Erziele einen Leistungsschub mit NVIDIA DLSS (Deep Learning Super Sampling). KI-spezifische Tensor-Cores auf GeForce RTX-GPUs geben deinen Spielen einen Geschwindigkeitsschub bei kompromissloser Bildqualität. Auf diese Weise kannst du die Einstellungen und die Auflösung für ein noch besseres visuelles Erlebnis erhöhen.
VIDEO VIDEO
Technische Daten
GeForce RTX 3090 Ti GeForce RTX 3090 GeForce RTX 3080 Ti GeForce RTX 3080 GeForce RTX 3070 Ti GeForce RTX 3070 GeForce RTX 3060 Ti GeForce RTX 3060 GeForce RTX 3050
NVIDIA CUDA-Recheneinheiten
10.
10.496
10.240
8.960 / 8.704
6.144
5.888
4.864
3.584
2.560 / 2.304
Boost-Taktung (GHz)
1,86
1,70
1,67
1,71
1,77
1,73
1,67
1,78
1,78 / 1,76
Speichergröße
24 GB
24 GB
12 GB
12 GB / 10 GB
8 GB
8 GB
8 GB
12 GB
8 GB
Speichertyp
GDDR6X
GDDR6X
GDDR6X
GDDR6X
GDDR6X
GDDR6
GDDR6
GDDR6
GDDR6
Hinweis: GeForce RTX 3080, 3070 und 3060 Ti mit LHR (Lite Hash Rate) werden ab Mai 2021 ausgeliefert.
Weitere Merkmale und Vorteile
Alle Produkte der GeForce RTX 30-Serie ansehen
GeForce Grafikkarten und Desktop-PCs der RTX 40-Serie
GeForce Laptops der RTX 40-Serie
Grafikkarten und Desktop-PCs der GeForce RTX 30-Serie
Notebooks der GeForce RTX 30-Serie
GeForce RTX 3060
GeForce RTX 3050
Bezugsquellen
© 2021 NVIDIA Corporation. NVIDIA, das NVIDIA-Logo, GeForce, GeForce Experience, GeForce RTX, G-SYNC, NVLink und ShadowPlay sind Marken bzw. eingetragene Marken der NVIDIA Corporation in den USA und in anderen Ländern. Alle anderen Marken und Copyrights sind das Eigentum der jeweiligen Rechteinhaber.
GeForce
GeForce
GeForce
GeForce
GeForce
GeForce
GeForce
GeForce
GeForce
Technische Daten des Grafikprozessors
NVIDIA CUDA® -Recheneinheiten
10.
10.496
10.240
8.960 / 8.704
6.144
5.888
4.864
3.584
2.560 / 2.304 (1)
Boost-Taktung (GHz)
1,86
1,70
1,67
1,71
1,77
1,73
1,67
1,78
1,78 / 1,76 (1)
Basistaktung (GHz)
1,56
1,40
1,37
1,26 / 1,44
1,58
1,50
1,41
1,32
1,55 / 1,51 (1)
Technische Daten des Speichers:
Standard-Speicherkonfiguration
24 GB GDDR6X
24 GB GDDR6X
12 GB GDDR6X
12 GB GDDR6X / 10 GB GDDR6X
8 GB GDDR6X
8 GB GDDR6
8 GB GDDR6 / 8 GB GDDR6X
12 GB GDDR6 / 8 GB GDDR6
8 GB GDDR6
Breite der Speicherschnittstelle
384 Bit
384 Bit
384 Bit
384 Bit / 320 Bit
256 Bit
256 Bit
256 Bit
192-Bit / 128-Bit
128 Bit
Unterstützte Technologien:
Raytracing-Kerne
2.
