60Hz Processor Design in Factorio

2023-05-07

The following is written for my WKRPT-300 report.

  1. Summary
  2. Introduction
  3. Problem Definition
  4. Detailed Processor Design
  5. Conclusion
  6. List of Instructions and Registers
  7. Tools Used
  8. Code Demonstration
  9. Processor Blueprint String
  10. Related Projects
  11. References

Summary

The Turing complete circuit network in Factorio has inspired its players to make many fun creations over the years. One of the popular circuits that were created are general purpose computing circuits or CPUs. This report describes a CPU design in Factorio that achieves an operation speed of 60Hz (which is the fastest possible speed to run in game) using a custom EPIC style RISC instruction set and Harvard architecture.

Two primary constraints when designing are the game update speed and number of connections per combinator. Extra combinators often needs to be used to get around the connection limit without additional penalties in time.

The core of the CPU are registers. These registers can latch data from different sources and from each other using control signals. All the mathematical operations are done in the ALU. In this design, the ALU can be used by writing values to operand registers and reading the result of the operations back to a register.

Unconditional jumps is identical to a mov instruction with destination being jmp. A conditional jump requires additional circuits to determine of a jump should be performed. Both type of jump has a delayed branching, where the instruction immediately behind the jump instruction is always executed. This allows programmer or compiler to minimize pipeline stalls in absence of branch prediction.

The code of the CPU is stored in ROM constructed using constant combinators. Instructions are stored in them as decoded signals to remove the need of decoder in the circuit. The design follows the Harvard architecture where the instructions are on a separate address space as other peripherals. This allows instructions to be fetched in parallel to normal data access on the bus.

The CPU is connected to a bus controlled by a few registers. A RAM, a display, and a user input device are connected to the CPU via this bus. This allows the CPU to communicate with the user and also store data that does not fit within the limited registers.

An assembler written in Python is made for this CPU. It supports special syntax for specifying parallel instructions, as well as generating ROM for both instructions and data. The assembler also decodes the instructions into signals directly. The output of this assembler is a Factorio blueprint string containing constant combinators as ROM.

Introduction

The nature of Turing completeness in Factorio has invited many processor designs being created in the game [1][2][3], with each design having different design goals and constraints. This report introduces yet another processor design that focuses and execution speed, while keeping the instruction set usable for general purposes.

Factorio Circuit Network

Factorio is a game about building factories. In the game, the player need to manage resources and build processing facilities to achieve the end goal of the game [4]. One of the aspect in the game is its circuit network, allowing players to control the operation of their factories. The core part of the network are arithmetic combinators, decider combinators, constant combinators, and two colours of wires carrying multiple channels of data between combinators. A detailed explanation of circuit mechanics can be found in my previous report [5].

Being an open ended game, many players have dedicated majority of their play time creating logic circuits in the game. Creations such as Tetris, Pong, snake, and even a 3D renderer have all been made before [6][7][8]. The ability for the circuit network to simulate any sequential logic naturally leads to the development of general computing circuits, also known as CPUs.

CPU Basics

A central computing unit, CPU, are designed to handle wide variety of general computing tasks. CPUs are distinct from most sequential logics in that it does not require the connections between logic elements to be rewired to solve different problems. Instead, a series of instructions, or software, can be written for each purpose.

Clock Speed

A big factor that determines the speed of real life CPUs is the clock speed. A clock in digital circuit is used to synchronize data transfer between registers; higher the clock speed, the faster data can move inside the circuit. In real life, CPU clock clock speeds can go up to GHz with thermal dissipation being the main limiting factor alongside with other factors like transmission delays [9]. In Factorio, it is different, as the default update (or tick) speed in the game is limited to 60Hz only, the fastest possible CPU design in the game would consequently be limited to this speed. On a positive side, since all data gets updated synchronously at the same speed, it removes the need of an explicit clock signal.

Execution Cycle

A typical CPU follows a fetch-decode-execute execution cycle. Each cycle, the CPU fetches the next instruction from the memory, decode it into control signals, and execute the instruction based on the decoded control signals. For this Factorio CPU design, the capability of holding multiple signals in a single wire is utilized to store decoded instructions directly in memory storage, simplifying the cycle to be fetch-execute only.

