Has Taken Over the Consumer GPU Market Here’s How AMD & NVIDIA’s ‘Reprioritized’ Roadmaps Are Shifting Away From Gamers Muhammad Zuhair on at Key Features NVIDIA's RTX 50 SUPER not launching in 2026 NVIDIA's RTX 60 'Rubin' series delayed beyond 2027 AMD's RDNA 5 GPUs delayed to H2 2027 No GPU launch planned for 2026 At a Glance GPU manufacturers are shifting roadmaps ahead NVIDIA/AMD won't launch new models for years ahead GPU retail market situation is worsening with each day Timeline November 2025 RTX 50 SUPER delay rumors November 2025 NVIDIA no longer providing DRAM to AIBs January 2026 AMD's RDNA 5 now launching in H2 2027 January 2026 NVIDIA's Profit/GB model reports February 2025 RTX 50 SUPER & Rubin delayed February 2025 New RTX 50 flagship under works We have talked extensively about memory shortages, guiding gamers throughout the buildup to the situation, but when it comes to markets influenced by them, well, one of the most influential is, of course, the GPU industry. This isn't the first time we have seen the consumer GPU industry directly influenced by a so-called technological buildup, as crypto-mining brought a similar situation, but with AI, the dynamics are entirely different. Gamers are curious about how the AI buildout directly affects consumer GPU supply, given that several elements of the PC supply chain haven't yet been discussed. We'll dive deep into the potential reasons and how they are directly influencing manufacturers and AIB partners. More specifically, we have seen several reports in the last few months about both AMD and NVIDIA revising their GPU roadmaps, so we'll discuss that aspect as well. Overall, after this coverage, consumers should get a clear idea about how the situation is evolving in the GPU segment. The AI Frenzy Isn't Gobbling Up DRAM; It's Making the Client GPU Market Less Attractive For the Likes of AMD/NVIDIA One significant aspect of the GPU shortage is its association with ongoing memory supply constraints, and while this is an important factor, there is another aspect to discuss as well. With the infrastructure buildout that has accelerated since 2023, there has been a common trend among GPU manufacturers in how their revenue has evolved over the years. And when I explored how quarterly revenue across market segments has changed, the idea is simple. Capital from enterprise/DC growth is rising at an unprecedented pace and is far more attractive than ever. Image NVIDIA, and for AMD, the drop in client revenue share has been consistent as well. Now, the reason why I am diving into the finances is simple. When one segment delivers much more value than the other, it is the right move for businesses to divert their resources towards it, and for GPU manufacturers, well, enterprise is a much more solid bet. In the case of NVIDIA, well, before the AI frenzy, the company's market capitalization lay around $200-$300 billion, and once enterprise demand kicked in, it soared to $5 trillion, marking a phenomenal increase. Of course, market capitalization isn't a direct representation of how beneficial AI/DC has been to NVIDIA, but it does show that for them as a company, the only way to boost shareholder value is by giving priority to enterprise over client, and we have already seen the "great pivot" happening. Image AMD, but given the company's soaring CPU business, the drop in client revenue isn't as steep as NVIDIA's. It's only that, with DRAM shortages, the focus on enterprise has been much greater than on client, which is why delayed product roadmaps, a lack of GPU supply, and no mention of consumer GPUs in public shows are indicators that enterprise has taken over from gamers. Consumer GPU Shortages: Extensive Rundown on Delayed Launches & What the Future Holds When we talk about delayed launches, it is essential to note that there's no official disclosure from any GPU manufacturer; rather, it is our analysis based on the original roadmap and the fact that the launches haven't occurred on the defined timeline. One of the major talks in recent times was NVIDIA's RTX 50 SUPER series, given that we were anticipating an unveiling at this year's CES, or even a soft launch, but it appears the lineup is delayed. It's important to talk about the 'why' of the delay as well, and Gigabyte's CEO himself explained the reason. In a previous