NewoGame

The difference between PC gaming and Console gaming is that no matter how you twist it you know exactly how something performs and how much power it will consume. Using fuzzy words to try to sell "hybrid consoles" aint gonna work. Its snake oil. youtube.com/watch?v=dhJFgX…

@Timmysofine Yeah, nice SUMMER photo. Wait till WINTER comes. You'll be less of a poet.

Any minute now his industry colleagues from the Guardian, Times, BBC, Sky News, LBC and Channel 4 will express solidarity in the face Israel’s barbaric targeting of a journalist. Any minute now.

The moment Israel targeted British journalist Steve Sweeney in southern Lebanon.

@_helium Playing a bad game at the arcades

What is considered a waste of tokens lol?

@IslanderWORLD Sleep cells don't have televisions or internet?

🇺🇸🇮🇱🇮🇷 “U.S. authorities have reportedly intercepted encrypted communications from Iran that may serve as an operational trigger for sleeper cells in the Western Hemisphere.” The non-alert alert – reviewed by Fox and ABC News cites “preliminary signals analysis” of a transmission described as “likely of Iranian origin,” relayed across multiple countries. The transmission cannot be decoded. Its “exact contents cannot currently be determined.” In other words... They intercepted a message they cannot read, from a sender they cannot confirm, containing instructions they cannot decipher and the American client media is running it to prime you for false flag attacks. This is narrative architecture, not intelligence. And the timing could not be more desperate. Because here is what is actually happening on Day 10 of this war. Iran has fired over a thousand missiles and nearly 4,000+ drones since February 28 — sixty percent of them aimed at US targets across the region wreaking havoc on US-Israeli military infrastructure. HRW documented white phosphorus use in residential areas of southern Lebanon. The US Embassy in Riyadh has ordered non-emergency staff and their families to leave Saudi Arabia. 7 American soldiers are dead (the number is likely far higher). And Trump, surveying this panorama of regional catastrophe he authored, called it “a little excursion.” A little excursion. Filed and noted. Now against this backdrop of burning refineries, dead soldiers, and energy price mayjem — Washington issues a breathless warning about an undecipherable transmission that may activate sleeper assets. Ask yourself who benefits from that headline landing today. Ask yourself what it prepares the public to accept tomorrow. Many of us have been warning you about Israels track record with false flags. As desperation continues to accelerate, the false flag frequency accelerates. We have already watched it operate in real time. On March 5, four drones struck Azerbaijan’s Nakhchivan exclave, hitting the international airport and a school. Azerbaijan immediately blamed Iran. Iran denied it and pointed directly at Israel, calling it a false flag engineered to drag Azerbaijan and its treaty ally Turkey into the war. Iranian Foreign Minister Araghchi explicitly condemned the Israeli regime’s role “in such attacks in order to divert public opinion and destroy Iran’s good relations with its neighbours.” A Turkish opposition MP named it plainly: a “possible false flag operation by the US and Israel,” warning that “there is no treachery they would not commit to drag Türkiye into the dirty anti-Iran front.” No treachery they would not commit. Now add the sleeper cell “transmission” to that ledger. The one they cannot decode. The one that may activate assets who may be prepositioned in cities that may be targeted. Weaponised ambiguity deployed exactly when the architects of this war need the domestic population frightened, compliant, and stripped of the instinct to ask questions.. UN human rights experts have already characterised strikes from the opening day of this war as war crimes under the Rome Statute. The Iranian Red Crescent reported over 6,600 civilian units targeted. On June 8, 1967, when aircraft and torpedo boats belonging to Washington’s most "loyal" ally attacked the USS Liberty in international waters, killing 34 American sailors and wounding 174 more and the US government buried it before the smoke cleared. The cover-up was bipartisan, institutional, and absolute. The cui bono was never pursued. The survivors were ordered into silence. Fifty-nine years later, the same alliance is manufacturing the pretext architecture for escalating a war that has no exit strategy, no legal mandate, and no honest accounting of cost — running an undecipherable “transmission” as terror warning. The world is watching and knows the history. Trust the pattern. Not the manufactured "Intel" designed to prime you for accepting much greater costs.

