Size comparison between RTX 3090 and RTX 4090

The GeForce 40 series is a family of

graphics processing units A graphics processing unit (GPU) is a specialized electronic circuit designed to manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device. GPUs are used in embedded systems, mob ...

developed by

Nvidia Nvidia CorporationOfficially written as NVIDIA and stylized in its logo as VIDIA with the lowercase "n" the same height as the uppercase "VIDIA"; formerly stylized as VIDIA with a large italicized lowercase "n" on products from the mid 1990s to ...

, succeeding the

GeForce 30 series The GeForce 30 series is a suite of graphics processing units (GPUs) designed and marketed by Nvidia, succeeding the GeForce 20 series. The GeForce 30 series is based on the Ampere architecture, which feature Nvidia's second-generation ray trac ...

. The series was announced on September 20, 2022, at the GPU Technology Conference (GTC) 2022 event; the RTX 4090 was released on October 12, 2022, and the RTX 4080 was released on November 16, 2022. The cards are based on the Ada Lovelace architecture and feature hardware-accelerated raytracing ( RTX) with Nvidia's third-generation RT cores and fourth-generation Tensor Cores.

Details

Architectural highlights of the Ada Lovelace architecture include the following: *

CUDA CUDA (or Compute Unified Device Architecture) is a parallel computing platform and application programming interface (API) that allows software to use certain types of graphics processing units (GPUs) for general purpose processing, an approach ...

Compute Capability 8.9 *

TSMC Taiwan Semiconductor Manufacturing Company Limited (TSMC; also called Taiwan Semiconductor) is a Taiwanese multinational semiconductor contract manufacturing and design company. It is the world's most valuable semiconductor company, the world' ...

4Nprocess (custom designed for Nvidia) – not to be confused with N4 * Fourth-generation Tensor Cores with FP8, FP16, bfloat16, TensorFloat-32 (TF32) and sparsity acceleration * Third-generation Ray Tracing Cores, along with concurrent ray tracing, shading and compute * Shader Execution Reordering - needs to be enabled by the developer * Dual NVENC with 8K 10-bit 60FPS

AV1 AOMedia Video 1 (AV1) is an open, royalty-free video coding format initially designed for video transmissions over the Internet. It was developed as a successor to VP9 by the Alliance for Open Media (AOMedia), a consortium founded in 2015 th ...

fixed function hardware encoding * A new generation of Optical Flow Accelerator to aid DLSS 3.0 intermediate AI-based frame generation * No

NVLink NVLink is a wire-based serial multi-lane near-range communications link developed by Nvidia. Unlike PCI Express, a device can consist of multiple NVLinks, and devices use mesh networking to communicate instead of a central hub. The protocol was f ...

support * DisplayPort 1.4a and HDMI 2.1 display connections * Double-precision (

FP64 Double-precision floating-point format (sometimes called FP64 or float64) is a floating-point number format, usually occupying 64 bits in computer memory; it represents a wide dynamic range of numeric values by using a floating radix point. Fl ...

) performance of the Ada Lovelace chips are 1/64 of single-precision ( FP32) performance. * All the cards feature

Micron The micrometre ( international spelling as used by the International Bureau of Weights and Measures; SI symbol: μm) or micrometer (American spelling), also commonly known as a micron, is a unit of length in the International System of Un ...

GDDR6X Graphics Double Data Rate 6 Synchronous Dynamic Random-Access Memory (GDDR6 SDRAM) is a type of synchronous graphics random-access memory (SGRAM) with a high bandwidth, "double data rate" interface, designed for use in graphics cards, game conso ...

video memory.

Products

Desktop

Controversies

RTX 4080 12 GB controversy

When the 12 GB RTX 4080 was announced, numerous outlets, prominent YouTubers, reviewers, and the community criticized Nvidia for calling it an RTX 4080 instead of RTX 4070, given previous Nvidia GPU generations and the large gap in specifications and performance compared to the 16 GB card. The names RTX 4080 12GB and RTX 4080 16GB implied that the only difference between the two products was a difference in VRAM capacity. However, unlike other cases of the same-named product with differing memory configurations that were otherwise very close in performance, the 12GB RTX 4080 uses a completely different chip and configuration: among other differences, the 12 GB 4080, using the AD104 chip, was to feature 27% fewer CUDA cores, along with a cut down 192-bit memory bus, which in prior generations (e.g. GeForce 10, 20, 30 series) had been used for the xx60-class cards. This made the card up to 30% slower than the 16 GB RTX 4080 in raw performance while being priced significantly higher than previous xx70 cards ($900 vs. $500 for the RTX 3070, approximately 80% more expensive). On October 14, 2022, Nvidia announced that due to the confusion caused by the naming scheme, it would be "unlaunching"—i.e. pausing/cancelling the launch of—the 12 GB RTX 4080, with the 16 GB RTX 4080's launch remaining unaffected. Future marketing plans for the card have not been revealed.

12VHPWR connector failure controversy

Some buyers of the RTX 4090 reported that the 12VHPWR—also known as PCIe Gen 5 or 16-pin—connectors of their RTX 4090 were melting, which sparked several theories to explain it. After investigation, several sources reported that the main cause was the 12VHPWR connector not being fully socketed whilst being put under load that resulted in overheating of the connector's pins, which in turn caused the melting of the plastic housing.

PCI-SIG PCI-SIG, or Peripheral Component Interconnect Special Interest Group, is an electronics industry consortium responsible for specifying the Peripheral Component Interconnect (PCI), PCI-X, and PCI Express (PCIe) computer buses. It is based in ...

