First of all, NVIDIA RTX 40 mobile series GPUs will be released at 9:00 a.m. Pacific time on January 3, 2023, but it is not ruled out that Huang Renxun may temporarily change his mind.

According to reports, Nvidia will launch RTX 4080 Ti / 4090 Mobile (16GB), RTX 4080 Mobile (12GB), RTX 4070 Mobile (8GB), RTX 4060 Mobile (8GB), RTX 4050 Mobile ( 6GB) and an RTX 3050 with 6GB of video memory.

The final product naming may change, but the product itself will not change, so it cannot be ruled out that Nvidia will continue the 4080 16GB and 4080 12GB naming rules. But in the final analysis, the above models are SKUs such as GN21-X9, GN21-X6, GN21-X4, GN21-X2, etc. Please wait and see.

In addition to this, they also got performance data on the NVIDIA RTX 40 mobile graphics card, which is expected to bring about a 30% (average) performance improvement over the previous generation. In addition, Nvidia’s PPT also specifically pointed out that the RTX 4070 mobile version will be 15% higher than the RTX 3080 Ti.

In order to avoid revealing the identity of their informant, they did not give the original PPT picture but summarized it into a chart, let’s take a look together.

The above performance data is based on the TimeSpy benchmark test in the Nvidia PPT, but the RTX 4080 Ti (GN21-X11) is a bit strange, because this card does not appear in any single section of the PPT, and it is in the product line that Nvidia provides to board manufacturers It’s also never shown in the picture, so it’s possible this is the often-mentioned RTX 4090 Mobile, but it could also be a completely different SKU.

It is said that major manufacturers will be the first to launch models equipped with Intel Raptor Lake-H processors and Nvidia RTX 4080 graphics cards and GN21-X11 in early February, and there will be other mobile SKUs in the future, which means that Intel will also announce at CES. Its Raptor Lake-H series of mobile processors is expected to be the first to be launched by the end of January/early February for a new generation of notebook computers, and the review will be lifted in mid-February at the latest.

The post NVIDIA RTX 40 mobile graphics card will be released in January with a performance increase of about 30% appeared first on TechGoing.

Unusually, this cooler is still the familiar X-shaped style, but according to the internal structural design, it has an even thermal plate rotated by 90 degrees so that the graphics card PCB board is vertical and parallel to the motherboard when inserted into the case.

In this way, the layout of the graphics card’s cooling ducts and power cables will be more logical, but the design will be more complex and the output connectors will have to be rearranged.

However, the semi-professional Titan Ada, which was rumored to draw up to 600-800W, has been canceled, and it’s still up in the air whether the RTX 4090 Ti will be available – who is such a behemoth for? Or has it been aborted together?

The post NVIDIA RTX 40 4-slot giant cooler that can suppress 900W power consumption appeared first on TechGoing.

(via WCCFTech)

According to the information listed on Zotac’s website, the company recommends that users replace the 4×8-pin to 1×16-pin PCIe 5.0 graphics card power connector after plugging and unplugging 30 times.

Cards such as the RTX 4090 and 4080 will come with one of these adapter cables in the package. But for PC DIY enthusiasts who like to try things out, is there any point in changing a cable after the 30 nominal lifecycles have been exhausted?

For reference, the PCIe 5.0 cable can handle 600W of power. But as NVIDIA states, a fourth 8-pin is completely optional, just to provide more room for overclocking (previously claimed to have been overclocked to 3 GHz+ in the lab).

Update] WCCFTech has confirmed with NVIDIA that the new 16-pin connector’s 30-times nominal life is actually the same as the old 8-pin connector (mini-fit Molex) specification that AMD and other GPU manufacturers have been using for the past 20+ years.

For PC DIY users updating from the old standard to ATX 3.0 power supplies, the RTX 40 Series Founders Edition graphics cards require only one cable to meet auxiliary power needs, thus effectively eliminating cable management headaches.

Finally, major manufacturers will be launching new ATX 3.0 power supplies starting in October this year. Even if you’re not in a hurry to update your power supply, it’s easy to use the three groups of 8-pin adapter cables for a single 16-pin adapter.

