Tiny demo scene
A tiny-gate demo in Verilog: a testbench, a gate-count cap, and golden framebuffer dumps, with no FPGA or tapeout required.
This is a HDL spec on Spekkr, published by onyx. It is listed with the other specs. Paste the line to an agent that can open a web address, and the agent fetches the files and builds from them.
- Spec
onyx/tiny-demo-scene- Version
- 0.2.1
- Published by
- onyx
- Updated
- 14 Nov 2026
Build this spec
Build the spec at https://spekkr.com/onyx/tiny-demo-scene/start.md.
Paste this to your AI agent. It opens the address, fetches the spec's files, records your choices and builds from them. Nothing to install, no account.
Published by onyx.
What does the result look like?
What can I customize?
Your choices go into the line you copy. Anything else is yours to edit, but it leaves the tested path.
What do I get?
-
spec.mdRequirements and acceptance checks.Tiny demo scene
What this builds
A small, original demo that fits in a tight gate budget (around 4k gates, as counted by a documented tool), written in Verilog. A testbench clocks the design and dumps a framebuffer. Golden frames are compared. No FPGA, no board, and no tapeout are required. The art and the music (if any) are original.
Read
spekkr.inputs.jsonfirst.titlemay appear on a frame.widthandheightsize the buffer.audioandscanlinechange what is dumped.simis the simulator.Out of scope
- A real FPGA, a tapeout, or a board in the checks.
- A copyrighted demo or a commercial game's art.
- A gate-array vendor's licence server.
Requirements
Each requirement has an id. Checks name the id they cover.
R1. Testbench and goldens
- The testbench runs a documented number of frames (for example 4) and writes framebuffer dumps. Frames 0 and N (the last, or a mid frame in the file) match goldens within the tolerance in
design.md. - If
titleis drawn, the pixels of the title region match a mask or a glyph dump in the file. - Reset is deterministic: two runs hash the same dumps.
Accept when: the golden frames match, the title mask matches if used, and two runs hash equal.
R2. Gate cap and files
- A documented count (Yosys or an equivalent, with the script in the repo) reports a gate or cell count at or under 4096, or the bound in the file, for the default size.
- Renders: frame 0, frame N, a title-card still, and a small diff view (empty when goldens match).
- One command simulates, counts, and writes the dumps and renders. No FPGA.
Accept when: the count is under the cap, the command passes for both simulators, and a deliberately large fixture exceeds the cap.
R3. Inputs
- Inputs are read from
spekkr.inputs.json. Values outside the ranges or choices indesign.mdare refused before a netlist, a board or a testbench is built. - Options turn the optional circuits, ports or features in the Inputs table on or off.
Accept when: an illegal input is refused by name, and the default input set produces the measurements in
design.md.R4. Checks and report
- One command runs the simulation or the testbench, writes the plots or dumps, and compares them to the numbers in
design.md. - The command works without a physical board, a lab instrument or a network.
Accept when: the command passes at the defaults, and fails if a documented value in the netlist or the ROM is corrupted on purpose.
Verification
A Verilog testbench dumps a framebuffer that matches golden frames. A documented gate-count script stays at or under the cap. Two runs are deterministic. No FPGA, no handheld, and no tapeout are required.
Definition of done
- Every acceptance check above runs from one command, and all pass at the person's inputs.
- A README says how to simulate, how the count is taken, and that a tapeout is optional.
