look what the prusa dragged in
chain test! for a crossbody bag chain
calibration for the mad overhangs was a bitch, two totally fubared prints before I figured it out
my only problem now is that if I make this bonkers crazy cool bag there will be nothing on youmagine that can top it
so yeah I'm learning blender
@roxmsauce love the keyboard :D
@roxmsauce @Rtzq0 tell me more! I am nubbin addicted and would love one.
RIGHT RIGHT AMIRITE
so this baby is a 60% mechanical keyboard by Tex, a quick web search yields
https://mechanicalkeyboards.com/shop/index.php?l=product_detail&p=3094
idk though, I remember these being impossible to find like a year ago? Hopefully they'll stay in production. I seriously cannot function without a nubbin and this thing is just WAY TOO NICE
@roxmsauce @Rtzq0 OK, it's the one I already knew of. I want a full-size or TKL, not a 60%...so I passed on the Yoda II. Lenovo did a limited run of those with their badging even.
How's the nubbin for you compare to a e.g. ThinkPad keyboard? Heard some reports it was a bit stiff.
It is, now that you mention it. I figured I just hadn't dialed in the calibration yet - I want to change the nubbin sensitivity (YEAHHHHHHHH) and also change some of the key mappings. Still, works fine right out of the box, linux woo! I have an insane number of input devices plugged in right now and they all work, hard to believe I don't have more calibration issues honestly
@roxmsauce @wohali as many frustrations as I many have about the stuff coming out of RH these days, the libinput autodetection/hotplug defaults have gotten pretty good.
@Rtzq0 @roxmsauce Haven't explored it. Definitely systemd averse (that's putting it mildly...)
@wohali @roxmsauce I may at some point move away from arch to something openrc-based. Unfortunately linux in the business space will continue to involve systemd, since it's now the init in all the major server distros. It's possible to use libinput and wayland w/o systemd I've been told, but I haven't had the free time yet to put in the legwork.
@Rtzq0 @roxmsauce I've moved most of my business deployments to fbsd, as much as possible. Feels nice leaving SVR4 behind, oddly. Like a homecoming.
...yes i am old shut up
@Rtzq0 @roxmsauce Spent a bunch of time with SmartOS. I'm not going back to Solaris any time soon. I can get into this at length but preferably not in a public forum.
Mostly freebsd for commercial customers. A few friends work on netbsd which is my choice when I need bsd on a weird or old platform. Theo is primarily why I won't use openbsd ;)
@wohali @roxmsauce that last sentence is fair.
There are many things that exist in Solaris that I wish the Linux community would prioritize, but realistically it seems that people don't want them (support for good fACLs being high up there.)
@Rtzq0 @roxmsauce Right, and fbsd (well, ZFS, really) has pretty good ACLs overall. You can zfs-on-linux if you want, but the licensing bothers me, possibly more than it should.
@wohali @roxmsauce zfs-on-linux doesn't get you NFS v4 fACLs (still an open issue https://github.com/zfsonlinux/zfs/issues/4966). On Linux I still use primarily btrfs for the "zfs-like" use cases, as it ticks most of the boxes and doesn't risk module misalignment (which is a concept that is outright terrifying to me for a filesystem driver).
@Rtzq0 @roxmsauce I'm still seeing high exposure data loss with btrfs in QA environments so I won't touch it for, oh, let's say another year at least.
Yup, when people come to me and say "i really want ZFS" I say "you really want FreeBSD." :)
@wohali @roxmsauce I would *love* to hear about btrfs data loss experiences outside of the RAID 5/6 case. I keep hearing people say that it's common but I've yet to see it.
@Rtzq0 RAID 5 is a pretty common use case for what I do, so unfortunately, I can't give you that data point. Sorry :)
Couple of interesting pointers here. https://www.reddit.com/r/archlinux/comments/4f5xvh/saving_power_is_the_btrfs_dataloss_warning_still/
@wohali RAID56 on btrfs is still the one area of absolutely do not use so I'll agree with you there https://btrfs.wiki.kernel.org/index.php/Status
@Rtzq0 basically, we deploy a sharded DB where speed matters over redundancy, since the DB itself ensures at least 3 separate machines each hold a copy of the data stored.
So RAID5 and sometimes RAID0 are the weapons of choice.
@wohali the use case you are describing sounds similar-ish to most elasticsearch deployments. And yes in such cases I typically go RAID0 and use something else for my safety net. But a lot depends on what the database is *for* and the acceptable failure stories.
@Rtzq0 @roxmsauce Weird that someone on your link is complaining about fbsd's network stack not being good enough for high throughput/bw. I saturate multiple Infineon 40gb links with it regularly... *shrug*
@wohali @roxmsauce TCP mystifies many...at least that's my experience from performing interviews.
@wohali @roxmsauce hashtag unixisnotlinux *but*....
If you're from a Solaris background curious if you've spent any time with illumos? Also curious about your choice of BSDs if you're interested in sharing.