2. Generation
2. Generation
2. Generation
2. Generation
2. Generation
2. Generation
2. Generation
2. Generation
Tensor-Recheneinheiten
3. Generation
3. Generation
3. Generation
3. Generation
3. Generation
3. Generation
3. Generation
3. Generation
3. Generation
NVIDIA-Architektur
Ampere
Ampere
Ampere
Ampere
Ampere
Ampere
Ampere
Ampere
Ampere
Microsoft DirectX® 12 Ultimate
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA DLSS
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Reflex
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Broadcast
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
PCI Express Gen 4
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Resizable BAR
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA® GeForce Experience™
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Ansel
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Freestyle
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA ShadowPlay
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Highlights
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA G-SYNC®
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Game Ready-Treiber
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Studio-Treiber
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Omniverse
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA GPU Boost™
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA NVLink™ (SLI-Ready)
Ja
Ja
–
–
–
–
–
–
–
Vulkan RT API, OpenGL 4.
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
NVIDIA Encoder (NVENC)
7. Generation
7. Generation
7. Generation
7. Generation
7. Generation
7. Generation
7. Generation
7. Generation
7. Generation
NVIDIA Decoder (NVDEC)
5. Generation
5. Generation
5. Generation
5. Generation
5. Generation
5. Generation
5. Generation
5. Generation
5. Generation
CUDA-Fähigkeit
8.6
8.6
8.6
8.6
8.6
8.6
8.6
8.6
8.6
VR Ready
Ja
Ja
Ja
Ja
Ja
Ja
Ja
Ja
–
Unterstützte Anzeigeoptionen:
Maximale digitale Auflösung (2)
7680 x 4320
7680 x 4320
7680 x 4320
7680 x 4320
7680 x 4320
7680 x 4320
7680 x 4320
7680 x 4320
7680 x 4320
Standard-Bildschirmanschlüsse
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 3x DisplayPort(4)
HDMI(3) , 2x DisplayPort(4)
Multidisplayfähig
4
4
4
4
4
4
4
4
4
HDCP
2,3
2,3
2,3
2,3
2,3
2,3
2,3
2,3
2,3
Abmessungen der Karte
Länge
12,3 ” (313 mm)
12,3 ” (313 mm)
11,2 ” (285 mm)
11,2 ” (285 mm)
10,5 ” (267 mm)
9,5 ” (242 mm)
9,5 ” (242 mm)
Breite
5,4 ” (138 mm)
5,4 ” (138 mm)
4,4 ” (112 mm)
4,4 ” (112 mm)
4,4 ” (112 mm)
4,4 ” (112 mm)
4,4 ” (112 mm)
Steckplatz
3 Steckplätze
3 Steckplätze
2 Steckplätze
2 Steckplätze
2 Steckplätze
2 Steckplätze
2 Steckplätze
Temperatur- und Leistungsdaten:
Max.
92
93
93
93
93
93
93
93
93
Leistungsaufnahme der Grafikkarte (W)
450
350
350
350 / 320
290
220
200
170
130
Erforderliche Systemleistung (W) (5)
850
750
750
750
750
650
600
550
550
Zusätzliche Stromanschlüsse
3x PCIe-8-polige Kabel (Adapter im Lieferumfang) ODER PCIe-Gen-5-Kabel mit 450 W oder mehr
2x PCIe 8-polig
2x PCIe 8-polig
2x PCIe 8-polig
2x PCIe 8-polig
1x PCIe 8-polig
1x PCIe 8-polig
1x PCIe 8-polig
1x PCIe 8-polig
Alle Technischen Daten Anzeigen
Nvidia Graphics Cards List In Order Of Performance
TABLE OF CONTENTS
1
A list of Nvidia Graphics Cards in order of performance is something I keep looking for myself.
That is why I have put together this page for you with the most recent and some older but still widely used Nvidia Graphics Cards in a list that you can sort to your liking .
Also check out our related GPU and CPU Lists:
AMD Graphics Card List in order of performance
AMD Desktop CPU List in order of performance
Intel Desktop CPU List in order of performance
AMD Desktop CPU List in order of performance
List of Desktop CPUs with the highest single core performance
Nvidia Graphics Cards List In Order Of Performance
Graphics Card
Render Score
Gaming Score
Performance Total
Performance / Dollar
Price (MSRP)
Nvidia RTX 4090
1271
370
1640
1.02
1599
Nvidia RTX 4080
952
290
1242
1.03
1199
Nvidia RTX 3090 Ti
675
278
953
0.
