site banner
ZorbaTHut  1yr ago (text post)   3245 thread views

Friday Fun Thread for September 30, 2022

Be advised; this thread is not for serious in depth discussion of weighty topics, this thread is not for anything Culture War related. This thread is for Fun. You got jokes? Share 'em. You got silly questions? Ask 'em.

4
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Something really fascinating, at least to me: I was walking in front of a bee hive doing evening orientation flights, where workers take their first flights outside the hive to learn navigation landmarks. Even to my stuffed-up nose there was a strong lemony smell of the Nasonov beacon pheromone they use when swarming or when you've housed or moved the hive, with a few bees standing in the entrance fanning it out to guide returning trainees who might have gotten lost.

Been working with them for years, but I keep being amazed by how coordinated everything they do is. I've never smelled them using it except when swarming, but in retrospect of course they'd use it like this too.

Apparently people can only detect a few of the stronger pheromones, and some can't even smell those. It's nice to be able to smell them getting angry just before they all pour out and attack (tangy banana peel means "check your suit seals"), but it would be amazing to experience the subtler ones too.

Honestly, yeah, that's out of left field and I never thought about it, but it is fascinating.

I would post this in my town's subreddit but anything to the right of Bernie/AOC is down voted to oblivion.

During the election this fall, Oregonians will be presented with ballot measure 114.

A "yes" vote supports this ballot initiative to:

  • require permits issued by local law enforcement to buy a firearm;
  • require photo ID, fingerprints, safety training, criminal background check, and fee payment to apply for a permit; and
  • prohibit manufacturing, importing, purchasing, selling, possessing, using, or transferring ammunition magazines capable of holding more than 10 rounds and make violations a class A misdemeanor.

Signatures were collected for this immediately after the Uvalde shooting.

I find this measure rather offensively misguided. I don't own a gun, though if I needed one it would probably be because I'm feeling very threatened. The last thing I want to deal with in those circumstances is a hulking and dumb nanny state demanding that I prove competency to their standards and can impress upon local police that I deserve to have one.

This law is misguided because it actually would have failed to stop the Uvalde shooter from getting a gun unless it was so overly broad that the police had discretion to deny anyone for any reason.

I'm not sure how its proponents expect this to work, unless they would deny him simply on the basis of appearing obviously poor, Hispanic and not fitting in at school?

The people advancing this law almost certainly believe, e.g. that the police implement systemized racism but that black people should be allowed to own guns. What makes them think police won't blanket deny guns to all black people, then?

Additionally, I think people who support these laws are genuinely failing to understand that guns are the only thing that reliably prevent the rest of us from being at the mercy of the strongest and most dangerous people.

There has been a noticeable rise in home intrusions in this town lately where women awaken to deranged men in their houses. Usually they're homeless meth addicts who are distracted enough by the woman waking up and screaming at them that they leave. But there was a case recently where a woman had acid thrown on her, then again where she woke up to the same guy setting her on fire. The police treated her like she was crazy and making it up until the media elevated the story.

If my daughter lived alone I would 100% want her to have a loaded handgun by her bedside. It seems outrageous that she'd have to prove to the police that she's worthy of keeping this gun, when the people breaking into houses don't have to meet any such standards.

Again, I don't own a gun, but the last thing I want to do, in the society we live in, where police take 20 minutes to respond (they didn't used to, they stopped responding quickly after 2020 for some reason) is make it more difficult for law abiding people to arm themselves.

Is this too "boo outgroup"? I feel like these are strong racial justice and feminist arguments for not making it difficult to get and keep handguns.

"getting a gun unless it was so overly broad that the police had discretion to deny anyone for any reason."

There's your answer. Selective enforcement.

Post in your town's sub and organize a bunch of your friends to brigade it. They win on- and off-line by enforcing the appearance of consensus and conformity.

I'm just wondering how this proposed law doesn't run afoul of the second amendment. As far as I'm aware, the argument for existing gun control measures (w.r.t the second amendment) is that they aren't really making it so you can't own a firearm, they simply make it harder to obtain certain types or introduce hoops to jump through. I don't really agree with that argument, but that's my understanding. This measure, though, would make it so some people (however small a group) straight up cannot have a firearm, period.

