Rectus

針妙丸ちゃん

"From the height of the port crane" superimposed over a promotional image of Mechwarrior 5 so that it appears that the woman in the picture is operating a battlemech instead of a port crane

I'm sorry i need to leave this call, something came up. My dog just sat down on the couch and made an adult human fart noise

need some of that australian caulk

happy international women's day

just needs some caulk and it will be as good as new

ニナ

An anti-gender manifesto.

For International Women's Day, I, a female-presenting tranny dyke, am asserting certain things. I am beseeching people to stop saying that women would end all wars if they were in charge. That women are not violent.

📚Kosuzu🔔
#東方Project #touhou

#深夜の真剣お絵描き60分一本勝負
#touhou
#東方Project
——上海人形
第1293枚
魔

蓮メリの通う大学にも農学部があって、学生がゼミの飲み代目当てに作ってたりする野菜を売ってたりするんだ。

nug

From the gamedev community on Reddit Explore this post and more from the gamedev community

Unpopular opinion, but a lot of this "doing it right" mentality is how we wind up with games all feeling the exact same.

Difficulty spikes, weird controls, ropey onboarding all of it critical to prevent games feeling like they're ground down and pumped out safe mush.

www.reddit.com/r/gamedev/s/...

imo playtesting should be in service of your vision, not deciding what the vision should be

they're so hard o visualize, but i always found it fascinating that a non-linear projection is a linear transform in a higher dimension matrix

雷鼓です
Skebありがとうございました～

Bad dragon outbidding cult of the lamb for the Jim Carrey birthing rhino was not on my 2026 bingo card, but boy am I glad it happened.

i think people are forgetting how good photos you can take when not using a phone camera

#BirdArt
ﾂﾋﾟﾂﾋﾟ

more early counter strike vibes

Jersey City Parking Authority van with the sliding back door open: JERKING AUTHORITY

MYTH: As commander-in-chief, Trump has the authority to take military action. FACT: It is unconstitutional for a U.S. president to declare war without the approval of the Knesset.

Trump’s War On Iran: Myth Vs. Fact https://theonion.com/trumps-war-on-iran-myth-vs-fact/

some neat photos from Vyborg in 2020

"Mountains by Moonlight" glass ----- [from the Smithsonian American Art Museum] ----- Mark Peiser has made many sculptures that show landscapes with a sun or moon hovering above. He forms the initial image by pouring layers of molten glass into a mold, then quickly adds a dollop of a different color to the hot mixture for the floating sphere. In Mountains by Moonlight a simple twist of frosted glass creates the illusion of rolling hills below a clear blue sky. The highly polished planes of glass on the outside refract the image so that we see a slightly different view from each angle. -- [artist biography] -- Mark Peiser started his career as an industrial designer, but after working briefly in Chicago he left his job to study at Penland School of Crafts in North Carolina. Glassblowing fascinated him, and within a few years he had built his own home and studio on the school grounds. Peiser experimented with a wide variety of effects, adding different materials to the molten glass such as ceramic glaze powders, household products, and even rat poison, which made"beautiful bubbles." Peiser has continued to play with established formulas and techniques throughout his career, creating and combining new and unusual colors in his glass sculptures.

Mark Peiser (1984)

My photo shows a museum display with colourful Minoan pottery cups arranged on three clear shelves, one above the other. These cups, known as Kamares Ware, are from Phaistos, Crete. They were made in palace workshops, c. 1800-1700 BC. The cups range in shape and size from conic and cylindrical cups (top and middle shelves) to hemispherical and carinated shaped cups (bottom shelf). They are decorated with multi-coloured geometric motifs; with spirals and swirls painted in red and white pigment on black.

Sipping my coffee ☕️ and thinking about these marvellous Minoan cups!

They look so modern it’s incredible to think they were made during the Bronze Age some 3,800 years ago!

Heraklion Archaeological Museum, Crete. 📷 by me

#Archaeology

ナズーリン🐭

time to put vim expert on salads cv

dampf: Old-school Steam skin for Dear ImGui A 2004-era Steam/VGUI2 tracker skin for Dear ImGui

i've been fiddling with an old-school Steam skin for Dear ImGui: seidt.quest/s/dampf/. i open it several times a day to click and drag stuff around therapeutically because it's… grounding? :D imgui gallery thread post with source and references coming.

oh no, my spicy markov chain isn't god after all

Nav Toor @heynavtoor 🚨BREAKING: OpenAI published a paper proving that ChatGPT will always make things up. Not sometimes. Not until the next update. Always. They proved it with math. Even with perfect training data and unlimited computing power, AI models will still confidently tell you things that are completely false. This isn't a bug they're working on. It's baked into how these systems work at a fundamental level. And their own numbers are brutal. OpenAI's o1 reasoning model hallucinates 16% of the time. Their newer o3 model? 33%. Their newest o4-mini? 48%. Nearly half of what their most recent model tells you could be fabricated. The "smarter" models are actually getting worse at telling the truth.

Here's why it can't be fixed. Language models work by predicting the next word based on probability. When they hit something uncertain, they don't pause. They don't flag it. They guess. And they guess with complete confidence, because that's exactly what they were trained to do. The researchers looked at the 10 biggest AI benchmarks used to measure how good these models are. 9 out of 10 give the same score for saying "I don't know" as for giving a completely wrong answer: zero points. The entire testing system literally punishes honesty and rewards guessing. So the AI learned the optimal strategy: always guess. Never admit uncertainty. Sound confident even when you're making it up. OpenAI's proposed fix? Have ChatGPT say "I don't know" when it's unsure. Their own math shows this would mean roughly 30% of your questions get no answer. Imagine asking ChatGPT something three times out of ten and getting "I'm not confident enough to respond." Users would leave overnight. So the fix exists, but it would kill the product.

This isn't just OpenAI's problem. DeepMind and Tsinghua University independently reached the same conclusion. Three of the world's top AI labs, working separately, all agree: this is permanent. Every time ChatGPT gives you an answer, ask yourself: is this real, or is it just a confident guess?

⚠️Told you so moment:

OpenAI published a paper proving that ChatGPT will always make things up [...] Always.

They proved it with math.

[...] This isn't a bug they're working on. It's baked into how these systems work at a fundamental level.

More: x.com/heynavtoor/s...