1999
Nvidia RTX 4070 Ti
706
240
946
1.18
799
Nvidia RTX 3090
652
260
912
0.61
1499
Nvidia RTX 3080 Ti
642
232
874
0.73
1199
Nvidia RTX 4070
646
190
836
1.40
599
Nvidia RTX 3080
554
238
792
1.13
699
Nvidia Quadro RTX A6000
540
207
747
0.16
4650
Nvidia Quadro RTX A5000
511
185
696
0.19
2599
Nvidia RTX 3070 Ti
432
165
597
0.99
599
Nvidia Titan V
396
171
567
0.19
3000
Nvidia RTX 3070
399
158
557
1.11
499
Nvidia RTX 4060 Ti
410
145
555
1.
399
Nvidia RTX 3080 [Laptop]
399
130
529
0 (N/A)
0 (N/A)
Nvidia RTX 2080 Ti
346
172
518
0.43
1199
Nvidia RTX Titan
326
191
517
0.19
2700
Nvidia RTX 3060 Ti
366
132
498
1.24
399
Nvidia RTX 3070 [Laptop]
371
111
482
0 (N/A)
0 (N/A)
Nvidia RTX 3060
284
98
382
1.16
329
Nvidia Titan Pascal
250
124
374
0.23
1650
Nvidia RTX 2080 SUPER
234
131
365
0.48
750
Nvidia RTX 3060 [Laptop]
268
87
355
0 (N/A)
0 (N/A)
Nvidia GTX 1080 Ti
221
129
350
0.
700
Nvidia RTX 2080
226
121
347
0.43
799
Nvidia Quadro RTX A4000
239
94
333
0.28
1149
Nvidia RTX 2070 SUPER
220
112
332
0.63
520
Nvidia RTX 2070
210
107
317
0.71
450
Nvidia RTX 2060 SUPER
203
99
302
0.71
420
Nvidia RTX 2060
170
85
255
0.72
350
Nvidia GTX 1080
148
104
252
0.50
500
Nvidia RTX 3050
179
72
251
1.01
249
Nvidia GTX 1070 Ti
153
94
247
0.61
400
Nvidia RTX 3050 Ti [Laptop]
171
58
229
0 (N/A)
0 (N/A)
Nvidia GTX 980 Ti
136
79
215
0.
230
Nvidia GTX 1070
132
78
210
0.60
350
Nvidia GTX 1660 Ti
132
74
206
0.68
300
Nvidia GTX 1660 SUPER
136
70
206
0.82
250
Nvidia RTX 3050 [Laptop]
151
54
205
0 (N/A)
0 (N/A)
Nvidia GTX 1660
117
64
181
0.86
210
Nvidia GTX 1060
95
57
152
0.61
250
Nvidia GTX 980
93
59
152
0.95
160
Nvidia GTX 1650 SUPER
82
58
140
0.82
170
Nvidia GTX 1650
78
41
119
0.74
160
Nvidia GTX 1050 Ti
54
32
86
0.57
150
Nvidia GTX 960
46
34
80
0.
90
Nvidia GTX 1050
46
28
74
0.67
110
Graphics Card
Render Score
Gaming Score
Performance Total
Performance / Dollar
Price (MSRP)
The Nvidia GPU Comparison List above includes the Nvidia GTX Series and RTX Series GPUs. You can rank this GPU tier list by sorting it to your liking.
You can use all Nvidia Graphics Cards for many different workloads such as Gaming or Rendering and 3D Modeling. If you need RayTracing Cores for your work or games, though, you will have to buy an RTX GPU.
The amount of Video Memory, or VRAM, depends greatly on the type of workloads you are running. Be sure to give our VRAM guide a read to find out how much you’ll need.
Best Nvidia Graphics Card for the money
The currently best Nvidia GPU for the money is the Nvidia RTX 3060. It comes in at under 350$ and leads the list in the top value spot.12GB of GDDR6 VRAM that clock at 1875MHz with a bandwidth of 360GB/s.
