345
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 01 Jan 2024
345 points (94.8% liked)
Technology
59623 readers
1446 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
Moore's law has been dead for a long long time.
E: if you're downvoting this it's because you don't have a clue what you're talking about. Moore's law was the observation that transistor density would double every ~2 years. That's not happening and hasn't for a long time.
No need to downvote this. It's an insidery technically correct statement. We've redefined how we measure Moore's law several times to make it "keep working" and some people designing chips, not selling them, think it's not only outlined it's usefulness but also not true anymore.
In my experience, a lot of people incorrectly conflate Moore's Law with "computers get faster"
So when you say Moore's Law is dead and it's unrealistic to expect it not to be, they get upset and jump to the conclusion that you're defending tech companies for giving paltry upgrades, which obviously isn't what I'm doing.
There are other things to PCs getting faster in a post Moore's Law world. Architecture improvements, hardware acceleration, advanced packaging such as AMD's chiplet technology, etc - these are all commonplace and have replaced the idea of "let's just double transistor counts every two years"
We've gone through die size, clock speed, instructions and operations, the transistors count. All are stand-ins for "complexity" which is why some people question if the law ever existed.
That said, regardless of the "real" law, until recently the colloquial usage has always been a stand in for how "quick" a processor is. In that sense, you really need to do some hand waving around core counts and even then it doesn't really work.
Maybe more importantly, one of the most important processor markets are mobile and servers which are largely focused on less complex more efficient processors like arm.
So outside of marketing, it's very easy to see why a lot of people think Moore's law is dead and we're all better for it. We can continually make better processors without trying to meet some arbitrary metric that didn't really mean anything useful to start with.
E: aggressively agreeing
Moore's law hasn't died, if you mean number of transistors per area. Linear scaling to transistor counts has.
It absolutely has. Transistor count in an area absolutely is not doubling every two years.
I just checked. Yup, you're right. Funny though that Pat of all people claims it's not dead lol