, the

standards organization A standards organization, standards body, standards developing organization (SDO), or standards setting organization (SSO) is an organization whose primary function is developing, coordinating, promulgating, revising, amending, reissuing, interpr ...

responsible for the creation of the 12VHPWR connector has decided to make changes to the connector's specifications following the recent failures. A

class-action lawsuit A class action, also known as a class-action lawsuit, class suit, or representative action, is a type of lawsuit where one of the parties is a group of people who are represented collectively by a member or members of that group. The class actio ...

has been filed against Nvidia over melting 12VHPWR cables which the lawsuit states is "a dangerous product that should not have been sold in its current state." The plaintiff who brought the suit claims that Nvidia unjustly enriched itself, violated the product's warranty and engaged in fraud and they are demanding that Nvidia pay damages to affected customers as compensation. Following its own investigation and testing, Nvidia officially offered a statement on the melting connectors. They determined that the melting connectors are a cause of user error from not inserting the 12VHPWR connector properly, causing partial contact. They have offered an expedited RMA process for any RTX 4090 affected by the melting connectors. PCI-SIG later said in a statement that Nvidia and their partners were still responsible for testing their products to account for user error.

Reception

RTX 4080

The RTX 4080 received more mixed reviews when compared to the RTX 4090. ''TechSpot'' gave the RTX 4080 score of 80/100. Antony Leather of ''Forbes'' found that the RTX 4080 consistently performed better than the RTX 3090 Ti. The GPU's power efficiency was positively received with ''Digital Trends'' finding that the GPU had an average power draw of 271W despite its rated 320W TDP. The RTX 4080's $1199 price received criticism for its dramatic increase from that of the RTX 3080. In his review for ''RockPaperShotgun'', James Archer wrote that the RTX 4080 "produces sizeable gains on the RTX 3080, though they're not exactly proportional to the price rise". In another critique, ''RockPaperShotgun'' highlighted that AIB models can significantly exceed the base $1199 Founders Edition price, creating further value considerations. The Asus ROG Strix AIB model they reviewed came in at $1550 which is $50 less than the RTX 4090 Founders Edition. Tom Warren of ''The Verge'' recommended waiting to see what

AMD Advanced Micro Devices, Inc. (AMD) is an American multinational semiconductor company based in Santa Clara, California, that develops computer processors and related technologies for business and consumer markets. While it initially manufactur ...

could deliver in performance and value with their RDNA 3 GPUs. AMD's direct competitor, the Radeon RX 7900 XTX, comes in at $999 compared to the $1199 price of the RTX 4080. The RTX 4080 received criticism for reusing the RTX 4090's massive 4-slot coolers which are not required to cool the RTX 4080's 320W TDP. A smaller cooler would have been sufficient. The RTX 3080 and RTX 3080 Ti with their respective 320W and 350W TDPs maintained 2-slot coolers while the 320W RTX 4080 has a 3-slot cooler on the Founders Edition and 4-slots on many AIB models. It was reported that RTX 4080 sales were weak compared to the RTX 4090, which had sold out during its launch a month earlier. The global cost of living crisis and the RTX 4080's generational pricing increase have been suggested as contributing factors for poor sales numbers.

RTX 4090

Upon release, the RTX 4090's performance received praise from reviewers with ''Tom's Hardware'' saying "it now ranks among the best graphics cards". A review by ''PC Gamer'' gave it 83/100, calling it "a hell of an introduction to the sort of extreme performance Ada can deliver when given a long leash". Tom Warren in a review for ''The Verge'' said that the RTX 4090 is "a beast of a graphics card that marks a new era for PC gaming". Aside from gaming performance, ''PCWorld''s review highlighted the RTX 4090's benefits for content creators and streamers with its 24GB VRAM and

encoding ability and was positive towards the Founders Edition's quiet operation and cooling ability. However, it also received criticism for its value proposition given that it begins at $1,599. Analysis by ''TechSpot'' found that the RTX 4090's value at 1440p was worse than the RTX 3090 Ti and that the RTX 4090 did not make much sense for 1440p as it was limited by CPU bottlenecks. Power consumption was another point of criticism for the RTX 4090. The RTX 4090 has a TDP of 450W compared to the 350W of its last generation equivalent. However, Jarred Walton of ''Tom's Hardware'' noted that the RTX 4090 has the same 450W TDP as the RTX 3090 Ti while delivering much higher performance with that power consumption. Aftermarket AIB RTX 4090 variants received criticism in particular for their massive cooler sizes with some models being 4 PCIe slots tall. This made it difficult to fit an RTX 4090 into many mainstream PC cases. YouTube reviewer JayzTwoCents showed an Asus ROG Strix RTX 4090 model being comparable in size to an entire PlayStation 5 console. Another comparison showed Nvidia's RTX 4090 Founders Edition next to the

Xbox Series X The Xbox Series X/S are home video game consoles developed by Microsoft. They were both released on November 10, 2020, as the fourth generation Xbox, succeeding the Xbox One. Along with Sony's PlayStation 5, also released in November 2020, ...

for size. It was theorized that a reason for the massive coolers on some models is that the RTX 4090 was originally designed to have a TDP up to 600W before it was reduced to its official 450W TDP.

References

External links

Official website

Nvidia GeForce RTX 40 Series announcement

Nvidia Ada GPU Architecture Whitepaper
* {{NVIDIA Computer-related introductions in 2022

4000 series The 4000 series is a CMOS logic family of integrated circuits (ICs) first introduced in 1968 by RCA. It had a supply voltage range of 5V to 20V, which is much wider than any contemporary logic family. Almost all IC manufacturers active during th ...

Graphics processing units Graphics cards