pny geforce rtx 5070 oc spec:
CUDA Core: 6144 Base Clock Speed: 2.16GHz Boost Clock Speed: 2.51GHz VRAM: 12GB GDDR7 Power Supply: 250W Recommended System Power: 650W Price: £540/$550
I knew the RTX 5070 was cheating on me. Parked next to the luxurious stupidity of the 2-grain RTX 5090, this £539/$549 graphics card looks like a very comfortable deal, offering the same DLSS 4 and multi-frame generation as a bigger and more expensive brother. It also upgraded to the RTX 4070 Super, a GPU that can handle 4K without looking too out of place on a premium 1080p rig. But tragically, the RTX 5070 breaks the sacred covenant. This is a mutual understanding between PC owners and parts manufacturers that have been held strong for decades, and if you buy a newer version, it should be faster than the older version of that thing. Watching past the fantasies of MFG, it’s not too often.
Running through the specs does not cause any clear reason for this lack of performance improvement. The RTX 5070 CUDA co-account appears in the 6144 and is placed only between the RTX 4070 (5888) and the RTX 4070 Super (7168), but the clock speeds on both the base and boost have increased to compensate. Meanwhile, VRAM remains at 12GB, but with a switch from GDDR6X to GDDR7, it gives a lot of bandwidth. Next is the power rating. This is the Delaed Founders Edition and the Triple Fan PNY GEFORCE RTX 5070 OC set here at 250W-50W than the entry-level RTX 4070 Super. It also sounds like a decent generation update.
nevertheless…
Nvidia GeForce RTX 5070 Review: 4K Benchmark
The RTX 5070 cuts as a 4K candidate over 60fps in most super-quality games without upscaling, but the only time it produced a visible faster frame rate was, elsewhere, outperformed the RTX 4070 Super only the super frames of second frames for second and second for second frames frames frames of and second frames frames a super. Process ti.
RPS Test PC:
CPU: AMD Ryzen 7 9800X3D RAM: 32GB Trident Z5 NEO RGB DDR5 Motherboard: MSI MPG X870E Carbon WiFi PSU: NZXT C1000 Gold
Obviously, few potential buyers will upgrade directly from last year’s RTX 4070 Super to the RTX 5070. The latest models offer at least the meat improvements of the RTX 3070. The problem is that the improvements have already been made in the final generation.
It is also worth addressing certain claims made by NVIDIA during the announcement of the original RTX 50 series that the RTX 5070 will deliver “RTX 4090 Performance” on MFG. This depends on how generous you want to be, either lacking in context or full tosh. For example, it is true that 4x MFG can drag the RTX 5070 to 71FP on a fully pass-trace Alan Wake 2, but if you reject the old GPU’s own tools, it is equal or faster than the RTX 4090. As expected, otherwise you only need to generate two frames that are old-fashioned with DLSS 3 at an average of 82fps with the same settings. In short, the only way the RTX 5070 is really running at the same pace as the RTX 4090 is if they are in the same delivery van.
MFG can still be considered a party trick on the RTX 5070, but specifically in 4K it’s not something that’s different with Barier cards like the RTX 5070 TI or RTX 5080. In fact, in the latter it is the slowest in a narrow lot. As a result, when the 4x MFG kicks in, it generates more numbers at the framerate counter, but in reality it doesn’t help the game run more and be smoother or more responsive than the DLSS 3-limited 40 series model. As far as your PC knows, it’s still only running at 26/21fps, and the frames generated by all AI in the world will not remove the dull feeling of aiming and camera control suffering in these conditions.
Nvidia GeForce RTX 5070 Review: 1440p Benchmark
The 1440p is a more comfortable environment for the RTX 5070 frame Gen Tech, but familiar performance issues remain. If anything, this is even more troublesome here, as the higher framerate makes it even more difficult to convey the difference in naked eyes between this and its predecessor.
In Assassin’s Creed Mirage, the RTX 4070 will finish evenly first if it’s a single frame, but even in games where the RTX 5070 is relatively successful (CyberPunk 2077, F1 24) only looks like a 10% increase. There’s a gap at the same time as the RTX 5070 TI, suggesting that there’s no need to stretch it to the TI version for high quality quad HD. Also, those who upgrade from the RTX 3070 will enjoy a significant speed boost. Still, these benefits could apply to the RTX 4070 Super just as much as the RTX 5070.
To be fair, the same cannot be said about Frame Gen’s performance. These tests suggest that the RTX 5070 is actually slightly worse at dealing with path trace/full ray tracing than those nasty RTX 40 GPUs, unlike 4K, but it can generate enough pre-occurring frames to support 4x MFG executable applications.
I’m not saying this is a good alternative to traditionally rendered frames, but at least at this resolution, you can see where Nvidia is coming from. Eliminating the need to compensate for the frame rate of a sub-30FPS, Frame Gen simply illuminates the visual smoothness of a game that is already running properly.
Still, it’s hard not to peek into the towering RTX 5070 Ti and RTX 5080 bars, and I wonder if the standard RTX 5070 could have done a little more in the traditional rendering department.
Nvidia GeForce RTX 5070 Review: 1080p Benchmark
To raise the value proposition of the RTX 5070, there are few drops to even less demanding screen Rez. Again, it is haunted by the past ghosts of the XX70 GPU, especially in the Assassin’s Creed Mirage. You get some extra frames above the RTX 4070 Super in the shadow of the Tomb Raider, but it already runs so fast that it requires a 300Hz monitor and a really inhuman observation skill.
That means we must return to the world of the greatest lighting effects and AI frame generation in search of awards of comfort. Like the 1440p, the RTX 5070 is inexplicably worse than the RTX 4070 Super for the path traces involved, but in both games it forms enough basics to apply 4x MFG without typing input lags in the trekle.
You’ll have to allow these results not to be shared with the hilarious enthusiasm worthy of a multi-colored bar chart. I know that the RTX 5070 is not regressing to the point that it is a bad GPU in the general sense. Like this PNY model, you’re getting it in RRP. And what I’ve said before, is the RTX 4070 Super Going cheap and faster? Reader, I’m worried that it was a speculation. Currently, these are still on sale in the range of £600 to £700. Many board partners’ RTX 5070s certainly fill the same space, but there’s no compelling reason to stick to the old ones when these two GPUs cost the same. There may be debates in favor of the power efficiency of the RTX 4070 Super. It peaked at 219W during testing against the 251W of the RTX 5070, but for all the drawbacks, the MFG has more MFG than the 30W of storage.
Still, it is completely fair and reasonable to expect something meaningful rather than an entirely dependent improvement from each new generation. The RTX 5070 is simply not delivered. Honestly, I’m a bit worried that frame generation will be the only way future GPUs can upgrade “performance” if we simply shrug and accept. As the 4K pass trace results for this card show, this approach is insufficient as the game side of the industry continues to ask for more and more hardware.
This review is based on retail units provided by the manufacturer.
