Skip to content
RU

GeForce GT 640 vs GeForce GTX 650 Ti: specs and benchmark comparison

ANVIDIA · Kepler

NVIDIA GeForce GT 640

  • 2 GB DDR3
  • 65 W

June 5, 2012 · $99 at launch

BNVIDIA · Kepler

NVIDIA GeForce GTX 650 Ti

  • 1 GB GDDR5
  • 110 W

October 9, 2012 · $149 at launch

Short verdict

Not enough data for a performance conclusion: the database has no comparable benchmarks for this pair. Below are the key differences in official specifications.

Compare with AI

Advantages

Only measured and rated values: a noticeable benchmark gap and specs where more is better.

Where GeForce GT 640 is better3

  • Video memory 1 GB more2 GB vs 1 GB

  • Board power 45 W lower65 W (TDP) vs 110 W (TDP)

  • Launch price $50 lower$99 vs $149

Where GeForce GTX 650 Ti is better2

  • Shader units 384 more768 (CUDA cores) vs 384 (CUDA cores)

  • Memory bandwidth 57.9 GB/s more86.4 GB/s vs 28.5 GB/s

Benchmarks

Results are compared within one benchmark, one version and the same conditions.

  • Blender Benchmark 2.93.1

    GPU, bmw27 (CUDA) · s, lower is better

    No advantage conclusion: small sample in one of the results.

    About the measurement and source
    A · NVIDIA GeForce GT 640

    1,298.5 s · community measurement, sample 1 · November 14, 2026

    Blender Open Data (Blender Foundation)

    Median of 1 submissions (0 outliers excluded), interquartile range 1298.45–1298.45; submissions 2022-01-07 to 2022-01-07; CUDA backend. Snapshot 2026-11-14.

    B · NVIDIA GeForce GTX 650 Ti

    559.7 s · community measurement, sample 1 · November 14, 2026

    Blender Open Data (Blender Foundation)

    Median of 1 submissions (0 outliers excluded), interquartile range 559.68–559.68; submissions 2021-08-07 to 2021-08-07; CUDA backend. Snapshot 2026-11-14.

    difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.

    Render time of a single Cycles scene without synchronization. Median of submissions after filtering.

Reference results (21)
BenchmarkABDifference
Blender Benchmark 2.90GPU, bmw27 (CUDA)1,209.9537.8reference
Blender Benchmark 2.92GPU, bmw27 (CUDA)1,241.9537.1reference
Blender Benchmark 2.82GPU, classroom (CUDA)6,499.81,815.5reference
Blender Benchmark 2.90GPU, fishy_cat (CUDA)3,537.9—no pair
Blender Benchmark 3.5.0GPU, sum of scenes monster + junkshop + classroom (CUDA)36.7—no pair
Blender Benchmark 3.6.0GPU, sum of scenes monster + junkshop + classroom (CUDA)31.9—no pair
Blender Benchmark 2.81aGPU, bmw27 (CUDA)—552.5no pair
Blender Benchmark 2.82GPU, bmw27 (CUDA)—534.2no pair
Blender Benchmark 2.83GPU, bmw27 (CUDA)—555no pair
Blender Benchmark 2.81aGPU, classroom (CUDA)—1,929.5no pair
Blender Benchmark 2.90GPU, classroom (CUDA)—1,949.4no pair
Blender Benchmark 2.92GPU, classroom (CUDA)—1,917.2no pair
Blender Benchmark 2.82GPU, fishy_cat (CUDA)—1,077.9no pair
Blender Benchmark 2.92GPU, fishy_cat (CUDA)—1,321no pair
Blender Benchmark 2.82GPU, koro (CUDA)—1,946no pair
Blender Benchmark 2.92GPU, koro (CUDA)—1,995.7no pair
Blender Benchmark 2.82GPU, pavillon_barcelona (CUDA)—4,025.7no pair
Blender Benchmark 2.92GPU, pavillon_barcelona (CUDA)—4,047.5no pair
Blender Benchmark 3.1.0GPU, sum of scenes monster + junkshop + classroom (CUDA)—50.2no pair
Blender Benchmark 3.3.0GPU, sum of scenes monster + junkshop + classroom (CUDA)—91.6no pair
Blender Benchmark 3.4.0GPU, sum of scenes monster + junkshop + classroom (CUDA)—42no pair

Each benchmark version has its own scale: numbers are compared only within a row.

Specifications

Chip

  • GPU chip

    AGK107-300-A2

    BGK106-220-A1

    differs
  • Architecture

    AKepler

    BKepler

    equal
  • Shader units

    A384 (CUDA cores)

    B768 (CUDA cores)

    B: 384 more

    Shader unit counts are comparable only within one architecture.

Clocks

  • Base clock

    A902 MHz

    B928 MHz

    B: 26 MHz more

Memory

  • Video memory

    A2 GB

    B1 GB

    A: 1 GB more
  • Memory type

    ADDR3

    BGDDR5

    differs
  • Memory bus

    A128 bit

    B128 bit

    equal
  • Memory bandwidth

    A28.5 GB/s

    B86.4 GB/s

    B: 57.9 GB/s more

    Excludes cache (Infinity Cache / L2), which affects effective bandwidth differently.

Power and dimensions

  • Board power

    A65 W (TDP)

    B110 W (TDP)

    B: 45 W higher

    TGP (NVIDIA) and TBP (AMD/Intel) are calculated differently; higher is not better. For laptops, the chip maker sets the range and the laptop maker picks the value, so the same chip behaves differently in different laptops.

  • Interface

    APCIe 3.0 x16

    BPCIe 3.0 x16

    equal

“More” does not always mean “better”: higher power draw or size is a system requirement, not an advantage (such bars are dimmed).

Compare with AI

An AI model reviews the specs and benchmarks of this pair and gives a plain-language verdict on who each model suits.

AI verdict

Questions and answers

How much video memory do NVIDIA GeForce GT 640 and NVIDIA GeForce GTX 650 Ti have?

NVIDIA GeForce GT 640: 2 GB DDR3, 128-bit bus. NVIDIA GeForce GTX 650 Ti: 1 GB GDDR5, 128-bit bus.

Which PCIe interface do the cards use?

NVIDIA GeForce GT 640: PCIe 3.0 x16. NVIDIA GeForce GTX 650 Ti: PCIe 3.0 x16.

Data sources (2)
  • Wikipedia (English) — specifications of NVIDIA GeForce GT 640, specifications of NVIDIA GeForce GTX 650 Ti, release date and launch pricechecked November 14, 2026
  • Blender Open Data (Blender Foundation) — benchmark Blender Benchmark 2.90, benchmark Blender Benchmark 2.92, benchmark Blender Benchmark 2.93.1, benchmark Blender Benchmark 2.81a, benchmark Blender Benchmark 2.82, benchmark Blender Benchmark 2.83, benchmark Blender Benchmark 3.5.0, benchmark Blender Benchmark 3.6.0, benchmark Blender Benchmark 3.1.0, benchmark Blender Benchmark 3.3.0, benchmark Blender Benchmark 3.4.0checked November 14, 2026

How we compare: methodology. Where the data comes from: sources.

TelegramVK