I've recently been getting into RSS reading and well, I usually just went with whatever was given in a forum (like Tildes for example). Although, I've recently been looking into news organizations I follow to see if I should actually trust them.

Factors that came to mind to be important was looking at past controversies regarding them to see where they might fail in the future and who owns them. It made me realize that most sources I had actually might not be who I want to follow for news but then well, not many are left and while I do want to cut down on the amount of news I get because it's overwhelming, I also don't want to miss important news.

So how do you pick what sources of news you listen and what are some news you trust and why?

Hey, all. I'm back four weeks to the day after you guys gave me a lot of great advice about potentially making a 2D RPG out of my tabletop RPG. I decided to try both Godot and Unity given what people told me and I completed two tutorials for each over the last few weeks. After completing these two tutorials, I have some questions that I hope maybe some of you can answer to help me choose between the two.

TL;DR at the bottom. This is a long post.

For context, here's the tutorials I did:

Godot - https://www.davidepesce.com/godot-tutorials/
Unity - https://learn.unity.com/project/ruby-s-2d-rpg

To be frank, the Unity tutorial wasn't really an RPG. There were no stats, no quests, XP. It was much more of an adventure game. That's fine, it still gave me a lot of time inside the engine to learn a lot of basics.

So, working with each one had it's own up and downs.

Unity's use of an external scripting program seemed to hurt me quite a bit, from simple things such as forgetting to save before going back to Unity (I did this way too much) to having to declare public variables in the script and then filling them in the Unity GUI rather than just doing it all by script. The editor itself also seems to be kind of heavy, I was get the spiral beach ball for a second or two every time I went between the script editor and Unity and I have a machine that can edit 8K video without proxies. These general load times and stuff like that seemed to come up regularly. Tilemapping in the tutorial didn't include autotiling, I assume Unity has this somewhere built in? Or do you need to purchase an asset to get this functionality?

On the plus side, Unity overall seemed easier to use for a non-programmer. A lot of things are done through the GUI. Animations seem easier to handle for sure. The Unity tutorial was also more written for someone that hasn't coded much as it explained what specifically the code was doing (so I assume more resources for Unity will be helpful in that way that perhaps Godot will not).

For Godot, GScript was easier to use than C#, but I do feel like it was easier to get my head around prefabs in Unity than the Node system in Godot. The Godot tutorial took almost twice as long as the Unity one, but I don't know if that's because Godot is more difficult or the combination of the Godot tutorial being more thorough (I feel like I mad an actual, if very uncomplicated game, plus I did Godot first, which probably helped me just learn more about scripting and thinking like a programmer that I took into Unity). The node/scene system seemed more difficult to get my head around than game objects and prefabs. That said, my Godot program felt very tight. There weren't things happening that I was having a tough time explaining or figuring out why they weren't working quite right, at least at the graphical level (this might have more to do with the Godot tutorial using 8-bit graphics and Unity using a more modern sprite look). Having the scripts in the editor meant I never ran into a case like in Unity where I couldn't attach code to a game object because it was failing to compile, but it was failing to compile because it wasn't attached to a game object (that headache took at least a half an hour to sort out).

Overall, I was able to finish both tutorials mostly understanding what the code I was given was doing and was able to edit it to get some different affects and kind of just play around. So, on that level, I'd say they're about equal.

One big thing I want for sure out of the engine we use is to be able to handle a lot of conversations and variables there from. We're hoping to make a "choices matter" (TM) game, and very story/dialog heavy. Ink seems like a good plug-in to do this in Unity, but implementation doesn't seem easy (though I did find a pretty good looking tutorial that may help de-mystify). Godot seems to have some assets available for handling dialog trees, but i haven't had a chance to really dig in to them yet. So, that could definitely be a decider for me: which engine has assets that make a dialog/choice heavy game easier to make.

While I had originally thought about making a tactics RPG for this project, looking around at both the Godot and Unity scenes, it seems like few people are making these types of projects that are giving out free advice on how to make them work in those engines. After talking with my team (I have a team!, see my post from a while back), it seemed like a good idea both to keep the game within the scope of a novice, but still tell the story we wanted, to do a skill role system instead. Since this came from a tabletop session anyway, seemed to make the most sense to do skill rolls rather than develop a whole combat system.

TL;DR - Looking for advice on which engine, between Godot and Unity, would be handle a 2D RPG that relies on a lot of dialog and choices along with skill rolls for the gameplay. Thanks in advance!

