NVIDIA GeForce GT 640 vs NVIDIA Quadro 2000 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA Quadro 2000 records an average benchmark score of 3898, while the NVIDIA GeForce GT 640 averages 3210. The Quadro 2000 sits in the 23rd percentile of all GPUs, compared to the GT 640's 20th percentile.

Q: How do the two cards compare in the Geekbench OpenCL test?

A: In the only head-to-head benchmark available, the Quadro 2000 scores 3898 against the GT 640's 3736, a 4.3% advantage for the Quadro. This is the sole direct comparison recorded in the database.

Q: What are the closest rivals to the Quadro 2000 in the database?

A: The AMD Radeon R5 Graphics is nearly identical at 3883 (0.4% behind), the NVIDIA Quadro K2000D scores 3919 (0.5% ahead), the Quadro 2000D scores 3930 (0.8% ahead), and the GeForce GT 745M scores 3953 (1.4% ahead).

Q: Which card has more shading units?

A: The GeForce GT 640 has 384 shading units, exactly double the 192 found on the Quadro 2000. Despite this, the Quadro still wins the OpenCL benchmark by 4.3%.

Q: What is the memory situation for each card?

A: The Quadro 2000 uses 1024 MB of GDDR5 with 41.60 GB/s bandwidth, while the GT 640 has 2 GB of DDR3 with 28.51 GB/s. Both use a 128-bit bus, but the Quadro's faster memory type compensates for its smaller capacity.

Q: Which card supports newer PCIe and API features?

A: The GT 640 uses PCIe 3.0 x16 and supports Vulkan 1.2.175, while the Quadro 2000 uses PCIe 2.0 x16 and has no Vulkan support listed. Both support DirectX 12 (11_0) and OpenGL 4.6.

Architecture Differences

The two NVIDIA cards come from different architectural generations. The Quadro 2000 is built on the Fermi architecture with the GF106 chip, fabricated on a 40 nm process at TSMC. The GT 640 uses the Kepler architecture with the GK107 chip, manufactured on a 28 nm process, also at TSMC. This process shrink is significant: the GT 640 packs 1,270 million transistors into a 118 mm² die, achieving a transistor density of 10.8M / mm². The Quadro 2000 contains 1,170 million transistors on a much larger 238 mm² die, resulting in a density of just 4.9M / mm².

The Kepler architecture in the GT 640 doubles the shading unit count to 384 versus the Quadro's 192. Both cards share 32 texture mapping units and 16 raster operation units, so the core difference lies in shader throughput. The GT 640 also has a higher pixel rate at 7.216 GPixel/s versus 5.000 GPixel/s, and a higher texture rate at 28.86 GTexel/s versus 20.00 GTexel/s. FP32 performance follows the same pattern: 692.7 GFLOPS for the GT 640 versus 480.0 GFLOPS for the Quadro.

Memory architecture differs substantially. The Quadro 2000 uses GDDR5 with a 128-bit bus and achieves 41.60 GB/s bandwidth. The GT 640 uses DDR3 on the same 128-bit bus but only reaches 28.51 GB/s. The GT 640 compensates with double the capacity at 2 GB versus 1 GB. The Quadro's memory clock is listed at 650 MHz (2.6 Gbps effective), while the GT 640's memory runs at 891 MHz (1782 Mbps effective). Despite the higher clock, the DDR3 type limits the GT 640's bandwidth.

Power and connectivity also differ. The Quadro 2000 has a TDP of 62 W, and the GT 640 is slightly higher at 65 W. Both are single-slot cards with no power connectors and a suggested PSU of 250 W. The Quadro is longer at 178 mm (7 inches) with a height of 111 mm (4.4 inches), while the GT 640 is shorter at 145 mm (5.7 inches) with no height listed. Display outputs show the Quadro's professional focus with 1x DVI and 2x DisplayPort, while the GT 640 offers 1x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The GT 640 supports PCIe 3.0 x16 and Vulkan 1.2.175, neither of which the Quadro offers.

Head-to-Head Benchmarks

The database contains one direct benchmark comparison between these two cards: Geekbench OpenCL. The Quadro 2000 scores 3898, and the GT 640 scores 3736. This gives the Quadro a 4.3% win. It is a close result, but it is decisive given the architectural differences.

The Quadro 2000's advantage comes despite the GT 640's higher theoretical specifications. The GT 640 has double the shading units, higher pixel rate, higher texture rate, and higher FP32 throughput. Yet in the OpenCL workload, the Quadro's GDDR5 memory with 41.60 GB/s bandwidth outperforms the GT 640's DDR3 at 28.51 GB/s. This suggests memory bandwidth matters more than raw compute for this particular test.

Looking at the broader performance context, the Quadro 2000's score of 3898 places it just above the AMD Radeon R5 Graphics at 3883 (0.4% difference). The Quadro 2000D slightly edges it out at 3930 (0.8% ahead), and the Quadro K2000D is 0.5% ahead. This clustering indicates the Quadro 2000 sits near the top of its immediate performance tier.

The GT 640's average benchmark score of 3210 is pulled down by its other test results. It scores 2086 in Geekbench Metal and 3809 in Geekbench Vulkan, but only 3736 in OpenCL. Its nearest rivals include the Intel Arc Pro B60 at 3182 (0.9% behind), the Quadro P1000 at 3163 (1.5% behind), the GeForce 920M at 3287 (2.3% ahead), and the GT 730M at 3316 (3.2% ahead). The GT 640's OpenCL score is actually its second-best result, but the Metal score significantly lowers its average.