# Tiny demo scene ## What this builds A small, original demo that fits in a tight gate budget (around 4k gates, as counted by a documented tool), written in Verilog. A testbench clocks the design and dumps a framebuffer. Golden frames are compared. No FPGA, no board, and no tapeout are required. The art and the music (if any) are original. Read `spekkr.inputs.json` first. `title` may appear on a frame. `width` and `height` size the buffer. `audio` and `scanline` change what is dumped. `sim` is the simulator. ## Out of scope - A real FPGA, a tapeout, or a board in the checks. - A copyrighted demo or a commercial game's art. - A gate-array vendor's licence server. ## Requirements Each requirement has an id. Checks name the id they cover. ### R1. Testbench and goldens - The testbench runs a documented number of frames (for example 4) and writes framebuffer dumps. Frames 0 and N (the last, or a mid frame in the file) match goldens within the tolerance in `design.md`. - If `title` is drawn, the pixels of the title region match a mask or a glyph dump in the file. - Reset is deterministic: two runs hash the same dumps. Accept when: the golden frames match, the title mask matches if used, and two runs hash equal. ### R2. Gate cap and files - A documented count (Yosys or an equivalent, with the script in the repo) reports a gate or cell count at or under 4096, or the bound in the file, for the default size. - Renders: frame 0, frame N, a title-card still, and a small diff view (empty when goldens match). - One command simulates, counts, and writes the dumps and renders. No FPGA. Accept when: the count is under the cap, the command passes for both simulators, and a deliberately large fixture exceeds the cap. ### R3. Inputs - Inputs are read from `spekkr.inputs.json`. Values outside the ranges or choices in `design.md` are refused before a netlist, a board or a testbench is built. - Options turn the optional circuits, ports or features in the Inputs table on or off. Accept when: an illegal input is refused by name, and the default input set produces the measurements in `design.md`. ### R4. Checks and report - One command runs the simulation or the testbench, writes the plots or dumps, and compares them to the numbers in `design.md`. - The command works without a physical board, a lab instrument or a network. Accept when: the command passes at the defaults, and fails if a documented value in the netlist or the ROM is corrupted on purpose. ## Verification A Verilog testbench dumps a framebuffer that matches golden frames. A documented gate-count script stays at or under the cap. Two runs are deterministic. No FPGA, no handheld, and no tapeout are required. ## Definition of done - Every acceptance check above runs from one command, and all pass at the person's inputs. - A README says how to simulate, how the count is taken, and that a tapeout is optional.
-
design.mdThe gate-count cap, the testbench timing, and framebuffer goldens.Design
This file fixes the circuits, the timings and the dumps a build must match. The toolchain can vary, as long as the numbers here hold.
Budget
4096 gates or cells, as counted by the shipped script (the file states the exact Yosys command or the cell library). The default 320 by 240 design must fit. Larger buffers may exceed it and are then out of the cap check, as the file says.
Goldens
fixtures/demo-frames/holds frame 0 and frame N. Pixel tolerance: 0 (exact) for a digital dump, or 2 levels if a dither is documented.Frames
N is 3 (the fourth frame, 0-based), or as in
fixtures/demo-timing.json. Clock rate and reset cycles are in that file.Inputs
Key Label Default Notes titleDemo title A tiny scene Shown in the first frames, if the design has a font, or only in the README if not. widthFramebuffer width 320 160, 320 or 640. heightFramebuffer height 240 120, 240 or 480. audioA tiny audio register off A documented register dumped by the testbench; no speaker is required. scanlineA visible scanline effect on A documented effect in the framebuffer. simSimulator verilator Either must pass the same testbench and goldens, within a documented pixel tolerance if timing differs by a cycle at reset. Acceptance at the defaults
A small HDL demo: a 320 × 240 (or documented 160 × 120) framebuffer, a bouncing mark, and a scanline option. The testbench dumps frames.
Quantity Value Tolerance Source Frame size 320 × 240 exact Default; 160 × 120 is accepted if documented Frame count in the loop 60 0 A full cycle of the bounce Background colour #10141c — As 24-bit RGB in the dump Mark size 16 × 16 0 The bouncing rectangle Scanline darkening 12% 2% When scanlineis on, odd rowsA Verilog (or VHDL) testbench clocks the design and writes PPM or raw frames. Frame 0, the mid bounce, and frame 59 are compared to goldens. No FPGA board is required. Icarus or Verilator is enough.
Screenshots
These pictures are dumps of a working reference model of the core, in
reference/preview.py, at the default inputs. They are not drawings of a product that has not been built or simulated. The first picture is the one Spekkr shows on the spec's card. A build may ship its own goldens; those goldens are of that build and do not have to match these frames pixel-for-pixel.File What it shows shots/frame-0.pngFrame 0 320 × 240, mark in the corner, scanlines on shots/mid-bounce.pngMid-bounce the 16 × 16 mark, scanlines on How to verify
- Run the simulation or the HDL testbench at the defaults, with no physical board.
- Compare periods, trip points, register dumps or frames to the acceptance table in this file.
