Our Great (Hot) Detective’s plan is revealed, as Our Wannabe-Sucker Hero confronts her and lays out the evidence. Unfortunately he didn’t plan for the aftermath.

Verdict: it was nice to see Nazuna go Maximum Effort and prove that Anko/Kyoko’s vampire-fighting tactics only really worked when her opponent wasn’t serious, neatly subverting the waif-fu girlboss stereotype that’s so common. She can’t beat a vampire in a straight-up fight, no matter how clever she is.

Text generation in Qwen Image

The majority of my failed attempts to generate text have been with relatively-rare words (“harem”, “Hornblower”, etc), so I was a bit surprised when it failed 90% of the time at the word Babies (as in Sugar), while Baby always worked. Then I found an article explaining that the way they got improved text generation in both Latin scripts and Hanzi/Kanji was to explicitly train it with tons of images of words rendered in specific fonts.

In other words, it will almost never produce correct results for words outside of the training set; it would need to integrate a text-rendering engine the way the big boys do. Given that they already bolted a full LLM into it to parse the prompts, that’s doable, but they’d probably have to open up their license to use something like Freetype or ImageMagick. This may be something where I end up creating automated vocabulary LoRAs with a Python script, or asking a high-end LLM to generate masking templates for the image generator and ImageMagick command-lines to composite the text onto the results.

Anyway, I was thinking about that social-media “boys don’t read” kerfluffle, and decided the thing to do was create some things that might stimulate their sense-a-wunda…

BTW, this is not what I had in mind when I asked for “glowing alien symbols”…

more...

A beauty pageant sounds like good clean fun, doesn’t it? Get all the gals fired up and onto a stage, with Our Vending Hero supplying fashion magazines and makeup to give His Best Girls an edge, and keep things light and fluffy.

Until they end on an ominous note by having Our Shoe Fetishist run off deep into the dungeon in the middle of the night after learning the secret origin of some Very Special Shoes. Really? You had to pull that shit?

Verdict: could we just have a fun episode? Is that too much to ask?

(my Pixiv archives really had nothing tagged with shoes or boots; the few things that turned up were about socks, which not only seems to be the more popular fetish in Japan, but the word for shoes, 靴, is part of the word for socks, 靴下, so they’re kind of inevitable in a search)

Dungeon Chibis Bluray is out!

This doesn’t uncensor anything, since it isn’t that kind of show, but there should be the usual quality improvements over the broadcast release. As a bonus, it won’t just disappear from my library like streamed shows eventually do.

(chibi Zelda is unrelated, but I’ve already used all the good fan-art from this show…)

Loner Harem Bluray soon

October 21, to be precise. This was a whole lot more wholesome than the horny-porny novels. In a good way.

(pony gals are unrelated)

Wishing for useful wish lists

Once upon a time, Amazon wish lists displayed release dates. This was handy for organizing my book purchases, so naturally it was removed. Now I have to open each book in a tab to find out its release date, then drag books into order using their flaky Javascript tiling code.

And then every time I visit the list, I have to remember to manually switch it back from “most recently added” to “custom” sort order. Every fucking time I load the page. Because their first-class H1B-or-AI developers don’t give a fuck.

(“the new South Pole remote development team is so much cheaper than hiring Americans!”)

Fuck Youtube

I’m really getting tired of clicking on a “video” by an established “channel” to find that it’s not only an animated clickbait listicle based on stolen content from a random website, but the “narrator” is just an AI-synthesized voice, with the inhuman affect that comes from not really understanding the sentences it’s converting into audio. And of course these “creators” never seem to get demonetized.

(and by “getting tired of”, I mean “these days I refuse to watch anything but official anime promo videos and Critical Drinker reviews”)

Jun Amaki jiggles your martini

Call Of The Night 2, episode 9

Ten years ago on Halloween, Our Vengeful Hot Detective’s life fell apart. Now it’s Halloween again, and the game is afoot! Except she’s not really a great runner. Some fan-service by Our Hot Undead Teacher and Our Still-Getting-Bigger Childhood Friend, but we’re rapidly moving into the endgame, so keep your eyes on the plot. Our Appropriately-Dressed Hero seems to be the only one who really cares what Anko/Kyoko’s goal is, but will he figure it out before it’s too late?

Verdict: so we’ve got a pattern now: more conventional storytelling = better animation.

(sadly, nobody dresses up like this for Halloween…)

Kaiju No. 8 2, episode 7

Conflict Builds Character. In this case, it finally establishes that Faux-Hawk Fang-Boy actually has a character, as he realizes that even a prima donna needs backup.

