CPU Infotech, a Bosnian retailer of computer hardware, recently posted a photo of their latest inventory entries on Facebook. The photo showcased the newly/received Gigabyte RTX 3060 Ti EAGLE graphics cards, one of Gigabyte's designs for this particular SKU. The RTX 3060 Ti EAGLE features a dual-slot, dual-fan cooler design that's the smallest seen on any Ampere graphics card to date. The retailer announces that the inventory should be for sale pretty soon - and all publicly available information points towards a December 2nd release date for the RTX 3060 Ti.

The RTX 3060 Ti is supposed to beat NVIDIA's previous RTX 2080 SUPER graphics cards in performance, whilst costing half of that cards' launch asking price at $399. This should make this one of the most interesting performance-per-dollar graphics cards in NVIDIA's lineup. The RTX 3060 Ti is reportedly based on the same 8 nm "GA104" silicon as the RTX 3070, with further cuts. It features 38 out of the 48 available streaming multiprocessors on "GA104". This amounts to 4,864 "Ampere" CUDA cores, 152 tensor cores, and 38 "Ampere" RT cores. The memory configuration is unchanged from the RTX 3070, which translates to 8 GB of 14 Gbps GDDR6 memory across a 256-bit wide memory interface, with 448 GB/s of memory bandwidth. This marks the first time in years NVIDIA has launched a Ti model before the regular-numbered SKU in a given series, showcasing just how intense AMD competition is expected to be.

It looks like NVIDIA will launch its 4th GeForce RTX 30-series product ahead of Holiday 2020, the GeForce RTX 3060 Ti, with VideoCardz unearthing a leaked NVIDIA performance guidance slide, as well as pictures of custom-design RTX 3060 Ti cards surfacing on social media. The RTX 3060 Ti is reportedly based on the same 8 nm "GA104" silicon as the RTX 3070, but cut down further. It features 38 out of 48 streaming multiprocessors physically present on the "GA104," amounting to 4,864 "Ampere" CUDA cores, 152 tensor cores, and 38 "Ampere" RT cores. The memory configuration is unchanged from the RTX 3070, which means you get 8 GB of 14 Gbps GDDR6 memory across a 256-bit wide memory interface, with 448 GB/s of memory bandwidth.

According to a leaked NVIDIA performance guidance slide for the RTX 3060 Ti, the company claims the card to consistently beat the GeForce RTX 2080 SUPER, a $700 high-end SKU from the previous "Turing" generation. The same slide also shows a roughly 40% performance gain over the previous generation RTX 2060 SUPER, which is probably the logical predecessor for this card. In related news, PC Master Race (OfficialPCMR) on its Facebook page posted pictures of boxes of an ASUS TUF Gaming GeForce RTX 3060 Ti OC graphics cards, which confirms the existence of this SKU. The picture of the card on the box reveals a design similar to other TUF Gaming RTX 30-series cards launched by ASUS so far. As for price, VideoCardz predicts a $399 MSRP for the SKU, which should nearly double the price-performance for this card over the RTX 2080 SUPER at NVIDIA's performance numbers.

NVIDIA today unveiled the NVIDIA A100 80 GB GPU—the latest innovation powering the NVIDIA HGX AI supercomputing platform—with twice the memory of its predecessor, providing researchers and engineers unprecedented speed and performance to unlock the next wave of AI and scientific breakthroughs. The new A100 with HBM2E technology doubles the A100 40 GB GPU's high-bandwidth memory to 80 GB and delivers over 2 terabytes per second of memory bandwidth. This allows data to be fed quickly to A100, the world's fastest data center GPU, enabling researchers to accelerate their applications even faster and take on even larger models and datasets.

"Achieving state-of-the-art results in HPC and AI research requires building the biggest models, but these demand more memory capacity and bandwidth than ever before," said Bryan Catanzaro, vice president of applied deep learning research at NVIDIA. "The A100 80 GB GPU provides double the memory of its predecessor, which was introduced just six months ago, and breaks the 2 TB per second barrier, enabling researchers to tackle the world's most important scientific and big data challenges."

