Raspberry Pi Drag Race: Pi 1 to Pi 5 – Performance Comparison • Hacker News Board

Re: Raspberry Pi Drag Race: Pi 1 to Pi 5 – Performance Comparison

#91

Post by andai » Sun, Jan 25, 2026, 8:02 PM UTC

tomcam wrote: post #57

andai wrote: post #53

Earlier quoted context omitted.

If it's low RAM and single threaded then I would expect every tier to perform identically. (Also I was just looking at the pricing, you don't get any extra CPU cores until the $84 tier!)

Thanks for pointing that out. I had thought maybe as the tiers went up you'd get faster CPUs, but upon reflection the pricing schedule seems to confirm your point. I guess they are completely virtualized because there is not mention of CPU type on the pricing page.

I'm not an expert on this but I recently read that "vCPU" doesn't actually mean anything, and there's indications that they might not even be cores but hyperthreads.

Which makes sense with the policies many providers have against actually utilizing all the CPU you paid for.

Re: Raspberry Pi Drag Race: Pi 1 to Pi 5 – Performance Comparison

#92

Post by Mr_Minderbinder » Mon, Jan 26, 2026, 12:31 AM UTC

encom wrote: post #51

zahlman wrote: post #40

> The Pi 2 managed to open the browser and actually started playing back a 1080P video, which was surprising, but playback was terrible. A Youtube tab, web browser modern enough for YouTube, and OS modern enough for that web browser, all fit in 1GB of memory? Wow.

Worth noting: A Pi 1 will decode 1080p h264 video flawlessly. I used one as a media center for a long time, running OpenELEC (now LibreELEC). Moving around the UI was a bit sluggish, but video playback was smooth. YouTube is an absolute clown show. It's so bad that I'm certain Google devs are actively making it terrible on purpose. I use Newpipe on an older (but not that old) tablet. Whenever Google breaks Newpipe an…

> YouTube is an absolute clown show. It's so bad that I'm certain Google devs are actively making it terrible on purpose.

Exactly, which is why I thought this was a terrible and meaningless benchmark. It completely obfuscates the actual video playback performance of these machines. It is more a measure of how awful and inefficient YouTube is. I am surprised that the author did not remark on or seem to be aware of this at all.

Re: Raspberry Pi Drag Race: Pi 1 to Pi 5 – Performance Comparison

#93

Post by tomcam » Mon, Jan 26, 2026, 3:26 AM UTC

mappum wrote: post #83

mappu wrote: post #73

Earlier quoted context omitted.

Fun! 126 Intel N5105, gcc -O2 (Debian) 98 Neoverse N1, gcc -O2 (Debian) - Oracle Cloud 63 Snapdragon 8gen3, clang -O2 (Termux) - Xiaomi 14 59 Ryzen 7600, clang -O2 (MSYS2 / Windows 11) 48 Ryzen 7600, clang -O2 (Debian) 45 Ryzen 7600, gcc -O2 (Debian)

Interesting username

Deep-ish cut

Re: Raspberry Pi Drag Race: Pi 1 to Pi 5 – Performance Comparison

#94

Post by RetroTechie » Wed, Jan 28, 2026, 6:48 PM UTC

cmovq wrote: post #48

Interestingly the original Pi had the same amount of memory as the PS3 which was available at the time of the Pi release. Still amazes me how much we did with only 512MB.

> Still amazes me how much we did with only 512MB.

Or the other way around: still amazes me how many GBs today's machines need to do conceptually simple things. Things that were done ages ago (successfully, if not better) on much, much, much lower-powered kit. Never mind CPU speed.

Eg. GUIs have been done on machines with a few 100s Kbytes of RAM. 'Only 512MB' is already >1000x that.