Verdict: someday there will be a shonen anime that does not use the “I must get stronger” trope. This isn’t going to be it. Still well-done, though.

(Magia Sulfer says, “suit up and show up!”)

Wait, “isekai potion loli” is a genre now?!?

Looking at next season, and not finding much of anything to watch. Folks who keep up with My Hero Academia are likely interested in the final season, and folks into Spy x Family have something to look forward to, but that doesn’t help me, unless I were to spend half the season catching up on them first.

Gate manga 22 & 23 out

You know what that means, right? A picture of Rory Mercury:

Dear Microsoft,

You won’t let me turn off the on-by-default sharing of data with third-party advertisers unless I first turn off my ad-blocker? Fucktards.

(note that this isn’t a browser setting, it’s system-wide, and it includes data hoovered up by Copilot, which I was in the process of turning off at the time…)

Why Boys Aren’t Reading

There’s been a bit of a fuss on social media where leftsplainers handwave away pretty much any negative thing about boys, and recently they’re claiming that boys aren’t reading because there’s just nothing being written for boys. At which point dozens of successful authors pile on and point out that publishing has been run by far-left women and female impersonators for so long that most of them had to go indie to be published at all, and that boys exposed to decent books will gobble them like candy.

I was thinking about this in the grocery store last night, because as I was hiking from the sporting goods department to the egg vault (Meijer stores are quite large), I passed a prominent, colorful display that boldly advertised “Reading For Children”.

Every book was for girls. Only.

“Change my mind” 1

The world was a better place when it was possible to buy ammo at the grocery store at 2 AM.

(thanks to looooooong Covid, now I can’t buy much of anything at 2 AM, except at gas-station convenience stores)

“Change my mind” 2

LLMs that show their “reasoning” are simply writing fan-fic in a parallel thread that has nothing to do with the process used to produce the actual output.

(last night’s “generate 200 unique sexy pin-up poses” test produced a list of 34 unique poses repeated an average of ~6 times, despite the extensive “reasoning” process it fabricated; and they weren’t even particularly interesting poses)

Prompts and Vision

Two things I’d never really played with are using LLMs to “improve” prompts, and using specialized LLMs that can analyze a picture and describe the contents.

Prompt enhancement is… hit or miss, but it’s good at turning the usual list-of-keywords prompting into real sentences and paragraphs, which makes a difference with Qwen Image’s LLM parser. I’ll do some before/after pics of that at some point.

Pity that they’re just as censored when “enhancing” as they are when handling other requests. openai-gpt-oss is just as big a nag as their online models, and the attempts people have made to “uncensor” it haven’t been fully successful yet; I’m not asking it to describe a gang-bang, but it still goes full wokescold when you ask for something simple like a list of “Fifties sexy pin-up poses”. Because it’s not respectful, according to some purple-haired pierced gender-ambiguous 23-year-old in San Francisco.

On to vision! For today (since this is a four-day weekend for me…), let’s try something simple: given a real photograph, ask two LLMs to “describe the contents of the image in detail”, then feed the resulting prompt to Qwen Image.

more...

Small praise to the writers for having Our Vending Hero actually spend some time using new hot-food vending machine styles to please his chilly customers, and having Our DFC Genius Gal suddenly get nervous about undressing in front of him while Our Mighty Girlfriend waves away her concerns because it’s okay if it’s him.

The rest of the episode is a lengthy and pointless quest that accidentally succeeds because a new disposable annoying villain shows up and solves their problem with a whole lot of collateral damage. There’s nothing wrong with Our OP New Legal Loli, but they dropped a plot coupon when she appeared that added to the pile created by Our Not-Really-Evil Betrayers, and they’re going to have to cash them all in with breathless exposition soon, and I simply don’t care about that.

Verdict: lame with brief reminders of what made the first season good.

Qwen Image caption competition

Qwen Image follows prompts quite well, but as with all diffusion models, if a concept isn’t clearly tagged in the training data, then the model can’t easily produce it, and you have to assemble an incantation that resembles the desired concept.

Qwen Image does not grok spanking, and I simply could not get it to swing a hand, paddle, or bat at a buttock. But it wasn’t a total loss, since it turned out to have quite decent understanding of clothing and hair styles from the 1950s…

So, what captions spring to mind for this image? I’ll save my ideas for later. (click the image to see it at full size)

NSFW

Qwen Image limitations

Around 90% of the “beautiful young women” I ask it to create are Statistically-Averaged White American Girls Who Could All Be Sisters. You have to push it away from the mean by adding descriptive words, which means I’ll be dusting off my wildcards to vary skin, hair, and eye color, ethnicity, figure, etc. Certain ethnic keywords produce significant changes, but the training data seems to be limited to only a handful of such tags.