AMD's Smart Access Memory (SAM) is a new technology that AMD decided to launch with its Ryzen 5000 series CPUs and Radeon RX 6000 series GPUs. The technology aims to solve the problem where a CPU can only access a fraction of GPU VRAM at once, making some bottlenecks in the system. By utilizing the bandwidth of PCIe, the SAM expands its data channels and uses all the speed that the PCIe connection offers. However, it appears that AMD might not be the only company offering such technology. Thanks to Gamer's Nexus, they got a reply from NVIDIA regarding a technology similar to AMD's SAM.

NVIDIA responded that: "The capability for resizable BAR is part of the PCI Express spec. NVIDIA hardware supports this functionality and will enable it on Ampere GPUs through future software updates. We have it working internally and are seeing similar performance results." And indeed, it has been a part of the PCIe specification since 2008. This document dating to 2008 says that "This optional ECN adds a capability for Functions with BARs to report various options for sizes of their memory mapped resources that will operate properly. Also added is an ability for software to program the size to configure the BAR to." Every PCIe compatible device can enable it with the driver update through the software.

According to the unknown manufacturer (AIB) based in Taiwan, NVIDIA is preparing to launch the new GeForce RTX 3000 series "Ampere" graphics card. As reported by the HKEPC website, the Santa Clara-based company is preparing to fill the gap between its top-end GeForce RTX 3090 and a bit slower RTX 3080 graphics card. The new product will be called GeForce RTX 3080 Ti. If you are wondering what the specification of the new graphics card will look like, you are in luck because the source has a few pieces of information. The new product will be based on GA102-250-KD-A1 GPU core, with a PG133-SKU15 PCB design scheme. The GPU will contain the same 10496 CUDA core configuration as the RTX 3090.

The only difference to the RTX 3090 will be a reduced GDDR6X amount of 20 GB. Along with the 20 GB of GDDR6X memory, the RTX 3080 Ti graphics cards will feature a 320-bit bus. The TGP of the card is limited to 320 Watts. The sources are reporting that the card will be launched sometime in January of 2021, and it will come at $999. This puts the price category of the RTX 3080 Ti in the same range as AMD's recently launched Radeon RX 6900 XT graphics card, so it will be interesting to see how these two products are competing.

INNO3D, a leading manufacturer of pioneering high-end multimedia components and innovations brings you the new INNO3D GeForce RTX 3090 / 3080 iCHILL Frostbite. Following the huge success of its predecessor iCHILL Frostbite from the previous generation RTX 20 Series, we have now also armed our powerhouse RTX 3090 / 3080 graphics cards with an updated version of the iCHILL Frostbite.

Founded in 1998 with the vision of developing pioneering computer hardware products on a global scale. Fast forward to the present day, INNO3D is now well-established in the gaming community known for our innovative and daring approach to design and technology. We are Brutal by Nature in everything we do and are 201% committed to you for the best gaming experience in the world.

GIGABYTE, the world's leading premium gaming hardware manufacturer, today announced the latest GeForce RTX 30 Series WATERFORCE graphics cards powered by NVIDIA Ampere architecture—AORUS GeForce RTX 3090 XTREME WATERFORCE WB 24G, AORUS GeForce RTX 3080 XTREME WATERFORCE WB 10G, AORUS GeForce RTX 3090 XTREME WATERFORCE 24G, and AORUS GeForce RTX 3080 XTREME WATERFORCE 10G. AORUS is the world's first manufacturer to include the patent-pending "Leak detection" in the WATERFORCE WB open-loop graphics cards. The built-in leak detection circuit covers the entire fitting and water block and can promptly alert users by flashing light at the first sign of leak, so users can deal with the leakage early and prevent any further damage to the system.

