Photo2PatternFree Anchor to DMC converter
Anchor to DMC conversion, three candidates a row
Paste a pattern’s whole Anchor thread list and get the DMC substitutions back together, each with the two nearest alternatives and the distance to them. It is free and asks for no account. No official Anchor-to-DMC table exists — the 464 rows here come from two published charts, they reach only 360 distinct DMC codes, and 104 of the numbers never convert back to themselves, so every row that cannot be answered cleanly says so.
- 100% free
- No signup
- 464 Anchor rows
- 3 candidates each
- Whole key at once
Paste the whole key at once — commas, spaces, semicolons and line breaks all separate, and up to 120 numbers are read. Converting a list rather than a number is the point: it is the only way to see which of them are about to become the same skein.
6 of 6 recognized
6 distinct DMC skeins to buy, 5 carrying a warning
| Anchor | What the chart says | Three candidates, and how far apart they are |
|---|---|---|
| Anchor 049 | DMC 3689Mauve Light |
The two published charts disagree here: the second sends Anchor 049 to DMC 3869, which this palette has no row for, so it cannot be drawn or measured against the first. Take both numbers to the shop. |
| Anchor 888 | DMC 3045Yellow Beige Dk |
The two published charts disagree here: one sends Anchor 888 to DMC 3045 and the other to DMC 3046, 0.114 away in Oklab. Both are somebody's eye on a shade card and neither is a measurement. |
| Anchor 028 | DMC 956Geranium |
DMC 956 is also what this chart offers for Anchor 038, 040, 054. Those 4 Anchor shades are separate colors on Anchor's own card and one DMC skein is being asked to cover all of them. |
| Anchor 001 | DMC 5200 | DMC 5200 is Snow White, printed on the shade card as B5200. The conversion chart drops the letter, which is why it looks like a missing number rather than the whitest thread in the range. |
| Anchor 1200 | DMC 62 | DMC 62 is a variegated thread: the color changes along the skein, so no single swatch can stand for it and none is drawn. It is stitched exactly like a solid, and the pattern of the change is the point of buying it. |
| Anchor 403 | DMC 310Black |
|
The skein count above is exact and it is the figure worth carrying: 6 recognized Anchor numbers become 6 DMC purchases, because 189 of the chart's 464 rows share a DMC code with another row. Everything else on this page is ranked opinion — the distances are computed from digitized RGB, and the digitization is not official.
Screen color is an approximation of floss, not a match. Every monitor is calibrated differently, and the published color data for dyed and molded goods varies between sources — for embroidery floss the disagreement between two independent color tables runs to a tenth of the range. Use the chart to pick candidates and the physical shade card, bead or brick to decide between them.
How to convert a pattern's thread list without losing a shade
Paste the whole key rather than one number; that alone is why this page beats a lookup table.
Paste the pattern key rather than one number
Take the whole thread list off the chart legend and drop it in — commas, spaces, semicolons and line breaks all separate, leading zeros are optional, and up to 120 numbers are read in one pass. Converting the list together is what makes the interesting problem visible, because 189 of the chart's 464 rows send their Anchor number to a DMC code that another Anchor number also claims. One at a time you never see two of your threads arriving at the same skein.
Read the row, not the first cell of it
Each converted line gives the code the published chart names, then the two DMC shades nearest to it with the perceptual distance to each, then whatever is wrong with the row. A distance under about 0.02 means the alternative is closer to the chart's answer than most adjacent shades in the DMC range are to each other, so it is a real option and not a consolation prize. Rows the two published charts genuinely disagree about carry both answers side by side — there are 2 of those, Anchor 049 and Anchor 888.
Buy against the skein count, then settle it in the shop
The figure at the top of the plate is the number of distinct DMC skeins the list actually needs, which is routinely fewer than the number of Anchor numbers you pasted, and it is exact because it is a count of physical objects rather than an estimate of one. Download it as CSV or print it, take it to the counter, and use the alternatives column there: a substitution nobody official signed off is a decision to make in front of real thread, not in front of a monitor.
