RX 7900 XT Vs RTX 3080 Ti: Which Is Better?

After AMD released their latest RX 7000 series GPUs in December, enthusiasts have conducted many benchmarking tests. While NVIDIA’s RTX 3000 series is a generation older, they hold the same price as AMD’s RX 7000 series GPUs. Therefore, keeping the price in mind, we’re testing the AMD Radeon RX 7900 XT Vs RTX 3080 Ti. We shall conduct several gaming benchmark tests to find out which has the better performance per dollar value.

Key Takeaways

The NVIDIA GeForce RTX 3080 Ti is an upgrade from the RTX 3080 from NVIDIA’s older 3000-series lineup. On the other hand, the AMD Radeon RX 7900 XT is one of AMD’s latest GPUs.
The AMD Radeon RX 7900 XT is ~24.76% faster in 1440P resolution and ~30.06% faster in 4K resolution.
The AMD Radeon RX 7900 XT is ~5% cooler than the NVIDIA GeForce RTX 3080 Ti at 4K resolution.
The NVIDIA GeForce RTX 3080 Ti is ~33.37% more expensive than the AMD Radeon RX 7900 XT in MSRP.
Due to adequate RT capabilities, the NVIDIA GeForce RTX 3080 Ti might be better with raytracing and DLSS enabled.

Architectural Differences

Before we jump onto the gaming benchmarks, we must know the differences between both graphics cards. Because both GPUs are from separate companies, their architectural developments vary significantly. Additionally, the architecture of a GPU affects its performance greatly. Therefore, we must analyze how the AMD Radeon RX 7900 XT and the NVIDIA GeForce RTX 3080 Ti have differentiated. You might be interested in checking out our RTX 4080 Vs RTX 3080 Ti comparison to see how the card performs against Nvidia’s latest lineup.

For example, AMD has taken a different route for upscaling technology from NVIDIA’s DLSS. Similarly, unlike NVIDIA’s “CUDA Cores,” AMD uses “Stream Processors.”

RDNA 3 — AMD Radeon RX 7000 Series Architecture

AMD has steadily upgraded its GPU arsenal while staying close to NVIDIA’s tail. After the failed “Vega” graphics cards, AMD has completely turned its situation around with the influx of RDNA 2 GPUs. Now, with its latest RDNA 3 graphics cards, a few things have changed within the core architecture of the GPUs.

Also Read: RTX 3060 Vs GTX 1660 Ti

For example, with the implications of the forthcoming FSR 3.0 and the latest multi-chiplet design, AMD’s Radeon RX 7000 series GPUs have a lot to offer.

Memory Cache Die

New architecture in RX 7900 XT — Unveiling AMD’s latest multi-chiplet design in AMD’s Radeon RX 7000 series GPUs. Source: AMD.

For its Radeon RX 7000 series GPUs, AMD has taken refuge from its Ryzen processors. To explain, the Radeon RX 7000 series GPUs come with a new chiplet design, particularly a multi-chiplet. Additionally, the RDNA 3 architecture offers multiple dies, coined “Memory Cache Die” (MCD), each containing 16MB of L3 Cache. So, for the AMD Radeon RX 7900 XT, that equates to 80MB of L3 Cache.

Because AMD uses two dies in its RDNA 3 architecture, one die houses all of the GPU cores, the Graphics Compute Die (GCD), while the other is the MCD. The GCD is based on the latest 5nm process node, whereas the MCD is based on the 6nm process node.

Due to the MCD utilizing older technology, it is more affordable to mass-produce, cutting down on costs. Therefore, consumers might see AMD Radeon RX 7000 series GPUs cost less than its competitor. Also, as the GCD houses all the GPU cores, the 5nm’s higher transistor density will help improve performance.

Radiance Display Engine

For a long time, graphics card manufacturers have not touched the refresh rate of their GPUs significantly. However, with AMD’s Radeon RX 7000 series GPUs, we see a new display engine, coined “AMD Radiance Display Engine.” Not only does the Radiance Display Engine showcase absurd refresh rates, but it also improves the 12-bit per-channel colors.

To explain, the AMD Radiance Display Engine allows up to 68 billion colors in its 12-bit per-channel colors. So, you shall be able to view even more realistic colors than ever. To top it off, the AMD Radiance Display Engine allows up to 165 FPS at 8K and 480 FPS at 4K resolutions. While the refresh rates go above and beyond the norm, we should easily surpass 165 FPS at the highest 8K resolution with future iterations of GPUs.