Processor Architecture

Reduced instruction set computer (RISC) and complex instruction set computer (CISC) are the two instruction architectures used in CPUs. RISC CPUs support less instructions, but each instructions takes less cycles, and requires less resources to complete decode. As opposed, CISC CPUs have more complicated instructions, requiring more resource to decode and cycles to complete, but reduces program size. Since instruction decoding is offloaded to the assembler for this Factorio CPU, either architecture can be implemented without any hardware change.

Regardless, a RISC architecture is still used for this design. Since RISC is a load-store architecture, where accessing memory is done through explicit instructions rather than implicit addressing mode, the program can better optimize memory access in the code [10]. This is advantageous particularly in this system, as the memory speed is also limited to game's 60Hz update speed, unoptimized memory access can slow down the program significantly.

Explicitly Parallel Instruction Computing

EPIC style architecture is a design philosophy to achieve instruction level parallelism (ILP). The code idea of EPIC is to move the complicated work of ILP, like instruction reordering, instruction dependency, to software (ie, the compiler), greatly reduce the complexity in the hardware [11]. For this CPU design, since the program in ROM is already decoded, executing non-conflicting instructions can be stored in the same address to be executed in parallel. This way, ILP can be easily be achieved using special assembler syntax. For example, the inc a instruction and mov b 0 do not share the same decoded control signal, therefore both instructions can be stored in together, and gets loaded and executed simultaneously.

Problem Definition

This section describes the problem definition of the report, including the design goal, constraints, and criteria of the design.

Design Goal

The goal of this report is to demonstrate a CPU design implemented in Factorio circuit network, optimizing cycles (ticks) per instruction and parallelism to minimize the impact of the slow game update speed. The design does not attempt to emulate any existing ISAs, or have high level programming support. Only a low level assembler will be implemented to create simple programs that can be run on the CPU.

Constraints and Criteria

There are two main constraints when implementing digital circuit in Factorio.

First is the update speed. As mentioned in Clock Speed, the default update speed in game is only at 60Hz. It is the fastest rate any object in game can can change state, including the circuit network. While the game speed can be increased through game editor (and it is a useful feature to run the game tick by tick for debugging), doing so does not change the relative speed between the game environment and the circuits. As the CPU is required to run at this fastest speed, every part of the circuit needs to be designed with this speed in mind. For more, the number of combinators a signal goes through should also be minimized to reduce latency.

Second is the limited number of wire colours. In the game, when two wires of the same colour connected together, they carry the same signals; on contrary, two wires of different colour can connect together while carrying independent colour. This is especially limiting when signals coming from different sources needs to be fed into one combinator. In low speed logic, this can be easily solved using another combinator to isolate the signals, but for high speed circuits, often duplicated combinators and tricks are needed to implement the same logic without time penalties.

Detailed Processor Design

The processor design is inspired by Ben Eater's Youtube playlist Building an 8-bit breadboard computer! [12]. The core of the processor consists of numerous registers, some of them are used to control other components of the processors, like the arithmetic unit and data bus. The instruction ROM is on a separate bus and address space than the data, allowing instructions to be read continuously in parallel with memory access from the code.

Figure 1 shows the screenshot of the overall design in the game.

Figure 1 Overall Processor in Game

Registers

The core of the processor is made of registers. All of the instructions specify how the data is moved around these registers. Each register has their own control signal to specify conditional loading from different sources, like other registers, bus data, arithmetic result, etc. A list of registers and all supported data source can be found in Table 1 and Table 2.

Registers a, b are arithmetic operands. They are connected to the input of the ALU, allowing arithmetic operations and comparisons to be done on these two registers.

wa, wd, ra are bus registers write-address, write-data, and read-address. By setting appropriate values to these registers the code can read and write values to the bus, and through it access peripherals like the RAM, ROM, display, etc.

x, y, z are general use registers and can be used for anything.

jmp is the jump register, also known as program counter. This register is different from others that ever tick it automatically increments when no control data is present. This register is connected directly to instruction ROM to fetch the next instruction to execute.

