Work with blocks

A block is one production unit. You choose its goals, recipes, machines, and constraints; PyOps calculates the rates, building counts, power, pollution, and flows across its boundary.

Blocks do not need to match one physical building. A block can represent one recipe, a complete production chain, or a section of the factory you want to build and operate as a unit.

Organize blocks

Use the folder button in the Blocks sidebar to group related production units. Drag an existing block onto a folder to move it. To start in the right place, right-click a folder and select New block here. The main New block button continues to create blocks under Ungrouped. After creation, the sidebar clears its filter, expands that section, and scrolls the new block into view.

Set goals and rates

Select New block, then use + goal to choose an item or fluid.

• The first goal starts at 1/s, names the block, and becomes its scaling anchor.
• Select a displayed rate to edit it. Select its unit to work in /s, /min, or /h.
• Enter a negative rate when the block should consume that item or fluid rather than produce it. For example, -10/min makes the block consume at least 10 per minute.
• Add more goals when the same production unit must guarantee several outputs.
• Drag a goal by its grip to reorder the goal tiles.
• Right-click a non-primary goal and select Move to front (names the block) to make it the primary goal.

Use Copy goals in the Goal heading, open another block, and select Paste goals to reuse a set of targets. Pasted goals keep their rates, stock amounts, refill windows, and source order. They are appended after the destination's existing goals, so its primary goal does not change; goods already present are skipped. Recipes, machines, modules, and other block settings are not copied.

Right-click a goal and select Keep in stock instead (buffer, not throughput) when the intent is to refill a quantity over time rather than sustain a continuous rate. Stock-only production appears separately in Factory. The solver rate is always the stock amount divided by its refill window; an older saved rate cannot turn a buffer goal back into continuous production.

In the Goal heading, select the ∞ ongoing-mode control (tooltip: Ongoing throughput — click to make this a temporary campaign) when the block should make a finite amount and then stop, such as the first creature in an uplifting chain. The icon changes to a timer while the campaign is active. Set Make on each goal and the shared clock window. Select the window unit to cycle through seconds, minutes, and hours, using the same compact interaction as goal-rate units. PyOps derives ordinary per-second recipe, machine, fuel, power, and import rates from those values, so the temporary block participates in Factory exactly like other enabled production while it is active. Block balance shows the expected campaign total as the primary value for every import and export, with the smaller derived average rate directly beneath it in the same chip. Item logistics remain beneath the chip when enabled; fluids do not show belt logistics.

Use Expected when average yield is enough. Choose 90% confidence or 95% confidence to add an operational reserve for random finite results; the goal tile shows both the amount requested and the larger expected amount PyOps plans. Variable recipe rows show their result range and chance, and hovering the warning shows the expected total over the campaign. This reserve is a probability estimate, not a guarantee: Factorio can still produce an unusually unlucky run.

Select the checkmark (tooltip: Mark complete and remove this campaign from factory planning) after the finite target is crafted. PyOps disables the block so it no longer contributes to Factory or Factory Scenario, while retaining the campaign as history. Select the history icon to reactivate it if more are needed, or select the timer mode control to return to ongoing throughput while keeping the derived rate.

A negative goal is already the block's visible import contract, so Block balance does not repeat that good under Imports. Other ingredients the block needs still appear there. Electricity is listed first, followed by the remaining imports from highest to lowest rate. An alert beside an import means no other enabled block currently exports it. This is a planning cue rather than an error: raw resources and intentionally external supplies can remain unsourced. Right-click an import to jump to the blocks that do produce it.

Add recipes

Select a positive goal icon to open Recipes that make goal. A negative consume goal instead opens Recipes that consume goal. Ctrl+Click (or Command+Click) a goal icon to immediately add its highest-ranked currently unlocked recipe. This shortcut uses the same logistic-cost ordering as the picker and ignores barrel fill/empty recipes. Within a recipe row:

• Select an ingredient chip to find recipes that make that ingredient.
• Select a product chip to find recipes that consume that product.
• Alt+Click any item or fluid icon to open the Recipe explorer without leaving the current page. Switch between Recipes and Uses to inspect every producer or consumer, including availability, machines, and filters; select an ingredient or product inside the dialog to keep walking the graph.
• Hover recipe, technology, and item details when you need the precise inputs, outputs, or unlock status.
• A probabilistic product shows its chance as a small percentage beneath its expected rate. Hover the product for its amount on success and expected amount per craft. The flask icon by the recipe rate summarizes every variable result in one tooltip.

The picker puts choices already unlocked in the synced save first, then recipes available later in the current planning horizon, and finally locked choices. It sorts each group by ascending estimated cost, so the first row is normally the cheapest recipe you can build right now. A recipe is available only when both its own research and at least one compatible building are available. Locked rows are disabled and name the missing recipe research or building research; recipes replaced by a TURD upgrade remain visible with their explanation. PyOps never adds a full chain without your choice.

Organize recipes into sub-blocks

Right-click a recipe name and select New sub-block from this row to start a named, collapsible group. Drag another recipe onto the sub-block header or one of its indented rows to add it. Drag the sub-block header to move the whole group together.

Grouping is visual by default and does not change the solve. Use the module control on the sub-block header when the grouped recipes should instead solve as a separate module with its own internal goals and a boundary contract for the parent block.

Choose machines, fuels, and modules

Each recipe row shows its selected machine and required building count.

• Select the machine to choose another compatible option.
• For a fuel-burning machine, select its fuel indicator to choose an allowed fuel.
• Use the module control to fill module slots or override the block's preferred defaults.

