oliwary

Or the Grey men in Momo (Michael Ende) who offer time savings through efficient appliances, while stealing the time made free through their use. Their main for is Momo, who encourages her fellow children to play for the sake of playing, which terrifies them.

It's a beautiful story, more and more relevant.

The most important graph for me on Artificial Analysis is this: https://artificialanalysis.ai/?agentic-speed=intelligence-vs...

It shows the time to complete a task versus intelligence index, which is much more important for my use-case as I do quick back and forths usually. Someone on here suggested it to me a while ago. It also captures speed and token efficiency nicely, since both of these decrease time per task.

I agree. There is no reason we have to be at work 9-5 or eat dinner at 7, it's just convention.

In fact, I have considered whether it may not make sense to just have a single timezone. Would make planning internationally much easier.

Well yes, but it's also slightly smarter compared to other models of similar intelligence.

This looks great! I also think speed should be part of the metric (i.e. how long does the model take to actually solve a task). For me, I prefer to run expensive models such as Sol on light reasoning, which usually gives me good answers with quick responses.

For my style of coding (quick back-and-forths and corrections) it makes a big difference if a model comes back in 1-2 minutes compared to 5-10, and I am happy to pay a bit extra for that.

Cool! It's really fun what browsers can do these days in terms of real time image processing even.

I made a set of games controlled by waving in front of the camera a while back at https://motionparty.net This reminds me that I should finish them!

Thank you for trying it! :)

Thank you very much!

A daily estimation game: https://estimatle.com

Every day, players answer 5 questions that require estimating some quantity, such as distance, year, cost etc. Then you can compare against answers of other players and the real answer.

Working on finding the right balance of questions that are both fun and also help understand facts about the world.

Thank you very much! :) That's a good idea, just need to find a good source for the correct answer

Hey!

Today I launched a new daily game called estimatle. Every day, you guess/estimate 5 different numerical questions (such as "When did the Titanic sink?" or "How many keys are on a standard piano?"). Then, you can compare the results to the other players and the real answer. The idea of the game is to test your general knowledge and guesstimation skills, while also hopefully teaching some facts about the world.

I am interested in any feedback!

Can likewise mirror the support for quasar, always the first framework I reach for when starting a new project. The amount of stuff you get for free in terms of setup and content is insane.

Wow thanks for the heads up, will have to look into that Firefox issue.

Thank you, much appreciated! :) And great ideas as well! Whack a mole would fit perfectly

https://motionparty.net/ - motion controlled web games, by analyzing the Webcam feed locally. In the vein of eye toys for ps2. Works quite well I think, but some elements need improvement. Want to add more games as well.

The name is really clever given that the character in Suits is called Mike Ross. :)

https://motionparty.net/ - A collection of games you control by waving in front of your camera, similar to playstation eyetoy back in the day. It supports 1-4 players.

I think it works quite well so far, but need to tweak the camera algorithm a bit to make the buttons work better. Thinking about more games to add as well.

Another game that I find has very impressive draw distance is Just Cause 2. You can see objects very far away when flying etc, but they look very detailed and do not change when moving closer. Definitely blew me away the first time playing it.

Impressive! Do you include puzzles released before the training data cutoff date?

I think it's akin to self driving cars prioritizing nornal roads rather than implementing new infrastructure. Tricky, but if you get it right the whole world opens up, since you don't depend on others to adapt your system.

What's next? Claude Iliad?

This is super cool! Ran into an issue though, the first time it boots perfectly, after the first refresh it loads for a bit (downloads the image again instead of from cache) and then a cachebuster URL is added and loading starts over, without ever finishing. Ideally it would just load from cache on refresh.

I have one of the TTGO T-watches: https://www.tinytronics.nl/en/development-boards/microcontro...

For me it is mainly a decoration of my drawer, but maybe comes close to what you are looking for. :)

This is insane if true - could be super useful for data extraction tasks. Sounds like we could be talking in the cents per millions of tokens range.

Cool, similar to helicopters which can also control direction indepentendent of thrust, which leads to RC helicopters being able to pull of crazy, physics-defying moves like here: https://www.youtube.com/watch?v=QSiwyoQldfo

I have found them to work quite well for frontend (most recently on https://changeword.org), although it sometimes gets stuff wrong. Overall, LLMs have definitely improved my frontend designs, it's much better than me at wrangling CSS. Two things that have helped me:

1) Using the prompt provided by anthropic here to avoid the typical AI look: https://platform.claude.com/cookbook/coding-prompting-for-fr...

2) If I don't like something, I will copy paste a screenshot and then ask it to change things in a specific way. I think the screenshot helps it calibrate how to adjust stuff, as it usually can't "see" the results of the UI changes.

I am working on my first multilingual word game: https://changeword.org

My previous games have all been exclusively in English, but this one also has Spanish, French, German, Dutch and Swedish. It's a take on the classic word ladder game, with golf scoring mechanics.

Still needs a bit of fine-tuning for the word lists and puzzle generation, but I think it's already pretty playable. :)

In https://squareword.org (2D variant) I was also running into this problem. It's a bit different though, since I need to find valid 5x5 squares, with 5 words down and 5 across. Surprisingly, there is quite a limited number of such squares.

Ive been able to solve it by slowly injecting more challenging words over time, which has the side effect of also introducing a difficulty gradient. Players seem happy so far :)