Figure 2 shows a general construction of a register.

Figure 2 Sample Register Construction

Both the red and green wire on the output side contains the value of the register. Some decider combinators, like A = -1 in this example, are duplicated to work around the limited number of wires. This control increments the value stored in the register, thus requiring a feedback input and an external D ⇐ 1 input. On top of the control input, 3 total inputs would be required, and since the number of wires per combinator is limited to 2 only, a second combinator is needed to achieve the desired behaviour without additional delay.

The duplication of only some combinators causes a problem. For starter, the control line can also carry the data signal D. Without proper filtering, undesired value can "leak" through the D ⇒ D conditions in the combinators and corrupting the register value. Filtering can be easily implemented by adding a signal D = -D arithmetic combinator in parallel, effectively subtracts the undesirable data out. However, with some combinators being duplicated, the "leaked" amount would vary depending on the value of control signal A. To solve this a second D = -D filter and a A ≥ 0 decider combinator are added. This way, non-duplicated control signals can be assigned with values A > 0, canceling the effect of the second filter. For duplicated control signals assigned to have A < 0, the second filter becomes in effect and filter out the additional leaked data correctly.

One exception to the above is the load-immediate control, A = 1 in this example. In this case, the data signal D on the control line is desired, therefore duplicated combinators are used but it has a non-negative control signal value. It can also be viewed as the first combinator cancels the filtering, while the second combinator outputs the data.

An absence of control signal indicates the data in the register should be held the same. This is easily achieved by one decider combinator with A = 0 condition with a feedback input.

The register data is available on both red and green wires on the right. Being able to access the register from either wire is important to the system, as it allows direct connections from the register to other components of either colour as input without additional combinators.

Some typical operations for this example register:

Note that the control signal for this example register is A, but each different registers would have a different signal. The signal D is used as data for all registers. For more, there can be more data sources than this example by adding more decider combinators in parallel with different control signal conditions.

Arithmetic Logic Unit

The arithmetic logic unit (ALU) handles all the mathematical operations in the processor. All the operations supported by Factorio arithmetic combinators are implemented, this includes addition, subtraction, multiplication, division, remainder, exponentiation, bitwise and, bitwise or, bitwise xor, and comparisons (><=≠≥≤). Unlike most instruction sets where there are dedicated instructions for these operations, in this design the result of each operation are computed constantly from registers a, b, and their results can be read back to any register when finished. This design makes it easy to pipeline same mathematical operations in the code, minimizing stalls from waiting for calculations to finish.

A sample construction of the ALU can be found in Figure 3.

Figure 3 Sample ALU Construction

For most operations, a minimum of 2 combinators are needed, one to convert data signal D from the two input into different signals A, B, so they can be fed into a second combinator to perform the actual operation. This introduces a 2 tick delay to perform these operations. Some optimization can be done using Factorio's implicit addition. Addition can be done without any additional combinator by having the destination register load both register a and b together. For subtraction, one of the input signal would need to be inverted, and the other delayed to match the timing. Adding the resulting signals together effectively produces the difference between the two. This results in one tick delay for subtraction, which is still better than the 2 ticks delay of other operations.

Jump Logic

Jump instructions allows control flows, usually in the forms of if statements, loops, function calls at the higher level language, to be implemented. They can be categorized into two types: conditional and unconditional jumps.

Unconditional jumps are simple move instructions to the jump register. The implementation of it is identical to any register load as described in Registers. Any value loaded into the register would get used next tick to fetch the next instruction.

Conditional jumps requires more complexity. The processor would need to select the condition to use, evaluate it based in registers, and use the result to determine if a jump should be performed. Figure 4 shows its construction.

Figure 4 Jump Logic

In this example, signal J is used for jmp register control, so for example, loading J = 1, D = 10 would load 10 into the jump register, effectively jump the program execution to address 10. A construction similar to other register, using signal K, can be seen in the figure. The difference being that it operates on the jmp register control signal J instead of the usual data signal D, and the output of combinators are connected directly to the control line. The input of this construction are connected to an array of combinators that outputs J = 1 when different conditions are met. Together, the signal J=1 is conditionally asserted to the control line, which in turns creates conditional jump.

