I started with computers in the cassette tape/5,25 floppy diskette era. RAM was measured in K’s, not Megas, not Gigas. A flight simulator (white pixels on green background, yeah) fit in a diskette, 360 KB.
Code was ultra efficient, as RAM was tiny.
With time, and as RAM got proportionally cheaper, code got sloppy, because, hey, we got space.
It’s true that programs today are several orders of magnitude more capable, but it is also true that code now is hideously RAM inefficient and bloated.
I’m not saying that the RAMcopalypse isn’t real, I run a Ryzen 9 with 64Gb in my desktop, bought before the bubble, but as the Microsoft backpedal to 8 Gigs for W11, and the 8 Gig Macbook show, software should be rethought.
AND RAM prices should go back to normal. That’s the win-win
Going back to the 8-bit era muddies the water. Those computers and programs really were so limited. However, comparing what we were doing 20 years ago on 2 or 4GB of ram vs now and you see the sloppness of the software.
You are expressing an illusion. The reason programs and operating systems use more ram is that they are doing more. If somebody stopped caring about resources it is web devs and Microsoft. Real-time and high end apps, like all games, have optimized frame time and the floppy era games is what looks sloppy in comparison. Except for sawyers insane games. Those are still somehow the cutting edge and a marvel of engineering to learn from 30 years later.
It’s true that programs today are several orders of magnitude more capable, but it is also true that code now is hideously RAM inefficient and bloated.
The entire codebase of software in production, yes. But one thing that people might not realize is that some of our software algorithms themselves, at least for certain algorithm families, are also much more efficient than before.
The focus isn’t on memory usage, but the same general trend applies there: when someone discovered a new algorithm that improves on old ones, it can change the amount of memory used to accomplish a specific result.
I started with computers in the cassette tape/5,25 floppy diskette era. RAM was measured in K’s, not Megas, not Gigas. A flight simulator (white pixels on green background, yeah) fit in a diskette, 360 KB.
Code was ultra efficient, as RAM was tiny.
With time, and as RAM got proportionally cheaper, code got sloppy, because, hey, we got space.
It’s true that programs today are several orders of magnitude more capable, but it is also true that code now is hideously RAM inefficient and bloated.
I’m not saying that the RAMcopalypse isn’t real, I run a Ryzen 9 with 64Gb in my desktop, bought before the bubble, but as the Microsoft backpedal to 8 Gigs for W11, and the 8 Gig Macbook show, software should be rethought.
AND RAM prices should go back to normal. That’s the win-win
Going back to the 8-bit era muddies the water. Those computers and programs really were so limited. However, comparing what we were doing 20 years ago on 2 or 4GB of ram vs now and you see the sloppness of the software.
You are expressing an illusion. The reason programs and operating systems use more ram is that they are doing more. If somebody stopped caring about resources it is web devs and Microsoft. Real-time and high end apps, like all games, have optimized frame time and the floppy era games is what looks sloppy in comparison. Except for sawyers insane games. Those are still somehow the cutting edge and a marvel of engineering to learn from 30 years later.
The entire codebase of software in production, yes. But one thing that people might not realize is that some of our software algorithms themselves, at least for certain algorithm families, are also much more efficient than before.
I thought this was an interesting overview:
https://futuretech.mit.edu/publication/how-fast-do-algorithms-improve
The focus isn’t on memory usage, but the same general trend applies there: when someone discovered a new algorithm that improves on old ones, it can change the amount of memory used to accomplish a specific result.
Having written both hand-coded assembler for microcontrollers and vibe-coded WebUI … things, I don’t know where my allegiance lies anymore.