Technical specifications
| Rows on the conversion table | 464 Anchor numbers, spanning 1 to 9575, sourced from two published charts and reconciled row by row |
|---|---|
| How the two sources compare | Agreement on 452 rows (97%), 10 rows differing only by PDF extraction damage in the second source, and 2 genuine disagreements: Anchor 049 and Anchor 888 |
| Collapse in the mapping | 464 Anchor numbers reach 360 DMC codes. 85 DMC codes are shared across 189 rows; the deepest collapse is 4 Anchor shades onto one code, at DMC 956, 3687, 798, 797, 738 |
| Reverse behavior | 104 of the 464 numbers do not come back to themselves through DMC, and each is flagged where it appears. In the DMC-first direction, 117 of the 454 shades have no Anchor row at all |
| Codes carrying no swatch | 23: 22 variegated threads in the 48-126 block plus DMC 5200 Snow White, which the shade card prints as B5200. Each is named in words rather than drawn |
| What the input accepts | Up to 120 numbers a pass, separated by commas, spaces, semicolons or line breaks. Leading zeros optional both ways; Blanc, White, Ecru and B5200 recognized by name in the DMC direction |
| Candidates per row | Three — the chart's own answer plus the two nearest DMC shades to it by Oklab distance, with the figure shown. Contested rows show both published answers and the gap between them |
| The file it produces | CSV carrying the input number, the chart's answer with name and hex, both alternatives with distances, and the full warning text for the row |
Frequently asked questions
Is there an official Anchor to DMC conversion chart?
No, and there never has been. Anchor and DMC are competing brands with separate dye recipes, separate mills and no incentive whatsoever to publish a table telling their customers which of the other firm's threads to buy instead. What circulates as the conversion chart is a substitution guide assembled by matching skeins by eye — useful, widely reproduced, and an opinion. Anywhere you see one of these tables presented as a fact, including in pattern software, it is the same class of document as this one.
Two charts agreed on 452 of 464 rows. Doesn't that prove the mapping is right?
It proves they are related, which is a different claim. Two published charts were compared row by row for this page and matched on 452 of 464, about 97%. Independent eyes comparing thousands of skein pairs would not converge that hard; a common ancestor — almost certainly the thread manufacturer's own substitution list — explains it in one line. Ten more rows differed only because one source's PDF text layer had dropped a leading digit, rendering Anchor 161 as DMC "83", a number DMC does not use. Two copies of one opinion, plus some scanning damage, is still one opinion.
Why do several of my Anchor numbers convert to the same DMC?
Because the mapping is many-to-one and the chart makes no apology for it: 464 Anchor numbers land on 360 DMC codes, so 85 DMC codes are each doing the work of several Anchor shades across 189 rows. The worst cases carry 4 Anchor numbers apiece — DMC 956, 3687, 798, 797, 738. On a design that used two of those Anchor shades to separate a highlight from the tone under it, converting flattens the pair into one thread and the shading disappears. That is the single most damaging thing conversion does, and it is invisible until you convert the whole key at once.
Why does converting back not return the number I started with?
Because a many-to-one map has no inverse, so the return trip has to pick one Anchor number out of several and it will not always pick yours. 104 of the 464 Anchor numbers fail that test, and the page says so on every row where it happens. The practical consequence: never verify a conversion by running it backwards, and never convert a chart from Anchor to DMC and back for any reason, because on those 104 shades the round trip silently changes the color.
Some rows have no swatch at all. What are those?
23 of the chart's DMC codes name a thread that no single square of color can represent. Twenty-two of them are DMC variegated threads — the block numbered 48 to 126, where the color changes along the length of the skein, so a swatch would have to be a gradient and would still be wrong. The last is DMC 5200, which is Snow White and which the shade card actually prints as B5200; the conversion chart drops the letter, so it looks like a missing number rather than the brightest white in the range. The page names each of them instead of drawing a hole or inventing a color.