For example, to jump to address 10 when register a is 0, one would send control signal K = 20. If the condition is true, J = 1 would be output by the D = 0 combinator, and gets passed to the control line by the K = 20 combinator. If in the following tick, another D = 10 is asserted to the control line, its value would be loaded into jmp and the jump succeeds. On the other hand, if the condition is false, then the K = 20 combinator would output J = 0, and a lone D = 10 on the control line would have no effect.

Both conditional and conditional jumps implements delayed branching, which comes inherently from the fetch-execute pipeline. The instruction immediately behind the jump instruction (the delayed slot) would be executed regardless if the jump is taken. This is also a common approach used by RISC many processors to allow the compiler or programmer to specify static jump prediction and minimize pipeline stalls in absence of a complex branch prediction circuit [13].

Program Storage

The program is stored in read only memory (ROM) constructed using constant combinators. By using the property that multiple signals can exist on the same wire or combinator, each constant combinator can store the decoded form of one or more instructions. This results in a compact and high instruction throughput ROM design. This design follows the Harvard architecture, where the instructions are on a separate address space to other peripherals [14], eg. RAM, with a direct connection from jmp register to the ROM addressing line. This way, the instruction can be constantly fetched every single tick in parallel to data access. Figure 5 shows the construction of ROM containing program that computes the Fibonacci sequence.

Figure 5 Sample Program Stored in ROM

This program consists of 6 instructions occupying 4 ROM cells (or 3, as the last noop is not really necessary). Three of the instructions do not have any conflicting signals, so they resides in the same ROM cell. The jmp register increments by one by itself each tick, loading decoded instructions in each cell sequentially to executed. The three instructions in the same cell would be executed in parallel.

The decider combinators are used to activate each cell based on the value of jmp. Note the conditions are not sequential like D = 1, D = 2, .... This is because the D signal from the constant combinator adds to jmp. This extra value needs to be compensated in the decider combinator conditions. For example, the first cell contains the signal D ⇐ 1, and the condition D = 2. This way, when jmp contains D = 1, it adds to the content of the constant combinator to D = 2 and activates the cell.

The constant combinator in the last cell contains no signal. It is to represent the noop instruction. In practice, it is not required and the program would run the same way without it.

Peripherals

The registers, ALU, and jump logic forms the main part of the computer, while other components, like RAM, display, and input device are peripherals. The processor controls and communicates with the peripheral through a peripheral bus. The bus operates full-duplex at one read/write per tick (60Hz) using 4 wires. Three of the wires connects to the wa, wd, ra registers, and the 4th wire connects one of the register inputs. Values can be written to peripherals by writing address to wa, then data to wd, and read from them by writing address to ra. The specific values and timings differs between each peripheral.

Memory

An exact copy of the instruction ROM described in Program Storage is also present on the data bus. This ROM can be used by the assembler to store data that are alongside the program. Since ROMs are read only, only the read address and read data lines are connected.

For read-write memory, a random access memory (RAM) is available. The RAM design used is based on my previous report Multi-port Random Access Memory in Factorio [5]. The RAM is wired to all 4 write address, write data, read address, and read data lines, providing interface for the CPU to access the data. Figure 6 shows a RAM construction, including the first and last memory cell.

Figure 6 Sample RAM Construction

On the top, a series of A + 1 ⇒ A combinators connected together to automatically assign each cell a unique address, which can be offset using the first constant combinator (10 in this case). This combinator string makes expanding the RAM as easy as appending the same cells to the end.

As with registers and ROM, a -D ⇒ D combinator is used in the circuit to filter out undesired data signal D. To ensure that the filter only applies to the address range of the RAM, two D < A and D > A decider combinators are connected to the first and last address of the RAM. The two combinators cancels the effect of the filter when non of the RAM cells are addressed thus no undesired data is emitted.

The core of each cell is a W ≠ A combinator with a looped-back input, which forms a register. Data is kept in this register as long as the W signal is not the address of the cell. When it is, the register is reset and new data is loaded via W = A decider combinator.

