Wikipedia

GeForce 30 series: Difference between revisions

Browse history interactively
← Previous edit
Content deleted Content added
VisualWikitext
m no uppercase stylization in this context
use more general term, as Nvidia does not only design its GPUs but also does market research, prototyping, product testing, marketing, etc., which is all part of their development
 
(43 intermediate revisions by 20 users not shown)
Line 1:
{{short description|GPU series by Nvidia}}
{{pp-vandalism|small=yes}}
{{Use American English|date=January 2023}}
{{Use mdy dates|date=January 2023}}
{{Infobox graphics processing unit
| name = GeForce 30 series
| image = [[File:RTX 30 series logo with slogan.svg|300px|]][[File:RTX 3090 Founders Edition!.jpg|300px]]
| img_w =
| caption = GeForce RTX 3090 Founders Edition
| date = {{Start date and age|2020|9|17}}
| created =
| manufacturer = [[Samsung Electronics|Samsung]]
| designfirm = [[Nvidia]]
| marketed_by = Nvidia
| codename = GA10x
| architecture = [[Ampere (microarchitecture)|Ampere]]
| model = [[GeForce RTX]] RTX series
| cores-nothread =
| numcores =
| process =
| fab = Samsung [[10 nm process|8LPU8LPH]]
| transistors1 = 8.7B (GA107)
| transistors2 = 13.25B (GA106)
| transistors3 = 17.4B (GA104)
| transistors4 = 22B (GA103)
| transistors5 = 28.3B (GA102)
| enthusiast entry = {{ubl|GeForce RTX 3080|GeForce RTX 30803050 (126 GB)|GeForce RTX 3080 Ti3050|GeForce RTX 3090|GeForce3050 RTXTi 3090(laptop Tionly)}}▼
 
| entry midrange = {{ubl|GeForce RTX 30503060 (8 GB)|GeForce RTX 30503060|GeForce TiRTX (laptop3060 only)Ti}}
| midrangehighend = {{ubl|GeForce RTX 3070|GeForce RTX 3070 =Ti|GeForce {{ublRTX 3080|GeForce RTX 30603080 (812 GB)|GeForce RTX 3060|GeForce RTX 30603080 Ti}}
| highend enthusiast = {{ubl|GeForce RTX 30703090|GeForce RTX 30703090 Ti}}
| d3dversion = [[Direct3D#Direct3D 12|Direct3D 12.0 Ultimate]] ([[Feature levels in Direct3D#Direct3D 12|feature level 12_2]])<br />[[High-Level Shader Language|Shader Model 6.78]]▼
▲| enthusiast = {{ubl|GeForce RTX 3080|GeForce RTX 3080 (12&nbsp;GB)|GeForce RTX 3080 Ti|GeForce RTX 3090|GeForce RTX 3090 Ti}}
| vulkanapi openclversion = [[VulkanOpenCL#VulkanOpenCL 13.30|VulkanOpenCL 13.30]]<ref>{{cite web|url=https://developer.nvidia.com/vulkan-driveropencl|title=VulkanOpenCL Driver Support &#124; NVIDIA Developer|publisher=NVIDIA|website=developer.nvidia.com|date=1024 FebruaryApril 20162013 |access-date=September 15, 2022}}</ref>{{efn|name="OpenCL3.0"|In OpenCL 3.0, OpenCL 1.2 functionality has become a mandatory baseline, while all OpenCL 2.x and OpenCL 3.0 features were made optional.}}▼
 
