Compare approaches and devices

The site answers two kinds of comparison question. One is about approaches: how does a tokamak differ from a stellarator? The other is about machines: is one device larger, or has it gone further, than another? The first is answered on /compare/. The second is answered on /devices/ and /leaderboards/.

Comparing two approaches

/compare/ lists the comparison pairs. Each card gives the pair’s title, the question it answers, and how many devices and organisations the record holds across both sides. There are three pairs:

• Tokamak vs stellarator: two ring-shaped magnetic machines that twist their magnetic field in different ways.
• Magnetic vs inertial confinement: holding a thin plasma steady for seconds, against crushing a fuel capsule in nanoseconds.
• Molten salt reactor vs fluoride salt-cooled reactor: two fission designs that both use molten salt, only one of which dissolves the fuel in it.

Each comparison page has four parts, in this order.

The two sides. A card for each approach gives its definition from the glossary, the number of devices and organisations tagged with it, and the number of measurements those devices hold.

Differences. A table with one row per point of difference, such as where the magnetic twist comes from, or how gain is quoted. Each row gives the answer for each side. This table is written text, not a query of the record, and it carries no citations on the page.

Devices. Up to 14 devices on each side, those with the most measurements first. The number after each name is its measurement count. If there are more, the list ends with “and N more”. Each name links to the device’s own page.

Organisations. Every organisation tagged with each approach, alphabetically, each linked to its page.

What the counts measure

The counts are drawn from the record each time the site is built, so they change as the record grows. A device or organisation is counted on a side only if it carries that side’s approach tag. On the tokamak vs stellarator page, the tokamak side counts only records tagged “tokamak”. Spherical tokamaks carry their own tag and are not included. The magnetic side of the magnetic vs inertial page does include them, along with the other magnetic approaches.

A larger count shows that more has been attempted with an approach. It does not show that the approach works better. On the molten salt vs FHR page, the FHR side has a single developer in the record. The pair is included despite the imbalance, because the two are so often confused.

Comparing two machines

There is no page for choosing two devices and setting them side by side. Three pages do the job between them.

The device table. /devices/ has one row per fusion device that has a published set of design parameters. The columns are Device (with operator), Approach, Status, Country, and five parameters:

• R₀, major radius, in metres: the distance from the centre of the ring to the centre of the plasma
• a, minor radius, in metres: the radius of the plasma’s cross-section
• Iₚ, plasma current, in megaamperes
• B₀, the magnetic field on the plasma’s axis, in tesla
• τ, plasma duration, in seconds

Hover over a parameter heading for its full definition. Select the Device heading or a parameter heading to sort by it, and select it again to reverse the order. A bar under R₀ shows each machine’s major radius against the largest in the table. Above the table you can search by device or operator, and filter by approach and by status. A line shows how many devices match.

These are design figures: what a machine was built to do. They are not results.

The leaderboards. For what machines have actually achieved, use /leaderboards/. Each board ranks one quantity, such as triple product or ion temperature, in a single unit, one row per device at its best figure. Targets and design figures are excluded. How to read a benchmark row explains the columns.

Device pages. Each device’s own page, at /facility/<name>/, lists every figure held for it, with value, unit, full metric name, status, date and shot number. Open two in separate tabs to compare them figure by figure. Check that both figures use the same metric and the same status. An achieved figure and a planned one are not comparable, and neither are two gains with different denominators.

Sources

• Comparisons, Devices, Leaderboards and Glossary on this site