I'm not a lawyer, so maybe I'm missing some nuance. But a law which makes it so some people cannot own a firearm legally sure seems like it will get struck down on second amendment grounds.

Edit: I didn't notice what thread I was in at first, I think you want to post this in the culture war thread and not the fun thread.

Felons cant own guns.

Oh whoops I'm dumb. Will move this when I'm not on mobile.

This is an interesting topic but I don't think it belongs in the Friday Fun Thread (but it definitely belongs in the next Culture War Roundup).

What entrance music should I imagine in my head when reading your comments on here?

https://youtube.com/watch?v=6W8URO1sM9k

I read everyone’s comments in an upper-middle class general American accent even when I know what country you’re from

https://youtube.com/watch?v=ZIniljT5lJI

(The bleeped version is way better, but this one is mine)

https://youtube.com/watch?v=YeM2e8-PtAQ

https://youtube.com/watch?v=BzNzgsAE4F0

https://youtube.com/watch?v=C06jAb-e_iU

https://youtube.com/watch?v=1CIimgUq7ic

https://youtube.com/watch?v=B9MShtCg4fk

https://youtube.com/watch?v=SMhwddNQSWQ

https://youtube.com/watch?v=NDSYJATGXEQ

https://youtube.com/watch?v=5txHGxJRwtQ

You should not do that.

Has anyone else noticed that podcasts are such crap these days? so many ads. It's ridiculous. Not just YouTube. Any format. Ads for first 3 minutes , random ads in the middle, more ads. The entire cloud/SAS architecture is dependent on podcast ads it seem.

Which podcasts? The only ones I listen to are the wsj editorial podcast, manifold podcast and macro voices podcast none of which seems to have that many adds.

Macro voices probably has the most valuable users but they have empty space right now. I think they’re picky though

darknetdiaries . com , tim ferris (really bad) anything biz or tech related

No, I haven't noticed, on account of not listening to any podcasts.

No, haven't noticed. I tend to listen to comedy podcasts, and many of them seem to have trouble keeping big name advertisers.

probably because comedy is a not a high ROI niche. Anything biz, tech, entrepreneur related probably has no problem filling inventory

Apparently no advertiser other than boner pills or online casinos will touch Cumtown. They are add free, other than the in-show boner pill endorsements.

Five of my favorite podcasts have zero ads and are consistently high quality: Econtalk, the Fifth Column, Blocked and Reported, the Glenn Show, and Chapo Trap House.

No.

Podcasts on youtube don't even have ads because all the good ones are demonetised.

Isn't part of it the current state of the economy? Even YouTube and Twitch are flirting with more ads.

Google was a cosmic mistake.

I usually use Youtube via a heavily uBlocked and uMatrixed Firefox, which means no ads.

Whenever my wife asks me to play a song via the Youtube program on my phone, I almost drop the thing in fright - people actually listen to music or watch videos that are preceded by several advertisements, plus interruptions, plus another round of ads in between videos? It's intolerable. I'd rather not.

Firefox mobile can install adblock addons & play videos in the background.

Can't do the skipping-sponsored-content afaik

SponsorBlock is an extra Firefox addon I found recently that automatically skips over the "first a quick message about our sponsor" sections in Youtube videos.

Use YouTube Vanced for android. No ads, skips over sponsored content and "ring that bell" spiels for any decently sized channel. Can also skip intro music.

The only downside is being the last person in the world to hear important news about RAID: SHADOW LEGENDS!!

I have just seen someone watch (quite focussed) an entire minute of woke sportswear ads before a song started on their youtube tab in the library. If this is how the average person reacts to ads we are all so screwed

Nice, will try!

Edit: Can't find it.

It's not in Google's app store, you have to install the APK. Hilariously there are YouTube videos explaining how that haven't been banned.

It appears to have been killed a few months back by Google making legal threats. Which isn't too surprising I guess, but is a shame.