| openglversion = [[OpenGL#OpenGL 4.6|OpenGL 4.6]]▼
▲| d3dversion = [[Direct3D#Direct3D 12|Direct3D 12.0 Ultimate]] ([[Feature levels in Direct3D#Direct3D 12|feature level 12_2]])<br />[[High-Level Shader Language|Shader Model 6.7]]
| openclversion vulkanapi = [[OpenCLVulkan#OpenCLVulkan 31.03|OpenCLVulkan 31.03]]<ref>{{cite web|url=https://developer.nvidia.com/openclvulkan-driver|title=OpenCLVulkan Driver Support &#124; NVIDIA Developer|publisher=NVIDIA|website=developer.nvidia.com|date=2410 AprilFebruary 20132016 |access-date=September 15, 2022}}</ref>{{efn|name="OpenCL3.0"|In OpenCL 3.0, OpenCL 1.2 functionality has become a mandatory baseline, while all OpenCL 2.x and OpenCL 3.0 features were made optional.}}
| predecessor = {{Unbulleted list|[[GeForce 16 series]]|[[GeForce 20 series]]}}
▲| openglversion = [[OpenGL#OpenGL 4.6|OpenGL 4.6]]
| variant = =▼
▲| vulkanapi = [[Vulkan#Vulkan 1.3|Vulkan 1.3]]<ref>{{cite web|url=https://developer.nvidia.com/vulkan-driver|title=Vulkan Driver Support &#124; NVIDIA Developer|publisher=NVIDIA|website=developer.nvidia.com|date=10 February 2016 |access-date=September 15, 2022}}</ref>
| predecessor successor = {{Unbulleted list|[[GeForce 16 series]]|[[GeForce 2040 series]]}}
▲| variant =
| successor = [[GeForce 40 series]]
 
| support status = Supported
}}
 
The '''GeForce 30 series''' is a suite of [[graphics processing unit]]s (GPUs) designed and marketeddeveloped by [[Nvidia]], succeeding the [[GeForce 20 series]]. The GeForce 30 series is based on the [[Ampere (microarchitecture)|Ampere]] architecture, which featurefeatures Nvidia's second-generation [[Ray tracing (graphics)#Interactive ray tracing|ray tracing]] (RT) cores and third -generation [[Deep learning super sampling#Architecture|Tensor Cores]].<ref>{{Cite web|title=NVIDIA GeForce RTX 30 Series GPUs Powered by Ampere Architecture|url=https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/|access-date=2020-09-02|website=NVIDIA|language=en-us}}</ref> ThroughPart of the [[Nvidia RTX]] series, [[Ray-tracing hardware|hardware]]-enabled [[real-time ray tracing]] is possible on GeForce 30 series cards.
 
The lineup, designed to compete with [[Advanced Micro Devices|AMD]]'s [[Radeon RX 6000 series]] of cards, consists of the entry-level and previously laptop-exclusive RTX 3050 and laptop-exclusive RTX 3050 Ti, mid-range RTX 3060, upper-midrange RTX 3060 Ti, high-end RTX 3070, RTX 3070 Ti, RTX 3080&nbsp;10&nbsp;GB, RTX 3080&nbsp;12&nbsp;GB and enthusiast RTX 3080 Ti, RTX 3090, and RTX 3090 Ti. This is the last generation from Nvidia to have official support for Windows 7 and 8.x as the latest drivers available for this generation require Windows 10.<ref>{{cite web|title=Support Plan for Windows 7 and Windows 8/8.1 &#124; NVIDIA|url=https://nvidia.custhelp.com/app/answers/detail/a_id/5201/related/1 |website=nvidia.custhelp.com|access-date=November 25, 2021}}</ref>
 
The GeForce 30 series began shipping on September 17, 2020.<ref>{{Cite web|url=https://nvidianews.nvidia.com/news/nvidia-delivers-greatest-ever-generational-leap-in-performance-with-geforce-rtx-30-series-gpus|title=NVIDIA Delivers Greatest-Ever Generational Leap with GeForce RTX 30 Series GPUs|first=NVIDIA|last=Newsroom|website=NVIDIA Newsroom Newsroom}}</ref><ref name="geforceSpecialEvent">{{cite web |title=GeForce Special Event |url=https://www.nvidia.com/en-us/geforce/special-event/ |website=[[Nvidia]] |access-date=September 1, 2020}}</ref> The initial launch, consisting of the RTX 3070, RTX 3080, and RTX 3090, occurred during the ongoing [[Global chip shortage (2020–2023)|2020–2023 global chip shortage]], resulting in widespread and notable shortages of the series as a whole that lasted from the series' launch until 2022.
 