report, we discussed how NVIDIA is currently prioritizing profit/GB figures, meaning that only a select set of RTX 50-series models is worth keeping in stock. Given that, with the SUPER series in general, the primary focus is on higher VRAM capacity, it is evident that, for Team Green, launching more SKUs with additional memory onboard certainly isn't the right choice for profits. And here's how we described what configurations are profitable for NVIDIA: Image NVIDIA so much that the company has reportedly stopped bundling VRAM with its GPU for AIB partners, meaning vendors will now have to source capacity themselves and negotiate DRAM agreements. And ultimately, such developments have started to influence next-gen GPU launches as well, with the RTX 50 SUPER series being the first indicator. At the same time, NVIDIA's Rubin series, originally slated for FY2027-FY2028, has been pushed to an 'indefinite' timeline, indicating that launches for several years ahead are under threat. Now, at Team Red's camp, the situation is similar as well. The company introduced the RDNA 4 GPU series last year, with the RX 9070 and RX 9060, but AMD's plans for this year are uncertain. We do know that Team Red doesn't plan to launch high-end RDNA 4 GPUs anytime soon, and since there isn't an "RDNA 4.5" launch in the works, it appears we'll need to stick with existing models for several years. The next major release for AMD in the GPU segment would be the RDNA 5 lineup, but unfortunately, that too is delayed. AMD's RDNA 5 series would now launch in H2 2027, around the same time as Rubin's gaming GPU. It won't be wrong to say that GPU manufacturers are currently 'clueless' about how to handle client revenues from this segment, since the prospects are diminishing by the day. Our best estimate is that AMD/NVIDIA might launch one or two high-end SKUs as a 'placeholder' launch, and we've already seen rumors of NVIDIA launching a model more powerful than the RTX 5090. Having such models launched makes sense, simply because they are more attractive in terms of the profit/GB figures we discussed above, and at the same time, the use case for these SKUs is limited among gamers; instead, they target workstation and AI-focused workloads. NVIDIA's RTX 50 SUPER - Delayed to 2027 NVIDIA's RTX 60 "Rubin" Lineup - Delayed beyond its original 2027-2028 timeline AMD's RDNA 5 GPUs - Delayed to H2 2027 You might think I am missing out a name in the list, and well, I never miss anything. Intel's plans for Battlemage and Celestial GPUs are under uncertainity as well. We were expecting the Arc B770 to be unveiled at this year's CES, but unfortunately, that didn't arrive either. Xe3 'Celestial' did see a launch with Panther Lake, but for discrete GPUs, we have no definite timeline. Intel did hire a key Qualcomm executive to bolster its GPU development plans, but again, the efforts would be directed entirely towards enterprise/AI customers. Consumer GPU Current Supply Chain Situation: Extensive Price Hikes, Little Supply & No Signs of Recovery We do know that major GPUs have seen extensive price hikes over the past few months, and several reports have revealed that GPU manufacturers have started raising prices on AIBs. The best way to show our viewers how prices are rising is through PCPartPicker's tracking tools. We have attached an analysis of major SKUs that shows the average price increase over the past few months. Image NVIDIA/AMD, getting GPUs today is impossible, at least at reasonable prices. TimelineFebruary2026Feb 08NVIDIA’s Replacement for the RTX 50 SUPER Series Could Be a “Mega” RTX 5090 GPU, but It’s Unlikely to Change Anything for GamersFeb 05NVIDIA Reportedly Won’t Launch the RTX 50 SUPER Series This Year; GeForce RTX 60 “Rubin” Also Delayed as Memory Shortages Disrupt the PC IndustryJanuary2026Jan 16NVIDIA’s Strategy Amid GPU Memory Shortages Is to Prioritize “Profit/GB”, Turning 8GB GPUs Into Hot Picks Over Higher-VRAM ModelsJan 11AMD’s Next-Gen RDNA 5 Gaming GPUs Might Launch After NVIDIA’s RTX 60 Series In 2027November2025Nov 27NVIDIA Reportedly Halts Bundling VRAM Chips With GPU Dies For Board PartnersNov 07NVIDIA GeForce RTX 50 SUPER GPUs Reportedly Postponed To Q3 2026 Read all on AI Has Taken Over the Consumer GPU Market – Here’s How AMD & NVIDIA’s ‘Reprioritized’ Roadmaps Are Shifting Away From Gamers
