Benchmark Numbers

Our Test Rig

Now for the fun part – performance! We have bumped up our graphics card test rig and comes with the following specifications:

• 10^th Gen Intel Core i9-10900K (3.7GHz base clock, 20MB Intel Smart L3 cache)
• ASUS ROG Maximus XII Formula (Intel Z490)
• WD Black SN750 2TB SSD
• Windows 10 Home 64-bit
• Acer Predator X27

Naturally, we’ll be stacking the RTX 3080 Founders Edition against the class of “80” cards from the RTX 20 Series. This includes the full suite of the RTX 2080 Founders Edition cards, including the original 2080, 2080 Super and the current reigning king of the hill RTX 2080 Ti. For good measure, I’ve also included an older GTX 1080 Ti Founders Edition to see how far a leap the RTX 3080’s performance compares.

• NVIDIA GeForce RTX 3080
• NVIDIA GeForce RTX 2080 Ti
• NVIDIA GeForce RTX 2080 Super
• NVIDIA GeForce RTX 2080
• NVIDIA GeForce GTX 1080 Ti

All graphics cards were installed with the yet-to-be-released ForceWare 456.16 driver.

Test cards compared

	NVIDIA GeForce RTX 3080 Founders Edition	NVIDIA GeForce RTX 2080 Ti Founders Edition	NVIDIA GeForce RTX 2080 Super Founders Edition	NVIDIA GeForce RTX 2080	NVIDIA GeForce GTX 1080 Ti Founders Edition
Core Code	GA102-200	TU102	TU104	TU104	GP102
GPU Transistor Count	28.3 billion	18.9 billion	13.6 billion	13.6 billion	12 billion
Manufacturing Process	8nm (Samsung)	12nm FinFET	12nm FinFET	12nm FinFET	16nm
Core Clock	Base: 1,440MHz Boost: 1,710MHz	Core: 1,350MHz, Boost: 1,635MHz	Core: 1,650MHz, Boost: 1,815MHz	Core: 1,515MHz, Boost: 1,800MHz	1480MHz (Boost: 1582MHz)
Stream Processors	8,704	4,352	3,072	2,944	3584
Texture Mapping Units (TMUs)	272	272	192	184	224
Raster Operator units (ROP)	96	88	64	64	88
Memory Clock (DDR)	10GB GDDR6X 19,000MHz	11GB GDDR6 14,000MHz	8GB GDDR6 15,500MHz	8GB GDDR6 14,000MHz	11000MHz
Memory Bus width	320-bit	352-bit	256-bit	256-bit	352-bit
Memory Bandwidth	760GB/sec	616GB/s	496.1GB/s	448GB/s	484.4 GB/s
PCI Express Interface	4.0 / 3.0	3.0	3.0	3.0	PCI Express 3.0
Power Connectors	1 x 12p-pin	2 x 8-pin	1x 8-pin, 1x 6-pin	1x 8-pin, 1x 6-pin	1 x 8-pin, 1x 6-pin
Multi GPU Technology	No	Yes (2-way SLI)	Yes (2-way SLI)	Yes (2-way SLI)	SLI
DVI Outputs	No	No	No	No	—
HDMI Outputs	1	1x	1x	1x	1
DisplayPort Outputs	3	3x	3x	3x	3
HDCP Output Support	Yes	Yes	Yes	Yes	Yes
Stream Processor Clock	—	—	1,650MHz	—	1480MHz

Benchmarks

We have also updated our benchmarks used, including the removal of some older titles for more recent ones and also those that supports ray tracing and/or DLSS:

• 3DMark
• Horizon Zero Dawn
• Wolfenstein: Youngblood
• Hitman 2
• Deus Ex: Mankind Divided
• Shadow of the Tomb Raider
• Metro: Exodus
• Red Dead Redemption 2
• Death Stranding
• Doom Eternal
• Control

We used the Fire Strike Extreme benchmark and stress test in 3DMark for our temperature and power consumption tests respectively.

3DMark

3DMark is a synthetic benchmark that tests graphics and computational performance at different resolutions, starting at 1080p and going all the way up to 4K. A series of two graphics test, one physics test, and then a combined test stresses your hardware in turn to assess its performance.

Unsurprisingly, the GeForce RTX 3080 knocks the wind out of the rest of the cards here. Interestingly, in Firestrike (based on DirectX 11) runs we see the GTX 1080 Ti matching up to the GeForce RTX 2080 yet on the other end of the spectrum, the RTX 2080 Ti is no match for the GeForce 3080. It’s only with TimeSpy, a DirectX 12 benchmark, do we see the GTX 1080 Ti lagging behind all RTX cards with the RTX 3080 clearly leading the way again.

Since the RTX cards support hardware ray tracing through its RT Cores, I’ve also used 3DMark’s Port Royal Ray Tracing benchmark to size up each card’s ray tracing capabilities. The GTX 10 Series cards run ray tracing too, but without dedicated RT Cores like on the RTX cards, it just can’t run playable frame rates with it enabled. What’s eye-catching here is how much better the RTX 3080 fares with ray tracing as compared to all three RTX 20 Series cards. The performance difference between it and the nearest RTX 2080 Ti is quite an impressive feat. We’ll find out more about this with real-world games benchmarks next page. In the meantime, let’s look at how the cards fared with DLSS turned on.

For the uninitiated, deep learning super sampling or DLSS uses artificial intelligence and machine learning to produce an image that looks like a higher-resolution image, without the rendering overhead. The idea is to make games rendered at 1440p look like they’re running at 4K, or 1080p games to look like 1440p. This is all possible thanks to NVIDIA’s Tensor cores, which are only available in RTX GPUs. Although RTX 20 series GPUs have Tensor cores inside, the RTX 30 Series come with NVIDIA’s second-generation Tensor cores, which offer greater per-core performance – and the results here isn’t surprising as much as it was expected of the RTX 3080 card.

Now let’s look at how the card performs in games.