The GeForce 30 series was succeeded by the [[GeForce 40 series]], powered by the [[Ada Lovelace (microarchitecture)|Ada Lovelace]] microarchitecture, which first launched in 2022.<ref>{{cite web|date=September 20, 2022|title=NVIDIA Delivers Quantum Leap in Performance, Introduces New Era of Neural Rendering With GeForce RTX 40 Series &#124; NVIDIA Newsroom|url=https://nvidianews.nvidia.com/news/nvidia-delivers-quantum-leap-in-performance-introduces-new-era-of-neural-rendering-with-geforce-rtx-40-series|website=nvidianews.nvidia.com}}</ref>
Line 56 ⟶ 52:
 
== Release and availability issues ==
{{See also|2020–present2020–2023 global chip shortage}}
Launch day for the RTX 3080 was September 17, 2020.<ref name=":1"/> The lack of pre-order functionality and high demand exacerbated by the [[COVID-19 pandemic]], and boost in scalping, resulted in a large number of online retailers struggling with the sheer number of purchases.<ref name=":1">{{Cite magazine|last=D'Anastasio|first=Cecilia|date=November 13, 2023|title=Desperate Gamers Camp Out in the Pandemic for $700 GPUs |url=https://www.wired.com/story/nvidia-amd-demand-pandemic-lines/|magazine=Wired}}</ref><ref>{{Cite web|last=Ridley|first=Jacob|date=September 24, 2020|title=Here's the latest on RTX 3080 stock, according to retailers and manufacturers|url=https://www.pcgamer.com/nvidia-rtx-3080-back-in-stock-check/|website=PC Gamer}}</ref> [[Newegg]] had completely sold out as expected on [[Black Friday (shopping)|Black Friday]]. Long lines formed outside physical stores with stock, such as [[Micro Center]] in the United States,<ref>{{Cite web|last=Freedman|first=Andrew |date=September 17, 2020|title=Nvidia, Newegg Address Nearly Non-Existent RTX 3080 Availability|url=https://www.tomshardware.com/news/nvidia-newegg-rtx-3080-apologize|website=Tom's Hardware}}</ref> and {{Ill|Thirdwave Group|ja|サードウェーブ|lt=Dospara}} in Japan.<ref>{{Cite web|last=株式会社インプレス|date=2020-09-18|title=GeForce RTX 3080の夜間販売に自作PCファン集結、120名以上の人で賑わうドスパラにはキャンセル分を求める行列も|url=https://akiba-pc.watch.impress.co.jp/docs/news/news/1277860.html |access-date=2020-09-27|website=Akiba PC Hotline!|language=ja}}</ref> Twitter users reported that they used bots to buy large numbers of cards to resell for higher prices.<ref name=":2">{{Cite web|last1=Kan|first1=Michael |date=September 17, 2020|title=How a Bot Bought Dozens of RTX 3080 Units Before Consumers Could Grab Them|url=https://uk.pcmag.com/graphic-cards/128682/how-a-bot-bought-dozens-of-rtx-3080-units-before-consumers-could-grab-them|website=PCMag UK}}</ref>
 
Line 77 ⟶ 73:
Architectural improvements of the Ampere architecture include the following:
* [[CUDA]] Compute Capability 8.6<ref>{{Cite web|url=https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#compute-capability-8-x|title=<!--CUDA C Programming Guide -->I.7. Compute Capability 8.x|website=docs.nvidia.com|language=en-us|access-date=September 23, 2020}}</ref>
* [[Samsung Electronics|Samsung]] 8&nbsp;nm 8N (8LPH) process (custom designed for NVIDIA)<ref name="NVIDIA-ampere-GA102" />
* Doubled FP32 performance per SM on Ampere GPUs
* Third-generation Tensor Cores with FP16, bfloat16, TensorFloat-32 (TF32) and sparsity acceleration
Line 88 ⟶ 84:
 
=== Desktop ===
* Only the RTX 3090 and RTX 3090 Ti support 2-way [[NVLink]].▼
* All the RTX 30 GPUs are made using the 8&nbsp;nm Samsung node.<ref>{{Cite web|last=James|first=Dave|date=September 1, 2020|title=Nvidia confirms Samsung 8nm process for RTX 3090, RTX 3080, and RTX 3070|url=https://www.pcgamer.com/nvidia-ampere-samsung-8nm-process/|access-date=June 1, 2021|website=PC Gamer|language=en-US}}</ref>
▲* Only the RTX 3090 and RTX 3090 Ti support 2-way [[NVLink]].
* RTX 3050 feature limited 8 lanes for the [[PCI Express#PCI Express 4.0|PCIe 4.0]] bus interface. All other cards support the full ×16 bandwidth.
* Double-precision ([[Double-precision floating-point format|FP64]]) performance of the [[Ampere (microarchitecture)|Ampere]] chips are 1/64 of single-precision ([[Single-precision floating-point format|FP32]]) performance.
 
