FSR 4: AMD RDNA 3 & Older GPU Support – Update Status

0 comments

AMD is playing a dangerous game with its customers, and the latest chapter in the FSR 4 saga proves it. While the company touts the advancements of its FidelityFX Super Resolution 4 (now simply FSR 4) technology, it’s deliberately withholding functionality from older Radeon GPUs – even though the capability demonstrably exists. This isn’t about technical limitations; it’s about artificially creating value tiers and pushing consumers towards the latest hardware. The core issue? FSR 4’s full potential relies on 8-bit floating point (FP8) instructions, found in the new RDNA 4 architecture, while older cards lack this hardware support. But a leaked version utilizing 8-bit integer (INT8) formats proves older GPUs *can* run a version of FSR 4, albeit with a performance trade-off.

  • Artificial Scarcity: AMD has a working FSR 4 implementation for older GPUs but is choosing not to release it, likely to incentivize upgrades.
  • Performance Trade-offs: The INT8 version of FSR 4 offers a balance between native resolution quality and FSR 3.1 performance, but isn’t as fast as the FP8 version.
  • Product Segmentation: AMD is increasingly segmenting its features, reserving cutting-edge tech like Ray Regeneration for its newest RX 9000 series cards.

Let’s be clear: this isn’t a new tactic. The tech industry thrives on planned obsolescence and feature gating. However, AMD’s approach feels particularly cynical. FSR, initially positioned as an open-source, broadly accessible upscaling solution to combat Nvidia’s DLSS, is now becoming increasingly locked down. The shift began with FSR 3 and its frame generation technology, and now FSR 4 is following suit. The leaked INT8 version, tested by ComputerBase, shows a performance hit – around 16% slower than FSR 3.1 in Cyberpunk 2077 at 4K Ultra settings – but still delivers an 11% performance gain over native rendering. That’s a worthwhile improvement for many users, and AMD is simply choosing to deny them access.

The underlying reason is simple: AMD wants to justify the price premium of its RX 9000 series GPUs. By reserving features like Ray Regeneration and the full FP8 FSR 4 experience for its latest hardware, it creates a compelling reason for consumers to upgrade. This strategy mirrors Nvidia’s, which AMD previously criticized. The company’s official line of “no updates to share at this time” regarding FSR 4 support for older GPUs is a carefully crafted non-answer, designed to buy time and maintain the illusion of exclusivity.

The Forward Look: Expect continued pressure from the community for AMD to release the INT8-based FSR 4. The leak has already demonstrated its viability, and the negative PR from withholding a functional feature could outweigh the benefits of artificial scarcity. However, AMD is unlikely to budge unless forced to. The more interesting development will be how this strategy impacts future FSR iterations. Will AMD continue to segment features, creating a tiered system where older GPUs are perpetually left behind? Or will it eventually relent and offer broader support, potentially sacrificing some revenue in the process? I suspect we’ll see a slow drip-feed of features to older cards, but only after the majority of enthusiasts have already upgraded. The real question isn’t *if* AMD will release the INT8 version, but *when* – and how much damage will be done to its reputation in the meantime.


Discover more from Archyworldys

Subscribe to get the latest posts sent to your email.

You may also like