Global preferred machines, fuels, fluid temperatures, and module-fill behavior live under Settings → Planning. A choice made directly on a recipe row overrides the relevant default for that row.

Choose a fluid ingredient temperature

A fluid ingredient with more than one known produced temperature shows its accepted range inside the ingredient chip, such as ≥15°. Select that range to choose a concrete produced temperature, such as 250°, when the recipe should connect to a specific factory-wide fluid line. Select Recipe range to restore range-based matching. Fluids with only one producible temperature do not show a temperature label or control.

Select the star beside a produced temperature to make it that fluid's preferred default. A newly added compatible recipe row or fluid goal starts pinned to that temperature—for example, favorite 250° Steam and 15° Water for ordinary lines. The favorite does not rewrite existing rows or goals, and a recipe whose accepted range excludes it stays on the recipe range until you choose another temperature.

A fluid goal with several producible temperatures has the same clickable temperature text beneath its rate. Choose an exact value to require that output variant, or choose Goal range to allow any temperature in the displayed range. Fluids with only one producible temperature need no goal control.

The choice is stored on that recipe row; it does not change the Factorio recipe or create a new fluid prototype. PyOps treats the selected flow as an exact (fluid, temperature) identity when solving the block and matching it to supplier blocks. The block's Imports and Exports chips show the resulting exact temperature or accepted range as part of the boundary contract.

Fractional building counts are exact capacity requirements. 0.5 means half of one machine's capacity; 5.2 means five machines are insufficient and six provide spare capacity. PyOps leaves the construction decision visible instead of rounding the solve.

When Logistics → Inserters / loaders is enabled, an amber hammer and a second count can appear beneath the solved building count. It is the suggested physical build count: the selected movers, item inputs and outputs, fuel, burnt results, and active fluid connections need more adjacent access positions than the capacity-rounded buildings provide. The estimate uses the selected machine's footprint and increases the whole-building count until those connections fit around its perimeter.

This is a conservative loading estimate, not a generated layout. It does not prove that belts and pipes can reach every access position, account for beacon spacing, or model direct insertion and circuit-controlled sharing. Hover the badge to see the footprint, position budget, and selected mover used by the estimate. Machine footprints populate during Settings → Data sync; an existing project may need one sync before the badge appears.

Read the result

The Block balance card summarizes:

• whether the selected goals and constraints solve;
• energy and pollution totals;
• goods entering as imports;
• goods leaving as goals, surplus, or byproducts.

Select an exported good to add a recipe that consumes its surplus inside the block. The selected consumer runs as part of the chain, including when one of its products feeds back into another recipe in the block.

The recipe table explains how that result was produced. Table is the editing view; Flow is a read-only diagram of the same solved recipes and goods.

Variable generators show their average planned output and their minimum–maximum range on the electricity product chip. Building counts and block balances use the displayed average; the range shows how far live generation can move as the underlying surface condition changes.

A solved block is not necessarily self-contained

A block can solve while importing ingredients. This is intentional: the block boundary defines what another block or the existing factory must supply. Continue with Block boundaries to control that behavior.

Plan science consumption

Research is a consumer, not a recipe: one bank of labs eats every science pack at once. Use the flask control beside New block to create the project's science block. There is one per project, so the button reports This project already has its science block once it exists.

The block asks for three things:

• Lab — which lab prototype the bank is built from. Its research speed and power draw size the pool.
• Lab-seconds per pack — how long one lab spends on a single pack. Pack rates alone cannot size the bank, because 10 / 5 / 1 could be one technology's ratio or three different ones running together.
• Science per minute — the rate you want for each pack. These are the plan.

The result reads back the pool size, the total science rate, power, and the productivity applied to it. Imports the factory must supply is what the rest of the factory has to deliver: your rates divided by productivity.

Labs are one pool. Asking for 40 automation science and 20 logistic science per minute needs the labs for 60 packs a minute, not the two figures added as separate banks.

Derive rates from a technology

Select Derive from a technology when you would rather describe the target as research than as rates. Pick a technology, set how many minutes it should take, and the dialog previews the per-pack rates and the lab-seconds that follow from its cost. Use these rates writes those numbers into the fields.

This is a calculator. No technology is stored on the block, so the plan does not change when that research finishes—come back and derive again to move to the next target.

Apply modules and beacons to the bank

Select the Effects control to open the same module and beacon picker recipe rows use. A lab with no module slots of its own can still be covered by a beacon, so the picker stays available; the palette is empty because nothing can be inserted directly.

Set shared by to the number of labs each beacon building actually covers. Every lab still receives the full effect—the setting divides the buildings you place and their power, and it is what makes one beacon serving a whole lab row cost one building instead of dozens.

Where a beacon has to be fed to run, its consumption appears as Upkeep in items per minute alongside the beacon buildings.

Disable a block without deleting it

Use the power control in the block toolbar to disable a block. Its recipes and settings remain saved, but Factory and other whole-plan views exclude it. Re-enable it when the production unit should participate in the plan again.

When the solve is infeasible

Read the diagnosis in Block balance before adding more recipes. Typical conflicts are:

• a goal with no recipe that makes it;
• a good marked for in-block production without an enabled producer;
• an exact building-count pin that conflicts with a goal;
• multiple constraints demanding incompatible rates in a cyclic chain.

Use the suggested fixes in the diagnosis. Removing constraints one at a time is usually more informative than adding every candidate recipe.