<!-- * Only the RTX 3050 has [[PCIe 4.0]] ×8 interface, rest of RTX 30 GPUs have ×16. <Reported by TechPowerUp, it might need better source. -->
{{Row hover highlight}}
{| class="mw-datatable wikitable sortable" style="font-size:85%; text-align:center;"
Line 99 ⟶ 96:
! rowspan="2" | Launch<br />[[List price|MSRP]]<br />(USD)
! rowspan="2" | [[Code name|Code<br />name]](s){{efn|name="LHR"|Models with GA1xx-x02 chips feature improved Lite Hash Rate (LHR) limiter for cryptomining.<ref>{{Cite web |last=Wuebbling |first=Matt |date=May 18, 2021 |title =A Further Step to Getting GeForce Cards into the Hands of Gamers |url=https://blogs.nvidia.com/blog/2021/05/18/lhr/ |website=Nvidia |language=en-US |access-date=October 29, 2022}}</ref><ref>{{Cite web |author=WhyCry |date=June 7, 2021 |title=Inno3D launches GeForce RTX 30 LHR series |url=https://videocardz.com/newz/inno3d-launches-geforce-rtx-30-lhr-series |website=VideoCardz |language=en-US |access-date=October 29, 2022}}</ref>}}
! rowspan="2" {{vert header|va=mid|Transistors (billion)|nb=y}}
! rowspan="2" {{vert header|va=mid|Die size ([[Square metre|mm<sup>2</sup>]])}}
! rowspan="2" | Core<br />config{{efn|name="Core Config"}}
Line 109 ⟶ 106:
! colspan="4" | Processing&nbsp;power&nbsp;([[FLOPS|TFLOPS]]){{efn|name="Processing Power"}}
! rowspan="2" | [[Thermal design power|TDP]]<br />([[watt]]s)
! rowspan="2" | Bus<br/>interface▼
|-
! Core<br />([[Hertz|MHz]])
Line 146 ⟶ 142:
! style="background-position:center" |
! style="background-position:center" |
|- ▼
! style="background-position:center" |
! rowspan="34" style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3050<ref name="nvidia-30-series">{{Cite web |title=GeForce RTX 30 Series |url=https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/ |website=Nvidia |language=en-US |access-date=May 15, 2022}}</ref><ref>{{cite web|title=NVIDIA GeForce RTX 3050 8 GB Specs|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-8-gb.c3858|website=TechPowerUp|access-date=June 3, 2022}}</ref><ref>{{cite web |title=GeForce RTX 3050 OEM Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-oem.c3915 |website=TechPowerUp |language=en-US |access-date=September 19, 2022}}</ref><ref>{{Cite web |title=NVIDIA GeForce RTX 3050 8 GB GA107 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-8-gb-ga107.c3880 |website=TechPowerUp |language=en-US |access-date=January 11, 2023}}</ref><ref>{{Cite web |last=Shilow |first=Anton |date=February 2, 2024 |title=Palit Releases Fanless Version of NVIDIA&#39;s New GeForce RTX 3050 6GB |url=https://www.anandtech.com/show/21253/palit-launches-fanless-geforce-rtx-3050-6gb-with-dvi-connector |access-date=February 2, 2024 |website=Anandtech}}</ref>▼
▲|-
| {{dts|2024|February|2|format=mdy|abbr=on}}
▲! rowspan="3" style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3050<ref name="nvidia-30-series">{{Cite web |title=GeForce RTX 30 Series |url=https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/ |website=Nvidia |language=en-US |access-date=May 15, 2022}}</ref><ref>{{cite web|title=NVIDIA GeForce RTX 3050 8 GB Specs|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-8-gb.c3858|website=TechPowerUp|access-date=June 3, 2022}}</ref><ref>{{cite web |title=GeForce RTX 3050 OEM Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-oem.c3915 |website=TechPowerUp |language=en-US |access-date=September 19, 2022}}</ref><ref>{{Cite web |title=NVIDIA GeForce RTX 3050 8 GB GA107 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-8-gb-ga107.c3880 |website=TechPowerUp |language=en-US |access-date=January 11, 2023}}</ref>
| 169
| GA107-325
▲!| rowspan="2" | Bus<br/>interface8.7
| rowspan="32" | 2200▼
| 2304<br />72:32:18:72
| 18
| rowspan="34" | 14.02▼
| 1042<br />(1470)
| rowspan="94" | [[GDDR6 SDRAM|GDDR6]]14▼
| 33.34<br />(47.04)
| 75.02<br />(105.8)
| 6
| 168
| rowspan="10" |[[GDDR6 SDRAM|GDDR6]]
| 96
| 4.802<br />(6.774)
| 4.802<br />(6.774)
| 0.075<br />(0.105)
| 30.1<br />(60.2)
| 70
|-▼
| {{dts|2022|December|16|format=mdy|abbr=on}}
| $249
| GA107-150
| 8.7
| 200
| 2560<br />80:32:20:80
| 20
▲| rowspan="3" | 2
| 1552<br />(1777)
▲| rowspan="3" | 14.0
| 49.66<br />(56.86)
| 124.2<br />(142.2)
| rowspan="3" | 8
| rowspan="3" | 224
▲| rowspan="9" | [[GDDR6 SDRAM|GDDR6]]
| rowspan="3" | 128
| 7.946<br />(9.098)
Line 170 ⟶ 182:
| 36.4<br />[72.8]
| 115
| rowspan=3 | [[PCI Express#PCI Express 4.0|PCIe 4.0]]<br />×8
|-
| {{dts|2022|July|18|format=mdy|abbr=on}}
| {{partial|OEM}}<ref>{{cite web|last=Klotz |first=Aaron |date=July 18, 2022 |title=OEM Exclusive RTX 3050 Confirmed With Cutdown Specs |url=https://www.tomshardware.com/news/oem-exclusive-rtx-3050-neutered-core-specs|website=Tom's Hardware |language=en-US |access-date=September 19, 2022}}</ref>