Also Read: Gigabyte M27Q-P Gaming Monitor Review

Fidelity FX Super Resolution 3.0

Upscaling for RX 7900 XT and RTX 3080 Ti — Fidelity FX Super Resolution 3.0 announcement. Source: AMD

Since AMD’s Radeon RX 6000 series GPUs, we have seen the endorsement of its upscaling technology, “Fidelity FX Super Resolution.” Unlike NVIDIA’s Deep Learning Super Sampling (DLSS), FSR relies on software to upscale images. With the inception of FSR 2.0, we saw AMD tackle NVIDIA’s DLSS 2.0 head-to-head with almost the same performance.

However, AMD has now gone one step ahead and has announced its futuristic FSR 3.0, along with the Radeon RX 7000 series GPUs. According to AMD sources, FSR 3.0 will be twice as fast as FSR 2.0. So, unless NVIDIA can overcome such a massive wall with its DLSS 3.0, FSR will easily take the upscaling crown.

Moreover, because AMD utilizes software to upscale images, the quality of the images does not match the quality of DLSS. Therefore, AMD has also announced AI accelerators and Fluid Motion Frames to support its FSR 3.0 technology.

To explain, AI accelerators are AMD’s version of NVIDIA’s Tensor Cores. In contrast, not much information has been released regarding Fluid Motion Frames. Although, speculations suggest Fluid Motion Frames will help FSR 3.0 produce unique frames, like DLSS 3.0.

Ampere — NVIDIA RTX 3000 Series Architecture

Moving towards NVIDIA’s older RTX 3000 series architecture, “Ampere,” we see a different architecture utilized. As we are comparing the AMD Radeon RX 7900 XT Vs RTX 3080 Ti GPUs, knowing the difference in the architecture of the entire generation will help differentiate performance.

Although the AMD Radeon RX 7000 series GPUs are newer, the pricing and thermals have confused consumers. Moreover, the usage of raytracing and upscaling technologies adds to the confusion.

Process Node

When talking about next-gen hardware, many enthusiasts always anticipate improvements in the process node. Not only does a smaller process node improve the thermal efficiency of a GPU, but it also increases the transistor density. Although, when it comes to process nodes, NVIDIA is usually a bit slower when updating its lithography.

However, with its GeForce RTX 3000 series GPUs, NVIDIA has shifted to Samsung’s 8nm process node. In comparison, NVIDIA’s GeForce RTX 2000 series GPUs utilized TSMC’s 12nm process node. Similarly, because the process node has significantly shrunk down, Ampere GPUs have a remarkably higher transistor count.

For example, the NVIDIA GeForce RTX 2080 Ti has a transistor count of 18.6 billion transistors. On the other hand, the NVIDIA GeForce RTX 3080 Ti offers 28.3 billion transistors. While the NVIDIA GeForce RTX 3080 Ti barely has half the transistors of the AMD Radeon RX 7900 XT, housing 57.7 billion transistors, its performance is unparalleled.

Resizable BAR

When the NVIDIA GeForce RTX 3000 series GPUs launched, the company introduced a new feature named “NVIDIA Resizable BAR.” To explain, the NVIDIA Resizable BAR technology allows the transfer of data such as shaders, textures, and geometry without limits. The data is transferred from the VRAM to the CPU.

The concept vaguely resembles AMD’s Smart Access Memory (SAM) technology. However, unlike AMD’s SAM, Resizable BAR is only limited to AMD Ryzen 5000 series processors and above and Intel 10th gen processors and above.

While Resizable BAR and SAM are amazing concepts, the performance increment is negligible. For example, at most, you can gain 2-4% extra performance with Resizable BAR. But any form of free performance equates to a better experience.

Deep Learning Super Sampling 2.0

After launching its GeForce RTX 2000 series GPUs, NVIDIA launched the first-ever advanced upscaling technology, “Deep Learning Super Sampling.” Although DLSS 1.0 was lackluster, it showed a lot of promise for the future.

In that regard, with its GeForce RTX 3000 series GPUs, NVIDIA launched the next iteration of DLSS 1.0, the DLSS 2.0. Due to the low quality of upscaling with the DLSS 1.0, gamers preferred to keep it off. Although disabling DLSS and enabling raytracing means a significant performance drop, the technology wasn’t worth it.

