Some believe AMD might be missing the mark by not releasing X3D chips at launch.
Some believe AMD might be missing the mark by not releasing X3D chips at launch.
I've been considering this situation, and AMD's September 27th release feels quite disappointing. The lack of x3d chips really hurt the overall impression. Now that the 5800x3d is available, it's clear how much extra memory capacity can boost performance across various tasks, especially gaming. I'm not surprised if the reaction to the Zen 4 announcement was muted—just a modest increase in single-threaded speed and decent IPC, but the cost and complexity are too high. Back when Intel was still pushing 14nm+ technology, this would have been acceptable, but AMD is now a few generations ahead. I was leaning toward the Red Team build, but the whole reveal felt underwhelming. In a market where Intel was still leading, AMD's delay of Zen 4 from September 13th to 27th seemed strategic. They positioned Raptor Lake as a stronger alternative, despite the single-core gains being modest. Board prices are now disclosed. To assemble a solid AM5 system, you'd need to spend around $249 on the board, $179 for DDR5, and $299 for the CPU. Plus, cooling will be a major challenge—expect a massive tower or AIO cooler for the 7600x. While Raptor isn't as power-efficient, it still outperforms Zen 4 on single cores and will crush them in multi-core scenarios. The 13600k vs 7600x comparison will be striking. The 13600k is expected to nearly double multi-threaded performance, plus the chance to use a cheaper Z690 board and DDR4 RAM. I suspect early sales of Raptor will outpace Zen 4's, mainly because of the missing X3d chips. The 5800x3d already shows impressive potential with more cache and DDR5 support. If AMD had released X3d at launch, Raptor would likely fail. It almost feels like a delayed Zen 4 reveal, with the real launch coming at CES.
Alternatively, they might not use X3D chips at all in this generation. Intel did the same—once they made a single chip with a large cache and then stopped trying. Late-stage products like the 5800X3D also tend to appear later in life. High-end tower coolers are readily available for around $60. However, if the boards cost $250, that would be a significant issue. It seems likely they’ll find better options at lower prices or struggle to compete even with improved chips, as seen with Intel’s 12900K/KS versus the $130 B660 board.
The boards come with a premium price tag. The MSI X670E Gaming Carbon costs $480. Comparing its price jump to the original Z690 Gaming Carbon at $380 shows a $100 rise overall for all AM5 models, not just the Z690 ones. Even their older AM4 versions would be more than double that amount. The B660 tomahawk is priced at $189, meaning the B650 tomahawk would be around $289—making it a top choice for balanced value in mid-range builds. That’s a significant jump.
It's not about fabricating anything. Intel confirmed this morning that the 13700k, 13600k, and 13900k models are real. The 13600k will feature 14 cores compared to the 6 cores on the 7600x. It's puzzling how anyone would think multicore performance won't improve. The Zen 4x3d chips seem to outperform standard Zen 4 by about 30% in gaming. This suggests they're prioritizing their own products if that's the case. The 7900x will become less relevant once the 7800x3d is released. It's their strategic move.
The outlook seems adjusted. Roadmaps suggest 3D v-cache chips might be mid-cycle, possibly due to development or business factors rather than immediate launch. AMD has evolved beyond being the underdog, so their approach now aligns more with competitors. If Gaming Carbon delivers strong features, it could be acceptable. For a top SKU board, this is typical in recent releases. It's unlikely to be a flat scalar, and while lower chipset/feature boards might see less change, some improvements are expected with DDR5 and PCIe 5.0 support.
The most affordable X670 boards are expected to be around $300 or less from all vendors. B650 models are the only ones average players should think about, while X670 boards seem overpriced unless you're willing to pay a lot. Unfortunately, B650 boards won't be available until about a month after their release. The main expense on X670E boards is the large amount of Gen 5 PCIe and dual chipsets. B650 boards likely won't include those features, so the price jump for a B650 should be closer to $20-30 rather than $100. We'll find out for sure once they're officially released.
I completely agree with this perspective. It seems unclear how the original poster is suggesting Zen 4 X3D chips will replace Raptor Lake since no release date has been announced yet. We should observe how Raptor Lake evolves under different developers rather than relying on Intel's claims. Chip development is a lengthy process, so thinking of X3D chips as an evolution similar to Intel's Tick-Tock cycle—where each phase improves performance and efficiency—is reasonable. That explains why these technologies are cutting-edge. DDR5 is still in the early stages for consumers, with high costs and limited availability. The new architecture means you're investing now for future advantages rather than waiting a few years after launch.
Fair observation, even with just a few cores they still add value to multicore setups, making the tables seem more balanced. Intel remains stronger for productivity, though this time they also perform well in singlecore compared to Zen and Zen+. The 13600 non-K and lower models are essentially rebranded Alder Lake; if they’re not affordable, a 12600K model might be better. Still, reviews will matter—Intel is likely to dominate decisively, with solid ECORE numbers and solid single-core performance plus budget-friendly options. Cheap options with DDR4 chips won’t stand a chance, but pricing could be the issue. 300$ vs 7600x doesn’t seem to alarm them much.