Can I use this to convert DMC numbers into Anchor?
You can switch the direction and the page will answer, but it answers a question the chart was not built for and it says so. The table was written to send Anchor stitchers to DMC, so it covers only 360 DMC codes and leaves 117 of the 454 shades with no Anchor number pointing at them at all. Where a DMC number has no entry the page lists the nearest shades that do have one, with distances, so at least the substitution is explicit. Where it has several, all of them are shown, because nothing in the data justifies choosing between them.
How close are the alternatives you offer to the chart's own answer?
Close enough to matter and it is measured on every row rather than asserted. The two alternatives beside each result are the nearest DMC shades to the one the chart named, ranked by distance in Oklab — the same perceptual space every pattern page on this site snaps cells with, chosen because Euclidean distance in RGB matches dark shades to each other far too eagerly. Where the chart's own answer and a neighbor sit within a few thousandths, the substitution has effectively been made for you by whoever assembled the table, and picking the other one costs nothing.
About substitution charts, and what they quietly throw away
A conversion between two floss ranges is a substitution, and substitution is the word worth holding on to. Anchor and DMC are dyed by rival firms to recipes that are trade secrets, in different mills, on cotton from different lots; there is no shared standard they are both approximating, so there is no sense in which an Anchor number and a DMC number can be the same color. What exists instead is somebody’s judgment that one is the closest thing on the other shelf. That judgment is worth having — it is how every stitcher who has ever worked a European chart with an American thread box got through it — but it is a recommendation, and a page that renders it as an equals sign has misrepresented what it knows.
Two independently published tables were reconciled row by row to build the data behind this page, and the shape of the result is the whole story. They agree on 452 of 464 rows, which sounds like corroboration and is not: nobody matching skeins by eye reproduces another person’s 97% of the time, so the agreement is inheritance from a single ancestor table rather than confirmation. Where they part company is more informative than where they meet. Ten rows differ only because one source was scraped out of a PDF whose text layer had lost a leading digit — Anchor 1023 comes out as DMC 312, a blue sitting between two salmons, which is the signature of a broken extraction rather than of a disagreement. That leaves 2 rows where two people genuinely saw different threads, and this page shows both answers on those rows rather than picking one. Everything else about the mapping is arithmetic on top of that: 464 numbers reach 360 codes, the deepest collapse puts 4 Anchor shades onto DMC 956 alone, and 104 numbers never find their way home.
The practical damage is concentrated in that collapse, and it lands on shading. A designer who used two adjacent Anchor shades to lift a highlight off the tone beneath it has, after conversion, two identical skeins and a flat patch — which is why this tool insists on taking the whole key at once, since the collision is invisible when numbers go through one at a time. When the substitution matters, put the candidates against real thread: the DMC color chart explains why the swatches here are approximations in the first place and will not stand in for a shade card. If the chart you are converting came out of a machine file rather than off paper, the thread stops are already listed in it — the DST file viewer reads them out of a Tajima stitch file, the embroidery file converter moves that file between the formats a machine will accept, and the free embroidery digitizer is where artwork becomes stitches to begin with. For a design that has no thread list yet because it started life as a photograph, the cross stitch pattern maker assigns DMC numbers directly and skips the conversion entirely, which is always the better route when it is available.
Product and company names shown here are the trademarks of their respective owners. Photo2Pattern is an independent tool and is not affiliated with, sponsored by or endorsed by any of them.
Where the conversion runs
Nothing you type here leaves the tab. The arithmetic runs in JavaScript on your own machine — there is no server to send it to, no account to sign in to and nothing stored between visits. Close the tab and nothing of it remains.
The 464-row table and all 454 DMC shades are compiled into this page, so a pasted thread list is matched against data already sitting in your browser. Nothing is looked up, nothing is logged, and a pattern key — which is a fair description of an unpublished design — never leaves the machine you typed it on.