GeForce GT 630 vs GeForce RTX 4090 D: specs and benchmark comparison
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 AIAdvantages
Only measured and rated values: a noticeable benchmark gap and specs where more is better.
Where GeForce GT 630 is better1
Board power 360 W lower65 W (TDP) vs 425 W (TDP)
Where GeForce RTX 4090 D is better3
Video memory 23 GB more24 GB vs 1 GB
Memory bus 256 bit more384 bit vs 128 bit
Memory bandwidth 979.2 GB/s more1,008 GB/s vs 28.8 GB/s
Benchmarks
Results are compared within one benchmark, one version and the same conditions.
Blender Benchmark 2.90
GPU, bmw27 (CUDA) · s, lower is better
GeForce GT 6301,248 sGeForce RTX 4090 Dno dataNo advantage conclusion: no comparable result for the second model.
About the measurement and source
- A · NVIDIA GeForce GT 630
1,248 s · community measurement, sample 1 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 1 submissions (0 outliers excluded), interquartile range 1247.99–1247.99; submissions 2021-05-16 to 2021-05-16; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 4090 D
- no comparable result
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 (8)
| Benchmark | A | B | Difference |
|---|---|---|---|
| Blender Benchmark 2.83GPU, bmw27 (CUDA) | 1,133.8 | — | no pair |
| Blender Benchmark 4.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 10,713 | no pair |
| Blender Benchmark 4.5.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 10,724 | no pair |
| Blender Benchmark 5.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 11,863.2 | no pair |
| Blender Benchmark 4.0.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 11,007.3 | no pair |
| Blender Benchmark 4.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 10,349.6 | no pair |
| Blender Benchmark 4.3.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 10,085.9 | no pair |
| Blender Benchmark 4.4.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 10,675.2 | no pair |
Each benchmark version has its own scale: numbers are compared only within a row.
Specifications
| Specification | A · GeForce GT 630 | B · GeForce RTX 4090 D | Difference |
|---|---|---|---|
| Chip | |||
| GPU chip | GF108-400-A1 | AD102-250-A1 | differs |
| Architecture | Fermi | Ada Lovelace | differs |
| Shader unitsShader unit counts are comparable only within one architecture. | 96 (CUDA cores) | 14,592 (CUDA cores) | not directly comparable |
| Clocks | |||
| Base clock | 810 MHz | 2,280 MHz | B: 1,470 MHz more |
| Boost clockClocks of different architectures do not reflect performance directly. For laptop chips, the clock depends on the laptop’s power limit. | no data | 2,520 MHz | no data for A |
| Memory | |||
| Video memory | 1 GB | 24 GB | B: 23 GB more |
| Memory type | DDR3 | GDDR6X | differs |
| Memory bus | 128 bit | 384 bit | B: 256 bit more |
| Memory bandwidthExcludes cache (Infinity Cache / L2), which affects effective bandwidth differently. | 28.8 GB/s | 1,008 GB/s | B: 979.2 GB/s more |
| Power and dimensions | |||
| Board powerTGP (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. | 65 W (TDP) | 425 W (TDP) | B: 360 W higher |
| Interface | PCIe 2.0 x16 | PCIe 4.0 x16 | differs |
Chip
GPU chip
AGF108-400-A1
BAD102-250-A1
differsArchitecture
AFermi
BAda Lovelace
differsShader units
A96 (CUDA cores)
B14,592 (CUDA cores)
not directly comparableShader unit counts are comparable only within one architecture.
Clocks
Base clock
A810 MHz
B2,280 MHz
B: 1,470 MHz moreBoost clock
Ano data
B2,520 MHz
no data for AClocks of different architectures do not reflect performance directly. For laptop chips, the clock depends on the laptop’s power limit.
Memory
Video memory
A1 GB
B24 GB
B: 23 GB moreMemory type
ADDR3
BGDDR6X
differsMemory bus
A128 bit
B384 bit
B: 256 bit moreMemory bandwidth
A28.8 GB/s
B1,008 GB/s
B: 979.2 GB/s moreExcludes cache (Infinity Cache / L2), which affects effective bandwidth differently.
Power and dimensions
Board power
A65 W (TDP)
B425 W (TDP)
B: 360 W higherTGP (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 2.0 x16
BPCIe 4.0 x16
differs
“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.
Questions and answers
How much video memory do NVIDIA GeForce GT 630 and NVIDIA GeForce RTX 4090 D have?
NVIDIA GeForce GT 630: 1 GB DDR3, 128-bit bus. NVIDIA GeForce RTX 4090 D: 24 GB GDDR6X, 384-bit bus.
Which PCIe interface do the cards use?
NVIDIA GeForce GT 630: PCIe 2.0 x16. NVIDIA GeForce RTX 4090 D: PCIe 4.0 x16.
See also
Data sources (2)
- Wikipedia (English) — specifications of NVIDIA GeForce GT 630, specifications of NVIDIA GeForce RTX 4090 D, release date and launch pricechecked November 14, 2026
- Blender Open Data (Blender Foundation) — benchmark Blender Benchmark 2.83, benchmark Blender Benchmark 2.90, benchmark Blender Benchmark 4.2.0, benchmark Blender Benchmark 4.5.0, benchmark Blender Benchmark 5.2.0, benchmark Blender Benchmark 4.0.0, benchmark Blender Benchmark 4.1.0, benchmark Blender Benchmark 4.3.0, benchmark Blender Benchmark 4.4.0checked November 14, 2026
How we compare: methodology. Where the data comes from: sources.