ASUS Prime GeForce RTX 5070 Ti Overview: Midrange Blackwell Examined

	ASUS PRIME GeForce RTX 5070 Ti: MSRP $749 The GeForce RTX 5070 Ti brings NVIDIA’s newest Blackwell GPU structure all the way down to a $749 beginning value level, with out sacrificing any options, however it’s not an enormous improve over the earlier gen.
	Newest Structure And Characteristic Set Nice Overclocker Comparatively Energy Pleasant Cool And Quiet DLSS 4 + RTX Neural Rendering Upgraded Media Engine	Value Premium Small Improve Over The RTX 4070 Ti

NVIDIA continues its onslaught of Blackwell-based GeForce RTX 50 sequence playing cards immediately, with the discharge of the GeForce RTX 5070 Ti. Technically, playing cards go on sale tomorrow from NVIDIA’s board companions, however over the subsequent couple of days we will present you what a few the playing cards can do by way of efficiency, beginning with the ASUS PRIME GeForce RTX 5070 Ti we’ll be that includes right here.

Like the entire different GeForce RTX 5070 Ti playing cards coming down the pipeline, the ASUS PRIME contains a proprietary design with ASUS’ personal tackle NVIDIA’s upper-midrange GPU, however this explicit mannequin ought to be arriving at NVIDIA’s specified $749 MSRP – an ASUS PRIME GeForce RTX 5070 Ti OC Version, with considerably larger increase clocks can also be coming, and shall be considerably dearer. The identical is true for plenty of different producers that shall be providing a number of fashions at or above MSRP.

Earlier than we dive in and get up-clos-and-personal with the ASUS PRIME GeForce RTX 5070 Ti although, let’s lay a few of the groundworks and clarify the place the RTX 5070 Ti lives in NVIDA’s line-up. Specs and a fast comparability up first…

NVIDIA GeForce RTX 5070 Ti Specs

The above desk compares the principle options and specification of the final three “70” class GPUs from NVIDIA. As you possibly can see, by way of its specs, the GeForce RTX 5070 Ti is an improve over the previous-gen GeForce RTX 4070 Ti in most methods. The GeForce RTX 5070 Ti has extra cores, that are additionally primarily based on a more recent structure, extra reminiscence with a lot larger bandwidth, extra L1 cache, a bigger register file, an upgraded media engine, a local PCIe Gen 5 interface, and better pixel and texture fill charges, regardless of a considerably decrease increase clock. Just like the RTX 3070 Ti, a 256-bit reminiscence interface returns, which ends up in almost 900GB/s of peak bandwidth when paired to its speedy 16GB of 28Gbps GDDR7 reminiscence.

Just like the GeForce RTX 5080, the brand new RTX 5070 Ti is constructed across the 45.6 billion transistor GB203 GPU. However in contrast to the GeForce RTX 5080, which contains a full-fat chip, the RTX 5070 Ti is scaled again considerably. The total GB203 as it’s configured on the RTX 5080 options 11 GPCs, 42 TPCs, 84 SMs, and eight 32-bit reminiscence controllers, for an mixture 256-bit bus width. Every SM incorporates 128 FP32 CUDA Cores, for a grand complete of 10,752, with 84 4th Gen RT Cores, 336 fifth Gen Tensor Cores, 336 Texture Models, and 112 ROPS. The GB203 additionally features a complete of 10,752 KB L1 cache, a 21,504 KB Register File, and 65,536 KB L2 cache. On the GeForce RTX 5070 Ti, nevertheless, solely 6 GPCs, 35 TPCs, and 70 SMs are enabled, which ends up in 8,960 complete CUDA cores, 280 Tensor cores, and 70 RT cores. The L1 cache commensurately drops down to eight,960KB, with 49,152KB of L2 as properly.

Whereas the GB203 powering the GeForce RTX 5070 Ti is scaled again versus the extra highly effective GeForce RTX 5080, or GeForce RTX 5090 for that matter, its options and capabilities stay the identical. RTX Neural Rendering, RTX Mega Geometry, DLSS 4 with Multi-Body Gen, the brand new AI Administration Processor, and up to date media engine with {hardware} acceleration for 4:2:2 video, are all a part of the GeForce RTX 5070 Ti’s function set. You’ll be able to study all of these options in our earlier NVIDIA Blackwell GPU structure protection, if you have not already learn it.

