The Powder Toy

Shift Unit (60Hz) by mark2222

mark2222
mark2222
17 / 0
19th Oct 2016
23rd Oct 2016
Subframe bitshift unit designed with the Column design pattern. Anyone has a traditionally implemented (DRAY-based) bitshift unit to compare with? The one in the R1 is too over-featured.
900d g0od go0d spark bitshift subframe electronics electronic 60hz

Comments

  • mark2222
    mark2222     23rd Oct 2016
    Talk about errors. I had set a column of ORs to XORs by mistake; fixed as of 23 Oct.
  • Self_Destruct
    Self_Destruct     21st Oct 2016
    @mark2222 and @LBPHacker: Lol. So true, tho. I tried attempting even the mechanism to carry the FILT ctype values up to your FRAM module, just to see if I could do it. Particle order keeps on biting me in the butt!
  • mark2222
    mark2222     20th Oct 2016
    @ChaozAirflow I find subframe easier, actually, cos it's more structured (when I was making LightPC, stuff like moving parts around or just plain modifying components would cause timing bugs to appear out of nowhere as the particle order gets messed up). I guess the difficult part of subframe is that you have to remember to reset particle order every time, and that it's hard to analyze stacked particles in vanilla TPT, both of which my mod takes care of (*shameless self-promotion*).
  • LBPHacker
    LBPHacker     20th Oct 2016
    @ChaozAirflow: Subframe tech usually either works or fails horribly, there are not many things that can cause subtle misbehaviour. Once you get a knack for it, you'll also be able to tell the difference between what you expect and what actually happens, and that's enough to find bugs. Well, most of the time. Other times you just run into TPT limitations, lol.
  • ChaozAirflow
    ChaozAirflow     20th Oct 2016
    @mark2222 I tried subframe, and it's really difficult. So I think the work of debugging will take a long time. It's really hard to find where the mistakes are, isn't it? OoO
  • mark2222
    mark2222     20th Oct 2016
    @Self_Destruct You can refer to the R1, or older CPUs if you're interested in CPU building. The method that the R1 uses, iirc, is to DRAY the result FILT into a bus. That DRAY can be activated at will. But really any method works; use your imagination. A calculator isn't really a CPU, but it's a great place to start :) Btw you might want to consider PMing ("conversations" on the website). We don't get notifications when mentioned on comment threads.
  • Self_Destruct
    Self_Destruct     20th Oct 2016
    @LBPHacker Also, if I wanted to create a simple CPU for, for example, a calculator, how would I begin and what would I do? Maybe have three diff. filt for "storage," one of which holds LHS, one holds operator, and one holds RHS, and then set up a set of electronics to redirect the FILT through a series of ops? After that, of course, I would have a 7-seg decoder output.
  • Self_Destruct
    Self_Destruct     20th Oct 2016
    @LBPHacker I suppose that makes a decent bit of sense. Would that just be a simple SWCH for the colored BRAY to pass through? Or, if that doesn't work, perhaps just a set of FILT which are changed to activate or deactivate the incoming BRAY?
  • mark2222
    mark2222     20th Oct 2016
    How did this get on fp o.O @ChaozAirflow What makes you say that?
  • ChaozAirflow
    ChaozAirflow     19th Oct 2016
    You really spent time correcting errors +1