A D ⇒ W register at the beginning of the cell converts D signal on the wa line to W signal. This is required so the address signal does not add to the wd write data signal, corrupting them both. As a side effect, when writing to the RAM, write address will need to be available one tick before write data. Since loading immediate values to registers often contain conflicting control signals, not requiring two registers to be loaded in the same tick can actually simplify the code.

The value of the core register connects to a D + A ⇒ B arithmetic combinator. The B signal is used to compensate cell addressing for the output. Similar to the ROM, the D = B combinator that outputs the cell value based on read address receives D signal from both the address line and the stored data. By using the sum of stored data and cell address for comparison, the cell can be correctly addressed. A dummy D ⇒ D combinator is connected in parallel to synchronize the stored data with this sum, ensuring the stored data and the new corrected address value reaches the output combinator at the same time.

Graphical Display Output

The graphical output of this computer are built using arrays of lamps. There are two parts: the display controller and the lamp array. The controller contains a bitmap decoder, bitmap RAM, and colour decoder. The lamp arrays consists of 64 memory cells with 29 lamps connected to each. Both the RAM and lamp array registers can be written to from the bus, but can not be read.

A 32 bit integer can store any glyph that needs to be displayed in a cell. The bits are decoded into signals 0, 1, ..., 9, A, B, ..., Z, where the sign bit (which is also the MSB) of each signal determines whether a lamp should turn on or off, and all other bits are ignored. This reduces decoding into single bit shift combinator for each signal. Testing the sign bit can be done by checking if it is less than 0.

The decoded bitmap is stored in a multi-signal RAM [5]. This RAM variant allows multiple signals to be stored at the same address. When data is written to the display cells, the lower 28 bit of that data is used as address to look up corresponding bitmap from the RAM. The upper 4 bit of display data is used to carry colour information. By shifting the bits down and compare it to a lookup table, the special colour signals are output to change the colour of the cell.

In addition, the beginning of the RAM contains a register to hold an address offset. This is useful for storing ASCII bitmaps, where the first printable character does not have a value of 0.

Figure 7 shows the display controller.

Figure 7 Video RAM and Graphics Controller

On the top left of the diagram, the arithmetic combinators with bit shifts are the bitmap decoders. They shift the input data such that the sign bit of each output signal contains each bit of the input data.

The second column at the top of the diagram is the RAM offset register. It contains a writable register with its value negated and added to the address line of the bitmap RAM, effectively offsets the base address of the RAM.

The right part of the diagram is the bitmap RAM. It is similar to the general use RAM described in Memory, except it operates on the special ALL signal, with two additional combinators -X ⇒ X and -W ⇒ W filters to cancel the undesired data in the output from the use of ALL.

The lower part of the diagram contains combinators used to delay and synchronize write address from the bus, so the decoded bitmap and display address output at the same tick for the display cells.

Figure 8 shows one cell of the display.

Figure 8 Lamp Display Cell

The display cell is a simple writable register with hard-coded address (0 in this example). The register uses the ALL signal and can hold multiple signals, thus a -W ⇒ W filter is added. The register is connected to 29 lamps, each of them tests the sign bit of different signals by checking if they're negative. The ALL signal also forwards the special colour signals as well, which the lamps reads and emit the corresponding colour in game.

Figure 9 shows the display working in game.

Figure 9 Lamp Display in Game

User Input

There are many ways in Factorio for players to interact with the circuit network, including going through gates, rotating inserters, and dropping items on belts or chests, or editing the combinators directly. Currently, a chest based solution is included that allows the player to trigger input values by dropping items in different chests. The in-game view of this chest input is shown in Figure 10.

Figure 10 User Input with Chests

For this design, active provider chests are arranged in a grid. A roboport is placed nearby containing logistic bots to remove items dropped by the user. Red wires connect chests in each column and green wires connect chests in each row. The chests are set to output signal corresponding to their contents. Each row and column are connected to a rising edge detector. This way, when the number of items in a chest increases, signal is emitted containing information on the row and column of the chest. The design of the edge detector is shown in Figure 10.