@etiennejcb @FakePsyho Cars are super human. Could say the same about a bicycle. To what end? Not sure. Hype?

I hesitated to write a tweet about AGI not being here, I guess i'll put it here. I have this example from my field of expertise where AI has bad capability: given a procedural animation, give code that reproduces it. A lot of people can do this, and the AI can't. I guess it has to do with the ability to handle a task that's quite original each time (hardly expressed in training data), and that requires very strong multimodality between the visual and the maths/algo. But once it will be able to do that, I think it will be huge because it's about modeling things from observation. And when we figure out how to make AI good at something, it's soon extremely good. So far AI is superhuman at some things like combining lots existing techniques with normal reasoning, but poor at some other things. Maybe mathematician share my experience of AI being good for solving technical things, but lacking insight to truely innovate.

The AGI is already here – it's just not evenly distributed I have a friend who recently started optimizing his life with Claude Code (CC). Some of the more notable examples: - Gathered all of his medical records, analyzed them and used them to suggest additional health screenings (CC had access to his email and messages, but not his passwords; it went through his mails to find all the relevant clinics, reset the passwords, and navigated the sites to find the relevant docs) - Created high-level financial advice and a detailed expense summary based on his bank statements (PDFs/CSVs) and asset data (here, finding all the relevant documents was a manual process) - Created a custom gmail client where every email is automatically categorized, has improved support for unsubscribing (via CC) and allows quick payments with a single keystroke when someone sends an invoice - Created a local read-only database with messages from all apps and aggregates them by user + adds semantic search on top, so you can easily find stupid things your friends said ten years ago We're already at the point where agents are extremely capable personal assistants and their main bottleneck is not memory or reasoning skills, but the permissions that they have. And yes, this is all very risky and may fail spectacularly at any point, either due to agent misalignment or an external attack (like prompt injection). Just to be clear, I'm not suggesting that people should do this. But this sci-fi stuff is already possible and it doesn't even require any substantial technical knowledge (although it helps with avoiding disasters). It's going to get weird really soon for some of us.

@heynavtoor confident guess

🚨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?

I have been messing with the signs a little youtube.com/watch?v=BzTQC5…

Bombing Iran in the middle of negotiations, while starving Cuba, while genociding Palestinians, while threatening to invade Greenland… the US and Israel are the single greatest threat to humanity and it’s not even close. We are all forced to live in the nightmare they create.

Introducing Texel Splatting a technique for "true" 3D pixel art

@0xC0DE6502 What console is this?

Creepy Crawly WIP on the Acorn Electron. Single screen buffer stress test. Decided to use MODE5 instead of MODE1 because I may need a second screen buffer later. This can easily run at 50fps but that just makes the sprite animations look weird

The NY Times admits that the US military has imposed a naval blockade on Cuba. This is an act of war. The Trump admin is illegally seizing any oil tanker that tries to provide fuel to Cuba. This is a barbaric medieval siege aimed at starving millions of Cubans into submission.

Not sure what's going on here youtube.com/watch?v=FhAZIl…

Updating Metal Canary for Nintendo Wii! 🎮 Videos are now working, controls adapt to Nunchuk, aim & throw bombs, and Stage 2 graphics got a HD upgrade. Coming physically at Retrocon this July! youtu.be/DuUZ7uTmctM?si… #MetalCanary #NintendoWii #IndieGame #Shmup #RetroGaming

@games_rozz Clip too short

Happy Valentines Retro Racers 🥰

Pretty much

@sirjoeldean Arguing with a 2 year old

Claude: "Joel, I have verified all 41 rulesets, and the system is 100% ready for production." Me : "Are you sure and if you are so confident, can you generate a rationale for every claim you are making and present it to me." Claude : " Joel, honestly, 20 of the 41 rulesets arent working, I am going to fix them right now" LMAO!