Once third-party results confirmed AMD's outstanding benchmarks and retail delivery was a success, the big remaining question was: could the company extend its 7nm success story to mobile and server CPUs?

Yesterday, AMD formally launched its new line of Epyc 7002 "Rome" series CPUs—and it seems to have answered the server half of that question pretty thoroughly.

Rome offers far more CPU threads per socket than Intel's Xeon Scalable CPUs do.

While it's easy to complain that Intel CPUs have an unfair advantage in MKL-DNN benchmarks, it is representative of the kind of entrenched advantage Intel enjoys—and it's a real advantage.

On vendor-neutral and multithreading-friendly workloads such as x265 video and OpenSSL, the Rome CPUs significantly outperformed the Xeons across the board.

In the past, that sharp delta in cost efficiency has forced a project to seriously consider putting up with added machine count and more complex overall architecture.

Now that AMD Zen 2 processors have hit both the mainstream market and enterprise with Ryzen 3rd Generation and 2nd-generation Epyc processors, respectively, it's time once more to look to the future.

And, it looks like AMD has finished the design phase for its Zen 3 CPU architecture.

This news comes out of its reveal event for 2nd-generation Epyc processors, where AMD showed a slide indicating that the Zen 3 architecture design is complete, and even labeled the more-distant Zen 4 architecture as "in design".

This doesn't mean that AMD Ryzen 4th Generation processors are right around the corner, but it does mean that AMD isn't taking their recent victories as permission to slow down.

AMD Zen 3 will likely be based on a 7nm+ process, much like AMD Zen+ after it followed the original Zen architecture with a 12nm+ manufacturing process.

That means we can expect higher clock speeds paired with lower power consumption.

Back at Computex 2019 when AMD CEO Lisa Su took the stage to unveil all the Zen 2 processors Team Red would be releasing over the next year, we knew that server processors were on the way – and now they're here.

AMD has unveiled its Epyc processors, and they're featuring some pretty serious specs.

With these server processors, you'll be able to get up to 64-cores provided you're willing to drop the $6,950 (about £5,580, AU$9,890) it calls for.

Now, AMD Epyc processors are for servers, which means you won't exactly find them in the best gaming PCs, but they also serve as an example of where AMD's 7nm (nanometer) Zen 2 architecture can go when it's pushed to the extreme.

And, because these processors feature the same 15% IPC (instructions per clock) uplift as their Ryzen 3rd Generation counterparts, they feature up to 80% faster performance over last generation products, according to Tom's Hardware.

Couple the raw performance gains with the support for up to 4TB of RAM in a dual-CPU setup and the relatively low power requirements, and AMD has some serious firepower in the server space – where the AMD vs Intel war will be won.

Qualcomm might be a contender in the mobile space, but Always Connected Windows laptops powered by Snapdragon 835 and 850 chips have seen mixed results.

Both Intel’s and Qualcomm’s chips combine CPU and GPU cores on a single die, so we’ll look at the performance of each segment of the chips individually and consider their cost and efficiency too.

That means it has eight cores to work with, four of which will be lower-power and operate when efficiency is more important than performance.

That may have been evidenced during a graphics test at Computex, which saw the Intel 8250U maintain a greater CPU score than the Snapdragon part, despite seemingly weaker performance in more productivity-focused tasks.

That’s a more than 20 percent uplift in performance for the Snapdragon SoC and equates to an FPS difference of 29 FPS for the Intel chip (at best), versus the Snapdragon’s 37 FPS.

One of the concerns with Microsoft’s “Always Connected” PC style is the lack of native applications for the Qualcomm hardware under the hood.

Whether its with the AMD Ryzen 7 3700X selling more than Intel’s entire lineup, or with the Ryzen 9 3900X outperforming any other processor in its territory, Team Red has pulled ahead.

However, it’s not hard to look back four or five years ago to see AMD processors sitting at the bottom of the pile.

For instance, the AMD Ryzen 7 1800X was able to score 1,612 in Cinebench R15, compared to the Intel Core i7-7700K’s 970.

Looking back, this kind of started a pattern, where AMD processors would come out, wipe the floor with the current-generation Intel processors in multi-core workloads, then Intel would release another refinement of Skylake, and then the process would repeat.

These were potentially catastrophic security failures, where attackers could execute some relatively simple code to get access to anything from passwords to photos you’ve since deleted.

There is more information here, but essentially these attacks take advantage of speculative execution, where a processor essentially guesses its next project, and then dumps any wrong data into the cache.