Test Out The ASUS PRIME GeForce RTX 5070 Ti

   

       

           

           

           

       

       

           

           

           

           

       

   

	NVIDIA GeForce RTX 5080 Founders Version: MSRP $999                         The NVIDIA GeForce RTX 5080 Founders Version appears similar to the flagship RTX 5090, and presents the identical options, however packs a smaller GPU at solely half the value.
	Newest Structure And Characteristic Set                 Straightforward Overclocker                 Comparatively Energy Pleasant                 Extra Compact Type Issue                 Runs Cool And Principally Quiet                 DLSS 4 + RTX Neural Rendering                 Upgraded Media Engine	Value Premium                 Gentle Improve Over The RTX 4080

NVIDIA is prepared with the second member of the GeForce RTX 50 sequence, the GeForce RTX 5080. Final week we have been in a position to present you NVIDIA’s personal flagship GeForce RTX 5090 and an superior companion board from MSI, the cool and quiet GeForce RTX 5090 Suprim Liquid SOC. Each of  these playing cards are loopy highly effective, however additionally they sit on the very, tippy-top of NVIDIA’s product stack, and as such, they command a hefty premium. The RTX 5090’s MSRP is 2 grand, and we’d be stunned if anybody apart from the courageous few already tenting out for playing cards get one for the prompt retail value. The brand new GeForce RTX 5080 is a unique animal although, and its prompt value is half of the RTX 5090’s, at just below a grand.

Though the RTX 5080 Founders Version card we’ll be displaying you right here appears primarily similar to the RTX 5090, there are many variations below the hood. In truth, the playing cards aren’t even constructed utilizing the identical GPU. Extra on that in a bit.

The NVIDIA GeForce RTX 5080 Founders Version’s packaging and bundle is strictly the identical because the RTX 5090’s. In stead of writing about the entire environmentally pleasant and sustainable goodness in NVIDIA’s newest packaging once more, we’ll let Davo let you know straight. Try the video above to see for yourselves – simply think about that 9 is an 8…

NVIDIA GeForce RTX 5080 Specs

The GeForce RTX 5080 is constructed across the 45.6 billion transistor GB203 GPU, however in contrast to the RTX 5090, which has a scaled again GB202 at its coronary heart, the GeForce RTX 5080 contains a full-fat chip. The GB203 is supplied with 11 GPCs, 42 TPCs, 84 SMs, and eight 32-bit reminiscence controllers, for an mixture 256-bit bus width. Every SM incorporates 128 FP32 CUDA Cores, for a grand complete of 10,752 CUDA Cores, with 84 4th Gen RT Cores, 336 fifth Gen Tensor Cores, 336 Texture Models, and 112 ROPS. NVIDIA’s GB203 additionally consists of 10,752 KB L1 cache, a 21,504 KB Register File, and 65,536 KB L2 cache. The GPU is linked to 16GB of GDDR7 reminiscence working at an efficient 30Gbps, for as much as 960GB/s of bandwidth.

Test Out The NVIDIA GeForce RTX 5080 Founders Version

Just like the GeForce RTX 5090, the GeForce RTX 5080 is far thinner than its direct predecessor. The cardboard measures 304mm in size, 137mm in top, and it’s actually two-slots (40mm) extensive – it’s not only a 2-slot case bracket, with a protruding cooler meeting. It is a dense package deal that makes use of premium supplies through-and-through. You’ll be able to really feel the construct high quality as quickly as you choose up the GeForce RTX 5080.

NVIDIA was in a position to shrink its top-end GeForce RTX 50 sequence playing cards this era by growing what it calls a “Double Movement Via” cooler design, which permits each cooling followers to blow air straight via the heatsink, for optimum efficiency with decreased noise.

To attain this, the GeForce RTX 5080 has a multi-part PCB setup, just like the 5090’s. The central PCB is dwelling to the GPU, reminiscence, and energy circuitry. It’s located within the heart of the cardboard, with solely small parts protruding beneath every fan. A separate daughterboard attaches to the central PCB to the PCIe x16 connector, and a 3rd, versatile board runs perpendicular alongside the underside edge with high-speed signaling connections from the central PCB to the show outputs. 