However, now with DLSS 2.0, we see better quality frames through upscaling and greater FPS gains. Even so, enabling raytracing with DLSS 2.0 might prove to be a bit difficult at higher resolutions. Therefore, NVIDIA has released DLSS 3.0 with its GeForce RTX 4000 series GPUs.

Similar to AMD’s FSR 3.0, NVIDIA’s DLSS 3.0 creates unique frames to increase FPS. But according to reviews, the unique frames have worse quality than DLSS 2.0, making them fairly visible. Therefore, staying on DLSS 2.0 might be better than upgrading to DLSS 3.0. So, aside from the GPU comparison, we’re also looking at the performance differences between FSR 3.0 Vs DLSS 2.0.

RX 7900 XT Vs RTX 3080 Ti Specifications

Through the analysis of the architectural differences between both graphics cards, we have found certain differences. Looking at their specifications will show us which GPU performs better on paper. Knowing the specifications makes calculating both GPUs’ performance per dollar ratio easier.

GPU Name	NVIDIA GeForce RTX 3080 Ti	AMD Radeon RX 7900 XT
Node	Samsung 8nm	TSMC 5nm (GCD), TSMC 6nm (MCD)
Architecture	Ampere	RDNA 3
CUDA Cores/ Stream Processors	10,240	5,376
Core Clock	1365MHz	1500MHz
Boost Clock	1665MHz	2394MHz
Memory	12GB GDDR6X	20GB GDDR6
Memory Clock	19 Gbps	20 Gbps
Memory Bandwidth	912.4GB/s	800GB/s
RT Cores	80	84
Tensor Cores	320	N/A
Transistors	28.3 billion	57.7 billion
TDP	350W	300W
Power Connector	1 x 12 pin	2 x 8 pin
Launch Date	June 3, 2021	December 13, 2022
MSRP	$1,199	$899
Current Price	$1,140	$900
Best Graphics Cards	Best RTX 3080 Ti Cards	Best RX 7900 XT Cards

Analyzing the table above shows a few differences between both graphics cards. For example, the newer AMD Radeon RX 7900 XT follows a multi-chip die with the 5nm and 6nm process, while the NVIDIA GeForce RTX 3080 Ti only has an 8nm process node. Want to see how it performs against Nvidia’s top card? Check out our RTX 4090 Vs RTX 3080 Ti comparison.

Furthermore, looking at the CUDA cores, the NVIDIA GeForce RTX 3080 Ti showcases double the amount the AMD GPU does. However, the AMD card takes back the lead in core clocks and VRAM. Furthermore, the NVIDIA GeForce RTX 3080 Ti requires 50W more wattage than the AMD Radeon RX 7900 XT.

Lastly, comparing prices, we notice that the NVIDIA GPU is ~26.6% more expensive than the AMD Radeon RX 7900 XT. So, the NVIDIA GeForce RTX 3080 Ti needs to showcase 27% more performance than its competitor to win the performance per dollar battle. However, looking at the raw specifications, it doesn’t seem easy to accomplish for the NVIDIA GPU.

Gaming Performance At 1440P

As stated above, conducting practical benchmarks is important when comparing two pieces of hardware. Not only do practical tests show how the hardware performs, but they also show the PC specs required to get the FPS shown. Therefore, the following RX 7900 XT vs RTX 3080 Ti 1440P benchmarks have been conducted by Benchmark PC Tech.

Testing Rig

Processor: Intel Core i5-13600K
Motherboard: ASUS STRIX Z790-F
RAM: Kingston Fury Renegade RGB DDR5 32GB 6400MHz
Storage: Samsung 980 SSD 1TB
Case: Cooler Master H500P
Graphics Card: AMD Radeon RX 7900 XT & NVIDIA GeForce RTX 3080 Ti

Looking at the testing rig specifications for the processor, we see the Intel Core i5-13600K being utilized. Although the GPUs require more power, the Intel Core i5-13600K should be enough at higher resolutions. Due to utilizing a bigger 1440P resolution monitor, the processor’s workload is decreased, making games more GPU-dependent. Therefore, you can pair slightly weaker CPUs with more powerful GPUs at 1440P and 4K resolutions.