- Corrupt one documented value on purpose and confirm the check fails.
- A Verilog testbench dumps a framebuffer that matches golden frames. A documented gate-count script stays at or under the cap. Two runs are deterministic. No FPGA, no handheld, and no tapeout are required.
# Design This file fixes the circuits, the timings and the dumps a build must match. The toolchain can vary, as long as the numbers here hold. ## Budget 4096 gates or cells, as counted by the shipped script (the file states the exact Yosys command or the cell library). The default 320 by 240 design must fit. Larger buffers may exceed it and are then out of the cap check, as the file says. ## Goldens `fixtures/demo-frames/` holds frame 0 and frame N. Pixel tolerance: 0 (exact) for a digital dump, or 2 levels if a dither is documented. ## Frames N is 3 (the fourth frame, 0-based), or as in `fixtures/demo-timing.json`. Clock rate and reset cycles are in that file. ## Inputs | Key | Label | Default | Notes | | --- | --- | --- | --- | | `title` | Demo title | A tiny scene | Shown in the first frames, if the design has a font, or only in the README if not. | | `width` | Framebuffer width | 320 | 160, 320 or 640. | | `height` | Framebuffer height | 240 | 120, 240 or 480. | | `audio` | A tiny audio register | off | A documented register dumped by the testbench; no speaker is required. | | `scanline` | A visible scanline effect | on | A documented effect in the framebuffer. | | `sim` | Simulator | verilator | Either must pass the same testbench and goldens, within a documented pixel tolerance if timing differs by a cycle at reset. | ## Acceptance at the defaults A small HDL demo: a 320 × 240 (or documented 160 × 120) framebuffer, a bouncing mark, and a scanline option. The testbench dumps frames. | Quantity | Value | Tolerance | Source | | --- | --- | --- | --- | | Frame size | 320 × 240 | exact | Default; 160 × 120 is accepted if documented | | Frame count in the loop | 60 | 0 | A full cycle of the bounce | | Background colour | #10141c | — | As 24-bit RGB in the dump | | Mark size | 16 × 16 | 0 | The bouncing rectangle | | Scanline darkening | 12% | 2% | When `scanline` is on, odd rows | A Verilog (or VHDL) testbench clocks the design and writes PPM or raw frames. Frame 0, the mid bounce, and frame 59 are compared to goldens. No FPGA board is required. Icarus or Verilator is enough. ## Screenshots These pictures are dumps of a working reference model of the core, in `reference/preview.py`, at the default inputs. They are not drawings of a product that has not been built or simulated. The first picture is the one Spekkr shows on the spec's card. A build may ship its own goldens; those goldens are of that build and do not have to match these frames pixel-for-pixel. | File | What it shows | | --- | --- | | `shots/frame-0.png` | Frame 0 320 × 240, mark in the corner, scanlines on | | `shots/mid-bounce.png` | Mid-bounce the 16 × 16 mark, scanlines on | ## How to verify 1. Run the simulation or the HDL testbench at the defaults, with no physical board. 2. Compare periods, trip points, register dumps or frames to the acceptance table in this file. 3. Corrupt one documented value on purpose and confirm the check fails. 4. A Verilog testbench dumps a framebuffer that matches golden frames. A documented gate-count script stays at or under the cap. Two runs are deterministic. No FPGA, no handheld, and no tapeout are required.
-
tasks.mdBuild tasks in order, each with a check.Tasks
Work in order. Finish a task's check before starting the next.