Working fine for me. Until Google changes the API to fuck it up I guess.

Well sure, but for new users that isn't much help. :P

Can't you still just DL the APK file? Nothing's changed about how it works, it's just no longer getting updates when Google periodically changes stuff to fuck with people's quality of life extensions.

More comments

I haven't listened for many podcasts lately but various new streaming subscription services are equally garbage. Why can't I watch an episode of Star Trek without being interrupted in the middle 8 times for several minutes to listen in the 20th time for the same moronic ad? Not enough that they want me to subscribe to a dozen of different ones, they also insist on shoving ads down my throat to milk my wallet a bit more.

That just makes me to unfold my old Jolly Roger flag that I hoped to retire for good one day, because there's a limit of how much abuse one can take while paying money for it.

What streaming service has ads?

Hulu, prime, many small shitty ones.

The ads on streaming services infuriate me. It's like, motherfucker I'm paying you. How the hell are you going to turn around and show me ads? If not for my wife (who unfortunately doesn't care as much as I do), I would immediately cancel any service that does that. But, you gotta pick your battles in marriage and this one just isn't worth it.

I’m fascinated by the female propensity to watch ads, to the extent that nearly half of all commercials during sporting events are aimed toward women. What’s going on there?

It's a trope as old as cable television, and a true one at that. Jerry Seinfeld joked about it back in the '90s: Men hunt and women nest. I remember my mum constantly complaining that my dad would change the channel any time a commercial came on, her concern being that she'd miss part of the program if he stayed away too long; my dad always countered that he knew the length of the breaks pretty much by heart at that point so no one would miss anything. A friend's mum complained that she'd be watching a show with her husband and would ask what happened to a specific character, only to be informed that that character was in a different show airing on another channel.

With sporting events this tendency is magnified because most sporting events are aired weekend afternoons and evenings, so there's plenty of other sports programming running counter to whatever you're watching, and the temptation is too great to take a quick look at another game. This tendency is magnified further if you're not watching any particular game and just flipping through channels to find something interesting. In my experience, women are more likely to watch a sporting event only if it's a local team or other event that they have a specific interest in, and while men will do the same, they're also more likely to watch a game because it's on television, so there's no loyalty to a specific broadcast. Networks base ad fees on total viewers, but the advertisers know that a certain percentage of viewers are casual fans who aren't going to watch any commercials, period, unless absolutely forced to. So the advertisers are going to tailor their ads to the demographic that's least likely to skip around.

I would also ad that the NFL seems to be the biggest culprit of this (though I don't watch NBA games). I see less of this in college football and national NHL and MLB broadcasts, and practically none in local hockey and baseball broadcasts or PGA and NASCAR. And the female-centered ads during PGA events, rare as they are, tend to be for women's-specific golf products, which makes sense. This seems to make sense—there are only 17 regular season NFL games a year per team, and they're almost all on weekends, so following the sport is much easier for the casual fan. Baseball has like 150 games on television and hockey 82, and some of those games have rather late start times if you're on the East Coast, so it's hard to keep track of things closely unless you're really dedicated. Golf and NASCAR don't have regionally-based fanbases so unless there's a big national star like Tiger Woods it's hard to even know who you're watching, and golf is particularly bad on this front because the field is huge, the events start during the day on Thursday, and most golfers only participate in select events, so by the time Saturday rolls around the leaders could be all guys who are well-known in the golf world but who most people have never heard of. College is better on this front but most college teams don't generate a lot of national interest.

At least that's my theory; it could be totally wrong.

It's the terminal state in the eventual capital-ization of everything. No one can do anything for pleasure anymore and share openly, all activities eventually must turn into a money generating scheme or they're ultimately not productive.

The open source software world feels like a peek into an alternate dimension of what the world could've been like: in few other places do people share so much toil and effort so openly with others.

What is your favorite ridiculous working story to tell? Personally my favorite is when I was working at a subway during college and tripped acid during a shift. People gave me a bunch of strange looks but I think I did decently well, although I did piss off one lady by putting tuna in a sub that was not supposed to have tuna...

