My new calculator
-
@billion57 (View Post)
Yes, unless it is organised into something that may visually resemble a circuit (space efficient, consistent and logical arrangement).
The only true drawback is the space. -
the only thing i don't really like how the data is allready programmed into it and just checks it when you ask it stuff
-
@Minishooz (View Post) thanks. I hoping that I can somehow make a more compact subtractor modeled after my adder that doesn't have to be so precise with timing so I'll be able to get rid of some stuff and have room to try to add multiplication (I found a multiplier online that uses logic gates), and it would hopefully let me use signed binary easier (atm i'm not gonna change it to signed binary)
@The-Con (View Post) The thing I don't like about using inst instead of wifi is that I'm not a very organized person so it wouldn't look very professional. I was also hoping that by labeling, that would help people follow what was going on
@vanquish349 (View Post) If you're talking about where it stores what's going to be calculated, I did that just because it was the first, and only, thing I could think of to ensure that when the numbers are sent to the subtractor, it's done simultaneously since if it's off by even a few frames, the whole thing wont work right. -
@Niven
Easy fix.
If you want the signals the subtraction to work without the inputs being simultanious (assuming the problem is that if the inputs are not simultanious, the subtracter initially outputs the wrong number, and then eventually the right one after it sorts the shit out) all you need to do is create a timer on the outputs of the subtractor. Instead of sending the outputs directly to the next part of the calculator, make an AND gate. Basically you have one AND gate on each output bit. The outputs from the subtractor should each connect to the AND gate (it should be a two input AND gate). Now all you need to do is make a wire that connects to each of the vacant inputs of all the AND gates. Connect a delay mechanism to that wire (water or delay element or something). The delay should be long enough so that the subtractor is able to finish calculating the result properly. In order to trigger the spark that starts the delay, just make a wire going from the very beginning of the subtractor that loops around and connects to the delay. This way, the second your calculator gets to the adder/subtractor, it will automatically spark the delay and the outputs won't pass onto the next part of the calculator until the delay is finished. -
@Synergy (View Post) Unfortunately, I'm not sure if that'll work. The problem you assumed isn't the right one. If the sprks aren't simultaneous then it just outputs the wrong number since the gate inputs will be off causing them to have wrong outputs(mostly just with the xor gates). That's why it took me so long to get this right (over 2 weeks) and why some parts of it may look a little weird compared to others. I had to test, modify, test, modiy, test, modify, test, modify and so on until every single possible subtraction problem was timed perfectly for the subtractor not to mess up. Like if any part of it slowed down by as little as 1 or 2 frames, then some problems will give the wrong answer. It's not so much as the sparks have to be simultaneous but that it's set up to where it works when they are. It could be set up otherwise but then the sparks would have to be however far away for it work.
-
Just a single spark.Edit:Just thought of something. I could have swch at each input so when the first number is sparked, the swch next to the A's that make that number are sparked, then when the second number is sparked, the inst next to that B's that make that number are sparked, then equal would send spark to all inputs and only the ones being used will recieve it. It could work. And since it would get rid of the place where the problem is stored, I might decide to add a bit to subtractor so that I could use signed binary instead of unsigned binary with the converter.
-
Yeah that would work for single sparks. The spark will turn on the switch at each input bit. And then you just trigger something that sends a spark into every switch. That spark will continue on if the switch was turned on earlier by the sparks. I have done this multiple times before.
I will make a quick save illustrating this in a sec.801530View Save 801530
This is what I meant. I wrote a little in the description also.
