MPW shuttle cost planner

How it works

A multi-project wafer (MPW) shuttle shares one mask set and one wafer lot between many customers, so a small prototype is billed by the area it occupies. This tool prices a block from the provider's price per square millimetre, its minimum area and its grid, adds samples, packaging, test and shipping, repeats it over your planned runs and finds where a dedicated mask set becomes cheaper.

1. Round the block to the provider's grid. Providers place blocks on a grid, so each side is rounded up to the next grid step. Several designs on one submission add their areas; designs submitted apart are billed one by one, each with its own minimum.side = ceiling(side / grid) x grid; block area = width x height; submission area = designs x block area (one submission) or block area (each design apart)
2. Find the billed area and the area charge. The provider bills at least its minimum area, and some bill in whole steps or blocks. Some price lists give a first square millimetre at one price and further area at a lower one; the tool supports that tier. The price is in euros and is converted at the rate you enter.billed area = max(minimum area, ceiling(area / step) x step); area charge = (min(area, tier) x price + max(0, area − tier) x price beyond) x USD per EUR
3. Add samples, packaging, test and shipping. A run delivers a standard number of unpackaged die. Wanting more means buying extra samples. Packaging and test apply to the die you package. Shipping and any extras you pay for each run are added.extra samples = max(0, wanted − delivered); cost per run = area charge + extra samples x extra price + packaged x (packaging + test) + shipping + extras per run
4. Repeat over the planned runs. Prototype work usually takes several iterations. The one-off costs (design kit, set-up) are paid once.MPW total = runs x cost per run + one-off costs; cost per delivered die = MPW total / (runs x die delivered per run)
5. Cost the dedicated alternative. A dedicated run buys a mask set and a lot of wafers. A later iteration repeats the wafers and a share of the mask cost (100% for a full new set, less for a metal-only fix; enter a multi-layer reticle price as the mask set if you use one). The die area gives the die on each wafer and so the die you receive.dedicated total = mask set x (1 + (runs − 1) x re-spin share) + runs x wafers per run x wafer cost + the same packaging, test, shipping and extras as MPW; die per wafer = pi x (d / 2)^2 / A − pi x d / sqrt(2 x A)
6. Find the break-even. Both totals are straight lines in the number of runs. Let A be the MPW fabrication cost of a run (area charge plus extra samples) and D the dedicated cost of a later run (re-spin share of the mask plus wafers). The break-even number of runs follows. The break-even area is the area for which the MPW area charge equals the dedicated cost over your planned runs.break-even runs = mask set x (1 − re-spin share) / (A − D); break-even charge = (dedicated fabrication over the runs − runs x extra samples) / runs, converted back to area through the price

Sources

The methods and default figures rest on these public sources. The wording here is our own; read the source for its full text.

• Prototyping & Fabrication Services: MPW schedules and prices 2026 (opens in a new tab), Europractice IC service, 2026. Standard and discounted prices per square millimetre or per block, with minimum areas, for the provider presets.
• Tiny Tapeout sees industrial boost as it recovers from eFabless closure (opens in a new tab), eeNews Europe (Nick Flaherty), 16 May 2025. Community shuttle pricing by tile and by 100 chips, as context for small prototypes.
• The Dark Side Of The Semiconductor Design Renaissance: Fixed Costs Soaring Due To Photomask Sets, Verification, and Validation (opens in a new tab), SemiAnalysis (Dylan Patel), 24 July 2022. Published mask set cost ranges by node, which set the scale of the dedicated alternative.