6mian
Are there reliable ways to figure out which pixels on subtracted frames are cloud movement and which are the few ones that are an aircraft?
Correct me if I'm wrong, but adding extra cameras doesn't seem to solve this problem, each source reporting "almost everything moves" would make solving intersections and tracking them impossible, because each target candidate can be assigned to many changes pixels in consecutive frames. Unless some additional pattern detection is done, but again it's hard for very small objects.
"Better" was referring to what OP wrote in the top comment. I guess 10x faster, 10x longer context, and 100x less prone to hallucinations would make a good "10k x better" than GPT-4.
I'm not assuming that something 10k x better than GPT-4 will be good or bad; I don't know. I was just curious what exactly to be worried about. I think in the current state, LLMs are already advanced enough for bad uses like article generation for SEO, spam, scams, etc., and I wonder if an order of magnitude better model would allow for something worse.
I'm glad to see so many senior engineers on the team; it must be an exciting lifelong journey to work on the mission.
I spent some great time playing with the base implementation. Making the rays act as particles* that bend their path to/away from objects, making them "remember" the last angle of bounce and use it in the next material hit etc. Most of them looked bad, but I still got some intuition what I was looking at. Moving the camera by a notch was also very helpful.
A lot of fun, great for a small recreational programming project.
* Unless there's an intersection with an object, then set the maximum length of the ray to some small amount, then shoot many rays from that point around and for each hit apply something similar to the gravity equation. Of course this is slow and just an approximation, but it's easy and you can implement a "black hole" type of object that will bend light in the scene.
Going through Andrej's makemore tutorials required quite a lot of time but it's definitely worth it. I used free tier of Google Colab until the last one.
Pausing the video a lot after he explains what he plans to do and trying to do it by myself was a very rewarding way to learn, with a lot of "aha" moments.
That's why it's a really good idea to collect 4 Scrolls of Mapping for levels 1-24.
If you were hand-held to the very end you'd only see a a narrow path you went through, which is also ok. It's just a difference in the journey, similar to Breadth-First Search and Deep-First Search in graphs.
Another thing is game fairness. In my opinion, SPD is fair, there's no "you stepped on something you couldn't detect if you tried and died". If you pay attention you will see enough information to identify threats after the initial die-a-lot period :)
I've been playing SPD for years now and for me the best strategy to win the game is to keep Scrolls of Upgrade until you get a hold of a weapon of at least tier 4, same goes for armor. This makes the initial levels much more risky, but if you survive it's a matter of not making mistakes or getting yourself into high risk situations. Using this method I get to the Amulet in about 30% of the runs. Since the recent updates getting back to the ground has become a much more fun challenge.
Besides postponing the use of Scrolls of Upgrade, here are a few things that work for me:
* You don't have to identify potions and scrolls right away, often you can keep them until you gather more information. For example, killed flies sometimes drop health potion and they are on early levels, so you can learn the color without using it.
* Before you get the Scroll Holder that protects the scrolls from fire, it's a good idea to stash all the scrolls in water at the beginning of each level and retrieve them back before going to the next one.
* Scrolls of Mapping are absolutely crucial on levels >20 because of the traps and they are relatively cheap in the first two shops.
* Summoning trap rooms are great to gather a lot of xp if you have potions of paralytic gas and fire/toxic gas. After the level is cleared, you just open the doors by throwing something at it, go as far away as possible, throw paralytic and then fire/toxic potions. And run to the next level.
If you like the game, consider donating to the project, you can do this directly in the game, at least in the Android version.I wonder how many others of "us" are there.