AMD Has Ryzen

Okay, I know that subtitle is a little bit on the cheesy side; it’s such an obvious line that it must have been used a thousand times already on different technology news sites, but it also happens to be entirely true. When the first generation of AMD Ryzen chips launched back in 2017, it sent shockwaves throughout the entire PC industry. Intel had been content to deliver high-priced quad core CPUs for years, with little increase in performance between subsequent generations. AMD took advantage of this scenario by launching better performing chips with higher specifications at a lower price, and soon began to eat into Intel’s position as the dominant player in the market.

These new Ryzen chips packed a hell of a punch – I had been using my 2nd Generation Intel Core 2500K for years when they were released, and can remember struggling just to run my regular applications whilst browsing the latest games on the best slots sites. After upgrading to the Ryzen 1700X, all of my problems immediately went away, and I’ve been back on “Team Red” ever since.

Intel’s Seemingly Endless Woes

Intel’s story over the past few years is the polar opposite of AMD’s. Every generation of AMD’s processors seems to offer a gargantuan leap over the previous one – core counts continue to increase, and further miniaturization of semiconductor components has allowed raw clock speeds to continue increasing as well. Despite this, the highest-performing Intel chip has usually managed to just hold the edge over the best-performing part from AMD, albeit at an insane price premium.

Take last years 12900K, for example; this part runs in a state of barely controller overload and requires an enormous cooling setup to run with stability. When properly configured and installed, however, the chip can outperform an AMD CPU with more cores and a higher clock speed, so how can this be?

There are differences in the way both companies count the size of the transistors on their chips. This effectively means that the 7nm parts that AMD was offering in previous years were roughly equivalent to the 10nm chips being offered by Intel. As ever, raw numbers just do not tell the whole story when it comes to computing.

The problem is that Intel charges a ridiculous premium for their highest performing part; they have no choice but to do so, as the yields on the 12900K are reportedly terrible, with more than half of the chips on each wafer having to be sold as lower-performing units or simply being scrapped altogether.

Intel is now constructing new fabrication plants in the USA which should allow to catch up with AMD eventually; these plants will use the latest Ultra Violet Lithography technology to produce chips of 5nm and below, but it could be years before these plants are fully operational. In the meantime, AMD are likely to maintain their advantage, as we are about to discover.

The Next Battle: 7th Generation Ryzen vs 12th Generation Intel Mk2

The new generation of AMD Ryzen chips is set to be launched late in 2022, although this is likely to be just a handful of early chips. An announcement expected at the Las Vegas Consumer Electronics Show (CES) in the 2nd quarter of 2023, where the full range is likely to be shown off for the first time. AMD has already revealed pricing for the first four chips in the range; there will be a new Ryzen 9 7950X at $699, as well as a 7900X at $549. The lower-performing Ryzen 7 and Ryzen 5 variants will be priced at $399 and $299 respectively.

These prices are extremely competitive, and obviously sent shockwaves throughout Intel when they were announced earlier this week. The high-end Ryzen 9 7950X will have 16 cores, 32 threads, a 5.7Ghz clock speed and 80Mb of on-chip cache. Intels’ Mk2 12900K features eight regular cores and eight lower-performance “efficiency cores”, which are great for laptop and mobile device users, but not so great for those who want maximum performance on the desktop. The top speed is also much lower at 5.2Ghz, whilst the 30MB of cache is near one third of that on AMD’s latest designs. Clearly, Intel has an awful lot of catching up to do.

Furthermore, Intel is still asking US $589 for their highest-performing chip – that’s only $100 less than AMD’s highest offering, and a higher price than the 5900X which still offers 12 full performance cores, a 5.6Ghz clock speed, and 76Mb of on-chip cache. Let’s be honest here – this generation is hardly a battle – Intel just has nothing that can truly compete with what AMD can currently offer.

Don’t expect this situation to last forever though – the past few decades have seen the performance crown move back and forth between Intel and AMD repeatedly, and the cycle is likely to continue as long as both of these companies feel that there is sufficient competition to keep their R&D budgets high.

It’s fascinating to wonder what the situation might be today if there will still half a dozen major semiconductor manufacturers producing chips, as there was in the 1990s, but as that is no longer the case, we can only hope that these two companies will continue to spar with one another for the foreseeable future.