AORUS WATERFORCE WB is ideal for those who wish to build open-loop liquid cooling systems. GIGABYTE specializes in thermal cooling solutions, providing optimal channel spacing between the micro fins for enhanced heat transfer from the GPU via stable water flows. The sunk-designed copper micro fins also shorten the heat conduction path from the GPU, so that the heat can be transferred to the water channel area quickly. Moreover, the cover and backplate of the new-gen WATERFORCE WB feature customizable RGB lighting as users can create their own PC styles and bring creativity into the liquid cooling systems.

AMD sent ripples in its late-October even launching the Radeon RX 6000 series RDNA2 "Big Navi" graphics cards, when it claimed that the top RX 6000 series parts compete with the very fastest GeForce "Ampere" RTX 30-series graphics cards, marking the company's return to the high-end graphics market. In its announcement press-deck, AMD had shown the $579 RX 6800 beating the RTX 2080 Ti (essentially the RTX 3070), the $649 RX 6800 XT trading blows with the $699 RTX 3080, and the top $999 RX 6900 XT performing in the same league as the $1,499 RTX 3090. Over the weekend, the company released even more benchmarks, with the RX 6000 series GPUs and their competition from NVIDIA being tested by AMD on a platform powered by the Ryzen 9 5900X "Zen 3" 12-core processor.

AMD released its benchmark numbers as interactive bar graphs, on its website. You can select from ten real-world games, two resolutions (1440p and 4K UHD), and even game settings presets, and 3D API for certain tests. Among the games are Battlefield V, Call of Duty Modern Warfare (2019), Tom Clancy's The Division 2, Borderlands 3, DOOM Eternal, Forza Horizon 4, Gears 5, Resident Evil 3, Shadow of the Tomb Raider, and Wolfenstein Youngblood. In several of these tests, the RX 6800 XT and RX 6900 XT are shown taking the fight to NVIDIA's high-end RTX 3080 and RTX 3090, while the RX 6800 is being shown significantly faster than the RTX 2080 Ti (roughly RTX 3070 scores). The Ryzen 9 5900X itself is claimed to be a faster gaming processor than Intel's Core i9-10900K, and features PCI-Express 4.0 interface for these next-gen GPUs. Find more results and the interactive graphs in the source link below.

Manli Technology Group, Limited is proud to announce the Manli GeForce RTX 3070. It will be based on the NVIDIA Ampere architecture, and offer great performance value. Building upon RTX, the second-generation of GPUs will feature new RT Cores, Tensor Cores, and streaming multiprocessors. The RT and Tensor Cores both have double the throughput as before. There are 5,888 CUDA cores onboard powering the 3070. It also possesses 8 GB of memory, and GDDR6 memory speeds of up to 14 Gbps. This makes the 3070 faster than the RTX 2080Ti and 60% faster than the original RTX 2070.

The twin fan front plate features an aggressive dual curved blade design. Four composite copper heat pipes and segmented heat sinks maximize cooling efficiency. The metal back plate lends structural rigidity. NVIDIA Ampere architecture will usher in a new era of computing power, and the thundering tempest on the packaging captures that excitement and energy.

A German website, Heise.de, has discovered a bug in HDMI 2.1 chipset that causes black screen issues on specific hardware. On AV chipsets sourced by Panasonic, and used by Denon, Marantz, and Yamaha HDMI 2.1 AV receivers, the chipset experiences a specific issue of a black screen. More specifically, the bug happens once you connect Microsoft's newest console, Xbox Series X, or NVIDIA's Ampere graphics cards. When connecting these sources at resolutions like 4K/120 Hz HDR and 8K/60 Hz HDR to Panasonic HDMI 2.1 chipsets, the black screen happens. This represents a major problem for every manufacturer planning to use the Panasonic HDMI 2.1 chipset in its AV receivers, meaning that the issue has to be addressed. The Audioholics website has reached out to Sound United and Yamaha to see what their responses were, and you can check them out below.