| rowspan="2" | GA106-150
| rowspan="5" | 13.25
Line 189 ⟶ 200:
|-
| {{dts|2022|January|27|format=mdy|abbr=on}}
| $249
| 2560<br />80:32:20:80
| 20
Line 202 ⟶ 213:
! rowspan="4" style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3060<ref name="nvidia-3060-family">{{Cite web |title=NVIDIA GeForce RTX 3060 Family |url=https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3060-3060ti/ |website=Nvidia |language=en-US |access-date=October 30, 2022}}</ref><ref>{{cite web |title=NVIDIA GeForce RTX 3060 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3060.c3682 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref><ref name="techpowerup-3060-GA104">{{cite web |title=NVIDIA GeForce RTX 3060 GA104 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3060-ga104.c3832 |website=TechPowerUp |language=en-US |access-date=May 15, 2022}}</ref>
| {{dts|2021|February|25|format=mdy|abbr=on}}
| rowspan="4" | $329
| GA106-300
| rowspan="4" | 3584<br />112:48:28:112
Line 210 ⟶ 221:
| rowspan="4" | 15.0
| rowspan="4" | 63.36<br />(85.29)
| rowspan="4" | 147.8<br />(199.0)
| rowspan="2" | 12
| rowspan="2" | 360
Line 219 ⟶ 230:
| rowspan="4" | 51.2<ref name="NVIDIA-ampere-GA102">{{Cite web |title=Ampere GA102 GPU Architecture Whitepaper v2.1 |url=https://www.nvidia.com/content/PDF/nvidia-ampere-ga-102-gpu-architecture-whitepaper-v2.1.pdf |website=Nvidia |language=en-US |access-date=February 26, 2021}}</ref><br />[102.4]
| rowspan="4" | 170
| rowspan=16 | [[PCI Express#PCI Express 4.0|PCIe 4.0]]<br />×16
|-
| {{dts|2021|May||format=mdy|abbr=on}}
Line 239 ⟶ 249:
! rowspan="4" style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3060 Ti<ref name="nvidia-3060-family" /><ref>{{cite web |title=NVIDIA GeForce RTX 3060 Ti Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3060-ti.c3681 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref><ref name="techpowerup-3060Ti-GA103">{{cite web |title=NVIDIA GeForce RTX 3060 Ti GA103 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3060-ti-ga103.c3872 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref>
| {{dts|2020|December|02|format=mdy|abbr=on}}
| rowspan="4" | $399
| GA104-200
| rowspan="4" | 4864<br />152:80:38:152
Line 245 ⟶ 255:
| rowspan="7" | 4
| rowspan="4" | 1410<br />(1665)
| rowspan="2" | 14.0
| rowspan="4" | 112.8<br />(133.2)
| rowspan="4" | 214.3<br />(253.1)
Line 251 ⟶ 261:
| rowspan="2" | 448
| rowspan="7" | 256
| rowspan="4" | 13.72<br />(16.202)
| rowspan="4" | 13.72<br />(16.202)
| rowspan="4" | 0.214<br />(0.253)
| rowspan="4" | 64.8<ref name="NVIDIA-ampere-GA102" /><br />[129.6]
Line 269 ⟶ 279:
| 22
| 496
| rowspan="2" | 14.0
| rowspan="2" | 448
| rowspan="2" | GDDR6
Line 275 ⟶ 285:
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3070<ref name="nvidia-30-series" /><ref>{{cite web|title=NVIDIA GeForce RTX 3070 Specs|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3070.c3674|website=TechPowerUp|access-date=June 3, 2022}}</ref>
| {{dts|2020|October|29|format=mdy|abbr=on}}
| $499
| GA104-300<br />GA104-302
| rowspan="2" | 17.4
Line 282 ⟶ 292:
| 46
| 1500<br />(1725)
| 144.0<br />(165.6)
| 276.0<br />(317.4)
| 17.66<br />(20.31)
| 17.66<br />(20.31)
Line 292 ⟶ 302:
! rowspan="2" style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3070 Ti<ref name="nvidia-30-series"/><ref>{{cite web |title=NVIDIA GeForce RTX 3070 Ti Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3070-ti.c3675 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref><ref name="techpowerup-3070Ti-GA102">{{cite web |title=NVIDIA GeForce RTX 3070 Ti GA102 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3070-ti-ga102.c3936 |website=TechPowerUp |language=en-US |access-date=November 1, 2022}}</ref>
| {{dts|2021|June|10|format=mdy|abbr=on}}
| rowspan="2" | $599
| GA104-400
| rowspan="2" | 6144<br />192:96:48:192
Line 302 ⟶ 312:
| rowspan="2" | 608
| rowspan="7" | GDDR6X
| rowspan="2" | 19.42<br />(21.707)
| rowspan="2" | 19.42<br />(21.707)
| rowspan="2" | 0.303<br />(0.34034)
| rowspan="2" | 87<ref name="anandtech-ti-models">{{Cite web |last=Smith |first=Ryan |date=June 1, 2021 |title=NVIDIA Announces GeForce RTX 3080 Ti & 3070 Ti: Upgraded Ampere Cards Coming in June|url=https://www.anandtech.com/show/16727/nvidia-announces-geforce-rtx-3080-ti-3070-ti-upgraded-cards-coming-in-june |website=AnandTech |language=en-US |access-date=December 12, 2022}}</ref><br />[174]
| 290
Line 316 ⟶ 326:
