736 against the Quadro 2000M’s 3434. The delta percentage is -8.1%, meaning the GT 640 outperforms the Quadro by that margin. This is the sole data point for direct comparison, and it decisively favors the GT 640.

To contextualize this win, look at the rival sets. The Quadro 2000M’s nearest rival is the GeForce GT 740, which scores 3431, just 0.1% behind. This means the Quadro is extremely close to the GT 740 in average performance. The GT 640’s nearest rival is the Intel Arc Pro B60 at 3182, which is 0.9% behind the GT 640’s average. So the GT 640 sits slightly above the Arc Pro B60 but below the GeForce 920M (3287, delta -2.3%) and GT 730M (3316, delta -3.2%). The Quadro, meanwhile, is above the Intel HD Graphics P4600 (3389, delta 1.3%) and Intel HD Graphics 530 (3332, delta 3.1%).

The GT 640’s average score of 3210 is lower than its OpenCL score of 3736 because of the Metal result of 2086. This Metal score is significantly lower, pulling the average down. The Vulkan score of 3809 is slightly higher than the OpenCL score. This suggests the GT 640’s performance varies by API, with Vulkan being its strongest showing. The Quadro has no such variation in the data, as its only benchmark is OpenCL. The head-to-head result therefore reflects only one API, and the GT 640’s win in OpenCL does not necessarily predict its performance in other workloads, though Vulkan appears even stronger for the GT 640 based on its score.

Specification Differences

The two GPUs differ across nearly every architectural and physical specification. The process node is a major divergence: the Quadro 2000M uses 40 nm, while the GT 640 uses 28 nm. Transistor counts are close but not equal, with 1,170 million for the Quadro and 1,270 million for the GT 640. Die size is dramatically different: the Quadro’s die is 238 mm², while the GT 640’s is 118 mm². This results in transistor densities of 4.9M / mm² for the Quadro and 10.8M / mm² for the GT 640.

Shading units are 192 for the Quadro versus 384 for the GT 640. Both have 32 TMUs and 16 ROPs, so those are the same. Pixel rate is 4.400 GPixel/s for the Quadro versus 7.216 GPixel/s for the GT 640. Texture rate is 17.60 GTexel/s versus 28.86 GTexel/s. FP32 compute is 422.4 GFLOPS versus 692.7 GFLOPS. Memory clocks are 1800 Mbps effective for the Quadro and 1782 Mbps for the GT 640. Bandwidth is 28.80 GB/s versus 28.51 GB/s. TDP is 55 W for the Quadro and 65 W for the GT 640.

The slot width differs: MXM Module for the Quadro versus Single-slot for the GT 640. The bus interface is MXM-A (3.0) versus PCIe 3.0 x16. Display outputs are "Portable Device Dependent" for the Quadro, while the GT 640 offers 1x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The GT 640 has a suggested PSU of 250 W, while the Quadro has none. The GT 640 has a length of 145 mm (5.7 inches), while the Quadro has no listed dimensions. API support differs: both have DirectX 12 (11_0) and OpenGL 4.6, but only the GT 640 lists Vulkan 1.2.175. The GT 640 also has a launch MSRP of 99 USD, which is the only price-related field in the data. Release dates differ: the Quadro launched on 2011-01-12, and the GT 640 on 2012-06-04.

DETAILED SPECIFICATIONS

SPECIFICATION
GT 640
Quadro 2000M
Core Specs
Shading Units
384
192 -50.0%
Shaders
384
192 -50.0%
TMUs
32
32 0.0%
ROPs
16
16 0.0%
SM Count
4
Clocks
GPU Clock
902 MHz
550 MHz
Shader Clock
1100 MHz
Memory Clock
891 MHz 1782 Mbps effective
900 MHz 1800 Mbps effective
Memory
Memory Size
2 GB
2 GB
VRAM (MB)
2,048
2,048 0.0%
Memory Type
DDR3
DDR3
Memory Bus
128 bit
128 bit
Bandwidth
28.51 GB/s
28.80 GB/s
Cache
L1 Cache
16 KB (per SMX)
64 KB (per SM)
L2 Cache
256 KB
256 KB
Performance
Pixel Rate
7.216 GPixel/s
4.400 GPixel/s
Texture Rate
28.86 GTexel/s
17.60 GTexel/s
FP32 (TFLOPS)
692.7 GFLOPS
422.4 GFLOPS
FP64 (TFLOPS)
28.86 GFLOPS (1:24)
35.20 GFLOPS (1:12)
Power
TDP
65 W
55 W
TDP (W)
65
55 -15.4%
Suggested PSU
250 W
Power Connectors
None
None
Architecture
Architecture
Kepler
Fermi
GPU Name
GK107
GF106
Generation
GeForce 600
Quadro Fermi-M (x000M)
Process Size
28 nm
40 nm
Transistors
1,270 million
1,170 million
Die Size
118 mm²
238 mm²
Foundry
TSMC
TSMC
Density
10.8M / mm²
4.9M / mm²
API Support
DirectX
12 (11_0)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.2.175
OpenCL
3.0
1.1
CUDA
3.0
2.1
Shader Model
6.5 (5.1)
5.1
Physical
Slot Width
Single-slot
MXM Module
Length
145 mm 5.7 inches
Outputs
1x DVI1x HDMI 1.4a1x DisplayPort 1.2
Portable Device Dependent
Bus Interface
PCIe 3.0 x16
MXM-A (3.0)
Other
Launch Price
99 USD
Production
End-of-life
End-of-life
Predecessor
GeForce 500
Quadro FX Mobile
Successor
GeForce 700
Quadro Kepler-M
View GeForce GT 640 Details View Quadro 2000M Details