At a Power Draw of 170W TDP , the chip clocks at 1320MHz and can boost up to roughly 1777MHz, depending on the variant of card you are looking at.
Some Brands such as Gigabyte, with its 3060 Gaming OC variant, overclock the GPU slightly to gain some extra performance.
Make sure you get this at or near its MSRP Price of 329$. A lot of GPUs are price-inflated right now and waiting for prices to return to normal might make sense for you.
Best Nvidia Graphics Card under 500$
The recently released Nvidia RTX 3070 is positioned in-between the RTX 2080 Super and RTX 2080 Ti. The performance of the RTX 3070 though tells a different story, as it is much closer, and in many benchmarks on par, with an RTX 2080 Ti at a significantly lower price.
At under 500$, the RTX 3070 features 8GB of GDDR6 VRAM that clock at 1750MHz with a bandwidth of 448GB/s.
The RTX 3070 is manufactured on an 8nm process node and its chip clocks in at 1500MHz base and 1725MHz Boost Clock. There are a multitude of partner-cards available that have different overclocks and coolers for you to choose from.
The RTX 3070 comes with 5888 Shading Units (CUDA Cores) and is rated at 220W of TDP .
Best Nvidia Graphics Card under 400$
Although the Nvidia RTX 3060 has surpassed its bigger brother, the Nvidia RTX 3060 Ti in terms of performance per dollar, the RTX 3060 Ti is still a great buy for those that want some additional performance over the RTX 3060.
The Nvidia RTX 3060 Ti sits among the top 2 Value-Based GPUs with serious Gaming and Rendering Performance.
It too is built on the 8nnm Process Node but, weirdly, has 8GB of VRAM – 4GB less than the weaker RTX 3060.
The Nvidia RTX 3060 Ti has 4864 CUDA Cores and a Chip that clocks in at 1410MHz Base and up to 1750MHz Boost.
Best Nvidia Graphics Card under 200$
There is an excellent Nvidia Graphics Card at every Price Point and the Sub-200$ Mark is no different. In this price-tier the Nvidia GTX 1650 is the clear winner, giving you excellent performance in both gaming and rendering at a budget.
The GTX 1650 sports 4GB of GDDR5 VRAM that clock at 2GHz on a 128bit Bus and bandwidth of 128GB/s. This card is rated at only 75W, making sure it runs extremely quiet and staying nice and cool.
896 CUDA Cores accelerate your gaming and rendering performance decently, and the chip that clocks in at 1485MHz Base and 1860Mhz Boost, will make sure you have a smooth experience on a budget.
Benchmark List Performance Metrics
The performance metrics that you see in the above Nvidia GPU ranking list cover different areas:
Nvidia Graphics Cards have lots of technical features like shaders, CUDA cores, memory size and speed, core speed, overclock-ability, to name a few.
The list could go on, but what I want to give you here is a quick and easy overview of Nvidia Graphics Cards in order of Performance throughout two of the most popular use cases on this site.
Rendering and Gaming.
I have taken the performance average of currently Popular gaming benchmarks such as Futuremark and assigned points depending on the benchmark score.
To find the best performing Nvidia Graphics Cards in Rendering I took the average of the three most popular GPU Render Engines: Redshift, Octane, and Vray-RT, and assigned points depending on the performance.
What you ultimately get from this list is an Nvidia Graphics Cards comparison.
Answers to frequently asked questions (FAQ)
Is RTX faster than GTX?
Yes, Nvidia RTX GPUs are faster than GTX GPUs. They come with Hardware-level Raytracing Cores and are segmented towards the high-end of the Graphics Card market often also featuring more VRAM, higher clocks, more CUDA Cores and a superior GPU architecture.
How do I choose a good graphics card?
The best way to choose a graphics card is to make note of the kind of workloads you’ll be running and then look for benchmarks that’ll show you how fast different GPUs perform in said workloads. All you need to do then is set a budget you’re willing to spend and get the fastest GPU for your workloads that fits your budget.