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3080<ref name="nvidia-30-series" /><ref>{{cite web |title=NVIDIA GeForce RTX 3080 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3080.c3621 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref>
| {{dts|2020|September|17|format=mdy|abbr=on}}
| $699
| GA102-200<br />GA102-202
| 8704<br />272:96:68:272
Line 334 ⟶ 344:
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3080 (12&nbsp;GB)<ref name="nvidia-30-series" /><ref>{{Cite web |title=NVIDIA GeForce RTX 3080 12 GB Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3080-12-gb.c3834 |website=TechPowerUp |language=en-US |access-date=April 30, 2022}}</ref>
| {{dts|2022|January|11|format=mdy|abbr=on}}
| $799
| GA102-220
| 8960<br />280:96:70:280
Line 352 ⟶ 362:
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3080 Ti<ref name="nvidia-30-series"/><ref>{{cite web |title=NVIDIA GeForce RTX 3080 Ti Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3080-ti.c3735 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref>
| {{dts|2021|June|03|format=mdy|abbr=on}}
| $11991,199
| GA102-225
| 10240<br />320:112:80:320
Line 359 ⟶ 369:
| 1365<br />(1665)
| 152.8<br />(186.5)
| 436.8<br />(532.8)
| 27.96<br />(34.101)
| 27.96<br />(34.101)
| 0.437<br />(0.533)
| 136<ref name="anandtech-ti-models"/><br />[273]
Line 367 ⟶ 377:
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3090<ref name="nvidia-30-series"/><ref>{{cite web |title=NVIDIA GeForce RTX 3090 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3090.c3622 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref>
| {{dts|2020|September|24|format=mdy|abbr=on}}
| $14991,499
| {{Nowrap|GA102-250}}<ref>{{Cite web |last=Hagedoorn |first=Hilbert |date=September 29, 2021 |title=GeForce RTX 3080 Ti GA102-250 GPU revealed and ends up in RTX 3090 |url=https://www.guru3d.com/news-story/geforce-rtx-3080-ti-ga102-250-gpu-revealed.html |website=The Guru of 3D |language=en-US |access-date=May 8, 2022}}</ref><br />GA102-300
| 10496<br />328:112:82:328
| 82
Line 374 ⟶ 384:
| 9.75
| 156.2<br />(189.8)
| 457.6<br />(556.0)
| rowspan="2" | 24
| 936
Line 384 ⟶ 394:
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3090 Ti<ref name="nvidia-30-series"/><ref>{{cite web |title=NVIDIA GeForce RTX 3090 Ti Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3090-ti.c3829 |website=TechPowerUp |language=en-US |access-date=June 3, 2022}}</ref>
| {{dts|2022|March|29|format=mdy|abbr=on}}
| $19991,999
| GA102-350
| 10752<br />336:112:84:336
Line 391 ⟶ 401:
| 10.5
| 174.7<br />(208.3)
| 524.1<br />(625.0)
| 1008
| 33.55<br />(39.99)
Line 442 ⟶ 452:
| rowspan="3" | 2
| rowspan="2" | 712&ndash;1530<br />(1057&ndash;1740)
| rowspan="3" | 12.0
| 48.96<br />(55.68)
| 97.92<br />(111.4)
Line 453 ⟶ 463:
| rowspan="3" | 35&ndash;80
|-
| {{dts|2022|July|6|format=mdy|abbr=on}}
| GA107<br />(GN20-P0-R)
| rowspan="2" | 2560<br />80:32:20:80
| rowspan="2" | 20
|
|
| 6
| 144
| 96
|
|
|
|-
! style="text-align:left;" | {{Nowrap|GeForce RTX}}<br />3050&nbsp;Ti<br />Laptop<ref name=":nvidia-30-series-laptops" /><ref>{{Cite web |title=NVIDIA GeForce RTX 3050 Ti Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-3050-ti-mobile.c3812 |website=TechPowerUp |language=en-US |access-date=April 30, 2022}}</ref><ref>{{Cite web |title=NVIDIA GeForce RTX 3050 Ti Laptop GPU |url=https://www.notebookcheck.net/NVIDIA-GeForce-RTX-3050-Ti-Laptop-GPU-Benchmarks-and-Specs.527430.0.html |website=NotebookCheck |language=en-US |access-date=June 3, 2022}}</ref>
Line 471 ⟶ 481:
| 735&ndash;1463<br />(1035&ndash;1695)
| 46.82<br />(54.24)
| 117.0<br />(135.6)
| 4
| 192
| 128
| 3.763<br />(5.299)
| 3.763<br />(5.299)
Line 482 ⟶ 492:
| {{dts|2021|February|25|format=mdy|abbr=on}}
| GA106<br />(GN20-E3)
| 12.0
| 276
| 3840<br />120:48:30:120
Line 488 ⟶ 498:
| 3
| 817&ndash;1387<br />(1283&ndash;1703)
| rowspan="4" | 14.0<br />(12.0)
| 66.58<br />(81.74)
| 166.4<br />(204.4)
| 6
| 336<br />(288)
| 192
▲|
| 6.912<br />(10.94)
| 6.912<br />(10.94)
Line 508 ⟶ 518:
| rowspan="4" | 4
| 780&ndash;1215<br />(1290&ndash;1620)
| 97.202<br />(129.6)
| 194.4<br />(259.2)
| rowspan="2" | 8
| rowspan="3" | 448<br />(384)
| rowspan="4" | 256
|
| 11.36<br />(15.97)
| 11.36<br />(15.97)
Line 526 ⟶ 536:
| 99.36<br />(142.6)
| 190.4<br />(273.2)
|
| 12.93<br />(14.62)
| 12.93<br />(14.62)
Line 540 ⟶ 549:
| 259.2<br />(328.3)
| 8<br />16
|
| 13.64<br />(18.98)
| 13.64<br />(18.98)
Line 554 ⟶ 562:
| 58
| 585&ndash;1230<br />(1125&ndash;1590)
| 16.0<br />(12.0)
| 118.1<br />(152.6)
| 285.4<br />(368.9)
| 16
| 512<br />(384)
|
| 14.43<br />(18.71)
| 14.43<br />(18.71)
Line 568 ⟶ 575:
 