Any other fun stories?

Only somewhat funny but I got dismissed from temp work for suggesting that maybe specialising would make some sense.

We were refurbishing old coke vending machines, mostly fairly routine work of cleaning and swapping broken parts, but us temps were expected to nicely spray paint them too. The employees who have been doing it for a long time found it easier. It's not trivial to get the coat right without bubbles and applied evenly.

Suggested this to the boss, was told to not come back tomorrow.

How many different instruments/parts do you prefer to hear in a piece of music?

I am inclined to feel that anything in excess of three, or maybe four, is overkill/bloat. But maybe I just listen to too much chiptune/MIDI stuff.

According to Wikipedia, Papa Was A Rolling Stone, a great piece of music, has far more: " A solo plucked bass guitar part, backed by hi-hat cymbals drumming, establishes the musical theme, a simple three-note figure; the bass is gradually joined by other instruments, including a blues guitar, wah-wah guitar, electric piano, handclaps, strings and solo trumpet; all are tied together by the ever-present bass guitar line and repeating hi-hat rhythm."

I can't even tell how many instruments are in a piece of music.

But maybe I just listen to too much chiptune/MIDI stuff.

Ironically(?), I think some of the best kind of "programmed music" is the 16+-channel tracker music (example). Some tracks, you have to pay attention to the complexity of the composition, and thankfully, most videos are recorded in tracker programs so that you can see each "note" scroll by.

MIDI and 8-bit chiptune, though, I can see where there are hard limits (though I think MIDIs could be elevated by the ability to use more instruments/channels).

MIDI and 8-bit chiptune, though, I can see where there are hard limits (though I think MIDIs could be elevated by the ability to use more instruments/channels).

Apparently you've never encountered the genre of Black MIDI.

I thought about mentioning that, but didn't. I figured that was obviously right out.

What is a part? Is a chord on a guitar six or one parts?

Is a choir one part, 4-8 or ~8-48?

If I have an orchestra using a bunch of different instruments to produce a single chord is that one or multiple parts?

Is the guitar and bass guitar different or the same part?

Maybe I should differentiate more clearly between overkill and bloat. When I listen to a piece of music that has overkill, I wonder why there are so many different sounds playing at once, when it would sound less confusing/overwhelming/bad to have just three or four sounds. When I listen to a piece of music that is bloated, I wonder why there are so many different sounds playing (whether simultaneously or in sequence), when it would have been so much simpler/cheaper for the composer to use just three or four sounds.

What is a part? Is a chord on a guitar six or one parts?

I guess that's one spot where the synthesized instrument and the physical instrument can diverge significantly in bloat, even though both compositions have exactly the same position on the overkill meter. Many physical instruments, of course, can create a chord standing alone—but an electronic tracker may require one synthesized "instrument" for each note in a chord.

Is a choir one part, 4–8, or ~8–48?

Similarly, a choir would (I assume) be around four instruments in terms of overkill regardless of its physical or electronic implementation, but a physical implementation could be horribly bloated, with dozens of members beyond the minimum.

If I have an orchestra using a bunch of different instruments to produce a single chord is that one or multiple parts?

I would say that, if an entire physical orchestra is being used to generate a single chord, it's horrendous bloat, but not necessarily overkill. MIDI, of course, has the "Orchestra Hit" instrument that can be used to eliminate the bloat.

Is the guitar and bass guitar different or the same part?

I don't know much about musical composition, but if they're playing the same note then I imagine I probably wouldn't be able to tell that there were two instruments. However, if they're playing the "same" note at different octaves, then I might be able to tell, and to wonder what the point is.

Three or four instruments in one entire piece or three or four instruments going at the same time? These are two very different things.

Personally, I would say it depends on the piece of music and how complex each of the individual parts are. A track can sound very, very dense and disorienting with quite little if the sequencing and/or sound design is sufficiently detailed. Here's an example.

Three or four instruments in one entire piece or three or four instruments going at the same time?