The post Nvidia clarifies the service life of 16-pin power connector in RTX 40 series not been shortened appeared first on TechGoing.

However, it should be noted that DLSS 3 is based on the optical flow analyzer on the RTX 40 series graphics card to realize the analysis of continuous frame and motion data.

This means that this new technology, which NVIDIA calls “cross-era”, will only be enjoyed by RTX 40 series graphics cards, while old graphics cards still only support DLSS 2.

At present, NVIDIA has announced the first batch of 36 games, 3 game engines, and 2 applications that support DLSS 3, including Unity and Unreal, the two largest commercial game engines on the market, which means that DLSS 3 will soon become a new technology favored by game developers, and it also meant that the gap between RTX 40-series and 30-series graphics cards was further widened.

The post NVIDIA DLSS 3 is exclusive to RTX 40 series graphics cards appeared first on TechGoing.

In an earlier official preview, NVIDIA had almost made it clear that it was going to release the Ada Lovelace architecture RTX 40 series graphics cards, and the “T.I or Tie” in the sticky note seemed to tell us that there would be no Ti graphics cards this time.

The latest details revealed by Kopite7kimi, the God of Explosives, show that the NVIDIA GeForce RTX 40 launch will include the 4090 24GB, 4080 16GB and 4080 12GB products, which will be released as follows.

NVIDIA GeForce RTX 4090 graphics card, October 2022 release

NVIDIA GeForce RTX 4080 graphics card, releasing November 2022

NVIDIA GeForce RTX 4070 graphics card, December 2022

GeForce RTX 4060 graphics card, unveiled at CES in January 2023

Videocardz gives a report that the NVIDIA GeForce RTX 4090 will use 128 SMs out of 144, for a total of 16,384 CUDA cores. The GPU will come with 96MB of L2 cache and a total of 384 ROPs and is expected to run at 21Gbps on a 384-bit bus interface using 24GB of GDDR6X capacity.

This will provide up to 1TB/s of bandwidth. This is the same bandwidth as the existing RTX 3090 Ti graphics card and in terms of power consumption, the TBP is said to be rated at 450W.

The post NVIDIA’s Next-Generation RTX 40 Series Graphics Cards Unveiled appeared first on TechGoing.

Specifications, AD106 core is expected to open 36 groups of SM units, that is, 4608 CUDA, the second level of cache increased to 32MB, the peak power consumption of 260W; AD107 is 24 groups of SM units, corresponding to 3072 CUDA, the second level of cache also increased to 32MB.

It must be said that the cache of this generation of GPU cores has changed a lot, AD102 core has as much as 96MB, AD107 also has 32MB, you know GA107 core is only 2MB ……

Performance, the same source gives the 3DMark TSE run score is, RTX 4080 to 15000 points, almost twice the RTX 3080, and even RTX 4070 10,000 points already on par with the RTX 3090.

It is inferred that the RTX 4060/4050 strength should not be underestimated, close to RTX 3080 or RTX 3070 Ti can fully expect under.

The post NVIDIA has begun testing the AD106 and AD107 GPU cores for its graphics products appeared first on TechGoing.

As for the timing of the RTX 4090, the fastest September release, October began to store for the market.

Previously the most controversy around RTX 4090 is power consumption, now the argument is that NV allows AIC to configure itself on 450/600W two-block TGP setting, quite spiritual.

In terms of price, the RTX 4090, which crosses the performance level of RTX 3090 Ti, will be cheaper, between $1499 (3090) ~ $1999 (3090 Ti). However, the RTX 30 Series has now broken the market price long after the fierce price cut, with the RTX 3090 already hovering near $1000 and the RTX 3090 Ti falling below $1500.

In this way, NVIDIA still attaches great importance to the RTX 30 Series inventory and is unwilling to allow any of the new cards that may have duplicate positioning to cause any interference with the old cards going into inventory.

Back to the specification level, RTX 4090 is expected to open 128 groups SM, integrated 16384 CUDA, 96MB L2 cache, matching 24GB G6X memory, etc., single precision floating point performance over 70T.

The post RTX 4090 or the only new 40-Series Graphics Card available this year: pricing is lower than 3090 Ti appeared first on TechGoing.