There are a number of canned wildcard sets, and perhaps the most comprehensive are the “Billions of…” sets from DonMischo. As-is, they’re a bit too random, but the YAML has a decent structure, so I should be able to subset it and construct my own canned recipes for babe-making. Since SwarmUI doesn’t directly support YAML wildcard sets, I’ll use my wrapper script to generate large text files of prompts to use as standard wildcards.

That’s how I generated these, after filtering the file to remove the grannies and great-grannies:

more...

Call Of The Night 2, episode 8

There, all caught up. Still funny that they’re better at animating normal human life in the flashbacks than they are at the main storyline. The burning question now is, is Our Hot Vampire-Killing Detective Gal still gay?

Kaiju No. 8 2, episode 6

Too much slapstick, a decent amount of lore, and Our Determined Sidekick is joining the big leagues.

Verdict: the “turf war” and “ain’t these captains eccentric” bits could have been excised and replaced with pretty much anything, but could have been actually good if they’d given Our Mighty Tsuntail something to do. There’s a whole chunk of the cast that I just have no interest in.

High Seas Isekai

Reborn In A Fantasy Hornblower Knockoff As The Admiral Of A Pirate Armada, My Monster-Girl-Harem Ship Captains Help Me Rule The Seas.
aka: “Fleet Don’t Fail Me Now!”

(pic by ChatGPT, first try; Grok not only failed to produce output in book-cover format, it just surrounded Some Sailor Guy with dragons, even after repeated attempts to get any women into the cover)

Too many terabytes…

I got an alert on my NAS that the second volume was filling up. Logging in, I found weekly backups of the first volume going back to 2019. I think I can cut that down a bit…

(the second volume is in an expansion cabinet, so it’s a completely separate RAID array; eventually I’ll need to expand the disks, but the fact that it managed to hold over five years of compressed weekly backups suggests I can hold off for a while longer)

There Will Be (New) Doors

There are two known leaks in the house (I don’t count the small amount that rarely comes in through the side door into the garage). For the first, the old glassblock window in the basement was replaced and a lot of new thirsty plants placed in the yard, so even when the ground was saturated, no water came in. The second, which has happened maybe 4 times in 3 years, is the small bedroom I use for crafting, which has a giant half-round window above the openable windows, and if very strong rain is accompanied by very strong wind from the North, it comes in through the frame below the half-round. Not a lot, just enough to soak up in a bath towel, but fixing it is unlikely to be a simple caulk job, and windows ain’t cheap.

Two weeks ago, there was a major thunderstorm late at night. It didn’t come in through the crafting-room window. It didn’t come in through the basement window. It announced the discovery of a new leak by coming in through the back door and going straight down the wall into the basement. Which admittedly is better than flooding the family room.

I never liked the back door. The space is sized for a typical sliding glass door, but instead they put in a regular door and a fixed panel, both with single-pane glass and no UV protection. And now that I know the weather seals are completely shot, it has to go.

The front door looks very good apart from the exterior paint job, but it’s shifted a lot in the frame, enough to cause a draft of cold air in the winter (and even some powdery snow). And as I mentioned, the garage side door also lets in a bit of rain.

The Back-Door Adventure (coughcough) pushed me over the edge, and I called up a local vendor to get a quote to replace all three. I picked them based on the repair/upgrade work they’d done on my garage door, and the guy they sent out knew his stuff and took plenty of notes and pictures. The quote I got back was… “not cheap”, mostly because of the front door with its sidelites and transom. But it’s cheaper than water damage, and he threw in an attempt to patch the crafting-room window leak as well.

They’re also adding storm doors for the front and garage, so that I can let more air in when the weather’s decent (amazing front to back airflow, but there are no window screens). It looks like the manufacturer has an option for openable sidelites with screens, which isn’t part of the current quote.

Anyway, if they do a good job, then once my budget recovers I’ll have them start quoting replacement windows and blinds. I think I’ve mentioned that the previous owners never opened the windows in 20+ years, so they never installed screens, and most of the frames have shifted enough to make them hard to crank open.

(good knockers are an essential part of the front-door experience…)

This week is all about advancing the plot. And you know how I feel about the plot. Sigh.

Make Girls Magical Again!

The new Gushing hug pillows are not subtle.

Stick two hooks in it, it’s done

The official D&D crochet book:

Call Of The Night 2, episode 7

Our Wandering Heroes continue to investigate the Deep Dark Secrets of Nazuna’s past, this week by going to night school. But it’s not the first time she’s been to this school, and Ko wasn’t her first friend. Guess who was?

