This board is for the discussion of technology and related topics. Reminder that instigating OR participating in flame/brand wars will result in a ban.Tech support threads should be posted to >>>/wsr/ Cryptocurrency discussion belongs on >>>/biz/ To use the Code tag, book-end your body of code with: [code] and [/code] The /g/ Wiki: https://wiki.installgentoo.com/
Satania uses an iPhone.Naruto uses a NinjaBook Pro.
>>100346266is her pussy lightning cable compatible
iNiggers use a designated shitting street
>>100350141Satania included?
There's lots of Macs in Death Note. Japan really loves Apple.
>>100350380Based L had a Quicksilver G4. Beautiful thing. Why can't Apple into aesthetics lately?
weird science editionprevious: >>100314443READ THE WIKI! & help by contributing:https://wiki.installgentoo.com/wiki/Home_server>NAS Case Guide. Feel free to add to it:https://wiki.installgentoo.com/wiki/Home_server/Case_guide/hsg/ is about learning and expanding your horizons. Know all about NAS? Learn virtualization. Spun up some VMs? Learn about networking by standing up a OPNsense/PFsense box and configuring some VLANs. There's always more to learn and chances to grow. Think you’re god-tier already? Setup OpenStack and report back.>What software should I run?Install Gentoo. Or whatever flavor of *nix is best for the job or most comfy for you. Jellyfin/Emby/Plex to replace Netflix, Nextcloud to replace Googlel, Ampache/Navidrome to replace Spotify, the list goes on. Look at the awesome self-hosted list and ask.Comment too long. Click here to view the full text.
I've been running a truenas scale box for about a year now as my first home server. I am starting to consider using a different option however because I am tried of going through the apps available through truecharts with how everything is fairly obfuscated and constant changes breaking things or forcing convulsed upgrade and migration paths. Going a few weeks without checking the truecharts blog page always results in a pain in the ass. My main use case is storing and serving all photos from mobile devices with Immich, torrenting, and hosting a music server. What would be some good options for to consider for these uses?
>>100348048here's the front, it's definitely old as fuck. from the date codes on the stickers it's maybe circa 1995? despite that it's in immaculate shape
>>100348048opened up. I took that door off to move it, that alone was like 40 lbs.
>>100350324>>100350348wow, that's pretty industrial. all the cable management seems to be on one side, so it's almost certainly from a datacenter that used all-IBM gear designed specifically for it, the racks were probably built-in, too.can't wait to see it filled
>>100348048a pic of the front with the door off while prepping to move it to the the basement. I did it with an appliance dolly, was super unsafe and I definitely should've had at least another person or two
Previous /sdg/ thread : >>100341895>Beginner UI local installFooocus: https://github.com/lllyasviel/fooocusEasyDiffusion: https://easydiffusion.github.io>Local installAutomatic1111: https://github.com/automatic1111/stable-diffusion-webuiComfyUI (Node-based): https://rentry.org/comfyuiAMD GPU: https://rentry.org/sdg-link#amd-gpuIntel GPU: https://rentry.org/sdg-link#intel-gpu>Use a VAE if your images look washed outhttps://rentry.org/sdvaeComment too long. Click here to view the full text.
Garbage output, but I kinda like the aesthetic.
>>100350303fun
>>100350298Let's see, with chrome open with some 50+ tabs I have about 14GB used up, then I load up comfyUI, which is only like 1GB. I have a bunch of scripts loaded onto it though, so maybe one of these is the culprit. Then when T5 loads I see 50 GB of usage but python is only 20GB, 20GB + 15GB = 35GB so I have no clue what's going on
Why did he do it?
If you've ever been laid off, it's for performance reasons, regardless of what management tells you. If you did your job better or the work you did was actually important, then you wouldn't get fired. People always try to rationalise it because they can't accept that they weren't doing enough to keep their jobs.
>>100346003he wants his new remuneration package.
>>100346441I can take my compiler and go elsewhere. Fuck your union it doesn't even make any sense when the workers already own the means of production.
>>100348037hes a nigger, literally south african
>>100348011You ain't gapping shit above 120-130mph in your RC car, fuckboy.
Dubs decide which nostalgiafag OS I'll install on this 86box VMRules:>no XP because 86Box is an emulator, not a hypervisor, ergo it's slow as shit
dubs and run magnets over your harddrives and leave /g/
/lmg/ - a general dedicated to the discussion and development of local language models.Previous threads: >>100342454 & >>100331862Awaiting phi-3-medium / no good l3 tunes edition►News>(05/06) Someone that actually does something good posted on /lmg/ >>100330552>(05/02) Nvidia releases Llama3-ChatQA-1.5, excels at QA & RAG: https://hf.co/collections/nvidia/chatqa-15-662ebbf6acc85f5c444029a8>(05/01) KAN: Kolmogorov-Arnold Networks: https://arxiv.org/abs/2404.19756>(05/01) Orthogonalized Llama-3-8b: https://hf.co/hjhj3168/Llama-3-8b-Orthogonalized-exl2>(04/27) Refusal in LLMs is mediated by a single direction: https://alignmentforum.org/posts/jGuXSZgv6qfdhMCuJ►News Archive: https://rentry.org/lmg-news-archive►FAQ: https://wikia.schneedc.comComment too long. Click here to view the full text.