The latter—though, on the other hand, the former can be considered bloat (as opposed to overkill) if they're real physical instruments with human players who are just sitting around doing nothing when the score doesn't call for them, rather than a single computer player swapping between instruments instantly.

the former can be considered bloat (as opposed to overkill) if they're real physical instruments with human players who are just sitting around doing nothing when the score doesn't call for them, rather than a single computer player swapping between instruments instantly.

I suppose by that standard you would consider most classical orchestral music bloated beyond belief, then?

Yes.

I like ska, jazz and orchestral music so I don't think there's really a limit for me.

Is https://youtube.com/watch?v=BtyWhHwCXIU too much or close to right?

Is https://youtube.com/watch?v=BtyWhHwCXIU too much or close to right?

Well, that's three or four different parts, right? Guitar, percussion, and one or two synthesized parts (I guess—I'm not an expert at decomposing music). The tune sounds a little generic to me (I have no knowledge of the Scott Pilgrim franchise, so I have no feelings associated with this video game), but I don't find the instruments objectionable.

What are y’all listening to?

I just discovered Lunar Society and I was blown away. The guest list is great but I’m most impressed by Dwarkesh’s ability to think through a statement and find something to push back. Very highly recommended.

I recently discovered Carpenter Brut and Perturbator. It never occurred to me that synth music could be good, much less seriously rock.

Roller mobster goes way too hard

Music: The latest album by DIM (or at least the latest I stumbled across): https://youtube.com/watch?v=CBtL8ogXZcw&t=1466s

Not sure if it's their best, I think nothing will top Compendium for me, but I still like it.

And if you meant audiobooks: An open source reading of Moby Dick from Librivox. Librivox is hit and miss in terms of quality, but this one is honestly pretty good. The reader is quite skilled and obviously had fun doing it. Puts a lot of variety into the different characters.

Podcasts: None, I hate the format. Why people like second-hand conversations is beyond me.

New DIM album, great news.

I just finished listening to The Stand by Stephen King on Audible yesterday. Still not sure how much I like the climax and ending, but at least in terms of the characters and intrigue, along with the quality of the prose, I thought it was easily as good as everything I'd heard of it (which, TBH, wasn't much beyond that it was fantastic). It tickled me to learn that a miniseries adaptation of it had come out in 2020, given that the plot is about a superflu that nearly wipes out all of America and what happens to people both during and immediately after it runs its course. Even more amusing was learning that Amber Heard was in it, playing Nadine Cross, who was a less than fully mentally stable character in the book, to say the least.

There was some [previous discussion], more culture war focused, on AI art and specifically AI pornography. 17 days ago, furry-specialized models were "currently a WIP and will be available soon".

The Yiffy specialized model has reached Epoch18 last night, following hot on the tail (hur hur) of Epoch13 on September 25th and Epoch15 a couple days later. While it's not quite up to my test case yet (or I'm not a good enough promptomancer to get it there), it's made a huge amount of progress toward it. And while I can't speak for Primaprimaprima's test of "a single high-quality AI image of two people having sex", there's absolutely the opportunity to generate images of two furries having sex, now. While it does PoV shots more easily, people have already found a few prompts that pretty consistently get common positions or even some kinks like exhibitionism going.

((Separately, we have separate Doe Biden and Buck Breaking jokes from Trace Woodgrains. Not sure if Trace was using Yiffy, or the less-porn-trained Furry model.))

Some somewhat surprising revelations:

  • Furry models seem to be doing better about anatomy like hands than conventional StableDiffusion. Which is kinda funny when someone wants paws, but potentially useful. Still not great, though, and probably only because the source images have such a restrained number of poses.

  • It's actually somewhat useful to train and tag for things you don't want. The Automatic1111 WebUI has the option of negative prompting. For adult content use, that can be useful for avoiding orientations or genders or other content you're not interested in. But that also useful if you don't want adult content at all; not only can you find the opposite of dicks, you can find the opposite of "bad anatomy". Which isn't necessarily going to make an output good, but it does point to some interesting options.

