efog.tech
Chinchillin'
59 posts
19 followers
19 following
Regular Contributor
Active Commenter
comment in response to
post
And*
comment in response to
post
github.com/efogdev/ploo...
comment in response to
post
LLMs are just great for SQL in particular. I know so much more now.
comment in response to
post
Cursed. At least no 2 yo girls around
comment in response to
post
On the other hand, you get a floating connection in case something panics after you begin the transaction. But I tend to agree.
comment in response to
post
Ah yes finally a service for stoners with no memory
comment in response to
post
Stored procedures are a really good compromise.
comment in response to
post
Yes
comment in response to
post
What air defence doin
comment in response to
post
No Code November, shell commands counts
comment in response to
post
It serves different purpose I believe
comment in response to
post
Omfg every time
comment in response to
post
Except for syntax, it's kinda messy
comment in response to
post
Yes threads count is hardcoded :D
comment in response to
post
gist.github.com/efogdev/f737... (+/- LLM-generated)
comment in response to
post
@skatkov.com zstd for the win (my interpretation).
comment in response to
post
I just shared this with a friend, and he sent me an 8MB 7z file. It contains over 2GB of scraped web data. What the fuck (zip/tgz = ~530MB)
comment in response to
post
For some reason, this got me really curious about modern compression algorithms on modern CPUs.
comment in response to
post
For some reason, this got me really curious about modern compression algorithms on modern CPUs.
comment in response to
post
What's the content? Binary, text, media? Or you want something general-purpose
comment in response to
post
Isn't 100% CPU utilization good for such tasks?
comment in response to
post
That was my main problem with it. Immediately closed and removed. I've had enough ugly fonts in 2010-s :)
comment in response to
post
Tech can't be unethical, same as science can't. It just exists
comment in response to
post
github.com/efogdev/apollo
Repo to deploy locally. I think even ~12Gb of VRAM is enough (for small videos, not sure however).
comment in response to
post
I like big balls tho