Did it improve the 5900X'es boost?
avidamoeba
Main site:
- 5950X on a GA-AB350-Gaming 3
- 64GB
- 1TB NVMe mirrored
- 24TB RAIDz1, using external USB 3 disks
- Ubuntu LTS
- 700Mbps uplink
- OpenWrt on Pi 4 router
- Home Assistant Yellow
Off site:
- ThinkCentre 715q
- 2400GE
- 8GB
- 256GB NVMe
- 24TB RAIDz1, using external USB 3 disks
- Ubuntu LTS
- 30Mbps uplink
- OpenWrt on Pi 4 router
Syncthing replicates data between the two. ZFS auto snapshots prevent accidental or malicious data loss at each site. Various services are running on both machines. Plex, Wiki.js, OpenProject, etc. Most are run in docker, managed via systemd. The main machine is also used as a workstation as well as games. The storage arrays are ghetto special - USB 3 external disks, some WD Elements, some Seagate in enclosures. I even used to have a 1T, a 3T and a 4T disk in an LVM volume pretending to be an 8T disk in one of the ZFS pools. The next time I have to expand the storage I'll use second hand disks. The 5950X isn't boosting as high as it should be able to on a chipset with PB2, but I got all those cores on a B350 board. ๐ Config management is done with SaltStack.
There may be PUE. However AI adds significant net new energy expenditure. Even if you get higher efficiency per compute unit, you're still burning more coal and gas, or using renewable capacity that could be going towards replacing coal and gas. And then in this particular example 7T is a huge amount of capital that could move the needle on so many difficult climate problems like decarbonizing steel, concrete and ag. I think spending that on AI that will likely only accelerate us towards planetary ecosystem destruction is .. not great to say it politely. ๐ Climate change is an existential problem for our species. The lack of AI advancement, even the lack of AGI is not. If there was any likely solution to climate change that could come out of AI, I might think differently. However we have all the computation methods needed to analyze and solve what we can about climate change. We just need much more resources in doing those things now.
Speaking of compute optimization, someone who's actually built chips just chimed in on the matter.
I don't know if OpenAI of all things is the contrast needed to show this. ๐
Oof. I'm gonna use that. ๐
Why not? Here's one reason - because the planet is burning from climate change and we need a shit ton more resources for solving it. It's unlikely that AI would help much with that. Of course the UAE isn't too interested in the world moving away from fossil fuels.
You're clearly not understanding what I'm saying and you're trying to mock. Not great but you do you. โบ๏ธ
NGL, was totally expecting a different last paragraph. ๐
Agreed and this is the reason I think was possible to hollow it out without it falling apart. Then some other guy with 5 extra brain cells and another robust system under their control is probably like - "I wonder how much labor I can hollow out, till shit starts cracking. So long as I don't pull any public stunts like Elon and I don't go beyond the cracks... ๐ค๐ธ"
BTW I've been at a hollowed out tech company. A very well known name. It's truly incredible how different things could look from the inside compared to the outside. Massive mountains of tech debt, people doing multi-week on-call shifts to keep shit running, rarely getting a full night's sleep, trustworthy household nameplate on the outside. The stock was doing okay. Still is actually.
I mean, the signs were there, Elon wasn't exactly silent about his plans for Twitter and investors gave him a lot of money to proceed. Clearly there is plenty stupid.
You can tell eh? ๐ Now consider the point of view that you can do the labor trimming without the additional massive mismanagement by Elon. Could that perhaps cause fewer advertisers and users to flee, could perhaps the cost savings vs lost revenue balance positively even if the service quality has decreased by some margin?
You play games on that server don't you. ๐