Improvements made based on feedback
English + Korean support has been applied. English is the default. There was a point raised in the feedback that the translator did not work properly, making it difficult to read. I apologize for not being considerate of you all.
A guide panel has been added. It was added because feedback suggested that it would be helpful to have text explanations of how it works. A brief explanation of the Enigma and its principles has been added using a translator.
A performance mode has been added. I received feedback that the browser was lagging, so I was struggling to figure out how to handle it. Depending on the mode, I lowered the pixel ratio or optimized the wiring calculations.
The logo engraving has been changed. The black background plate has been removed, and a transparent texture with a dark groove and a bright rim has been created using a dual pass to make it look like it was carved into wood. To be honest, it still hasn’t reached the level I wanted, but I have wasted so much time here that I am honestly exhausted and plan to postpone it until next time.
Rotation lock and camera presets have been added. I simply did exactly what was requested in the feedback, so there isn’t much to say about it.
The crack mode was created simply because I was caught up in the strange idea that if there is an original cipher, there should also be a decryption function. It was the absolute hell; I will never do it again.
Web workers use the same WASM engine as the simulator to [set] initial positions Scoring through brute force is difficult, involving the Interval of Concordance (IoC) as the primary axis plus the chi-square auxiliary for character frequency, plaintext previews for the top 5, and a button to apply directly to the machine…
Trial and error: When I scored only the first 48 characters of each candidate in the full scope, a completely wrong rotor order happened to get a higher IoC and beat the correct answer (0.0677 vs 0.0651). However, when I selected the top 8 from a 64-character sample and re-scored and reordered the entire sentence, the correct answer won by a landslide (0.058 vs 0.012).
Complete reconstruction of X-ray wiring. Honestly, I left the existing wiring as it was because it was a hassle, but it was tough working on it this time. I roughly standardized the thickness and connected the wires properly. Honestly, I still need to tidy it up a bit more, but I’m exhausted.
Lastly, since some people claim my devlog was written by AI, I intentionally didn’t write a title or use Markdown this time. However, I’m not sure what the best approach would be. Please give me lots of feedback in the comments.
Comments 3
I find your project to be very impressive. As far as the develop is concerned I believe your quality of devlogs was so high people considered it AI and also you have mentioned before that you use translator so it looks like AI but I believe your language is very polite, professional and soft spoken which makes it feel like AI but it is really nice. I personally prefer you having your own style of devlogs but if other people are saying that it looks like AI and you believe you don’t want that then you can change. Otherwise I would prefer your previous style
@Rezz Thank you so much for leaving feedback. I will stick to the previous style, but I will try to check the translation later as much as possible to make it look more human lol.
@reimunyancat Yes I really like your devlog style I am glad you chose to keep it. It just shows how much effort you put into it. Also your projects are very high quality already so having high quality devlogs is also a must.
Sign in to join the conversation.