  • Albeit at the cost that you've probably trained for things that you don't like. Both the Furry and Yiffy checkpoints were trained against datasets filtered both on quality (albeit by simple upvotes), but also by content, for a variety of very good reasons.

  • Hit rates are either not great or outstanding, depending on what perspective you're looking at. Some more simple 'pinup' style prompts have gotten 30%+ as what their creators consider 'acceptable', but more complicated prompts can be ~10%, or even never produce good results at the first pass.

  • Furries have, perhaps understandably, focused on the use of furry artists for style prompting, but you can get somewhat surprising results looking at things in unexpected ways. Furry porn by DaVinci ends up looking pretty cool! I've had better luck getting SFW noodly cartoon people from prompts involving braeburned (cw: gay) and zachary911 (cw: gaaaaaay) than prompts involving Rick Griffin, who's pretty much the king of that field. A number of non-furry artstation artists (eg Greg Rutkowski, Michael & Inessa Garmash, Ruan Jia, Pino Daeni) can augment the style of prompt that's already got furry artists included.

  • Prompt and name collision seem to be an issue. Perhaps moreso in the furry fandom than elsewhere, but I do think it's going to point to some general issues with the tokenizer. I'm not sure if this is an issue from the scale of the data, or if it's one of many wider problems in the CLIP tokenizer.

  • This isn't very advanced. There's some fantastic work happening in the field, but the Automatic1111 webui also is missing unit tests and breaks functionality every other commit. The Yiffy model was trained on the word 'explict', because typos. It's not unusual to develop a prompt's settings by dartboard.

  • It's still very involved. At the extreme end, there's people who have entire workflows of inpainting and outpainting to correct defects like hands and eyes, follow by resolution enhancements, followed by resolution enhancements. But even well before that, it's tricky to dial in the right denoising settings. But with the exception of that last photoshop touchup phase, it's far from clear that these could not, themselves, be automated, and even that much of that automation would be new technology rather than slapping together existing bits. Indeed, automating the underlying 'does this image look /right/' step was a major part of the filtering of the LAOIN dataset used to train StableDiffusion to begin with.

  • It's also a surprisingly small training dataset. Yiffy trained starting on 150k images, moving to 200k for later epochs. A different model was separately trained on latex, rubber, and 'goo' with 100k images and, while I've not experimented with it or part of its audience, seems to be fairly successful. Many of the very useful tagged styles have less than 500 pieces in the training data: this (cw: topless fox guy in a loincloth, probably nsfw, but nothing 'showing') compares the relative effects of artists with 1400 (ruaidri), 230 (snowskau), and 47 (garnetto) works in the training data. That doesn't necessarily say something about training size floors, and it's possible that the terms are coming from previous training from the original LAION data, but it does suggest ceilings.

It is very impressive how quickly the furry model has progressed, and it’s much better at showing interactions between two characters than any of the other models.

As people predicted, it does look like making a honest living drawing smut or just pinup style stuff is going to get far harder.

Stable diffusion can provide some pretty solid results.. Prompt was, I believe, viction style paintings of big anime boobs.

Overall, while it's impressive, as someone whose life is significantly worse due to a porn habit, I'm not happy about any of it. This is just going to make more affordable & widespread superstimuli.

To some extent, perhaps. I left an nVidia 2070 crunching on "(by hioshiru and kenket and dimwitdog, Michael & Inessa Garmash, Ruan Jia, Pino Daeni), solo, anthro, male, ((wolf)), ((intricate detail fluffy fur texture)), mane, gray fur, pinup, ((yellow eyes)), chest fluff, nsfw, nude, digitigrade, (tail), handsome, cinematic" overnight, and in the morning had 5 GB of hot (if not great-resolution) werewolf dude. And there's pretty clearly been a handful of people on the Discord spinning hours of their attention on such.