=== RTX 3090 Ti ===
According to ''[[TechRadar]]'', "the Nvidia GeForce RTX 3090 Ti is an incredibly powerful graphics card". However, the problem would be that the RTX 3090 offers similar performance while "it costs significantly less."<ref name=":3">{{Cite web |last=Thomas |first=Jackie |date=2022-03-29 |title=Nvidia GeForce RTX 3090 Ti review |url=https://www.techradar.com/reviews/nvidia-geforce-rtx-3090-ti |access-date=2023-12-31 |website=TechRadar |language=en}}</ref> In addition to its price, they criticized itsthe RTX 3090 Ti's high power consumption.<ref name=":3" /> The performance was praised, as the 3090 Ti "will likely be the go-to GPU for creative professionals that need brute force in their day-to-day work." In gaming, "the RTX 3090 Ti fares quite a bit better" compared to the RTX 3090, and even in 8K "will be able to hit a solid 60 fps in many games at high settings."<ref name=":3" />
 
=== RTX 3090 ===
''[[PC Gamer]]'' described the Nvidia GeForce RTX 3090 as a "frankly enormous graphics card", which is "supremely powerful" and "more worthy of its Titan credentials than the GeForce branding", as for the average gamer "it doesn't deliver enough over the RTX 3080 to make sense, but for the pro-creator it's a workload-crushing card."<ref>{{Cite web |last=James |first=Dave |date=2020-09-24 |title=Nvidia RTX 3090 Founders Edition review |url=https://www.pcgamer.com/nvidia-rtx-3090-review-founders-edition-benchmarks/ |access-date=2023-12-31 |website=pcgamer |language=en}}</ref>
 