Figure 10 Rising Edge Detector

The combinators at the bottom receive the latest signal from the chests and a delayed signal from the top combinator. When the chest input is incremented, it becomes greater than the delayed value and a signal A is emitted. The value of A is decremented by 1 per column, and by 8 per row. The value is always negative to avoid triggering the EACH > C condition by A signal. The input register is shown in Figure 11.

Figure 11 User Input Register

There are a few extra properties for this register: the signal stored is different than the signal input, the value stored is the negative of the input value, and it clears the old value automatically when new signal is available.

The A = 0 combinator with loop back input stores the value of the register. When signal A contains any non-zero value, the combinator is disabled and value is cleared. The -A ⇒ D combinator then emit the converted value in D signal which gets stored as new value when input A signal is deasserted.

The D = W combinator is used to reset the register. When the condition is triggered, a A = -1 signal is emitted and it disables the A = 0 register, effectively clears the register. This reset input is connected to the same addressing line as the read addressing combinators (D ⇒ R). This way, the register is cleared when the value is read. The constant combinator W ⇐ 1000 is used to set the address of the register.

Assembler

The purpose of an assembler is to takes a list of human readable instructions and convert them into machine readable format [15]. For this report, a simple assembler is written to generate a Factorio blueprint string containing decoded instructions in ROM cells as described in Program Storage.

Different sections of the assembly file can have different modes, separated by .<mode> directive. Two modes .data and .code are supported. In .data sections, each line is interpreted as constant values. These values are stored in the data ROM on the peripheral bus and can be accessed using the ra register.

In .code sections, each line is interpreted as instructions. Some instructions can contain operands to specify the source and destination of the operation. A list of instructions can be found in Table 1.

The assembler supports special syntax to better aid the EPIC philosophy the processor follows (see Explicitly Parallel Instruction Computing). All instructions with destination operands can have multiple registers as the destination. For example, the instruction inc a,b increments both register a, b in one operation. In addition, an instruction can be specified to run in the same tick as the next instruction by appending a backslash \ to the end, provided they don't contain conflicting control signals.

An example of the assembly program can be found in Code Demo.

Conclusion

The versatility of Factorio's circuit network makes it possible to create a general computing circuit. However, challenges still exist in implementations, considering the constraints given. Factors like clock speed and wire limit create interesting engineering problems. Techniques such as parallel processing and circuit duplication need to be employed to maximize the execution speed.

By using a custom processor architecture and instruction set, it is relatively easy to create a fast circuit with relatively few resources. As consequence, existing compilers can not be used for high level languages such as C. For more, it lacks hardware support for a lot of common features such as stack pointer or context switching. Overall, this report is created for demonstration and education, and the design may not be practical for any use in game or in real world.

For future work, defining an application binary interface (ABI) for the processor would be the foundation for any more complex programming. Only with a standardized interface for function calls, parameter passing, stack usage, etc., can different compilers tool chains target this custom architecture.

List of Instructions and Registers

Table 1 List of Instructions Supported by Assembler

Instruction Description Note
mov dst src Load from data source src to registers dst dst and reg can have multiple registers sepearted by commas
inc reg Increment value in register reg Same as above
dec reg Decrement value in register reg Same as above
j loc Unconditional jump to location loc
j[cond] loc Conditional jump to location loc.
cond can be z, nz, eq, ne, gt, lt, ge, le		 z, nz takes effect 2 ticks after register a is changed. The rest takes 3 ticks after register a, b are changed.

Table 2 List of Registers

Register Signal Function Note
a A Arithmetic Operand
b B Arithmetic Operand
x X General
y Y General
z Z General
ra R Bus Read Address
wa W Bus Write Address Write to wd in the cycle following corresponding write to wa
wd L Bus Write Data Same as above
jmp J Program Counter Used by assembler to create jumps. Not usually used directly.
cj K Jump Control Same as above
cd C Data Control Same as above

