NVIDIA has officially unveiled its brand new GeForce RTX 4070 graphics card which aims to offer 100+ FPS in 1440P Gaming at $599 US.
NVIDIA Unleashes Its GeForce RTX 4070 Graphics Card, Ada Lineup Now Starting At $599 US
NVIDIA’s GeForce RTX 4070 is the latest addition to the Ada Lovelace gaming lineup and the green team is stating it as a tremendous upgrade for existing GeForce GTX 1080 and GeForce RTX 2070 users. The graphics card also brings the pricing of the Ada Gaming lineup down to $599 US, being the most lowest priced next-gen card yet. Besides just adding more performance and efficiency, it adds a range of new features including:
- New Streaming Multiprocessors (SM) – The new SM delivers up to 2x performance and power efficiency
- 4th Generation Tensor Cores and Optical Flow – Enable and accelerate transformative AI technologies, including the new frame rate multiplying NVIDIA DLSS 3
- 3rd Generation RT Cores – Up to 2x ray tracing performance, delivering incredibly detailed virtual worlds like never before
- Shader Execution Reordering (SER) – SER improves ray tracing operations by 2x, boosting FPS up to 44% in Cyberpunk with RT: Overdrive Mode
- DLSS 3 – A revolutionary breakthrough in AI-powered graphics that massively boosts performance using AI to generate additional high-quality frames
- NVIDIA Studio – Unmatched performance in 3D rendering, video editing, and live streaming
- AV1 Encoders – The 8th generation NVIDIA Encoder (NVENC) with AV1 is 40% more efficient than H.264, enabling new possibilities for streamers, broadcasters, and video callers
The NVIDIA GeForce RTX 4070 uses a cut-down AD104 GPU configuration with 46 SMs for a total of 5888 CUDA cores The GPU will come packed with 36 MB of L2 cache, 184 Texture Mapping Units & a total of 64 ROPs. The clock speeds for the graphics card are rated at 1920 MHz base and 2475 MHz boost clocks. The card offers 29 TFLOPs of FP32, 67.4 TFLOPs of RT, and 466 TFLOPs on INT8 compute output.
NVIDIA states that increasing the L2 cache by 9 times on the RTX 4070 versus the RTX 3070 (36 MB vs 4 MB) improves performance, reduces latency, & increases power efficiency, as data access can remain on-chip (rather than having to rely on memory bandwidth).
As for memory specs, the GeForce RTX 4070 features 12 GB GDDR6X capacities that will be adjusted at 21.0 Gbps speeds across a 192-bit bus interface. This will provide up to 504 GB/s of bandwidth. This is the same memory configuration as used by the RTX 4070 Ti and 4 GB more memory versus the 3070 series graphics cards.
- NVIDIA GeForce RTX 4070 12 GB “Official” TBP – 200W
- NVIDIA GeForce RTX 3070 8 GB “Official” TBP – 220W
As far as the power consumption is concerned, the TBP is rated at 200W. The card will be powered by a single 16-pin connector which can theoretically deliver up to 600W of power. Custom models will be offering higher TBP targets but are also open to using standard 8-pin connectors.
2 of 9
NVIDIA GeForce RTX 4070 Graphics Cards Performance
As for the performance of these monster GPUs, NVIDIA shared the computational and gaming performance figures and it looks like the GeForce RTX 4070 will be sitting in between the 3080 10GB and 12GB models without DLSS 3.
2 of 9
Just for comparison’s sake:
- NVIDIA GeForce RTX 4090: 83 TFLOPs (FP32) (2.5 GHz Boost clock)
- NVIDIA GeForce RTX 4080: 49 TFLOPs (FP32) (2.5 GHz Boost clock)
- NVIDIA GeForce RTX 3090 Ti: 49 TFLOPs (FP32) (1.86 GHz Boost clock)
- NVIDIA GeForce RTX 4070 Ti: 40 TFLOPs (FP32) (2.6 GHz Boost clock)
- NVIDIA GeForce RTX 3090: 36 TFLOPs (FP32) (1.69 GHz Boost clock)
- NVIDIA GeForce RTX 4070: 29 TFLOPs (FP32) (2.48 GHz Boost clock)
- NVIDIA GeForce RTX 3080: 29 TFLOPs (FP32) (1.71 GHz Boost clock)
- NVIDIA GeForce RTX 3070 Ti: 21 TFLOPs (FP32) (1.77 GHz Boost clock)
- NVIDIA GeForce RTX 3070: 20 TFLOPs (FP32) (1.75 GHz Boost clock)
Based on a boost clock speed of 2.48 GHz, you get up to 29 TFLOPs of compute performance and you can definitely squeeze out a lot more with an overclock as we had demonstrated with the RTX 4090. One should remember that compute performance doesn’t necessarily indicate the overall gaming performance.