On the other hand... if you had the bandwidth, you could have gotten a terabyte of furry porn anyway. The specific combo of "solo_male, canine, score:>10" caps e621's number of pages (meaning literally tens of thousands of results). Before the tumblr pornocalypse, you could have a well-filtered stream of novel content faster than the server would successfully refresh. And they're much more superstimuli, and I mention them because I expect they're among those least likely to appeal.

((And... uh, I like buff naked werewolf dudes. A lot. But that 5GB gets pretty samey pretty quick.))

Which isn't to downplay the issues related to superstimuli, at least to the susceptible. But I think it points to a more general problem, of which AI-generated content is not really going to be much of a leader on either side of the battlefield.

Has anyone ever bothered to count furries?

It does seem rather prevalent in spite of being really pretty weird.

((And... uh, I like buff naked werewolf dudes. A lot. But that 5GB gets pretty samey pretty quick.))

Curse you, furries. Because of you every time I try to talk to someone about White Wolf's Werewolf RPGs I see the light in their eyes dim as they think it's a furry thing. Why must the coolest splat also be the one the furries love :(

Yeah, I'm... probably part of the problem on that. I can complain about Awakening's balance and contradictory lore, if it'd help?

Wow, Exalted references in the wild. Unexpected.

Those tits are absolutely massive.

The system works.

Standard sized big ones in ecchi and hentai. Thousands of such just over at DA. That's where the AI learned it I guess.

Even happens IRL naturally, there's a rare disorder that causes adolescent breast growth to never really finish.

Worst cases surgery required because they get too big, best case it kind of tapers off eventually and can be even lucrative as illustrated by this tasteful somewhat SFW picture of a sufferer in her marble bathroom - a rarity in Croatia, where median wage doesn't exceed €12k a year.

Yeah, unfortunately if you want big anime titties you tend to get ginormous anime titties. There's not a ton of art where it depicts a woman who is busty, but still within reason for a normal human. Which I would imagine affects the AI in turn, because that's what it could get trained on.

There's some of that, probably related to adaptation to superstimuli. Not just anime, affects all kind of racy art. Over at DA you can sometimes even see it in action in an individual artists's portfolio.

Although, it does seem the large proportions are liked more than people assume.

Assuming people who order sex dolls have preferences in regards to proportions similar to general population, it looks like the the median preferred boob size is between E and F. Somewhere in or near huge, especially on a slim woman.(sfw picture of a woman with a 34F bra size)

Wondering if we're going to see similar IRL runaway effect due to Instagram and such online social media.

It's already driving an increase in cosmetic surgery, but if some biotech firm found a way of re-starting adolescent development things could get out of hand.

What are they training it on? Only a few hundred k, so presumably they're not throwing the entire e621 catalog at it yet.

From my understanding, the major trainers have largely downloaded a subset of e621 data, filtered by upvote score, and then by content. Furry and Yiffy to both SFW and less-extreme-kink NSFW, with different thresholds and limits. Zach3d on 'texture' fetish and a few specific species mixed with a small subset of general pictures with a higher upvote score. I think most have also filtered out material that they think is likely to cause artifacting, either technical stuff like severe jpg compression, or many-panel comic pieces.

So it's relatively easy to grab sets and train a model, to the point that groups of amateurs can do it? That's good news: I had a vision of all art production being censored by the kind of political commissars who inevitably take over large projects.

The idea of being able to restrict artists at the canvas level must be making some of them drool.

As far as I know, each of these datasets has been curated by one person, to their respective tastes. Hasuwoof for Yiffy, DirtyApples for Furry, and Zach for Zach3d. e621 is well-enough tagged for high-score posts that it seems fairly automatable, and as long as you're not abusing the download process, it's hard to tell a normal user from an archiver, especially if you filter before download. And the code itself is... not fun, since it's poorly documented python in most parts, but it's nothing ridiculous.

((There's a My Little Pony-specific one that's supposed to have been released recently, but I know less about that.))

There's been some discussion of setting up teams for difficult heavy lifting (eg, improving tagging, building and parsing datasets with more eyes-on-curation), but the big issue for now are cost and technical accessibility. The core model is expensive because it took literally millions of steps in a large dataset, but further tuning is relatively cheap, with most epochs taking less than a day on a single (beefy) cloud GPU server. But getting the data together and onto that machine rapidly enough can be complex to do right, and easy to end up with a staggering AWS bill if done wrong.