Table 3 List of Data Source

Source Description Note
Immediate Immediate value Can be a literal, a constant, or a label
a Register a
b Register b
x Register x
y Register y
z Register z
rd Bus Read Data Different peripherals have different timings. For RAM, it is available 2 ticks after ra is changed. For chest input, 3 ticks.
add Result of a + b Available 1 cycle after a, b are changed.
sub Result of a - b Available 2 cycle after a, b are changed.
mul Result of a * b Available 3 cycle after a, b are changed.
div Result of a / b Same as above
mod Result of a mod b Same as above
exp Result of a^b Same as above
ls Result of a << b Same as above
rs Result of a >> b Same as above
and Result of a ∧ b Same as above
or Result of a ∨ b Same as above
xor Result of a ⊕ b Same as above
eq 1 if a = b else 0 Same as above
ne 1 if a ≠ b else 0 Same as above
lt 1 if a < b else 0 Same as above
gt 1 if a > b else 0 Same as above
le 1 if a ≤ b else 0 Same as above
ge 1 if a ≥ b else 0 Same as above
zero 1 if a = 0 else 0 Same as above
nero 1 if a ≠ 0 else 0 Same as above
inc Increment Used by assembler for inc and dec instructions.
For example, inc a is equivalent to mov a inc.
dec Decrement Same as above

Tools Used

The mod Better Lamp Contrast by Purpzie is used throughout the development and testing for this report. The mod can be found at https://mods.factorio.com/mod/lamp-contrast.

Python 3 is used throughout the development to automate some of the circuit building and to assemble and generate code as blueprint strings.

The in-game /editor command is used to debug the circuit and code, as well as providing infinite resource for building.

Code Demonstration

This simple chess program demonstrates the usage of different components in the system and assembler syntax. The program copies chess piece bitmaps and initial position to the VRAM and RAM, then polls for user input in a loop, checks and perform a move by reading and updating content in the RAM. A video demonstration of the program working can be found on YouTube.

The source code of the program follows:

; memory locations
#vram 800
#disp 900
#input 1000

; colour
#pink (5 << 28)

; chess piece definitions
#bk (1 | pink)
#bq (2 | pink)
#bb (3 | pink)
#bn (4 | pink)
#br (5 | pink)
#bp (6 | pink)
#wk 1
#wq 2
#wb 3
#wn 4
#wr 5
#wp 6

; variables
#pos 400

.data
initial:
#br #bn #bb #bq #bk #bb #bn #br
#bp #bp #bp #bp #bp #bp #bp #bp
0   0   0   0   0   0   0   0
0   0   0   0   0   0   0   0
0   0   0   0   0   0   0   0
0   0   0   0   0   0   0   0
#wp #wp #wp #wp #wp #wp #wp #wp
#wr #wn #wb #wq #wk #wb #wn #wr

.code
; reset regiters
mov a,b,x,y,z 0

; move bitmaps to vram
mov a 12
mov ra bitmap:
mov wa #vram+1
j memcpy:
mov z .

mov wa #vram
mov wd 1

; move initial position to ram
mov a 64
mov ra initial:
mov wa #pos
j memcpy:
mov z .

; show pieces
mov a 64
mov ra #pos
mov wa #disp
j memcpy:
mov z .

loop:

mov ra #input   ; read from input
mov ra 0        ; clear read register
mov b #pos-1    ; calculate selected address
mov a,y rd      ; save user input, check zero
mov a x \       ; check saved value is not zero
mov ra add      ; address of selected position
jz loop:        ; input is zero, skip
mov a rd        ; read value of selected position

jnz check_move: ; check if saved value is zero
mov z sel_fin:  ; go to sel_fin when returning from check_move

; saved value is zero
jz loop:        ; if selected position is empty, continue
noop

j loop:         ; else
mov x y         ; save input value

sel_fin:        ; saved value is not zero
noop            ; wait for jnz condition

jnz move_piece: ; move piece if valid
mov z .

j loop:         ; loop
mov x 0         ; clear saved value

; a count
; ra src
; wa dst (set right before j)
memcpy:
mov wd rd \
jz z
dec a
j memcpy: \
inc ra
inc wa