NVIDIA has today updated its video encode and decode matrix with references to the latest Ampere GPU family. The video encode/decode matrix represents a table of supported video encoding and decoding standards on different NVIDIA GPUs. The matrix has a reference dating back to the Maxwell generation of NVIDIA graphics cards, showing what video codecs are supported by each generation. That is a useful tool for reference purposes, as customers can check if their existing or upcoming GPUs support a specific codec standard if they need any for video reproduction purposes. The update to the matrix comes in a form of Ampere GPUs, which are now present there.

For example, the table shows that, while supporting all of the previous generations of encoding standards, the Ampere based GPUs feature support for HEVC B Frame standard. For decoding purposes, the Ampere lineup now includes support for AV1 8-bit and 10-bit formats, while also supporting all of the previous generation formats. For a more detailed look at the table please go toNVIDIA's website here.

NVIDIA today announced that the Italian inter-university consortium CINECA—one of the world's most important supercomputing centers—will use the company's accelerated computing platform to build the world's fastest AI supercomputer.

The new "Leonardo" system, built with Atos, is expected to deliver 10 exaflops of FP16 AI performance to enable advanced AI and HPC converged application use cases. Featuring nearly 14,000 NVIDIA Ampere architecture-based GPUs and NVIDIA Mellanox HDR 200 Gb/s InfiniBand networking, Leonardo will propel Italy as the global leader in AI and high performance computing research and innovation.

Ubisoft has today updated the PC system requirements for its Watch Dogs: Legion game. Set to release on October 29th this year, we are just a few weeks away from its release. With the arrival of NVIDIA's GeForce RTX 3000 series Ampere graphics cards, Ubisoft has decided to update the official PC system requirements with RTX-on capabilities. The inclusion of raytracing in the game requires a faster CPU, as well as an RTX-capable GPU. At 1080p resolution, you need at least an RTX 2060 GPU to play with high settings, and raytracing turned to the medium, including DLSS. Going up to 1440p, Ubisoft recommends gamers to use at least an RTX 3070 GPU for very high preset, raytracing on high, and DLSS set to quality. If you want to max everything out and play with the highest settings at 4K resolution, you will need an RTX 3080 GPU.

AMD's RDNA2 graphics architecture features hardware-accelerated decoding of the AV1 video format, according to a Microsoft blog announcing the format's integration with Windows 10. The blog mentions the three latest graphics architectures among those that support accelerated decoding of the format—Intel Gen12 Iris Xe, NVIDIA RTX 30-series "Ampere," and AMD RX 6000-series "RDNA2." The AV1 format is being actively promoted by major hardware vendors to online streaming content providers, as it offers 50% better compression than the prevalent H.264 (translating into that much bandwidth savings), and 20% better compression than VP9. You don't need these GPUs to use AV1, anyone can use it with Windows 10 (version 1909 or later), by installing the AV1 Video Extension from the Microsoft Store. The codec will use software (CPU) decode in the absence of hardware acceleration.

GIGABYTE, the world's leading premium gaming hardware manufacturer, today announced the highest level of AORUS GeForce RTX 30 series graphics cards powered by NVIDIA Ampere architecture. GIGABYTE launched 4 AORUS graphics cards - AORUS GeForce RTX 3090 XTREME 24G, AORUS GeForce RTX 3090 MASTER 24G, AORUS GeForce RTX 3080 XTREME 10G, and AORUS GeForce RTX 3080 MASTER 10G. The 4 graphics cards are all equipped with top-of-the-line overclocked GPU that certified by GIGABYTE GPU Gauntlet sorting technology.