That'll be less an issue if newer GPU generations continue to bulk up on VRAM; if done at home, it's mostly an energy (and/or cooling) bill thing. And that might be coming as soon as this winter for people willing to splurge on the higher-VRAM versions of the 4090.

This deepfake youtube series of Gandalf commenting on The Rings of Power seems particularly tuned to the interests of a significant number of users here.

https://youtube.com/watch?v=9POcNCMlP_s

I've been thinking about balancing Rampant Arsenal (a big mod with lots of new weapons) for PvP gameplay in Factorio. The game IMO has quite a potential as a cooperative-competitive RTS.

Basically, balance things out so there aren't any OP strategies such as just running around and chucking grenades at turrets (judging on having seen some pvp games), take the most complex manufacturing mod, something like Bob's & Angel's, tie better quality weapons (tanks Mk2, Mk.3 etc) to larger quantities of ever more exotic materials.

If you've ever tried making titanium in B&A, you probably know.

PVP factorio? Why not just play Foxhole?!

Seriously, the new facilities update has a lot in common with Factorio it seems.

I should get back into it. Trains at last.

Why not just play Foxhole?!

Does it have a fraction of the industrial complexity ? Can you mod it?

Does it have nerve gas and atomic landmines ?

EDIT:

Also, I want to dip my toe back into software dev, starting out with some light scripting seems like a relatively painless way.

My mid terms are going to end tomorrow. I had just 4 papers this semester, only two were technical and even they went better than the ones before. I will likely smoke with a few friends and roam around.

I used to think that I needed to get female validation to feel good about myself but in all honesty, it is these little moments that I will miss. I will be leaving college in may 2023, January if I get an internship and that means that I only have a few weeks left of this.

There is something about being with young people. I live with my parents as my house is a 30-45 minute drive from my uni. I was earlier gonna try for a date with some girl or go for pickup but fuck that. I will just chat till 4 am with my friends, sleep on the dorm floors again and maybe eat at some cheap food joint. Reminisce about my oneitis and yell cuss words. It is not exactly what most here would be doing but I just realized that I too will likely be posting what most here post in a years time given I will be employed.

I hated uni and it was my fault that I fucked up all that I did but I still made good friends so yeah, hoping I will get dead drunk tomorrow, make memories forever and stay in the moment.

I have always had a oneitis since I can remember, maybe since 4th grade where I would progressively spend more time thinking about girls, during 9th and 10th grade I had the most fun I ever did in school and I was a complete loser for all girls. I do not remember any of the girls I was obsessed with back then but will always fondly look bad at those years before cram school became serious, same for this year in uni. Spending that night with my friends was more memorable than losing my virginity was.

Do suggest some hangover cures lol, I do not drink but it is good to be a degenerate at times and enjoy these little things. I also got high this Tuesday after a few beers, it was my second time this year and it hit me like a ton of bricks, I was out of it and could barely walk. I think this is what people call a bad trip or something, anyway, tomorrow should be fun.

I hear my parents often talk about how important these little moments are and as a kid I never understood what they meant but now I do.

Like that song two minds by Nero (https://youtube.com/watch?v=KFWFJGfEaNo)

"I told you too many times,

It's the little things that count which can make someone feel special."

Why do you drive 30-45 minutes to get from your house to your house? That seems slightly longer than necessary.

As for hangovers: Drink water. A lot of water. Start before you booze up, keep drinking water while boozing, then drink some more water and keep drinking water until you go to bed. You trade the hangover for needing to piss twenty times in one evening and maybe some kidney problems, but that's the only thing that ever reliably worked for me.

I meant uni lol.

Also wierd night, my first one at a nightclub, the girl I was with nearly fucked me in the girls restroom but I was caught by the security.

Later made out with 3 other chicks, 2 blocked me immediately when I texted them, like wtf man.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats