Its amazing what a good night sleep can do. Not to mention I am getting the hang of all this logic programing.
I was working on my sram cell when it hit me (like a train:P) the most SIMPLEST method to save a bit in rom. Small, and stackable! (3x2x2). Here is a picture of it. Of course in all ram/rom technology's, its the decoder that is the pain. At the very least it makes decoding a 8 segment digt much easer.
Kind of funny but the more I look at the 4004 schematics, the more I keep thinking its posable:P Got to keep my pessimism.
Meh cannot figure out how to edit my post. Found a bug when I was stacking these things. You have to be sure that when you put down a torch, you put down a blocking block above it to block out the word enable lines.
Its funny but I have been digging around the old patents and been looking at the 4004 schematics. After fiddling with it, I figured out the basic NOR gate is a MOSFET transistor. The old wire, square, torch. Put a wire perpendicular to it (but don't break the wire) and your done.
Kind of interesting building some circuits with it.
Haven't worked on this for a while, but I got around to create a program memory and PC-incrementor. (Look back to page 39 or so to see the last version)
Just to give some credit to awesome programs: NTB Forge by aeomin MCEdit by codewarrior INVedit by copyboy
Haven't worked on this for a while, but I got around to create a program memory and PC-incrementor. (Look back to page 38 or so to see the last version)
Just to give some credit to awesome programs: NTB Forge by aeomin MCEdit by codewarrior INVedit by copyboy
How many registers is that? Have you found a more efferent and/or smaller d-latch?
It's 8x4bit registers and 16x8bit solid state program memory(switches). The D-latch is right out of the wiki, not my credit. I might change one of the registers for a 16x4bit RAM unit.
I will try not to give this up, but the remaining part is also the boring one. Making the instruction-decoder. We'll see how it goes.
Instead of building a huage ram array, you could have the computer use regesters and use rom for the instructions. I am going for a Harvard Architecture like they use in micro controllers and just registers to save on space.
Not as much time as I once had, so I thought I would scale down and just create models of ttl chips. I work with them much more than just gates so I can visualize it better.
First one, very simple too, half a 74L244. Used to turn on and off buss lines, this has protected outputs but no protected inputs. I assume you going to have line drivers latter on at any rate. Inverting the OE will change it to a 245 by adding another 4 blocks to the side, but your going to need an amp as it goes beyond the 15 block size.
Hey look people, you all need to be either working for Intel/AMD or the Cylons. This is scary... Who can I possibly trust now...
This is some seriously OCD/ADHD involved forum and as a recent survivor I offer you my services to talk over this.. Together we can discuss various treatment options including employment at high-profile companies and government agencies such as NASA and the NSA.
New clocked D latch! While technicaly its not much smaller than the wiki standard, I designed it so the clock line passes over the latch itself. This way you can stack these things in a row by copy and paste. Here are two in a row to see what I am talking about.
Humm. It looks easily enough to daisy chain a reset line as well. Maybe put a buffer at the end, the oe driving by the clock at the end of the flip-flops so at the end of all the clock cycles it will output a ready output line for the next.
Its too late right now and I need some sleep, but I swear I see a shift register in there somewhere with just an extra gate or two.
While testing my computer (page 42) , I find that the redstone is way to "noisy" to preform sequential logic in large structures like this.
If I can't make it work, I think I will try to come up with another cpu-arcitecture that relies more on combinational logic, and have a reeealy slow clock.
Similar to how DeMorgan's Law is used to reduce the number of inverters (~p^~q) = ~(pvq), I find it more efficient to use unclocked flip flops, and run the clock input through an edge trigger (turns into a pulse input) before being sent to the flip flops, since edge triggers inside clocked flip flops take up quite a bit of space when you're using a lot of flip flops.
Rollback Post to RevisionRollBack
My Youtube channel.
Contains Pachebel's Canon made with noteblocks, a working Rubik's cube made with pistons, and the ultimate TNT cannon.
You get the same noise in ttl chips. One misplaced gate or using a to slow of a chip and your whole logic can go to crap. Its why god invented buffers:) I put an add_on that puts an output buffer at the end. To be honest, the biggest challenge for me is to remember this is a chain of torches lighting up in individual clocks. With that in mine, I put a buffer on the outputs that follows the stack of flipflops. Its odd calling it an edge detector, but now I have created a row of flipflops with an edge detector:P Valid data appears only after the clock is low again and as always, data out is inverted.
You can also use the exiting clock, put another edge detector, to clock another flipflop after logic. Its how I am building my two bit adder:)
Hey guys. I'm trying to figure out how to wire something with four inputs and one output. It needs to change the output regardless of what the inputs are doing as long as one of them changed. Basically I want four doors with wired pressure plates outside with switches inside. The switches and pressure plates are connected to an AND gate so when the switch is flipped the pressure plate is disabled. It's easy enough to do it for each door but I want to connect it all together. That way any switch will disable all the pressure plates. I won't have to worry about what door I leave out of because any switch should be capable of enabling the pressure plates; regardless of the state that it and the other switches are in.
Do you want all the switches to do the same thing, or should any switch turn off all doors, but only turn on one door?
In the first case you can just connect all of the switches, adding repeters if needed.
In the second case you probobly would need a nor-SR latch with the set connected to each switch by an edge detector, and the reset connected to all of the switches by an inverted edge detector. Then make four of those, and connect Q on each of the nor-SR latches to an AND with the pressure plate, and then to the door.
That turned out to be longer than I had intended. Hope its useful if understandable!
