GeForce RTX 4070 is imminent and details continue to be spilled. Whether or not these are controlled leaks is debatable, but if not, Nvidia is doing a poor job of keeping its next major graphics card under wraps.
Hot on the heels of complete specifications and confirmed pricing, we now have our first look at the Founders Edition card courtesy of a Tweet from David Giannis.
That’s right, while Nvidia handed over the reins to partners for RTX 4070 Ti, the regular variant, thought to be launching Wednesday, April 13, will be championed by an FE board at the purported $599 price tag. We imagine that’s a figure add-in-board partners are going to struggle to match, and at first glance, there appears little reason to look beyond Nvidia’s reference design.
Leaked images show a familiar aesthetic, but guesstimating from the side profile, RTX 4070 FE looks set to be considerably smaller than RTX 4080 FE. Word on the grapevine suggests this to be a strict dual-slot form factor that’s roughly the same length as outgoing RTX 3070 Ti. Good news for those who’ve been patiently awaiting an RTX 40 Series fit for small-form-factor builds.
Display outputs are said to remain unchanged – meaning one HDMI 2.1 alongside three DisplayPort 1.4 – while power is sourced from a solitary 16-pin 12VHPWR connector roughly half way along the card.
Taking design cues from existing RTX 40 Series products, RTX 4070 appears more seamless in its construction than previous-generation equivalents, and going by initial images, looks to include larger axial fans that should deliver an optimal balance of low temps and minimal noise.
| GeForce | RTX 4090 | RTX 4080 | RTX 4070 TI | RTX 4070 |
|---|---|---|---|---|
| Launch date | Oct 2022 | Nov 2022 | Jan 2023 | Apr 2023 |
| Codename | AD102 | AD103 | AD104 | AD104 |
| Architecture | Ada Lovelace | Ada Lovelace | Ada Lovelace | Ada Lovelace |
| Process (nm) | 4 | 4 | 4 | 4 |
| SMs | 128 of 144 | 76 of 80 | 60 of 60 | 46 of 60 |
| CUDA cores | 16,384 | 9,728 | 7,680 | 5,888 |
| Boost clock (MHz) | 2,520 | 2,505 | 2,610 | 2,475 |
| Peak FP32 TFLOPS | 82.6 | 48.7 | 40.1 | 29.1 |
| RT cores | 128 | 76 | 60 | 46 |
| Tensor cores | 512 | 304 | 240 | 184 |
| Texture units | 512 | 304 | 240 | 184 |
| Memory size (GB) | 24 | 16 | 12 | 12 |
| Memory type | GDDR6X | GDDR6X | GDDR6X | GDDR6X |
| Memory bus (bits) | 384 | 256 | 192 | 192 |
| Memory clock (Gbps) | 21 | 22.4 | 21 | 21 |
| Bandwidth (GB/s) | 1,008 | 717 | 504 | 504 |
| L2 cache (MB) | 72 | 64 | 48 | 36 |
| Power (watts) | 450 | 320 | 285 | 200 |
| Launch MSRP ($) | 1,599 | 1,199 | 799 | 599 |
As a reminder, leaked specifications paint RTX 4070 as a frugal 200W GPU. There’s ample differentiation between RTX 4070 and RTX 4070 Ti – the latter maintains a 30 per cent increase in core count – and we’re intrigued to see exactly how the newcomer compares with respect to 1440p performance.
Benchmarks will be coming soon, but one thing’s for certain; AIBs will be hoping RTX 4070 FE will be produced in limited quantities.
