When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.
The PNY RTX 4070 mostly matches the 4070 Founders Edition, but it runs slightly hotter and has noisier fans. But if you think Nvidia's 8-pin to 16-pin adapters are stupid, PNY agrees, opting for a solo 8-pin connector.
Why you can trust Tom's Hardware
Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.
Go to page:We showed benchmarks of the PNY GeForce RTX 4070 Verto Dual in our initial Nvidia RTX 4070 Founders Edition launch review. Then a bunch of other stuff came along, like the RTX 4060 Ti and RX 7600 launches, but now we're getting around to various third-party reviews. These are some of the best graphics cards, for gaming and other purposes, and the RTX 4070 looks to hit the sweet spot in terms of specs and pricing.
At its core, the RTX 4070 performs a lot like the previous generation RTX 3080, despite having 2GB more VRAM but a much narrower 192-bit memory interface. That's thanks to the Nvidia Ada Lovelace architecture, and specifically the inclusion of up to 8MB of L2 cache per 32-bit chunk of memory interface. The additional memory can help in a few instances, but higher resolutions also tend to prefer the increased raw bandwidth of the 3080's 320-bit interface.
Let's start with the usual specs table before getting into the specifics of the PNY card.
Swipe to scroll horizontally| Graphics Card | RTX 4070 | RTX 4060 Ti | RTX 4070 Ti | RTX 4080 | RX 7600 | RX 7900 XT | RX 6750 XT | RX 6800 | RX 6800 XT | RX 6950 XT |
|---|---|---|---|---|---|---|---|---|---|---|
| Architecture | AD104 | AD106 | AD104 | AD103 | Navi 33 | Navi 31 | Navi 22 | Navi 21 | Navi 21 | Navi 21 |
| Process Technology | TSMC 4N | TSMC 4N | TSMC 4N | TSMC 4N | TSMC N6 | TSMC N5 + N6 | TSMC N7 | TSMC N7 | TSMC N7 | TSMC N7 |
| Transistors (Billion) | 32 | 22.9 | 35.8 | 45.9 | 13.3 | 45.6 + 5x 2.05 | 17.2 | 26.8 | 26.8 | 26.8 |
| Die size (mm^2) | 294.5 | 187.8 | 294.5 | 378.6 | 204 | 300 + 225 | 336 | 519 | 519 | 519 |
| SMs / CUs | 46 | 34 | 60 | 76 | 32 | 84 | 40 | 60 | 72 | 80 |
| GPU Cores (Shaders) | 5888 | 4352 | 7680 | 9728 | 2048 | 5376 | 2560 | 3840 | 4608 | 5120 |
| Tensor / AI Cores | 184 | 136 | 240 | 304 | 64 | 168 | N/A | N/A | N/A | N/A |
| Ray Tracing "Cores" | 46 | 34 | 60 | 76 | 32 | 84 | 40 | 60 | 72 | 80 |
| Boost Clock (MHz) | 2475 | 2535 | 2610 | 2505 | 2625 | 2400 | 2600 | 2105 | 2250 | 2310 |
| VRAM Speed (Gbps) | 21 | 18 | 21 | 22.4 | 18 | 20 | 18 | 16 | 16 | 18 |
| VRAM (GB) | 12 | 8 | 12 | 16 | 8 | 20 | 12 | 16 | 16 | 16 |
| VRAM Bus Width | 192 | 128 | 192 | 256 | 128 | 320 | 192 | 256 | 256 | 256 |
| L2 / Infinity Cache | 36 | 32 | 48 | 64 | 32 | 80 | 96 | 128 | 128 | 128 |
| ROPs | 64 | 48 | 80 | 112 | 64 | 192 | 64 | 96 | 128 | 128 |
| TMUs | 184 | 136 | 240 | 304 | 128 | 336 | 160 | 240 | 288 | 320 |
| TFLOPS FP32 (Boost) | 29.1 | 22.1 | 40.1 | 48.7 | 21.5 | 51.6 | 13.3 | 16.2 | 20.7 | 23.7 |
| TFLOPS FP16 (FP8) | 233 (466) | 177 (353) | 321 (641) | 390 (780) | 43.0 | 103.2 | 26.6 | 32.4 | 41.4 | 47.4 |
| Bandwidth (GBps) | 504 | 288 | 504 | 717 | 288 | 800 | 432 | 512 | 512 | 576 |
| TDP (watts) | 200 | 160 | 285 | 320 | 165 | 315 | 250 | 250 | 300 | 335 |
| Launch Date | Apr 2023 | May 2023 | Jan 2023 | Nov 2022 | May 2023 | Dec 2022 | May 2022 | Nov 2020 | Nov 2020 | May 2022 |
| Launch Price | $599 | $399 | $799 | $1,199 | $269 | $899 | $549 | $579 | $649 | $1,099 |
| Current Price | $585 | $400 | $790 | $1,108 | $257 | $780 | $330 | $490 | $500 | $630 |
The above specs table shows the ten GPUs we'll be using for this review. The PNY RTX 4070 Verto Dual and Nvidia RTX 4070 Founders Edition have identical specs, since both use the reference clocks. There's not much more we can add.
The RTX 4070 lands at a $599.99 MSRP, and there are even cards periodically available for less than that — like right now, you can pick up the RTX 4070 from Asus, Gigabyte, MSI, or Zotac with a $50 instant rebate using code "ZIPDADS" (for Fathers' Day, if that wasn't obvious). PNY's card unfortunately isn't part of that deal. It's also "while funds last," so you may or may not be able to take advantage of that offer.
That sort of deal does speak volumes about graphics card sales in general. The past several Black Fridays and Prime Days, for example, had very few if any noteworthy GPU deals, but that's a straight $50 discount from Newegg on a graphics card that released less than two months ago. It's a safe bet that Nvidia could also drop the prices on all of its GPUs $50 or so without incurring too much pain.
Short-term sales aside, we can't help but feel the RTX 4070 should be selling for a lot less than its MSRP. It's faster than the previous generation RTX 3070, but it also comes with a substantially smaller die size. 294.5 mm^2 on TSMC 4N compared to 392.5 mm^2 on Samsung 8N likely means the AD104 chips do cost more, but Nvidia should still get about 190 chips per wafer. Even at $19,000 per 4N wafer, that's only $100 per chip (and Nvidia probably pays less than that due to the number of wafers it orders).