Based on the previous generation of AORUS graphics card, GIGABYTE has released a new generation of more advanced MAX-Covered Cooling technology to meet the high-wattage cooling requirements of the NVIDIA GeForce RTX 30 GPUs. There is an embedded powerful LCD monitor on the side of the graphics card, which can display the status of the graphics card, customized GIFs, text and picture. With RGB Fusion 2.0, the lighting effects of the entire AORUS graphics card can be adjusted according to your preferences.

NVIDIA could update the higher end of its GeForce RTX 30-series "Ampere" product stack with two new additions in December 2020. Sources tell VideoCardz that the company is preparing to launch a 20 GB variant of the GeForce RTX 3080, and a 16 GB variant of the RTX 3070. The RTX 3080 20 GB will come with double the memory of the RTX 3080 the company debuted last month, over the same 320-bit wide GDDR6X memory interface, possibly by using two 8 Gbit memory chips per 32-bit path (which is how the RTX 3090 achieves 24 GB, over its 384-bit memory bus). The RTX 3070 16 GB will likely use a similar approach, albeit with GDDR6 memory. Meanwhile, the mid-range "RTX 3060 Ti" could debut in November, following the late-October introduction of the RTX 3070 8 GB. Much of NVIDIA's product stack adjustments could be in preparation for AMD's late-October reveal of the Radeon RX 6000 RDNA2 series.

Shortages in supply of GeForce RTX 3080 and RTX 3090 graphics cards could persist until 2021, according to NVIDIA CEO Jen-Hsun Huang, responding to a question in a Q&A session of the GTC 2020 (Fall) conference. "The 3080 and 3090 have a demand issue, not a supply issue," said Huang. "The demand issue is that it is much much greater than we expected—and we expected really a lot," he added.

Jen-Hsun predicts that the Holiday 2020 shopping season will only compound availability woes. "I believe that demand will outstrip all of our supply through the year. Remember, we're also going into the double-whammy. The double-whammy is the holiday season. Even before the holiday season, we were doing incredibly well, and then you add on top of it the "Ampere factor," and then you add on top of that the "Ampere holiday factor," and we're going to have a really really big Q4 season." He likened the demand of the RTX 3080 to that of the Intel Pentium in the mid-1990s. "Retailers will tell you they haven't seen a phenomenon like this in over a decade of computing. It hearkens back to the old days of Windows 95 and Pentium when people were just out of their minds to buy this stuff. So this is a phenomenon like we've not seen in a long time, and we just weren't prepared for it."

NVIDIA today unveiled its RTX A6000 professional graphics card, the first professional visualization-segment product based on its "Ampere" graphics architecture. With this, the company appears to be deviating from the Quadro brand for the graphics card, while several software-side features retain the brand. The card is based on the same 8 nm "GA102" silicon as the GeForce RTX 3080, but configured differently. For starters, it gets a mammoth 48 GB of GDDR6 memory across the chip's 384-bit wide memory interface, along with ECC support.

The company did not reveal the GPU's CUDA core count, but mentioned that the card's typical board power is 300 W. The card also gets NVLink support, letting you pair up to two A6000 cards for explicit multi-GPU. It also supports GPU virtualization, including NVIDIA GRID, NVIDIA Quadro Virtual Data Center Workstation, and NVIDIA Virtual Compute Server. The card features a conventional lateral blower-type cooling solution, and its most fascinating aspect is its power input configuration, with just the one 8-pin EPS power input. We will update this story with more information as it trickles out.Update 13:37 UTC: The company also unveiled the A40, a headless professional-visualization graphics card dedicated for virtual-GPU/cloud-GPU applications (deployments at scale in data-centers). The card has similar specs to the RTX A6000.

Update 13:42 UTC: NVIDIA website says that both the A40 and RTX A6000 a 4+4 pin EPS connector (and not 8-pin PCIe) for power input. An 8-pin EPS connector is capable of delivering up to 336 W (4x 7 A @ 12 V).