Tildes should support Latex using MathJax, or something similar. Having a standard implementation would also act as an easy way to use advanced formatting, while extensions beyond the _sub and ^sup tags are being made. It would greatly increase the ease of using ~math (actually ~science). Though I could be wrong, it doesn't look like $$\LaTeX$$, $$x^2$$ $\int_{x=0}^p e^{-x^2}dx$ is rendering.

Edit: Katex (not my MathJax recommendation) was a good suggestion, using a partial font download and server side html/css generation.

Edit: Also, to be clear, this is not a must-have-now feature (though that would be nice), but I would like to see support for this in the future, as latex is almost universal in the mathematical sciences.

Suppose I am trying to iteratively produce a completed image from some subset using a combination of convolutional/DNN methods. What Image norm is best?

The natural (for me) norm to ascribe to an image is to take the bitmap as a vector with L₂. If the input image is anime or something else, the uniform coloring makes this very likely to be a good fit in a low dimension - that is: no overfitting.

However: pictures of fur. Given a small square, the AI, set to extrapolate more fur from that single image, should be expected to get that stuff right next to the given subimage right, but further away, i want it to get the texture right, not the exact representation. So, if the AI shifts the fur far away from the image left by just the right amount, it could get an incredibly poor score.

If I were to use the naive L₂ norm directly, I would be guaranteed to overfit, and you can see this with some of the demo algorithms for image generation around the web. Now, the answer to this is probably to use a fourier or a wavelet transform and then take the L_N norm over the transformed space instead (correct me if I'm wrong.)

However, we get to the most complex class: images with different textures in them. In this case, I have a problem. Wavelet-type transforms don't behave well with discrete boundaries, while pixel-by-pixel methods don't do well with the textured parts of images. Is there a good method of determining image similarity for these cases?

More philosophically, what is the mathematical notion of similarity that our eye picks out? Any pointers or suggestions are appreciated. This is the last of two issues I have with a design I built for a Sparse NN.

Edit: For those interested, here is an example, notice how the predictions tend to blur details

So if there are any sailors or prespective sailors on here some questions. What region do you sail in? Captain or crew (I guess also own or crew)? What kind of boat? What level do you sail at?

This is a lot of questions I know, but I've gotten into sailing more and more over the last few years and I've found it's one of the main things that's kept me sane throughout the pandemic.

It seems like there are not that many new topics posted on Tildes, and that we could post a lot more. But I sometimes find myself reluctant to do so. Don't I post too much already?

Recently there was a survey and apparently many people think Tildes is too tech-oriented. I don't think it's all that tech oriented, not like Hacker News or lobste.rs, but that makes me a little more reluctant to post tech links. (Though, really, other people should post more of the kind of links they want to see.)

I suspect it's not just me. Periodic topics sometimes get a lot of comments. Periodic topics have been started specifically to avoid having too many top-level topics on one subject.

But, why are we avoiding this? What's wrong with posting more links? If this were a social bookmarking site, I'd be saving more links. Maybe I'd save a bunch of accordion links, without any regard for whether people are interested?

It seems like we need something like folders. When new links are posted in a folder, they don't get listed individually at top-level. You could drop a bunch of links in a folder if you felt like it, without feeling like you're monopolizing conversation, because people would have to open the folder to see what's there. Or maybe instead of folders it would be something like creating a playlist. You could start a topic that's basically a list of links, and then anyone can add links to it if they want.

It seems like groups don't really do this, somehow? They feel a bit too open and exposed. Everything shows up on the front page regardless of group. (I mean, you can filter or unsubscribe from groups, but many of us don't. Partly because they're too broad. Who's going to unsubscribe from music just because they aren't interested in some music?)

So instead we use topics and post links as comments. It sort of works, but it's given me a lot of practice at writing markdown-formatted links on a mobile keyboard, and they appear differently in search and aren't tagged.

It seems like links posted within a topic and posted top-level should be more similar in the UI. Maybe if there's some conversation about a link within a topic, a moderator could promote it to top-level? Maybe a lot of topics would start that way, and then the site would feel a bit more full.

What have you been listening to this week? You don't need to do a 6000 word review if you don't want to, but please write something! If you've just picked up some music, please update on that as well, we'd love to see your hauls :)

Feel free to give recs or discuss anything about each others' listening habits.

You can make a chart if you use last.fm:

http://www.tapmusic.net/lastfm/

Remember that linking directly to your image will update with your future listening, make sure to reupload to somewhere like imgur if you'd like it to remain what you have at the time of posting.

As part of a weekly series, these topics are a place for users to casually discuss the things they did — or didn't do — during their week. Did you accomplish any goals? Suffer a failure? Do nothing at all? Tell us about it!