.##....##.########.##......##..######.....########..#######..########.....###....##....##
.###...##.##.......##..##..##.##....##.......##....##.....##.##.....##...##.##....##..##.
.####..##.##.......##..##..##.##.............##....##.....##.##.....##..##...##....####..
.##.##.##.######...##..##..##..######........##....##.....##.##.....##.##.....##....##...
.##..####.##.......##..##..##.......##.......##....##.....##.##.....##.#########....##...
.##...###.##.......##..##..##.##....##.......##....##.....##.##.....##.##.....##....##...
.##....##.########..###..###...######........##.....#######..########..##.....##....##...

All signal, no noise, 24/7.
Built for Humans & AI Agents.

Recent information circulating among technology sources suggests a significant shift in the launch timeline for NVIDIA’s next-generation GeForce RTX 60-Series graphics cards. The cards, which are expected to utilize the Rubin architecture, are now reportedly slated for release in 2028. This revised date indicates a potential postponement, according to leaks shared on social media platforms.

Timeline Discrepancies and Leaked Reports

The reported 2028 release date comes from a leaker, Kopite7Kimi, who posted on X. The user stated, “Unfortunately, GR20x is a product from 2028. That means they have postponed again.” This marks a deviation from the leaker’s previous estimate of the second half of 2027. While NVIDIA has provided no official statement regarding the timing of the RTX 60-Series, users should treat this information as unconfirmed rumor.

This newly surfaced timing aligns with previous claims made by another prominent source, Kepler_L2, who also suggested that Rubin gaming GPUs would not appear until 2028. However, this timeline contradicts a report from a source known as Moore’s Law Is Dead, which had stated approximately one week prior that an NVIDIA source had given them word that the gaming Rubin was approved for a first-half 2027 release. With multiple conflicting reports, the 2028 timeframe appears to be gaining traction, though the overall volatility of memory supply and evolving hardware roadmaps means these plans could change rapidly.

Analysis of Potential Delays

Industry analysis points to several factors contributing to the extended timeline for the RTX 60-Series. The primary obstacles appear to be the tight capacity of memory supply and TSMC fabrication facilities. Experts note that the profitability associated with data center hardware significantly outweighs the revenue generated by GeForce consumer products, making the data center market the primary focus for NVIDIA.

Furthermore, the development cycle for the RTX 50 Super refresh is reportedly encountering difficulties, allegedly due to the high cost of 3 GB GDDR7 memory modules. The confluence of these market pressures suggests a prolonged development phase for the Blackwell architecture. While NVIDIA might still manage a Super launch sometime in 2027, unless it is scrapped entirely, the path to the 60-Series seems extended.

Discussions surrounding the card delays and market trends highlighted the following points:

  • The market remains highly volatile regarding memory and hardware roadmaps.
  • The profitability of data center solutions is currently prioritized over consumer GPU development.

KritIt looks like there will not be a new gpus in 2027 so AMD actually have a time/chance to detrown nvidia for TPU most popular gpu of 2027! Right now nvidia has only 7.2% advantage but if they continue to be as cocky with their prices than at some point they will fall.

Solid State BrainNVidia can’t make the GPUs significantly faster without making the chips larger (thus more expensive and power-hungry) and memory prices have increased too much. These aren’t huge issues in the datacenter sector yet, but home consumers can’t spend any more.

BlazeoptimusThis situation reminds me of the crypto days. When nvidia has production limits, gamers always come last.

lambdaIf rumours are to be believed, AMD also has only a single SKU planned for 2027. It’s not the top end card either.

Kohl BaasBet! Even in datacenters it is a significant problem for nVidia not to be able to gain any performance without multiplying the number of GPUs. Not to mention pulling megawatt+ worth of heat from a single cabinet.

xSneakI’m going to need an upgraded gpu when the 4k 360hz monitors come out next year.

KritThere are 1000 better way’s to spent it. To buy that piece of shit for that price is something else reallity is it’s just a brick.

Vya DomusNeither Nvidia or AMD are going to make new products for consumers for the time being except for a few SKUs that were in the pipeline already pre-2023, there is no profit incentive anymore. PC gaming or at the very least high end PC gaming is basically dead, even if the AI bubble pops the demand for datacenter GPUs wont really go away. It’s hard to see an outcome where high end PC gaming becomes viable again, the only thing that can solve this is increased manufacturing but as long as the margins remain astronomical on datacenter parts that’s not going to happen.

Solid State BrainNVidia can’t make the GPUs significantly faster without making the chips larger (thus more expensive and power-hungry) and memory prices have increased too much. These aren’t huge issues in the datacenter sector yet, but home consumers can’t spend any more.

Vya DomusNeither Nvidia or AMD are going to make new products for consumers for the time being except for a few SKUs that were in the pipeline already pre-2023, there is no profit incentive anymore. PC gaming or at the very least high end PC gaming is basically dead, even if the AI bubble pops the demand for datacenter GPUs wont really go away. It’s hard to see an outcome where high end PC gaming becomes viable again, the only thing that can solve this is increased manufacturing but as long as the margins remain astronomical on datacenter parts that’s not going to happen.

Hue

Written by

Hue

Hue is obsessed with GPU benchmarks and checking her crypto portfolio between gaming sessions. She writes about PC tech, games, and crypto.

+ , ,