Accelerator
NVIDIA GeForce GTX 780 Ti: 3 GiB of GDDR5, 336 GB/s peak bandwidth, and 250 W graphics card power.
NVIDIA's reference-card specification page supplies clocks, memory, bandwidth and board power. It does not state an explicitly labelled FP32 peak.
| Component | Role | Count | Clock | Capacity |
|---|---|---|---|---|
| NVIDIA Kepler (NVIDIA GeForce GTX 780 Ti) The lower declared clock; the separately named upper clock is recorded as a peak-clock measurement. | GPU | 1 | 875 MHz Nominal clock 928 MHz Peak clock | — |
| NVIDIA GeForce GTX 780 Ti frame buffer | Video memory | 1 | — | 3 GiB |
provisional A figure marked provisional is recorded but not accepted: its evidence is unsettled. It may rely on one source, an unconfirmed announcement, or a digit read from a damaged scan. The reason appears under the figure. The catalog shows the figure but never uses it in a multiplier. How records are judged.
| Metric | Value | Evidence | Basis | Method | Source |
|---|---|---|---|---|---|
| Memory bandwidth whole-system provisional | 336 GB/s | Confirmed | Derived | Theoretical peak Computed here from figures the vendor stated Adds the one fitted frame-buffer bandwidth. No host memory is included. |
provisional A figure marked provisional is recorded but not accepted: its evidence is unsettled. It may rely on one source, an unconfirmed announcement, or a digit read from a damaged scan. The reason appears under the figure. The catalog shows the figure but never uses it in a multiplier. How records are judged.
| Metric | Value | Evidence | Basis | Method | Source |
|---|---|---|---|---|---|
| Clock frequency gpu provisional | 875 MHz | Confirmed | Vendor-rated | Nominal clock Stated by the vendor The lower declared clock. The separately named upper clock is a different quantity and uses peak-clock. | |
| Clock frequency gpu provisional | 928 MHz | Confirmed | Vendor-rated | Peak clock Stated by the vendor The vendor’s upper boost clock, distinct from the nominal clock. | |
| Rated power consumption gpu provisional | 250 W | Confirmed | Vendor-rated | Rated power Stated by the vendor NVIDIA labels this a graphics-card power rating for its reference design, not a measured draw or a thermal-design target. |
| Metric | Value | Evidence | Basis | Method | Source |
|---|---|---|---|---|---|
| Peak FP32 rate gpu provisional | unknown The page gives CUDA-core and clock figures but no explicitly labelled single-precision throughput. | Not stated | Theoretical | Theoretical peak Stated by the vendor NVIDIA's reference-card page does not state an explicitly labelled FP32 peak. |
provisional A figure marked provisional is recorded but not accepted: its evidence is unsettled. It may rely on one source, an unconfirmed announcement, or a digit read from a damaged scan. The reason appears under the figure. The catalog shows the figure but never uses it in a multiplier. How records are judged.
| Metric | Value | Evidence | Basis | Method | Source |
|---|---|---|---|---|---|
| Memory bandwidth memory provisional | 336 GB/s | Confirmed | Theoretical | Theoretical peak Stated by the vendor Vendor-stated maximum transfer rate of this memory interface, not a benchmarked application result. | |
| Memory capacity whole-system provisional | 3 GiB | Confirmed | Vendor-rated | Design capacity Stated by the vendor The frame buffer fitted to the reference accelerator; host memory is excluded. |
336 000 000 000 B/s
a + b + … sum v1 Adds 1 memory-bandwidth figure obtained by theoretical-peak. Nothing in the catalog is left out of it. A total is only as trustworthy as the decision about what belongs in it, so both the terms and the exclusions are named records.
This accelerator’s whole-system memory is its fitted frame buffer; host memory is excluded. Partner cards may differ from this reference configuration.