Help Fixing An Old Save (Electronics)

  • cyberdragon
    5th Nov 2021 Member 0 Permalink

    One of my saves from 2013 is now boken as it seems SPRK can jump larger gaps than before (or something), causing it to short out. I've been trying to fix it, but I can only get it to work once before breaking again. The problem is the diode under production gets sparked when the titanium cray to put the first lead on it fires, then the CRAY producing the second titanium lead fires twice deleting the titanium so when the pistons move it, there's only insulator on one side. The diamond marks on the delay line are the offending signals, there was originally only one but I made a second pulse which allows it to work once, but then it ends up back in the same state.

     

  • jacob1
    6th Nov 2021 Developer 1 Permalink
    I looked into it. The code is working as intended. This portion of the save is just sloppy, which is causing you issues and giving you trouble determining the real problem.

    Stuff like this interests me, we rarely break saves and I'm always interested in finding out what caused a problem. So I compared current vanilla with a very old version of tpt I had on hand (89.2). The issue is that the WIFI is sparking the PSCN, as it should, but in older TPT versions for whatever reason it does not. I am unsure what was changed to cause this, but it is working fine today.

    Here is what it does in the original:
    image
    And here is what it does today:
    image

    You have the WIFI on the right box connected to your timer. If you remove the WIFI on the right CRAY box, the save works correctly. The right box is being sparked due to your timer mechanism which sparks all the WIFI in a row. In older TPT versions, because of the feedback loop connecting the first CRAY box to the second CRAY box, the PSCN doesn't get sparked when the timer tells it to, but in modern TPT it does.

    EDIT: Sorry I recorded the "today" version of the gif on a different version of your save as the other gif. In the original version of your save, you linked up the WIFI timer once, but in the most recent version, you linked it twice, with DMND below to denote it. This is why it's sparked twice. Everything I said still applies, if you were to remove these from the timer, the save would work fine. I recommend doing some cleanup to ensure the electronics do exactly as you intend.
    Edited 2 times by jacob1. Last: 6th Nov 2021
  • EnganK
    6th Nov 2021 Member 0 Permalink

    By the way, how was this GIF (and GIFs on the wiki) recorded?

  • jacob1
    6th Nov 2021 Developer 0 Permalink
    @EnganK (View Post)
    The majority of the gifs were recorded by @Catelite, and I can't say how those were done.

    Modern gifs were recorded by me, using peek on Linux. This is a Linux-only tool, so if you use Linux, then it's a great solution, otherwise I don't have a good answer.

    Several years ago I experimented with ffmpeg, with the intention of adding native gif recording into TPT. The plan was to do it like modern consoles, which let you record into the past. I experimented with ffmpeg's C API and was able to turn tpt's raw recording output into a gif. I never got around to implementing it into TPT though, it was all experimental.
  • cyberdragon
    11th Nov 2021 Member 0 Permalink

    @jacob1 (View Post)

     Fixed it. I moved the placers up one and changed the BRAY cleaning beam from piston driven to a static location (which cuts a whole bunch off the controller).