=== RTX 3080 ===
Line 590 ⟶ 597:
== Notes ==
{{reflist|group=lower-alpha|refs=
{{efn|name="Core Config"|[[Unified shader model|Shader Processors]], [[Texture mapping unit]]s : [[Render output unit]]s : [[Nvidia RTX|Ray tracing cores]] : [[Tensor Core]]s}}
{{efn|name="SMs"|The number of [[Graphics processing unit#Computational functions|Streaming multi-processors]] on the GPU.}}
{{efn|name="Boost"|Core boost values (if available) are stated below the base value inside bracket.}}
{{efn|name="Pixel Fillrate"|Pixel fillrate is calculated as the number of render output units (ROPs) multiplied by the base (or boost) core clock speed.}}
Line 604 ⟶ 611:
== External links ==
* [https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/ Official website]
* [https://www.nvidia.com/content/dam/en-zz/Solutions/geforce/ampere/pdf/NVIDIA-ampere-GA102-GPU-Architecture-Whitepaper-V1.pdf NvidiaNVIDIA AmpereAMPERE GA102 GPU ArchitectureARCHITECTURE whitepaper]
* {{commons category-inline|Nvidia GeForce 30 series video cards}}
 
Retrieved from "https://en.wikipedia.org/wiki/GeForce_30_series"