>>100349210Take a break. Work on your movie or game backlog.
Am I the only one worried about the over-reliance on Huggingface to host datasets and models? It's become the go-to due to pure convenience, but one shift in the winds and they could start nuking all sorts of stuff. Take this dataset for example: https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicQAFinalThere is some truly heinous stuff in there. Very valuable for decensoring models (for some reason people are doing this orthogonal meme instead, but I digress...). This dataset's very existence is far worse than GPT-4chan ever was, which HF removed as soon as a few people started making a stink about that model.I just feel like so much stuff is only up because it's completely flying under the radar. And that could change at any moment. But what is the alternative to HF?
>>100350181I'm not quite seeing how you could repeatedly stick on a fp16 Rank32 LoRa + train and arrive with a sane thing in that scheme, but perhaps I misunderstood.
>>100350389it'll be fine
>>100349571Just to be clear, the idea here is to take the model in Q2 (or whatever), which has lots of error introduced by the quant, and add a higher-precision LoRA on top to correct the error. The LoRA is higher bpw than the base model, but has fewer weights, so its total size comes out to 10% or less of the base model.One thing you could try with this is quantizing the LoRA to something like Q6 or Q8, which is hopefully close enough to full precision that it will still work. But if you quantize the LoRA too aggressively, you'd most likely end up bringing back the same quantization error issues that the LoRA was supposed to correct. I suppose you could do something goofy like correct the Q2 base model with a small Q4 LoRA, then correct the Q4 LoRA with an even smaller Q6 LoRA, but first, this would only save you a tiny amount of bpw, and second, each LoRA in this setup adds performance overhead at inference time (unlike normal usage of LoRAs, you can't merge the LoRA into the base weights here because the base weights are quantized to a much lower precision than the LoRA).>>100350181No, the improved model, which consists of a Q2 base and an FP16 LoRA, was 16% smaller than the Q3 quant of the same model, while having similar performance in terms of KL divergence. It was of course larger than the original Q2 I started with. There are two Q2 dots on the graph - the top one is the baseline from just running ./quantize Q2_K, and the bottom one is the result after adding on the LoRA.
Users of all levels are welcome to ask questions about GNU/Linux and share their experiences.*** Please be civil, notice the "Friendly" in every Friendly GNU/Linux Thread ***Before asking for help, please check our list of resources.If you would like to try out GNU/Linux you can do one of the following:0) Install a GNU/Linux distribution of your choice in a Virtual Machine.1) Use a live image and to boot directly into the GNU/Linux distribution without installing anything.2) Dual boot the GNU/Linux distribution of your choice along with Windows or macOS.3) Go balls deep and replace everything with GNU/Linux.Resources: Please spend at least a minute to check a web search engine with your question.*Many free software projects have active mailing lists.Comment too long. Click here to view the full text.
>>100350211sounds like a missing fonts or bad gpu drivers issue 2bdesu
>>1003502548th gen intel laptop graphicsfully patched 24.04vscode installed by snap or upstream deb makes no difference
>>100349777It didn't work...What's funny is if I boot on the bleeding edge kernel ( Arch Linux) that's when that behavior is observed. However if I boot on LTS this doesn't happen. It's been like that from before I installed arch, ( Was runnin a distro called EndeavourOS) it also uses plasma kde and I pretty much mounted my old home exactly as it was. Something on the cache perhaps?
>>100350270weird, try booting a live USB with a different desktop environment
>>100350094The main differences between Linux distros are their package repos and their kernels. Ubuntu's whole deal is that it's an LTS distro that doesn't update packages as fast as some other distros do like Arch, Gentoo or Fedora. It's always been touted as the "beginner's distro" though which it very much was back in the 2000s/2010s, though over time a spinoff of it called Mint basically took over that moniker after the company behind Ubuntu started to advertise it more to enterprise business then the average user.
https://restoreprivacy.com/protonmail-discloses-user-data-leading-to-arrest-in-spain/
>highlighting the complex interplaynice ai slop
>>100350259Is the only alternative Tuta? No way i'm going to waste time with all the kruft that goes along with antiquated email protocols. Even iRedmail as a turnkey solution sucks.
