BakedCatboy @ BakedCatboy @lemmy.ml

Posts

1

Comments

360

Joined

2 yr. ago

Sadly my password stopped working like 10 years ago and I haven't gotten a response on the forums :(

TL;DW: Small sample size aside, it looks like the brick layers could be marginally stronger in some ways, but it can be weaker if you don't also increase the extrusion multiplier since offsetting a column of circles down by half a row in a grid of touching circles will make the circles not touch (hopefully that's an intuitive explanation of the need to increase the extrusion multiplier).

During a boil water advisory that lasted more than a week we got a water drop brand RO system (the G3P800). It doesn't have a tank but it instead uses a pump to do RO under a higher pressure so it outputs water fast enough to fill up a cup on demand, and a cooking pot if you're not impatient. It came with a UV sterilizer, remineralization was an addon so we skipped it, especially since I prefer the taste that way and because I also use the RO water in our humidifiers and espresso machine where scale is a hassle. I'm a little biased because they accidentally refunded me so I got it for free, but it's worked perfectly since we got it, and everything needed to install it was included (we just separately got a drain line adapter that lets us connect the waste line to the dishwasher drain so we could avoid modifying anything in our rental as the included adapter wants you to drill into the drain pipe). I have several TDS meters and the faucet's included readout matches at usually 2-5, and I've used the water in my humidifiers 24/7 for months and still haven't accumulated more than a couple days worth of regular water scale in any of them.

Oh and it also has a 3:1 tap to drain ratio which is nice.

I used to have a simple 3 stage under sink filter, so unfortunately I can't really compare to other RO systems.

I'm pretty sure that's exactly the problem, the package name is the same but it likely has a different signature. If the package name was different then it would let you install both side by side even if signed by different keys.

I don't really have a recommendation atm, I used to use mullvad but for torrenting I feel like the lack of port forwarding (once they removed that feature) was hurting my ability to seed so I switched to proton. I also recently added Usenet into my mix and since many providers bundle a VPN subscription - and mine in particular supposedly also supports port forwarding (usenetdirect bundles a ghost path VPN subscription), I'm gonna try to get it to work with that so I don't have to pay for a VPN separately but I haven't tried it yet.

Sounds like their strategy is to force US companies to block access to piracy sites.

I already run my torrent client through a non-US VPN so this can literally be bypassed by adding this to my prowlarr docker compose:

network_mode: service:gluetun

Chat rooms were released in 2020 fyi

There's a ticket in the Firefox bug tracker that appears to be tracking this: https://bugzilla.mozilla.org/show_bug.cgi?id=1913601

Makes sense, because even in reader mode it shows https, and the issue happens on other major sites like Wikipedia.

That seems kind of like pointing to reverse engineering communities and saying that binaries are the preferred format because of how much they can do. Sure you can modify finished models a lot, but what you can do with just pre trained weights vs being able to replicate the final training or changing training parameters is just an entirely different beast.

There's a reason why the OSI stipulates that code and parameters used to train is considered part of the "source" that should be released in order to count as an open source model.

You're free to disagree with me and the OSI though, it's not like there's 1 true authority on what open source means. If a game that is highly modifiable and moddable despite the source code not being available counts as open source to you because there are entire communities successfully modding it, then all the more power to you.

It's worth noting that OpenR1 have themselves said that DeepSeek didn't release any code for training the models, nor any of the crucial hyperparameters used. So even if you did have suitable training data, you wouldn't be able to replicate it without re-discovering what they did.

OSI specifically makes a carve-out that allows models to be considered "open source" under their open source AI definition without providing the training data, so when it comes to AI, open source is really about providing the code that kicks off training, checkpoints if used, and details about training data curation so that a comparable dataset can be compiled for replicating the results.

It really comes down to this part of the "Open Source" definition:

The source code [released] must be the preferred form in which a programmer would modify the program

A compiled binary is not the format in which a programmer would prefer to modify the program - it's much preferred to have the text file which you can edit in a text editor. Just because it's possible to reverse engineer the binary and make changes by patching bytes doesn't make it count. Any programmer would much rather have the source file instead.

Similarly, the released weights of an AI model are not easy to modify, and are not the "preferred format" that the internal programmers use to make changes to the AI mode. They typically are making changes to the code that does the training and making changes to the training dataset. So for the purpose of calling an AI "open source", the training code and data used to produce the weights are considered the "preferred format", and is what needs to be released for it to really be open source. Internal engineers also typically use training checkpoints, so that they can roll back the model and redo some of the later training steps without redoing all training from the beginning - this is also considered part of the preferred format if it's used.

