System news
Intel launches its 10th lake. Mark Williams looks at why Intel is treading water in it.
The core wars between Intel and AMD continue to rage on. Intel has released its 10th Gen Core processor line-up (Comet Lake) and despite being on the back foot, still stuck on its 14nm++ process node (the number of pluses one can append to that is something of a humorous meme in the industry now), it has put out some products that are not only better than the outgoing 9th Gen parts but continue to go toe-to-toe with AMD’s parts.
With no architectural improvements, Intel is scraping the bottom of the barrel in terms of pushing what the silicon can do and is finding innovative ways of taking advantage of the one solid strength it is still king of the hill of: clock speeds.
Intel’s parts clock very high in stock trim, into the high 4GHz range and with the 10th Gen into the low 5GHz, and Intel has been working hard to allow its processors to boost higher and for longer to maintain those peak clocks by adding yet more turbo boosting algorithms. A list which now includes: Turbo Boost 1.0, Turbo Boost 2.0, Turbo Boost Max 3.0 and Thermal Velocity Boost. All of which work together to give you the highest clocks possible at any point in time. Combined, they’re not quite as advanced as AMD’s singular Precision Boost 2 technologym, which has that company’s processors clock almost like a GPU in that clocks mostly depend on temperatures, but it’s getting close and it’s enough to keep Intel well ahead in gaming titles.
Intel is also sanding down the tops of the CPU dies it’s making to further enhance thermal conductivity to the IHS and heatsink to improve thermals on what are very power hungry and hot processors.
Despite Intel being stuck in the mud, it is still swinging hard and AMD while broadly with the technical lead overall, isn’t so far ahead that it can dominate the market yet. It must make you wonder what Intel will pull out of its hat when it finally does figure out 10nm and drops its new IPC improved microarchitecture. AMD has got to be worried about that looming scenario.
Speaking with Richard from Aftershock PC, it’s obvious that AMD is still outselling Intel quite comfortably, and that Intel is still having a hard time with supply from its constrained 14nm fabs, with 10900K shortages a-plenty.
The two things that hurt Intel, still is pricing and segmentation within its own line-up. Intel is still selling locked and unlocked processors, and selling only one chipset – its most expensive one – that can overclock.
This leaves a situation where the somewhat reasonably priced i5-10600K must be mated with a Z490 motherboard to achieve its full potential, something which is double the price of an AMD B450 motherboard. At least the LGA1200 socket is guaranteed to offer one extra generation of CPU support – something Intel hasn’t done in the consumer segment for an awfully long time.
Intel is treading water at the moment and has done enough this generation to remain a relevant
– if relatively pricey – choice if its specialty meets your requirements. With AMD’s Zen 3 architecture due out later this year, however, Intel could soon find itself in deep water. Its 10nm node for desktop processors can’t come soon enough.