The head-to-head result is a single data point, but it correlates with the percentile rankings. The Quadro 2000 is in the 23rd percentile of all GPUs, while the GT 640 is in the 20th. Both are low-end performers by modern standards, but the Quadro holds a measurable edge in the one test where they are directly compared.

The Verdict

For compute workloads measured by OpenCL, the NVIDIA Quadro 2000 is the better card. The 4.3% head-to-head win is consistent with its higher memory bandwidth and its higher percentile ranking. Users prioritizing raw OpenCL performance over everything else should choose the Quadro 2000.

The GT 640 offers advantages in other areas. It has double the memory capacity at 2 GB, which matters for applications that need larger working sets. It also supports Vulkan 1.2.175, a capability the Quadro lacks entirely. Its PCIe 3.0 interface provides more bandwidth to the host system than the Quadro's PCIe 2.0. For workloads that leverage Vulkan or require more memory, the GT 640 is the practical choice.

The physical footprint also differs. The GT 640 is shorter at 145 mm, making it easier to fit in compact cases. The Quadro is 178 mm long and adds DisplayPort outputs, while the GT 640 includes HDMI. Both are single-slot with no external power connectors.

Professionals who need the Quadro's specific feature set, such as its dual DisplayPort outputs and its Fermi-based driver support, will find the Quadro 2000 adequate for basic tasks. Gamers or general users who want broader API support and twice the memory will prefer the GT 640. The data shows a narrow performance gap in favor of the Quadro, but the GT 640's feature set makes it more versatile for modern software.

The GT 640's launch MSRP was 99 USD, while the Quadro 2000 launched at 599 USD. The price difference is substantial, but the performance difference is small. For new purchases, the GT 640 offers more features per dollar on paper. For specific professional workflows that rely on the Quadro's validated drivers, the Quadro remains the safer option.

Specification Differences

The two cards differ across nearly every specification category. The process node shifts from 40 nm on the Quadro 2000 to 28 nm on the GT 640. Transistor count increases from 1,170 million to 1,270 million, while die size shrinks from 238 mm² to 118 mm². Transistor density more than doubles from 4.9M / mm² to 10.8M / mm².

Shading units double from 192 to 384, while TMUs remain at 32 and ROPs stay at 16. Pixel rate rises from 5.000 GPixel/s to 7.216 GPixel/s. Texture rate increases from 20.00 GTexel/s to 28.86 GTexel/s. FP32 performance grows from 480.0 GFLOPS to 692.7 GFLOPS.

Memory capacity doubles from 1024 MB to 2 GB. Memory type changes from GDDR5 to DDR3. Bus width stays at 128 bit, but bandwidth drops from 41.60 GB/s to 28.51 GB/s. Memory clock changes from 650 MHz (2.6 Gbps effective) to 891 MHz (1782 Mbps effective).

TDP rises slightly from 62 W to 65 W. Both are single-slot with no power connectors and a 250 W suggested PSU. The Quadro is 178 mm long (7 inches) and 111 mm tall (4.4 inches); the GT 640 is 145 mm long (5.7 inches) with no height listed. Bus interface upgrades from PCIe 2.0 x16 to PCIe 3.0 x16.

Display outputs change from 1x DVI and 2x DisplayPort to 1x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. API support differs: both have DirectX 12 (11_0) and OpenGL 4.6, but only the GT 640 lists Vulkan 1.2.175. Release dates are December 2010 for the Quadro and June 2012 for the GT 640.

Where Each One Wins

The Quadro 2000 wins in the OpenCL benchmark, the only direct comparison available. Its GDDR5 memory provides 41.60 GB/s of bandwidth, which is 46% higher than the GT 640's 28.51 GB/s. This bandwidth advantage likely explains its 4.3% OpenCL lead despite the GT 640's higher compute throughput. The Quadro also has dual DisplayPort outputs, which are common in professional display setups.

The GT 640 wins in areas that the OpenCL test does not measure. Its 2 GB memory capacity is double the Quadro's 1 GB, which helps with larger textures or datasets. Its Vulkan 1.2.175 support enables modern graphics APIs that the Quadro cannot use. The PCIe 3.0 x16 interface doubles the bandwidth available to the host system compared to PCIe 2.0. The shorter 145 mm length makes it easier to install in space-constrained builds.

For compute tasks that rely on OpenCL, the Quadro 2000 is the winner. For tasks that use Vulkan, need more memory, or require modern API support, the GT 640 is the better choice. The GT 640's higher FP32 performance (692.7 GFLOPS versus 480.0 GFLOPS) suggests it would win compute workloads that scale with shader count, but the recorded OpenCL test does not show that. The database only records one head-to-head test, and in that test, the Quadro wins.

Users with professional display requirements will appreciate the Quadro's dual DisplayPort outputs and its Quadro driver validation. Users with consumer display needs will prefer the GT 640's HDMI output. The GT 640's lower launch MSRP of 99 USD versus the Quadro's 599 USD makes it the more accessible option for general use, though the Quadro's higher price reflects its professional positioning.

In summary, the Quadro 2000 wins on memory bandwidth and the single recorded benchmark. The GT 640 wins on memory capacity, API support, PCIe generation, and physical size. Neither card is dominant across all criteria, so the choice depends on which factors matter most for the intended workload.