Nvidia RTX 60-Series Could Bring More Power, But Don’t Expect a Wild GPU Size Jump
Nvidia’s next GeForce generation is still not official, but PC Gamer has done the spreadsheet grind and looked back at 10 years of GPU data to make an educated guess at what the RTX 60-series could look like.
The big takeaway? Don’t expect Nvidia to simply make every chip massive and call it a day. Based on past trends, the future RTX 60-series may bring a serious transistor jump, stronger shader performance, and more AI muscle — but likely while keeping die sizes fairly close to today’s RTX 50-series.
That matters a lot for Malaysian and SEA PC gamers because GPU pricing here is always painful. Once you add local stock issues, currency conversion, retailer margins, and “early adopter tax”, even a mid-range card can become a serious RM decision. If Nvidia keeps chips smaller and more efficient, that could help production costs — though whether those savings reach us in Malaysia is another story lah.
Why the manufacturing node matters
Modern Nvidia gaming GPUs are built by TSMC, the Taiwanese chip giant behind many of the world’s most advanced processors. The current RTX 50-series gaming chips use a custom version of TSMC’s N5-class process called 4N, similar to the RTX 40-series.
PC Gamer’s analysis suggests Nvidia may not jump straight to TSMC’s bleeding-edge N2 process for future gaming GPUs. Instead, the prediction is that RTX 60-series chips could use TSMC N3, partly because Nvidia is now heavily focused on AI hardware too, where the newest manufacturing capacity is extremely valuable.
If Nvidia does use N3, transistor density could rise significantly versus Blackwell — potentially around 60% or more in broad terms. But that does not mean every part of the GPU scales equally. Logic blocks like shader cores benefit more from node shrinks, while cache and memory-related circuitry do not shrink as neatly.
In simple gamer terms: Nvidia may be able to pack in more CUDA cores, but it probably cannot just throw in a ridiculous amount of extra cache without making the chip bigger.
More shaders, but maybe not a monster leap
Looking back from the GTX 10-series through RTX 20, 30, 40, and 50, Nvidia has not always increased CUDA core counts in a straight line. Some generations delivered big gains through architectural changes and clock speed improvements instead.
PC Gamer expects RTX 60-series cards could see a shader count increase somewhere around 30% to 50% compared with Blackwell, rather than a full jump matching transistor density. That sounds realistic. Nvidia still wants strong margins, and smaller dies usually mean better yields and more chips per wafer.
For SEA gamers, this is the part to watch. If the future RTX 6060 or RTX 6070 delivers a proper performance uplift without needing huge power supplies or giant triple-slot coolers, that is more useful for typical Malaysian builds. Not everyone is running a full-tower case with premium airflow. A lot of us are building around budget boards, compact cases, and whatever GPU deal pops up on Shopee or Lazada.
Cache, AI, and ray tracing remain key
Since RTX 40-series, Nvidia has leaned hard into larger L2 cache to reduce pressure on VRAM bandwidth. According to the analysis, RTX 60-series GPUs may keep similar cache amounts to Blackwell because SRAM does not scale well on newer process nodes.
AI performance is another huge piece. Nvidia’s Tensor cores have become more capable every generation, but PC Gamer notes that headline AI TOPS figures can be misleading because each generation uses different data formats. The expectation is not necessarily a massive per-core improvement, but more Tensor capability overall thanks to higher transistor budgets.
That could mean better DLSS, frame generation, ray tracing support, and future AI-assisted rendering features. For competitive esports players, raw latency and stable FPS still matter more than fancy lighting. But for AAA gamers in Malaysia playing on 1440p monitors or big TVs, better upscaling and frame generation could be the difference between “playable” and “sedap gila”.
The bottom line
This is not a leak or official spec sheet. It is a data-based prediction built from Nvidia’s past GPU patterns. Still, the logic makes sense: RTX 60-series cards may use a denser TSMC process, keep die sizes controlled, increase shader counts moderately, and continue pushing AI-driven graphics features.
For Malaysian buyers, the real question will be price. If Nvidia gives us stronger mid-range cards with sensible VRAM and decent power efficiency, the RTX 60-series could be a meaningful upgrade. If the pricing goes crazy again, many gamers here will probably stretch their RTX 40/50-series cards a bit longer — no shame, bro.
Source: PC Gamer
