HDMI 2.2 on next-generation Radeons: 80 Gbps as a status symbol or a real technological leap?

If you take a sober look at the latest revelation about AMD’s upcoming GPU generation, the question quickly arises: Is HDMI 2.2 with up to 80 Gbps really a milestone – or rather a marketing bolide on the data highway whose actual goal remains unclear?

GFX13, UDNA and the bandwidth skirmish

AMD is working diligently behind the scenes on the upcoming GPU architecture “GFX13”, which is already being traded internally as the basis for the UDNA series – the logical successor to the RDNA-4 family. While the chips themselves have probably not even been finalized yet, the first details are already leaking out via Linux drivers and well-known Twitter leakers such as Kepler_L2 – this time about the interface policy. The magic word of the hour: HDMI 2.2 – according to the specification with up to 96 Gbps bandwidth, whereby AMD’s implementation should be designed for 64 Gbps or 80 Gbps. A huge leap compared to HDMI 2.1b, which previously stopped at 48 Gbps. Sounds like a revolution on paper, but as is so often the case, the devil is in the detail.

Technically strong – practically relevant?

HDMI 2.2 undoubtedly brings theoretical advantages: 4K at 240 Hz with full 10/12-bit color depth, 8K at 240 Hz or even 12K at 120 Hz – a feast for specification fetishists. Uncompressed formats such as 8K60 4:4:4 or 4K240 4:4:4 make the hearts of technology-loving monitor manufacturers beat faster. But hand on heart: who actually owns a display that supports these formats? And who really needs it outside of high-end studio environments? In practice, many features can be implemented using DSC (Display Stream Compression) anyway – a process that has hardly any visual disadvantages, even for demanding applications. The move to HDMI 2.2 therefore appears to be less of an absolute necessity and more of a “compulsory exercise in the theater of progress”.

HDMI2.2 64G and 80G 🧐

— Kepler (@Kepler_L2) June 19, 2025

DisplayPort – the silent side issue

While HDMI is in the media spotlight, DisplayPort remains the silent workhorse – but not without controversy. According to leaks, AMD is sticking with DisplayPort 2.1 with UHBR13.5 (54 Gbps) in the consumer segment – while professional GPUs in the PRO series are allowed to work with UHBR20 (80 Gbps). The reason: cost. In other words, the mass market is given just enough to keep it going – but not enough to excite it. Incidentally, NVIDIA is taking a two-pronged approach here: HDMI 2.1b and DP 2.1b with UHBR20 are a reality in the current lineup. This gives you a slight technical advantage, at least on paper. But here too, if you don’t operate a native 8K workstation, you won’t notice much of this.

Between technical necessity and diplomatic duty

You could almost think that the new specifications are intended less for gamers or workstations – and more for regulators, investors and industry partners. It shows: Look, we can keep up. Or even better: we are ready for what will become standard in three to five years. Whether the market is already there plays a subordinate role. HDMI 2.2 is becoming the symbol of an industry that is overtaking itself with ever more absurd figures. The real bottleneck is often the software anyway – or the thermal design of the GPUs themselves.

More numbers, less benefit – but at least future-proof

The upcoming UDNA generation with GFX13-IP and HDMI 2.2 is a clear signal: AMD not only wants to be a player in the field of display interfaces, it wants to dominate, at least on paper. As is so often the case, it is questionable whether this will result in real benefits for the end customer. Most people will probably be fine with HDMI 2.1b and DisplayPort 2.1 – provided the GPU delivers more FPS in the game than Gbps on paper. And as long as real 8K gaming monitors with 240 Hz are as common in the living room as unicorns on the instruction leaflet, HDMI 2.2 will probably remain just that: a nice promise for the future – but not a must for the present.

Source: Kepler_L2