; x src
; y dst
; a 1 if valid, 0 if not
; checks for colour only
check_move:
mov b #pos-1 \  ; calculate addresses
mov a y         ; dest address
mov ra add \    ; read dest value
mov a x         ; source address
mov ra add      ; read source value
mov a rd        ; store dest value to a
mov b rd        ; store source value to b

jz z            ; pass if dest has no piece
mov a 1         ; return 1

mov a xor \     ; check colour bits
mov b 0x7FFFFFF
noop            ; wait for calculation
noop

jle z           ; same colour, fail
mov a 0         ; return 0

j z
mov a 1         ; return 1

; x src
; y dst
; move the piece and update display
move_piece:
mov a x \       ; calculate source address
mov b #pos-1

mov ra,wa add \ ; read write from source
mov a y         ; calculate destination address

mov wd 0 \      ; write 0 to source
mov wa add      ; write to destination

mov wd,x rd \   ; read piece from source and write to destination
mov wa 0 \
mov a x         ; calculate display source address
mov b #disp-1

mov a y \       ; calculate display destination address
mov wa add      ; clear source cell
mov wd 0

mov wa add      ; write to display destination
j z             ; return
mov wd x        ; write piece

.data
bitmap:
0x01f711c4 ; king
0x01f77dd5 ; queen
0x01f728ce ; bishop
0x01f779c4 ; knight
0x01f73bf5 ; rook
0x01f73880 ; pawn

Processor Blueprint String

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

Related Projects

These are some related projects that I've came across when researching for this project.

References

  1. Lupoviridae, Factorio Computer, Jun 23, 2015. Available: https://forums.factorio.com/viewtopic.php?t=13154
  2. jade52blue, First functioning programmable computer!, Oct 21, 2020. Available: https://forums.factorio.com/viewtopic.php?t=90583
  3. Halke1986, Factorio RISC V implementation, 2021. Available: https://github.com/Halke1986/factorio-riscv
  4. Factorio. Accessed: Mar 12, 2023. Available: https://factorio.com/
  5. Liu, R., Multi-port Random Access Memory in Factorio, Nov 12, 2022. Available: https://blog.r26.me/C-factorio-ram.en.html
  6. YsVc, Rendering 3D image on Factorio CPU, Dev 2, 2017. Available: https://www.reddit.com/r/factorio/comments/7h2fes/rendering_3d_image_on_factorio_cpu/
  7. alcatraz_escapee, Playing Tetris on a general purpose Factorio CPU, May 2, 2022. Available: https://www.reddit.com/r/factorio/comments/ugv2w2/playing_tetris_on_a_general_purpose_factorio_cpu/
  8. Sjadfooey, I saw somebody else on here made pong, so I made snake!, Jun 3, 2022. Available: https://www.reddit.com/r/factorio/comments/v4e30i/i_saw_somebody_else_on_here_made_pong_so_i_made/
  9. fuzzyhair2, What limits CPU speed?. Available: https://electronics.stackexchange.com/a/122060
  10. Charles W. Kann, "Load and Store Architecture," in Introduction to Assembly Language Programming: From Soup to Nuts: ARM Edition. https://eng.libretexts.org/Bookshelves/Computer_Science/Programming_Languages/Introduction_to_Assembly_Language_Programming%3A_From_Soup_to_Nuts%3A_ARM_Edition_(Kann)/04%3A_New_Page/4.04%3A_New_Page
  11. M. S. Schlansker and B. R. Rau, "EPIC: Explicitly Parallel Instruction Computing," in Computer, vol. 33, no. 2, pp. 37-45, Feb. 2000, doi: 10.1109/2.820037. Available: https://www.cs.binghamton.edu/~dima/cs522_05/epic.pdf
  12. Ben Eater, Building an 8-bit breadboard computer!, 2016. Available: https://www.youtube.com/playlist?list=PLowKtXNTBypGqImE405J2565dvjafglHU
  13. DeRosa, J. A. and Levy, H. M., An Evaluation of Branch Architectures, Jun 1987, doi: 10.1145/30350.30352. Available: https://dl.acm.org/doi/pdf/10.1145/30350.30352
  14. Wikipedia, Harvard architecture. Accessed: May 03, 2023. Available: https://en.wikipedia.org/w/index.php?title=Harvard_architecture&oldid=1150624733
  15. Wienand, I., "Assembler" in Computer Science from the Bottom Up. Available: https://bottomupcs.com/ch07s04.html