As you can see in the image above, the cooling module features a 4-bay thick design and the fan design is basically the same as the other RTX 40 public versions, but with a new fan in the middle of the heatsink to enhance airflow. The seller claims that the card features 22 heat pipes that are vertically embedded in the bottom heat spreader plate. Unlike regular graphics cards, the PCB of this cooling module is orientated perpendicular to the backplate.

Reliable tipster kopite7kimi said that the RTX 4090 Ti may have been axed, so it’s basically impossible for this cooling module to be used on RTX 40-series models.

Earlier this year it was announced that the RTX 4090 Ti would feature 18176 CUDA cores with 96MB of L2 cache, 24GB of 24Gbps GDDR6X memory, and consume 600W. For comparison, Nvidia’s now-introduced RTX 4090 has 16384 CUDA cores with 72MB of L2 cache, 24GB of 21Gbps GDDR6X memory, and consumes 600W of power. For comparison, NVIDIA has now released the RTX 4090 with 16384 CUDA cores, 72MB of L2 cache, 24GB of 21Gbps GDDR6X memory, and 450W.

The post NVIDIA’s RTX 4090 Ti Prototype Graphics Card Cooling Module Revealed appeared first on TechGoing.

According to the information, RTX 4090 Ti will have 18176 CUDA cores with 96MB L2 cache, 24GB 24Gbps GDDR6X memory and 600W power consumption.

For comparison, NVIDIA’s RTX 4090 is now available with 16384 CUDA cores, equipped with 72MB L2 cache, equipped with 24GB 21Gbps GDDR6X memory, and consumes 450W.

In terms of memory bandwidth, the RTX 4090 Ti will increase from 1 TB / s to 1.1 TB / s with the faster 24 Gbps 24GB GDDR6X memory. performance, it is expected that the RTX 4090 Ti will be 10% to 20% higher than the RTX 4090 performance.

A few days ago, blogger MEGAsizeGPU released a physical image of a four-slot thick Nvidia public version graphics card, and called it “4090Ti / Titan ada”, which is expected to be one of the prototype cards.

The post NVIDIA RTX 4090 Ti graphics card latest news: 18176 CUDA and 24GB memory appeared first on TechGoing.

The NVIDIA GeForce RTX 4090 Ti is said to be a “chicken blood” configuration with all 144 SMs enabled for a total of 18432 CUDA cores. The GPU will come with 96 MB of L2 cache and a total of 224 ROPs, which is insane.

The card will retain the TSMC 4N process node and offer a nice clock speed boost, the card is rated at 2.75 GHz (default) and can hit a boost clock of 2.95 or even 3 GHz in typical gaming workloads. The default clock has been increased by +190 MHz.

In terms of memory, the GeForce RTX 4090 Ti will have 24 GB of GDDR6X capacity, which will likely be available at a faster 24 Gbps over a 384-bit bus interface. This will provide up to 1.152 TB/s of bandwidth.

And all these boosted specs will also lead to higher power consumption, with the flagship expected to run at around 600W of TBP. The card is said to have a TGP of 475W, which is a 25W increase over the RTX 4090, but custom variants could push that up into the 600W+ range again.

The NVIDIA GeForce RTX 4090 is a very efficient graphics card that can deliver the same performance at 70% less than its power budget. If the 4090 Ti does use a binned chip, that might explain the lower wattage numbers, as NVIDIA can further optimize the voltage.

The RTX 4090 Ti is said to perform 10-20% better than the RTX 4090, so it will definitely match or even exceed the 2x performance figures of the RTX 3090 Ti, which would be truly impressive. However, cards of this nature won’t be coming out anytime soon.

The post NVIDIA RTX 4090 Ti specification exposure: 20% higher performance than RTX 4090 appeared first on TechGoing.

According to reports, because it is an engineering sample, this public version RTX 4090 Ti appearance is relatively simple, the final public version of the graphics card will add more design elements.

Previous news said that the RTX 4090 Ti GPU core is AD102-450-A1, with 18176 CUDA cores, 11% more than the RTX 4090, video memory is 48GB 24Gbps GDDR6X, twice as much as the RTX 4090.

For comparison, the NVIDIA RTX 30 series graphics cards all feature a specially designed dual fan design, with the lower fan drawing airflow in and the upper fan expelling it, making it more efficient than the traditional dual fan design.

NVIDIA is expected to release the RTX 4090 graphics card in October this year, and this RTX 4090 Ti is expected to be released at a later date.

The post NVIDIA RTX 4090 Ti engineering samples revealed 3 fan design appeared first on TechGoing.