The GeForce RTX 5090 has a newly-designed 3D Vapor Chamber affixed to its GPU, reminiscence and energy circuitry for cooling functions, with a liquid metallic TIM to help in warmth switch. The GeForce RTX 5080, nevertheless, has a extra conventional chilly plate, with a dense array of heatsink thin-fins, linked by way of 5 warmth pipes — it’s not outfitted with the identical vapor chamber because the 5090. NVIDIA additionally makes use of a extra conventional phase-change TIM on the RTX 5080, so not one of the complexities related to liquid metallic are at play right here. Externally although, you possibly can’t actually inform. The identical twin axial followers are current, together with the identical directional shops on the heatsink and high and backside vents on the shroud.

Outputs within the GeForce RTX 5080 (and different RTX 50 sequence playing cards) mirror the RTX 5090 as properly, and embrace a trio of DisplayPorts (2.1b) and a single HDMI port (2.1b). Their orientations have been reversed versus previous-gen playing cards although, and the case bracket contains a stable entrance bezel and an anti-fingerprint coating.

Now that you simply’re intimately aware of the GeForce RTX 5080’s design, let’s get to the benchmarks…

NVIDIA might need been in a position to shrink its flagship GeForce RTX 5090 and the RTX 5080 Founders Version playing cards down right into a two-slot kind issue, however its companions are usually using a lot bigger designs on their RTX 50 sequence playing cards. Though it sits a number of rungs down the stack versus higher-end playing cards, the ASUS GeForce RTX 5070 Ti continues to be 2.5 slots extensive, 12” lengthy, and contains a fin-stack and fan configuration that’s a lot larger than the precise PCB housing the cardboard’s parts.

Dominating the entrance of the ASUS PRIME GeForce RTX 5070 Ti is triple-fan cooling answer, with curved edges and a partial pass-through design. The cardboard options twin ball fan bearings for added sturdiness, with what ASUS calls “Axial-tech” followers. That mainly means the fan hub is a bit smaller and the fan blades are a bit longer than typical axial followers, which reduces noise, will increase downward stress, and finally improves cooling efficiency.

The heatsink below these followers covers all the PCB and extends off the again a number of inches to allow the aforementioned partial pass-through design. The heatsink makes direct contact with the GPU, reminiscence, and voltage regulation circuitry on the ASUS PRIME GeForce RTX 5070 Ti, and the corporate claims the bottom of the heatsink is machined with larger precision than competing playing cards, to enhance contact space and warmth switch. ASUS additionally makes use of a phase-change GPU thermal pad, which the corporate claims improves thermal conductivity below heavy load. At any price, the cooler is greater than succesful and retains the GB203 on this card working cool, even whereas overclocking. The ASUS PRIME GeForce RTX 5070 Ti can also be good and quiet, however we’ll focus on thermals and acoustics in additional element a bit later. 

The ASUS PRIME GeForce RTX 5070 Ti additionally contains a vented metallic backplate, and a dual-BIOS swap labeled with a “P” and “Q”, we supposed for efficiency and quiet modes, however ASUS’ website lists “default” and “OC” modes. Regardless of the modes are literally known as, the cardboard’s default GPU increase clock is 2,452MHz (when the swap is ready to Q). That clock will increase barely to 2,482MHz when in P-mode, nevertheless.

Outputs within the ASUS PRIME GeForce RTX 5070 Ti mirror different GeForce RTX 50 sequence playing cards and embrace a trio of DisplayPorts (2.1b) and a single HDMI port (2.1b). These ports are nestled into a stainless-steel backplate, that options venting for some odd motive, as a result of the plastic fan shroud behind these vents is stable.

Included with the ASUS PRIME GeForce RTX 5070 Ti is a primary lit pack, a thanks card, a single strip of hook & loop fastener, and an influence adapter that takes three 8-pin PCIe Specific energy feeds and converts them to a single 12VHPWR connector.

And with all of that lined, let’s get to some benchmarks…