Do you want all the switches to do the same thing, or should any switch turn off all doors, but only turn on one door?
In the first case you can just connect all of the switches, adding repeters if needed.
In the second case you probobly would need a nor-SR latch with the set connected to each switch by an edge detector, and the reset connected to all of the switches by an inverted edge detector. Then make four of those, and connect Q on each of the nor-SR latches to an AND with the pressure plate, and then to the door.
That turned out to be longer than I had intended. Hope its useful if understandable!
What I really want to do is have the four switches operate like a two-way light switch. It shouldn't matter if one is on or off if one switch changes it's state all of the pressure pads should turn off or on.
I made what you diagrammed but I don't think I did it right. The AND GATE only receives input for half a second before it goes back.
Hello all. I'm new here, as you can probably tell.
For the past day I've been trying to create a minecart station of elaborate circuitry (not nearly as elaborate as a ALU), however, after I've failed to finish it, I've come here for help.
Basically, what I want is a set of 4 radio buttons, as in, only one can be 'active' at a time. I've been using a RS NOR latch to "remember" the buttons being pressed, and wired the output to the appropriate tracks and a redstone torch with an inverter above the button so you can see that it's "active".
However, the problem is finding a practical way to "reset" the other 3 gates when a button is pressed, without resetting itself.
Effectively, I need a way for when a button is pressed, to "block" the signal coming to it's own reset gate. I'm thinking an AND gate but I'm not sure.
I'd really appreciate some help with this, or if you're feeling really generous, a better way to do it, as I believe I am doing it wrong and I've tried multiple times.
I was working on my sram cell when it hit me (like a train:P) the most SIMPLEST method to save a bit in rom. Small, and stackable! (3x2x2). Here is a picture of it. Of course in all ram/rom technology's, its the decoder that is the pain. At the very least it makes decoding a 8 segment digt much easer.
Kind of funny but the more I look at the 4004 schematics, the more I keep thinking its posable:P Got to keep my pessimism.
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
Kind of interesting building some circuits with it.
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
Just to give some credit to awesome programs:
NTB Forge by aeomin
MCEdit by codewarrior
INVedit by copyboy
EDIT: Oh, and Minecraft by Notch
How many registers is that? Have you found a more efferent and/or smaller d-latch?
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
Don't give up on this like conandrummer gave up on his turing machine!
It's 8x4bit registers and 16x8bit solid state program memory(switches). The D-latch is right out of the wiki, not my credit. I might change one of the registers for a 16x4bit RAM unit.
I will try not to give this up, but the remaining part is also the boring one. Making the instruction-decoder. We'll see how it goes.
Now i'm going to bed...
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
First one, very simple too, half a 74L244. Used to turn on and off buss lines, this has protected outputs but no protected inputs. I assume you going to have line drivers latter on at any rate. Inverting the OE will change it to a 245 by adding another 4 blocks to the side, but your going to need an amp as it goes beyond the 15 block size.
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
This is some seriously OCD/ADHD involved forum and as a recent survivor I offer you my services to talk over this.. Together we can discuss various treatment options including employment at high-profile companies and government agencies such as NASA and the NSA.
Humm. It looks easily enough to daisy chain a reset line as well. Maybe put a buffer at the end, the oe driving by the clock at the end of the flip-flops so at the end of all the clock cycles it will output a ready output line for the next.
Its too late right now and I need some sleep, but I swear I see a shift register in there somewhere with just an extra gate or two.
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
If I can't make it work, I think I will try to come up with another cpu-arcitecture that relies more on combinational logic, and have a reeealy slow clock.
This might have been harder than I thought...
Contains Pachebel's Canon made with noteblocks, a working Rubik's cube made with pistons, and the ultimate TNT cannon.
You can also use the exiting clock, put another edge detector, to clock another flipflop after logic. Its how I am building my two bit adder:)
PS - Going to do this last night but burned out:P
My new version of Redstone Simulator
Main Code Site: http://code.google.com/p/red-stone-simulator/
Any help would be much appreciated.
In the first case you can just connect all of the switches, adding repeters if needed.
In the second case you probobly would need a nor-SR latch with the set connected to each switch by an edge detector, and the reset connected to all of the switches by an inverted edge detector. Then make four of those, and connect Q on each of the nor-SR latches to an AND with the pressure plate, and then to the door.
That turned out to be longer than I had intended. Hope its useful if understandable!
This is probobly not what you want though. It would anyway require alot of space, time and redstone!
What I really want to do is have the four switches operate like a two-way light switch. It shouldn't matter if one is on or off if one switch changes it's state all of the pressure pads should turn off or on.
I made what you diagrammed but I don't think I did it right. The AND GATE only receives input for half a second before it goes back.
Sorry if it's hard to make any sense of this.
For the past day I've been trying to create a minecart station of elaborate circuitry (not nearly as elaborate as a ALU), however, after I've failed to finish it, I've come here for help.
Basically, what I want is a set of 4 radio buttons, as in, only one can be 'active' at a time. I've been using a RS NOR latch to "remember" the buttons being pressed, and wired the output to the appropriate tracks and a redstone torch with an inverter above the button so you can see that it's "active".
However, the problem is finding a practical way to "reset" the other 3 gates when a button is pressed, without resetting itself.
Effectively, I need a way for when a button is pressed, to "block" the signal coming to it's own reset gate. I'm thinking an AND gate but I'm not sure.
I'd really appreciate some help with this, or if you're feeling really generous, a better way to do it, as I believe I am doing it wrong and I've tried multiple times.