>>100350281The only viable solution is to not use email for sensitive communications in the first place. It's not a secure protocol and never will be. If you absolutely need to do so, use PGP.
>>100350259*mild shock*../ \_/ \( * o *)
>>100350259It's nothing-burger like last time.
just got one of these for $8, how do i get video on this? just antenna, no input ports
>>100350206>broadcasthow, what do i buy
>>100350286literally just touch the coax to the antenna?
Host your own radio TV stations. Blonder tongue usually makes the big boy converters that will convert from VGA to RF. From there you can just plug a coax antenna into the RF output and it will broadcast on whatever station it comes with or that you configure yourself, tune to that on the box and you're good.
>>100350336got no idea on products. but found this, should be simple or there should be some ebay cheap board that does it. check your tv's receiving bands/frequencies https://www.circuits-diy.com/tv-video-transmitter-circuit-vhf-uhf/
>>100350246i recently read a cool blog post about some ham autist relaying tv signals from mongolia using a cold war era russian satellite, so thats how I know
mpv ffmpeg yt-dlp
>>100348582read about him you slop gobbling gorilla nigger retard, he was based90% of his music was satire making fun of the le npcs, leftoid cia hippies, gubbamint and glowies. he also did many good stravinskian classical compositionsyou dismissed it because it isn't accesible goyslop and sounds le hecking weird XDkyshttps://www.youtube.com/watch?v=dLuMPS-FGCo>>100348682i'm also a 0.29.1 schizo btw
does he glow?
>>100348910glows orange
mpv pro tip:dont forget to set dither-depth to no if you have a system that has proper compositor dithering. you are going to be applying dithering twice otherwise and it will result in unnecessary filtering!
unused filtering is wasted filtering
why does nobody talk about this? it's far and away the simplets/easiest database i've ever worked with, and the "design document" api for views/indexing is incredibly powerful.
RethinkDB made it obsolete
last time i heard of anyone using couchdb, obama was still president
>>100348764Why use anything when Mariadb just works
>>100349199why'd they stop?
>Huawei’s profits have surged more than six-fold after the Chinese technology champion overcame Western sanctions to steal market share from Apple.>The Shenzhen-based company reported a 19.6bn yuan (£2.1bn) profit in the three months to March, up 564pc on the same period a year ago.>This comes despite facing US export controls and bans on overseas sales of its networking equipment in the last year, including in the UK.The company released a new high-end smartphone in August that was made largely with home-grown microchips.
How do I get a Huawei in North America? I miss it, the only good android phone. This samshit piece of shit sucks.
What do you even need 1440p on a phone for? Why go beyond 1080p, which is more than enough, and sacrifice so much battery life? My phone has a res of 720p and looks fine.
In b4 >poorfag cope
>>100350283consooooming has destroyed the world of technology and bloated everything beyond a reasonable degree. Worse yet These consoooooooooooooooomers beg for their slavery and bark like dogs at the latest hardware. I was born in a horrible time period.
Every phone is fucking huge these days as well.I'd love for an iPhone 5 with an edge to edge display.
>>100350283Eh, I can physically see the difference between 1080p and 1440p with my eyes. The 1440p one looks nicer, and it's as simple as that for me.
>>100350335This OP. I mean just look at this thread, motherfuckers complaining about the graininess of 1080p on a 6 inch display lolhttps://www.reddit.com/r/Android/comments/klhbl3/1080p_displays_are_objectively_better_than_1440p/#:~:text=It's%20well%20documented%20that%201440p,for%20UI%20and%20game%20performance.>plebbitIt is more representative of the average smartphone user than this board is
MMCX Golden age EditionHow to request advice:>Budget>Intended use (media, source, environment)>Frequency response preference (basshead, female vocal, treble sensitive, etc.)>Past gear and your thoughts on themFAQ:>Where do I buy IEMs?Amazon, Aliexpress, Linsoul, Hifigo, Shenzhenaudio, Bloom Audio, MusicTeck, Elise Audio (UK)>Full Guide (IEMs, Cables, Ear Tips, etc.):https://rentry.org/c8gqcComment too long. Click here to view the full text.
>>100349073>katojesu->miadPFFFFFTBWHAHAHAHAHAHAHA TAKE A LOOK AT THIS FUCKING RETARD HAHAHA HOLY FUCKING SHIT PFFT HAHAHAHHAA
i just unearthed aria 2021s and starfields from my drawer of "random shit i'll sort out one day (never)" and i'm surprised by how fucking dead the arias sound in comparison, is that normal? they might sound worse than my shitty samshit earbuds
>>100349687Kiwi Ear Cadenza
>>100342700Than recommend something better, newer and in the same price range you worthless nigger
>>100350350check the OP and rentry you illiterate retard