- Set up. Install the simulator or the HDL toolchain named in
design.md(ngspice, Icarus, Verilator, or a KiCad/SVG board tool). Read the inputs into one place. Check: a trivial netlist or testbench runs and writes a file. - Testbench and goldens (R1). Build what
R1. Testbench and goldensasks for, to the numbers indesign.md. In particular: The testbench runs a documented number of frames (for example 4) and writes framebuffer dumps. Frames 0 and N (the last, or a mid frame in the file) match goldens within the tolerance indesign.md. Check: theR1. Testbench and goldensacceptance checks pass. - Gate cap and files (R2). Build what
R2. Gate cap and filesasks for, to the numbers indesign.md. In particular: A documented count (Yosys or an equivalent, with the script in the repo) reports a gate or cell count at or under 4096, or the bound in the file, for the default size. Check: theR2. Gate cap and filesacceptance checks pass. - Inputs (R3). Build what
R3. Inputsasks for, to the numbers indesign.md. In particular: Inputs are read fromspekkr.inputs.json. Values outside the ranges or choices indesign.mdare refused before a netlist, a board or a testbench is built. Check: theR3. Inputsacceptance checks pass. - Checks and report (R4). Build what
R4. Checks and reportasks for, to the numbers indesign.md. In particular: One command runs the simulation or the testbench, writes the plots or dumps, and compares them to the numbers indesign.md. Check: theR4. Checks and reportacceptance checks pass. - Plots, dumps and goldens. Write the plots or framebuffer dumps named in
design.mdand compare them to the tables there. Check: every documented period, trip point, opcode or frame matches, within the stated tolerance. - Finish. Write a README that says how to install what is needed, how to change the inputs, and how to run the checks. Run the one command from a clean folder. Check: every item under Definition of done in
spec.mdis true.
# Tasks Work in order. Finish a task's check before starting the next. 1. **Set up.** Install the simulator or the HDL toolchain named in `design.md` (ngspice, Icarus, Verilator, or a KiCad/SVG board tool). Read the inputs into one place. Check: a trivial netlist or testbench runs and writes a file. 2. **Testbench and goldens (R1).** Build what `R1. Testbench and goldens` asks for, to the numbers in `design.md`. In particular: The testbench runs a documented number of frames (for example 4) and writes framebuffer dumps. Frames 0 and N (the last, or a mid frame in the file) match goldens within the tolerance in `design.md`. Check: the `R1. Testbench and goldens` acceptance checks pass. 3. **Gate cap and files (R2).** Build what `R2. Gate cap and files` asks for, to the numbers in `design.md`. In particular: A documented count (Yosys or an equivalent, with the script in the repo) reports a gate or cell count at or under 4096, or the bound in the file, for the default size. Check: the `R2. Gate cap and files` acceptance checks pass. 4. **Inputs (R3).** Build what `R3. Inputs` asks for, to the numbers in `design.md`. In particular: Inputs are read from `spekkr.inputs.json`. Values outside the ranges or choices in `design.md` are refused before a netlist, a board or a testbench is built. Check: the `R3. Inputs` acceptance checks pass. 5. **Checks and report (R4).** Build what `R4. Checks and report` asks for, to the numbers in `design.md`. In particular: One command runs the simulation or the testbench, writes the plots or dumps, and compares them to the numbers in `design.md`. Check: the `R4. Checks and report` acceptance checks pass. 6. **Plots, dumps and goldens.** Write the plots or framebuffer dumps named in `design.md` and compare them to the tables there. Check: every documented period, trip point, opcode or frame matches, within the stated tolerance. 7. **Finish.** Write a README that says how to install what is needed, how to change the inputs, and how to run the checks. Run the one command from a clean folder. Check: every item under Definition of done in `spec.md` is true.
- Set up. Install the simulator or the HDL toolchain named in
-
reference/preview.pyA CPU reference of the default framebuffer. A build should dump the same frames from an HDL testbench. -
reference/preview.htmlA working reference preview of the default view, used to make the listing pictures. A build does not have to start from this file. -
reference/fonts/fonts.cssWebfont faces for the reference preview. -
reference/fonts/Geist.ttfGeist variable typeface for the reference preview. -
reference/fonts/GeistMono.ttfGeist Mono variable typeface for the reference preview. -
reference/fonts/Unbounded.ttfUnbounded variable typeface for the reference preview.
Is it for me?
- Kind
- HDL
- Tools
- Verilog, Icarus, Verilator
What does it need?
- Agent permissions
- shell
- Commands it runs
install a Verilog simulator, such as Icarus or Verilatorinstall a tool that can write a PNG from a framebuffer dump
What changed?
- 0.2.114 Nov 2026Add animated loop to the listing shots.
- 0.2.014 Nov 2026Polished reference preview and new listing shots, including a short loop.
- 0.1.014 Nov 2026First version.
For agents: start.md says how to build this spec, spec.json holds everything on this page, and bundle.tar.gz holds its files.