OpenR1, which is attempting to recreate R1, notes: No training code was released by DeepSeek, so it is unknown which hyperparameters work best and how they differ across different model families and scales.

I would call "open weights" models actually just "self hostable" models instead of open source.

I think it's normal to see some variation, I have an official Google replacement battery from ifixit, the marked capacity on the battery is 4050, in battery settings it says the design capacity is 4180, and my current capacity is 4042. Also accubattery says the design capacity is 4000.

If you're worried I would use something like accubattery and let it take measurements for a week (while trying to discharge down to a few percent a couple times and also charging uninterrupted to 100%) and then see if the estimated battery capacity measures up to what you expect. If you get close to or above 4000 (would make it about 90%) then I think it should be fine. If it's much less, then think about having it replaced.

No problem! You can tell I went deep down the rabbithole a while back lol - I had to rip my dad's CD collection and assure him that what came out of the toslink to his DAC was identical coming from a FLAC as would come from a CD player with optical out.

I'm pretty sure if you rip CDs directly to FLAC, it's a perfect copy assuming you're using good software. PCM isn't lossy or lossless because it's not a compressed format, it's an uncompressed bitstream. Think of it like the original data. If it was burned to a CD as digital MP3 data and then ripped that to FLAC, then yes you'd be going from lossy compressed to lossless, which would hide the fact that quality was lost when it went to MP3 in the first place.

Just as an example, you can rip a CD directly to FLAC (you should also find and use the correct sample offset for your CD drive), rip the cue sheet for track alignment, then burn the FLAC back to a new CD using the cuesheet (and the correct write offset configuration), and you'll get a CD with the exact bit for bit pattern of "pits" burned into the data layer.

You can then rip both CDs to a raw uncompressed wav file (wav is basically just a container for PCM data) and then you'll be able to MD5sum both wav files and see that they are identical.

This is how I test my FLAC rips to make sure I'm preserving everything. This is also how CD checksum databases (like CDDB) work - people across the globe can rip to wav or flac and because it's the same master of the CD, they'll get identical checksums, and even after converting the PCM/wav into a flac you are still able to checksum and verify it's identical bit for bit.

I basically don't notice that I don't have a headphone jack. My usb-c adapter is just permanently affixed to my wired IEMs and it basically makes no difference to me if the plug is round or usb-c shaped.

I definitely recommend biting the bullet and getting a good adapter. Since I have a pixel I use the Google one. I made sure my partner got an official apple one for their iPhone since I remember seeing rumors about a volume difference between them if mixed and matched. Aside from Apple shenanigans I haven't really had an issue with them. I also only charge at night so I never have the problem of needing to charge and listen at the same time.

I think they're making it more complicated than it needs to be. On any other social media site, you find people by their username. So just ask for your friends username (username@instance.com) and put it in the search bar and it'll come up. Using the URL can be convenient on desktop because you can just copy and paste it from the address bar when you're looking at someone's profile.

And if you want to discover new people where you don't already know their username, then I believe that is the same as any other social media as well, you can come across them in the comments of people you follow or go to the discover tab or search hashtags and you'll find new people that you can tap on and follow.

I feel like this basically covers how you would find people. A lot of people get hung up on how you know what instance other people are on but it doesn't usually matter. Either someone will give you their username which includes @instance.com, or if you don't know the instance you can search for their name and all known accounts with that name will show up.

For example if I just search my username "BakedCatboy" (not my real username), the search results show both my mastodon and Pixelfed accounts.

This is great to hear, dansup has been stretched across so many projects for so long so hopefully this really supercharges pixelfed (and friends) development and helps to increase the diversity of AP servers.

It was waitlist for a while, not sure if it still is but I got my welcome email like a week later.

Sounds likely, I haven't used port forwarding with my VPN since Mullvad stopped supporting it, so when I recently shared my own torrent I paid for 1 month of a seedbox just to make sure it seeds well and the seedbox uploaded ~50GB while my local setup on a VPN without port forwarding only uploaded 1.8GB (and it hardly showed any peers as if nobody was trying to download). So it seems peers had a much easier time connecting to the seedbox.

I have since setup port forwarding in gluetun for my local torrent client. I just wish there was more support for it because gluetun only has built in support for port forwarding for 2 providers (I guess automated requesting a forwarded port), and even then you still have to make your own script to automatically set the port in the torrent client when it's assigned / changed. It's possible that some providers do it more like Mullvad where you get assigned a port via the website that is tied to the VPN credentials, so you just have to plug the assigned port into the torrent client settings (that's how it worked with Mullvad so I could just enter the port once and forget about it) but I haven't checked other providers to see.