Secondly, moving toward the motherboard, we realize that the ASUS Strix Z790-F is utilized. Although the processor is a bit older, with a high-end motherboard, we ensure a better upgrade path. Also, with the Z790-F motherboard, we see better VRMs, allowing for better overclocking.

Furthermore, talking about the RAM, we have the Kingston Fury Renegade DDR5 32GB kit. While DDR5 RAM is much faster than DDR4, the adoption rate is slow. Due to the higher prices and low gaming benefits, not many gamers prefer DDR5 over DDR4. However, if you want to work on production tasks, DDR5 RAM is a promising investment. If you want more information regarding DDR4 and DDR5 RAM, we recommend reading through our DDR4 Vs DDR5 RAM comparison!

Lastly, talking about the case, we have the Cooler Master H500P case. Because the NVIDIA GeForce RTX 3080 Ti and AMD Radeon RX 7900 XT are massive cards, we require a big case. For that reason, the H500P should do an ideal job housing the GPUs and keeping the cooling in check. Moreover, with better cooling, the graphics cards will have better thermal efficiency, further improving performance.

Red Dead Redemption 2

RX 7900 XT and RTX 3080 Ti FPS test — Red Dead Redemption 2 performance benchmarks at 1440P resolution.

The AMD Radeon RX 7900 XT averages 129 FPS, while the NVIDIA GeForce RTX 3080 Ti averages 100 FPS.
The AMD Radeon RX 7900 XT is ~29% faster than the NVIDIA GeForce RTX 3080 Ti in Red Dead Redemption 2.
Although the NVIDIA GeForce RTX 3080 Ti is ~26.6% more expensive, it loses by a high margin.

Assassin’s Creed Valhalla

The AMD Radeon RX 7900 XT averages 127 FPS. On the other hand, the NVIDIA GeForce RTX 3080 Ti averages 84 FPS.
The AMD Radeon RX 7900 XT is ~51.19% faster than the NVIDIA GeForce RTX 3080 Ti in Assassin’s Creed Valhalla.

Shadow Of The Tomb Raider

RX 7900 XT Vs NVIDIA GPU — Shadow of the tomb raider FPS test at 1440P resolution, ultra settings.

The AMD Radeon RX 7900 XT averages 181 FPS, whereas the NVIDIA GeForce RTX 3080 Ti averages 156 FPS.
The AMD Radeon RX 7900 XT is ~16.02% faster than its competitor in the shadow of the tomb raider.

Also Check: RTX 3080 10GB Vs RTX 3080 12GB

Forza Horizon 5

The AMD Radeon RX 7900 XT averages 122 FPS. On the other hand, the NVIDIA GeForce RTX 3080 Ti averages 104 FPS.
The AMD Radeon RX 7900 XT is ~17.3% faster than the NVIDIA GeForce RTX 3080 Ti in Forza Horizon 5.

God Of War

The AMD Radeon RX 7900 XT averages 146 FPS, whereas the NVIDIA GeForce RTX 3080 Ti averages 112 FPS.
The AMD Radeon RX 7900 XT is ~30.35% faster than the NVIDIA GeForce RTX 3080 Ti in God of War.

Dying Light 2

RX 7900 XT and RTX 3080 Ti benchmark — Dying Light 2 FPS benchmarks.

The AMD Radeon RX 7900 XT averages 140 FPS. On the other hand, the NVIDIA GeForce RTX 3080 Ti averages 110 FPS.
The AMD Radeon RX 7900 XT is ~27.27% faster than the NVIDIA GeForce RTX 3080 Ti in Dying Light 2.

Elden Ring

The AMD Radeon RX 7900 XT averages 102 FPS, whereas the NVIDIA GeForce RTX 3080 Ti averages 93 FPS.
The AMD Radeon RX 7900 XT is ~9.6% faster than the NVIDIA GeForce RTX 3080 Ti in Elden Ring.
While the AMD GPU showed greater differences, the difference is significantly cut in Elden Ring.

Gaming Performance At 4K

After examining the gaming benchmarks at the 1440P resolution, we must see how the GPUs differ at 4K. Because we’re testing high-end GPUs, consumers might buy them for gaming at higher resolutions with a faster refresh rate. While the AMD Radeon RX 7900 XT dominated the NVIDIA GeForce RTX 3080 Ti in 1440P, we might see different results in 4K. Therefore, the following tests have been conducted by Mark PC.