Should I get an Nvidia or AMD GPU?
Both Nvidia and AMD offer GPUs that compete inside the same performance and price brackets. Whether you should pick Nvidia or AMD will also depend on if you need specific features that are only supported on one of them. GPU Render Engines, for example, are often tied to using CUDA cores, which are only available on Nvidia GPUs. Some Games might run faster on AMD GPUs, or vice versa.
Check the requirements of Application or Game you’re looking to run on the new Graphics Card to narrow down your choices.
Over to you
What Nvidia Graphics Card do you want to buy? Let me know in the comments or ask us anything in our expert forum!
CGDirector is Reader-supported.
GeForce RTX 20 Gaming Cards & Laptops
Accelerated with NVIDIA DLSS
Maximum AI FPS and Quality
NVIDIA DLSS (deep learning anti-aliasing) delivers huge performance boosts. The technology accelerates the frame rate and also provides a new level of graphics quality with the help of Tensor Cores in GeForce RTX graphics cards. Turn up the settings and resolution to the maximum and play with amazing realism.
VIDEO
*Recorded on GeForce RTX 30 in 4K.
Built for streaming
Enhance your streaming experience with GeForce RTX. GeForce RTX graphics cards are equipped with the NVIDIA encoder (NVENC), which will increase performance and ensure that the finest visual details of the best moments of the gameplay are conveyed.
VIDEO
Compare
Where to buy
Russia
Ukraine
Kazakhstan
Options
© 2022 NVIDIA Corporation.
NVIDIA Introduces GeForce RTX 4090 and Two GeForce RTX 4080 – New Generation Accelerators Starting at $899
3DNews Technologies and IT market. Video card news NVIDIA introduced the GeForce RTX 4090 and two…
The most interesting in the reviews
09/20/2022 [19:12],
Nikolai Khizhnyak
NVIDIA officially introduced the new generation of GeForce RTX 4000 graphics accelerators. As part of today’s presentation, its head Jensen Huang announced the first models of the new series – GeForce RTX 4090, GeForce RTX 4080 with 16 GB of memory and GeForce RTX 4080 with 12 GB of memory . In the reference version, new items largely repeat the design of previous models.
Image source: NVIDIA
GeForce RTX 4000 graphics cards use GPUs based on the new Ada Lovelace architecture. The chips have up to 76 billion transistors and are manufactured using TSMC’s 4nm N4 process. The key features of the new architecture are the 3rd generation of RT cores and the 4th generation of tensor cores.
The company announced that the new graphics cards based on the Ada Lovelace architecture support DLSS 3.0 image scaling technology, which significantly improves game performance.
The flagship GeForce RTX 4090 graphics card is based on the AD102-300 GPU with 16,384 CUDA cores. The declared frequency of its operation is up to 2520 MHz. The theoretical performance of the video card in single precision operations is 82.6 Tflops, which is 2.3 times higher than that of its predecessor based on the Ampere architecture (GeForce RTX 3090). The new video card received 24 GB of GDDR6X memory with support for a 384-bit bus, an effective frequency of 21 GHz and a peak bandwidth of 1 TB / s.
The GeForce RTX 4090 will be available on October 12th. Its recommended price is $1599.
Along with the new flagship accelerator, Huang introduced two cards with the same name but different specifications. The first is the GeForce RTX 4080 16GB. It is based on a halved AD103-300 processor with 9728 CUDA cores, clocked at up to 2505 MHz. The card received 16 GB of GDDR6X memory with a bus width of 256 bits, a speed of 22.5 Gb / s per contact (effective frequency of 22.5 GHz) and a bandwidth of 720 GB / s. The declared power consumption of the novelty is 320 watts. This model will go on sale in November. Its recommended price is $1199.
What was supposed to be the GeForce RTX 4070 model ended up being called the GeForce RTX 4080 12GB. This card uses the AD104-400 GPU with 7680 CUDA cores running at up to 2610MHz.
Raport / rapport 50 Years of Cutting and Pasting: Modernizing Canada’s Tax System