FP32 Compute Horsepower Comparisons (Higher is Better)
Compute Power
020406080100120
020406080100120
RTX 4090
82
7900 XTX
61
7900 XT
51
RTX 4080
49
RTX 3090 Ti
40
RTX 4070 Ti
40
RTX 3090
36
RTX 4070
29
RTX 3080
29
RX 6900 XTX
25
RTX 3070 Ti
22
RTX 4060 Ti (Theoretical)
22
RTX 3070
20
RTX 3060 Ti
16
RTX 3060
13
Xbox Series X
12.1
PlayStation 5
10.2
2 of 9
As for the gaming performance, NVIDIA heavily emphasizes the use of RTX & DLSS3 plus the new RT-overdrive mode when comparing the GeForce RTX 4070 graphics card with older RTX 30 GPUs.
NVIDIA GeForce RTX 4070 1440P Performance:
NVIDIA GeForce RTX 4070 DLSS 3 Performance:
NVIDIA GeForce RTX 4070 eSports Performance:
NVIDIA GeForce RTX 4070 Synthetic Benchmarks:
NVIDIA GeForce RTX 4070 Power Profiles:
NVIDIA GeForce RTX 4070 Graphics Cards Price & Availability
The NVIDIA GeForce RTX 4070 graphics card will be available starting tomorrow for a price of $599 US. Both Founders Edition and custom models will be available on sale the same day.
NVIDIA GeForce RTX 40 Series Official Specs:
| Graphics Card Name | NVIDIA GeForce RTX 4090 | NVIDIA GeForce RTX 4080 | NVIDIA GeForce RTX 4070 Ti | NVIDIA GeForce RTX 4070 |
|---|---|---|---|---|
| GPU Name | Ada Lovelace AD102-300 | Ada Lovelace AD103-300 | Ada Lovelace AD104-400 | Ada Lovelace AD104-250 |
| Process Node | TSMC 4N | TSMC 4N | TSMC 4N | TSMC 4N |
| Die Size | 608mm2 | 378.6mm2 | 294.5mm2 | 294.5mm2 |
| Transistors | 76 Billion | 45.9 Billion | 35.8 Billion | 35.8 Billion |
| CUDA Cores | 16384 | 9728 | 7680 | 5888 |
| TMUs / ROPs | 512 / 176 | 320 / 112 | 240 / 80 | 184 / 64 |
| Tensor / RT Cores | 512 / 128 | 304 / 76 | 240 / 60 | 184 / 46 |
| Base Clock | 2230 MHz | 2210 MHz | 2310 MHz | 1920 MHz |
| Boost Clock | 2520 MHz | 2510 MHz | 2610 MHz | 2475 MHz |
| FP32 Compute | 83 TFLOPs | 49 TFLOPs | 40 TFLOPs | 29 TFLOPs |
| RT TFLOPs | 191 TFLOPs | 113 TFLOPs | 82 TFLOPs | 67 TFLOPs |
| Tensor-TOPs | 1321 TOPs | 780 TOPs | 641 TOPs | 466 TOPs |
| Memory Capacity | 24 GB GDDR6X | 16 GB GDDR6X | 12 GB GDDR6X | 12 GB GDDR6X |
| Memory Bus | 384-bit | 256-bit | 192-bit | 192-bit |
| Memory Speed | 21.0 Gbps | 23.0 Gbps | 21.0 Gbps | 21.0 Gbps |
| Bandwidth | 1008 GB/s | 736 GB/s | 504 GB/s | 504 GB/s |
| TBP | 450W | 320W | 285W | 200W |
| Price (MSRP / FE) | $1599 US / 1949 EU | $1199 US / 1469 EU | $799 US | $599 US |
| Price (Current) | $1599 US / 1859 EU | $1199 US / 1399 EU | $799 US | $599 US |
| Launch (Availability) | 12th October 2022 | 16th November 2022 | 5th January 2023 | 13th April 2023 |
Written by Hassan MujtabaOriginal Article