Skip to main content

New RTX 4090 leak teases the design and the launch timeline

A new leak shows us more details about the upcoming Nvidia GeForce RTX 4090. The card may feature a 16-pin power connector, but it doesn’t seem to have USB Type-C display outputs.

The rumors also suggest that, given the predicted launch date of the next-gen RTX 40-series GPUs, Nvidia’s board partners may have limited time to get ready to launch their custom cards.

Nvidia Ada Lovelace PCB teaser.
Igor's Lab

The information comes from Igor’s Lab. Igor published an alleged picture of the PCB design of the upcoming AD102 GPU. Although Igor wasn’t allowed to publish the real PCB due to keeping his sources anonymous, the picture he posted still gives us some insight into the GPU. Nvidia itself hasn’t said much about its Ada Lovelace graphics cards just yet, so even little details from various sources add something to the pool of predictions we’ve heard so far.

Recommended Videos

The rumored RTX 4000 PCB looks a lot like the current generation of cards. As previously predicted, the AD102 PCB is similar to the GA102. The GA102 is used in Nvidia’s current-gen flagship cards, ranging from the RTX 3080 to the RTX 3090 Ti, and to that end, this checks out — the PCB is almost identical to the one found in the RTX 3090 Ti.

Next-gen Nvidia PCBs come in two different variants, divided by board numbers: PG139 and PG136. The PG139 is V-shaped and will likely find its way into Founders Edition models as well as custom designs. However, it’s the PG136 that we see pictured by Igor’s Lab, and it will likely be used in non-Ti graphics cards.

On the leaked PCB, we can see that the GPU will feature up to four display connectors, three of which are DisplayPort and one HDMI. The card also features NVLink, which is Nvidia’s tech that helps connect two GPUs to utilize them in a single system. Lastly, the GPU comes with a single 16-pin power connector.

Nvidia's development schedule.
Igor's Lab

The power requirements of the RTX 4090 are something we’ve already covered in the past, as they have been the subject of much speculation. With some sources reporting that the RTX 4090 and RTX 4090 Ti may require an up to 1,200-watt power supply unit (PSU), it’s clear that the GPUs will be exceptionally power-hungry. It’s very likely that we may see a 600-watt TDP in this generation of GPUs.

Igor also talks about the processes Nvidia still has to go through before the launch of the next-gen GPUs, which is currently rumored to be around September. There are still plenty of things to be done. According to a leaked image that shows the rumored development schedule of Nvidia Ada Lovelace, no working samples have been sent out to Nvidia’s board partners just yet.

While mass production is supposed to start around August, which doesn’t jeopardize the predicted September release, it’s hard to imagine how custom designs are going to fit into that timeline. Igor’s Lab cites its own sources as it says that the board partners are still designing their GPUs based on the GA102 GPU. It seems that it might be a tight race to get the GPUs released on time to compete against AMD’s next-gen RDNA 3 graphics cards.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
Nvidia’s RTX 5070 Ti could displace the RTX 5080 — but I’m worried about it
Two RTX 4070 Ti Super graphics cards sitting next to each other.

Nvidia's RTX 5070 Ti is mere days away. Set to launch on February 20, the GPU marks the beginning of Nvidia's mainstream RTX 50-series. With similar specs to the RTX 5080, the RTX 5070 Ti has every chance to rank high among the best graphics cards -- but it's also going to face a number of obstacles that might impact its success.

Thanks to some leaked benchmarks, we now have a better idea of how the card might perform, and that rough ballpark is really a bit of a mixed bag. Here's why I think the RTX 5070 Ti might have an uphill climb when it first hits the shelves, but also why it might completely displace the RTX 5080.
Promising benchmarks -- well, kind of

Read more
Nvidia’s AI obsession is causing delays in its PC business
Logo on the Nvidia GeForce RTX 5080.

Component manufacturer, Nvidia appears to be short on GeForce RTX 50 series GPUs for laptops, which were originally set to launch in January.

Reports indicate that the supplier has been so focused on developing AI Chips that other parts of its business have lapsed. This has caused delays in supplying the GeForce RTX 50 Series chips to computing partners.

Read more
Nvidia may release the RTX 5070 in March to counter AMD’s RDNA 4 GPUs
The RTX 5070 in a graphic.

Nvidia’s upcoming RTX 5070 may now be launching in early March, according to industry analyst MEGAsizeGPU (@Zed__Wang on X). Initially expected to debut in February, the source suggests that the mid-range Blackwell GPU has been pushed back—potentially as a strategic move to counter AMD’s upcoming Radeon RX 9070.

Unveiled at CES 2025, the RTX 5070 is currently the most affordable GPU from the RTX 50-series lineup, at least till the RTX 5060 series goes official. It is powered by the GB205 GPU, featuring 48 Streaming Multiprocessors (SMs), 6,144 CUDA cores, and is equipped with 12GB of GDDR7 memory. The GPU utilizes a 192-bit memory interface, delivering a bandwidth of 672GB/s.

Read more