When NVIDIA introduced its Ampere consumer graphics cards, they launched three models - the GeForce RTX 3070, RTX 3080, and RTX 3090 GPUs. Both the RTX 3080 and RTX 3090 have seen the light of the day as they are now available for purchase, however, one card has remained. The GeForce RTX 3070 launch was originally planned for October 15th launch, but it has officially been postponed by NVIDIA. According to the company, the reason behind this sort of delay in the launch is the high demand expected. Production of the cards is ramping up quickly and the company is quickly stocking up the cards. Likely, NVIDIA AIBs are taking their time to stock up on cards, as the mid-range is usually in very high demand.

As a reminder, the GeForce RTX 3070 graphics card features 5888 CUDA cores running at a base frequency of 1.5 GHz and boost frequency of 1.73 GHz. Unlike the higher-end Ampere cards, the RTX 3070 uses older GDDR6 memory on a 256-bit bus with a bandwidth of 448 GB/s. The GPU features a TDP of 220 W and will be offered in a range of variants by AIBs. You will be able to purchase the GPU on October 29th for the price of $499.

NVIDIA has today released the latest iteration of its Game Ready driver with the version number 456.55. Marked as a WHQL release, the driver is supposedly going to bring new advancements to the stability of the latest GeForce RTX 3000 series Ampere graphics cards. While the release notes don't officially mention anything on how it improves, it is already confirmed by a few Redditors that the new driver removes crashed experienced with the past version 456.38. In the latest revision, the support has been added for NVIDIA Reflex in Call of Duty: Warzone and Call of Duty: Modern Warfare, as well as support for Star Wars: Squadrons game. Below is the link to the driver download page redirecting to NVIDIA's site, and in no time the TechPowerUp download page will be updated as well.DOWNLOAD:NVIDIA GeForce 456.55 WHQL Game Ready Drivers

The change-log follows:

MonsterLabo, a maker of fanless PC cases, today announced its latest creation - The Beast. Featuring a design made from glass and 6 mm thick aluminium, the ATX case is resembling a design we usually could see only from the folks like InWin. The whole chassis is actually made up of two 3 KG aluminium heatsinks that feature ten 6 mm copper heat pipes each. All of this is used for heat dissipation and the case can accommodate up to 400 W of TDP in passive mode. When two 140 mm fans, running at 500 rpm, are added the case can cool more than 500 W of TDP. The Beast measures at 450 mm (L) x 380 mm (W) x 210 mm (H), making it for one large and heavy case. It supports graphics cards up to 290 mm in PCB length and is fully capable of supporting the latest NVIDIA GeForce RTX 30 series "Ampere" graphics cards. Pre-orders for The Beast are starting onOctober 9th, with an unknown pricing. You can expect it to be a high premium over 349 EUR price of The First case. Pre-orders will be shipping in Q1 2021.

A number of RTX 3080 users have been reporting crashes to desktop while gaming on their newly-acquired Ampere graphics cards. The reports have surged in numerous hardware discussion venues (ComputerBase, LinusTechTips, NVIDIA, Tom's Hardware, Tweakers and Reddit), and appear to be unlinked to any particular RTX 3080 vendor (ZOTAC, MSI, EVGA, and NVIDIA Founders Edition graphics cards are all mentioned).

Apparently, this crash to desktop happens once the RTX 3080's Boost clock exceeds 2.0 GHz. A number of causes could be advanced for these issues: deficient power delivery, GPU temperature failsafes, or even a simple driver-level problem (though that one seems to be the least likely). Nor NVIDIA nor any of its AIB partners have spoken about this issue, and review outlets failed to mention this happening - likely because it never did, at least on samples sent to reviewers. For now, it seems that manually downclocking the graphics card by 50-100 MHz could be a temporary fix for the issue while it's being troubleshooted. An unlucky turn of events for users of NVIDIA's latest and greatest, but surely it's better to face a very slight performance decrease in exchange for system stability.

