On this page
Where our numbers come fromHow we pick a number when sources disagreeHow we choose the formulasHow we handle metric and imperialHow often we checkHow the comparison scores are calculatedHow we generate substitute recommendationsFound a number that looks wrong?Last updated: 17 July 2026
This page explains, in plain words, where the numbers on Vriksai come from and how we keep them honest. If you ever want to know why a figure says what it says, this is the page to read.
1. Where our numbers come from
Every number on Vriksai — density, hardness, shrinkage, strength, durability — is taken from a published book or standard that anyone can check. We do not guess, and we do not copy numbers from random websites or forums.
Our main sources are:
- USDA Wood Handbook (GTR-190) — the big reference book from the US Forest Products Laboratory. We use it for density, strength and moisture behaviour.
- The Wood Database — used for Janka hardness and shrinkage figures.
- PROSEA — a plant resource reference we use for South-East Asian woods.
- EN 350 — the European standard for how well a wood resists rot.
- Finnish ThermoWood Association — for heat-treated wood.
- IS 875 — the Indian standard we use for wind load.
Each species page and reference page lists the source it used, right at the bottom.
2. How we pick a number when sources disagree
Sometimes two books give slightly different figures for the same wood. This is normal — a tree grown in one place is not exactly the same as the same tree grown somewhere else. When this happens, we do a few simple things:
- We prefer the number from the more trusted source (a national standard or the USDA handbook comes first).
- We use a typical or representative value, not the highest or lowest one. A representative value is a sensible starting point, not a design value — where a figure affects structure or safety, verify it against the governing standard for your specific application.
- Where a wood naturally varies a lot, we say so on the page instead of pretending there is one exact number.
3. How we choose the formulas
Every calculator uses a formula from a known, published source — a standard, a handbook, or a long-used trade rule. We do not invent our own maths. On each tool page we show the formula and name where it comes from, so you can check it yourself.
Where a formula is well known but not from a single official standard (some old log rules, for example), we say that clearly too.
4. How we handle metric and imperial
Wood is measured in different units around the world — cubic feet in India, board feet in America, cubic metres almost everywhere else. Our tools let you switch between them, and the same wood always gives the same real answer whichever unit you pick. We test this on purpose, because a unit mistake is the kind of error that looks correct but is quietly wrong.
5. How often we check
We re-check our data against the sources during site-wide reviews. The most recent full review was finished on 17 July 2026 and covered all 62 calculators and 60 species pages. When we find a mistake, we fix it on every page it affects, not just the one that was reported. Each page shows its own last-updated date so you can see how fresh it is.
6. How the comparison scores are calculated
Every comparison page carries a Library Score for each timber. It is computed, not assigned, and this is the single canonical description of the method — the comparison pages link here rather than each restating it.
The weighting
Each wood receives a weighted composite of the verified figures shown on that page:
- Durability — 22%
- Dimensional stability — 18%
- Hardness — 12%
- Bending strength — 12%
- Machining ease — 12%
- Stiffness — 8%
- Flame spread — 8%
- Lightness — 8%
What the number means
The Library Score is that composite’s percentile rank within our 60-species library — not an absolute grade. 50 is the library median. 90 means it outperforms nine of every ten species we cover. Change the library and the percentile changes; the underlying figures do not.
What it deliberately excludes
No price data is scored. Cost guidance on this site is qualitative, because timber pricing moves with origin, grade, section, season and freight, and a number that stale would be worse than none. Availability and lead time are excluded for the same reason.
The limit of it
A single composite cannot know your application. A timber that scores 40 may be the correct specification for an interior job where durability weighting is irrelevant, and a 90 may be wrong for it. The score summarises the data; it does not replace judgement about your specific application. Use it to shortlist, then read the individual figures.
7. How we generate substitute recommendations
Each species page carries an Alternatives to … block: step up, step down, more dimensionally stable, closest overall match, and — where the species is not exterior-rated — more durable. Those shortlists are computed, not written. This section is the single canonical description of how, so the species pages do not each repeat it.
What the engine reads
Only measured engineering properties from our species dataset: Janka hardness, density, radial and tangential shrinkage, T/R ratio, modulus of elasticity, modulus of rupture, and EN 350 durability class. Nothing else is an input.
What the engine never reads
Availability, price, regional supply, grade, brand and appearance are not factors and are never implied by a recommendation. The lists are not personalised, not exhaustive, and not a specification.
Rules the engine cannot break
- It never proposes a species that would lose you exterior durability you already have. A Class 1 timber is never answered with a Class 4 one.
- On the stability axis it never trades a stability rating down, and never returns a species with higher tangential movement than the one you are reading about.
- Hardwood/softwood crossovers are only allowed inside a narrow hardness band, so you are not offered a 300 lbf timber in place of a 1,700 lbf one.
- Returning nothing is a valid answer. Accoya has no stability alternatives in our library; Lignum Vitae has nothing harder. Those sections are simply absent rather than padded with weaker suggestions.
“Closest overall match”
A weighted distance across six standardised properties — hardness and density weigh heaviest, then movement, stiffness and bending strength — with an additional penalty for any gap in durability class or stability rating. A lower number is a closer match. Anything beyond our distance threshold is dropped rather than shown as a weak third option.
How much confidence to place in it
Published wood properties shift with growing region, tree age, moisture content at test and test method, and different reference bodies measure specific gravity and shrinkage on different bases. Treat every figure as a representative value, not an absolute for a given consignment. Two species separated by a few percent should be treated as equivalent until you have verified the actual material. The recommendation tells you where to look; it does not close the decision.
8. Found a number that looks wrong?
Please tell us. If you have a better source, share it and we will check it. Real corrections are made within a few days and the page date is updated. You can reach us on our contact page.