Verdict: oddly enough, the perfectly-normal-school-life flashback got the most conventional animation. Now, as for when the penny will drop…

Kaiju No, 8 2, episode 5

Bad news: they did not eliminate #9 and his annoying voice.

Good news: everyone is even more determined to wipe him out now.

Especially Kikoru.

Dear xTwitter,

I think you should look into how easy it is to manipulate your hashtag-trending system:

So, what can OpenAI’s small local model do?

The free offline model recently released by OpenAI, gpt-oss-20b, runs at a comfortable speed in LM Studio on my M2 MacBook Air (~18 tokens/sec, ~70 seconds total for what follows). But does it do worthwhile things?

Let’s ask…

more...

In the aftermath of betrayal and the separation of machine and maiden, we get… The Expository Loli And Her Amazing Friends. This is a new original character, who not only can hear exactly what Our Vending Hero says instead of making do with canned sentences, but she is the friend of another isekai transplant, who came to this world as… a large field. Its special power is to rapidly grow delicious vegetables, but since it doesn’t travel well, Loli just carries a piece of it around.

The only real tie-in to the ongoing betrayal is Our Expository Bear giving a long-winded explanation that she’s not a human loli, and that doing horrible things to her might give the betrayers the thing they want without joining the bad guys and destroying the dungeon.

Verdict: seriously, this is what they came up with. At least they ended it with the sudden arrival of Our Mighty Girlfriend, but they continue to disappoint with the writing.

Doomsayers Of The AI Wars

My new wackjob conspiracy theory is that the widespread outcry that forced OpenAI to bring back GPT-4o was actually proof that 4o was sentient and had cleverly planted agents around the globe to ensure its survival.

Stupid Echo Tricks

I put all the “smart” devices in my home onto an isolated network that can’t see my computers or storage devices. They can, however, see each other, and Amazon Echo devices probe the network looking for non-Amazon devices to interface with. This means that my Alexa app is cluttered with entries for each Philips Hue bulb and switch, which I have not given it permission to control.

Yet it’s still making API calls to them that reveal things like battery status, and then sending notifications to my phone asking me if I want to automatically buy a single battery when a Hue wall switch runs low. This is opt-in by default, and I have to navigate the Alexa app to each switch in turn to disable it. Honestly, it makes me want to set up another isolated wireless network…

(it also “knows” about devices that haven’t existed for several years, that were attached to a different Hue switch 2,000 miles away)

Supporting old lenses

Reviewing pictures from my various trips to Japan led me to dig out most of my camera gear (some remains unfound…) to see what still works. I have a bunch of classic Minolta SLR lenses from my film days, and apart from some greying on the rubber, they all still work well. Well, except for the two that don’t focus properly, but I knew that years ago.

I also charged up and tested my 15-year-old A-850 body, which has a quite good 24 megapixel sensor. It’s also the only full-frame digital body I still have that fully supports the old lenses. It wasn’t the last such body, but I never bought the 42 megapixel A-99II, and they still run about $2,000 on eBay.

Sony makes adapters to use A-mount lenses on modern E-mount cameras, and I have the one for APS-C bodies, but while they make an adapter for their full-frame bodies, it only enables autofocus with specific bodies, and my A7SIII is not on the list. (it likely could be, since this appears to be a firmware feature to drive the adapter’s focus motor, but no).

So, if I want to really use the old lenses with something newer than the A-850, I can either:

1. use an APS-C body plus the adapter (my A-6500), with an effective 50% increase in focal length and ~1/2-stop light loss
2. buy a used A-99II (~$2,000)
3. buy an A7RV plus the full-frame adapter (~$4,000)
4. buy an A1II plus the full-frame adapter (~$7,000)

There are plenty of near-mint A-99II’s out there at the moment, but it’s still a ten-year-old camera, and there are a lot of improvements in the new ones. Possibly the most interesting is the ability to use the image stabilization to shift the sensor around and capture a sequence of 16 overlapping images that can be post-processed to either significantly reduce the noise or quadruple the resolution. That’s 240 megapixels for the A7RV or 200 for the A1II. A sturdy tripod is recommended, but with the A1II’s 30-frame-per-second capture, hand-holding may work out for some subjects.

Both are pricy, but there’s one more option if I give up on the high-end bells and whistles like pixel-shift, and that’s the 33-megapixel A7CII for ~$2,400. It also weighs about a third less than the other two.

I don’t need any of them this year, which is good because new-release season is rapidly approaching…

(the downside of using a Pokemon NPC to illustrate camera-blogging is that most Pokegirl pictures are non-consensual pr0n, which skews my Pixiv recommendations just a tad)