Kruti Dev 010 vs 011 vs 016: Which Version Do You Have?
The families look nearly identical in print. What actually changes between them, how to identify yours, and what that means for conversion.
The family at a glance
| Family | Where it is usually found | Supported by this converter |
|---|---|---|
| Kruti Dev 010 / 10 | Government office documents, typing tests, Hindi newspapers and most DTP templates | Yes — this is the reference table |
| Kruti Dev 011 | Documents produced by offices or printers that standardised on 011; glyph shapes are lightly adjusted | Mostly — occasional characters may need a manual fix |
| Kruti Dev 016 | Another close variant with its own glyph set, used by some publishers and institutes | Mostly — same caveat as 011 |
| Kruti Dev 055 | Marathi-oriented variant with a different layout | No — not promised, conversion will produce wrong output |
One thing this page will not do is print a neat list of "the twelve glyphs that changed", because such a list has not been verified here. Inventing one would be worse than admitting it is unknown: every unverified claim about a legacy font becomes a wrong instruction later. What follows is limited to what can be observed and what the converter actually guarantees.
Three ways to identify your version
1. Read the font name in the document
Open the file in Word, place the cursor inside the Hindi text and look at the font box in the toolbar. The exact family name is shown — for example “Kruti Dev 010”, “Kruti Dev 10”, “Kruti Dev 011” or “Kruti Dev 016”. Do not trust the file name alone: documents are often renamed after conversion, and a file called 011 may well contain 010 text.
2. Check what is installed on the machine
In Word, open the font dropdown and type Kruti. Every installed variant appears there. If the document was created elsewhere and the matching family is missing, Word silently substitutes a different font, and the shapes you see may not be the shapes the author saw. In that case, install the family named in the document before judging its output.
3. Compare the same string in each font
Type one line with a mix of features — a conjunct, a reph word, a nukta letter and a digit, for example कर्म, क्षेत्र, ज़, २ — then apply each installed Kruti variant in turn. Differences between 010, 011 and 016 usually show up as slightly different glyph shapes rather than as missing characters. If the line renders identically in two variants, the practical difference for your document is small; if shapes shift noticeably, treat the document as its own version and check the output word by word.
What this means for conversion
The converter on this site is built on the Kruti Dev 010 table. That table defines which code position draws which shape, and it was tested against a fixed list of Hindi sentences in both directions. For 011 and 016 documents, the same table is applied because the families share most of their layout. In practice, the majority of text converts correctly; the exceptions are individual glyphs whose positions differ, and those surface as a handful of wrong characters in an otherwise correct paragraph.
For 055 and for unrelated legacy families such as Chanakya, DevLys and Preeti, the table does not apply at all. Converting them produces output that looks plausible and is wrong — which is more dangerous than an obvious error. If a document is in one of those families, do not use this tool for it.
The honest summary: 010 is guaranteed, 011 and 016 are best-effort, everything else is out of scope.
Convert first, then fix — the practical workflow
- Convert a sample, not the whole file. Take one paragraph with the awkward features (conjuncts, reph, nukta, digits) and run it through the Kruti Dev to Unicode converter.
- Compare it with the source. Read the Unicode output against the original in Word. If the sample is clean, convert the rest in one pass.
- Fix by character, not by paragraph. When a few glyphs are wrong, note them and correct them after conversion. Retrying the conversion does not help — the table is deterministic, and the same input always yields the same output.
- Keep the original file. Unicode output changes the underlying encoding. If the document has to go back to a press that requires Kruti Dev, keep the legacy source, or convert back with the reverse converter.
For long documents, a line-by-line comparison is usually faster than trying to verify the result in bulk. A paragraph that contains a reph and a conjunct will expose almost every version-related discrepancy.