John Carmack Says The Silicon CPU Days Are Numbered
Mike Sanders / 10 months ago
Over the last 5 years, we have seen some pretty huge advancements in CPU designs. This has largely been driven by smaller node designs and AMD is certainly pushing this hard with their 7nm products such as their 3rd-generation Ryzen processors and 5700 graphics cards. This has, however, brought various problems with it and lord knows its hardly a secret that Intel is really struggling to move away from the 14nm platform.
In a report via PCGamesN, however, John Carmack (video game developer/programmer best known for ID Software) has said that he believes that the silicon-based CPU is approaching the end of its practical life-span.
Silicon CPU is “Going to Get an End of the Line”
We should clarify that he is 100% not saying that silicon-based CPUs are dead. Such a statement would be ridiculous at the moment largely because there’s nothing even close to replacing it. He does, however, believe that beyond a couple of further node shrinks, the technology might go as far as it practically can.
“They will have full confidence that we’ll see a couple more node shrinks. So it’ll still make chips cheaper, somewhat faster, more cores on them, but it’s going to get an end of the line.
But I hold out hope for potential other things, you know. There are directions that maybe you have your carbon nanotube wires. Or you’re starting to be able to do some things with photonic processing in different ways. There are possible outs for it, but I don’t know that any of them are a sure enough things to really be counting on at this point.”
Speaking of carbon nanotube alternatives, a working design was revealed earlier this week. The worlds first CPU to not use silicon. You can learn more about via the link here!
What Do We Think?
Creating smaller node designs is a very difficult process. Not just in terms of developing the technology, but in more basic terms of expense. While we are currently transitioning to 7nm node designs, most industry experts agree that the best we would hope from the technology would get this down to 3nm. Even then, the practicality might see an alternative route taken long before that happens. In terms of a 3nm design, however, even if it does happen, we’re looking at it being many, many, years away. At the bare minimum, probably 10-15.
We do, therefore, probably think he’s right. Until the alternative is invented, however, silicon-based CPUs are set to stick around for a good while to come yet!
What do you think? – Let us know in the comments!