An alleged event by GALAX targeted at distributors in China revealed up to three upcoming SKUs in NVIDIA's RTX 30-series. This comes as yet another confirmation from a major NVIDIA AIC partner about the 20 GB variant of the GeForce RTX 3080. The RTX 3080 originally launched with 10 GB memory earlier this month, and it is widely expected that NVIDIA fills the price-performance gap between this $700 SKU and its $1,500 sibling. The RTX 3080 uses twenty 8 Gbit GDDR6X memory chips (two chips per 32-bit data-path), much like how the RTX 3090 achieves its 24 GB memory amount.

Elsewhere we see GALAX mention the RTX 3060, a performance-segment SKU positioned under the RTX 3070. You'll notice that the product-stack graph by GALAX suggests performance comparisons to previous-generation SKUs. The RTX 3080 and RTX 3090 are faster than everything from the previous generation, while the RTX 3070, which is coming next month, is shown trading blows with both the RTX 2080 Ti and the RTX 2080 Super. In this same graph, the RTX 3060 is shown matching up to the RTX 2080 (non-Super), a card NVIDIA originally launched at $700.

Here is the first picture of an alleged next-generation Quadro RTX graphics card based on the "Ampere" architecture, courtesy YouTube channel "Moore's Law is Dead." The new Quadro RTX 6000-series shares many of its underpinnings with the recently launched GeForce RTX 3080 and RTX 3090, in being based on the 8 nm "GA102" silicon. The reference board design retains a lateral blower-type cooling solution, with the blower drawing in air from both sides of the card, through holes punched in the PCB, "Fermi" style. The card features the latest NVLink bridge connector, and unless we're mistaken, it features a single power input near its tail end, which is very likely a 12-pin Molex MicroFit 3.0 input.

As for specifications, "Moore's Law is Dead," shared a handful of alleged specifications that include maxing out of the "GA102" silicon, with all its 42 TPCs (84 SMs) enabled, working out to 10,752 CUDA cores. As detailed in an older story about the next-gen Quadro, NVIDIA is prioritizing memory size over bandwidth, which means this card will receive 48 GB of conventional 16 Gbps GDDR6 memory across the GPU's 384-bit wide memory interface. The 48 GB is achieved using twenty four 16 Gbit GDDR6 memory chips (two chips per 32-bit wide data-path). This configuration provides 768 GB/s of memory bandwidth, which is only 8 GB/s higher than that of the GeForce RTX 3080. The release date of the next-gen Quadro RTX will depend largely on the supply of 16 Gbit GDDR6 memory chips, with leading memory manufacturers expecting 2021 shipping, unless NVIDIA has secured an early production batch.

When NVIDIA announced its next-generation GeForce RTX 3080 and 3090 Ampere GPUs, it specified that the memory found in the new GPUs will be Micron's GDDR6X variant with 19 Gbps speed. However, being that there are faster GDDR6X modules already available in a 21 Gbps variant, everyone was left wondering why NVIDIA didn't just use the faster memory from Micron. That is exactly what Igor's Lab, a technology website, has been wondering as well. They have decided to conduct testing with an infrared camera that measures the heat produced. To check out the full testing setup and how they tested everything, you can go here and read it, including watching the video embedded.

Micron chips like GDDR5, GDDR5X, and GDDR6 are rated for the maximum junction temperature (TJ Max) of 100 degrees Celsius. It is recommended that these chips should run anywhere from 0C to 95C for the best results. However, when it comes to the new GDDR6X modules found in the new graphics cards, they are not yet any official specifications available to the public. Igor's Lab estimates that they can reach 120C before they become damaged, meaning that TJ Max should be 110C or 105C. When measuring the temperature of GDDR6X modules, Igor found out that the hottest chip ran at 104C, meaning that the chips are running pretty close to the TJ Max they are (supposedly) specified. It is NVIDIA's PCB design decisions that are leading up to this, as the hottest chips are running next to voltage regulators, which can get pretty hot on their own.