Same machine, two numbers.

PRINTING United Expo closed in Las Vegas on 25 September with 27,000 registered attendees on the organiser’s preliminary count, against more than 30,000 a year earlier, and an equipment-weight figure its two releases do not reconcile. On the floor, the apparel machines that came with numbers were presses, not printers: Stahls’ Hotronix Carousel is rated at up to 300 garments an hour on its own page and was quoted at 300 to 400 on the stand. Several printers that looked new had been announced in April or July. Section 01 reports the show the way the rest of this issue reads every number: what it counts, and whose page says it.

Sources: PRINTING United Alliance, 26 September 2026 · Stahls, Hotronix Carousel · ASI, four trends in apparel decoration

A DTG print survives “50+ cycles” at 60°C, or “60 wash cycles”, or earns a fastness grade of 4 to 5 out of 5 with no count at all. One maker’s white ink cures in 35 seconds at 356°F, while a blog calls 90 seconds at 330°F the standard for DTG ink. DTF powder should sit at 35% humidity or lower, or anywhere up to 60%. Each of those numbers is quoted accurately by its source. They do not reconcile, and the cause is rarely that someone made one up. Each leaves out the method, the unit, what it counts, or the date that would turn it into a measurement.

Sources: Winnerjet · Firebird · INKBANK · Image Armor · Winnerjet (cure temperatures) · DTF Station · VisionSub

That is the argument of this issue, and each section checks it against the page a claim came from. After the show, section 02 puts cost, cure and wash life next to each other. Section 03 reads what the certificates behind the adjectives cover. Section 04 takes the promise of rights a seller signs on every upload and holds it against the Copyright Office. Section 05 does the same for embroidery speeds and colour tolerances, and section 06 for the EU and UK rules that began to apply or fall due this autumn.

Same question, different answers

Four things a number needs

Method comes first. Winnerjet’s 50+ cycles arrives with one test condition, a 60°C machine wash, and neither a standard nor a definition of fading. Verdict 50 rates it partly true: an outcome claim from one vendor, not a benchmark.

Then the unit. Wash life comes as cycles, as a 0 to 5 grade, or as a vendor’s own test. Colour differences come as Delta E ab, as an unnamed “CIELAB”, or as a weighted delta-L. Embroidery speed comes by frame type. A number in one unit cannot be set against a number in another.

Then scope, meaning which object carries the claim. ECO PASSPORT belongs to the ink and STANDARD 100 to the garment (verdict 89). A General Certificate of Conformity is not the Children’s Product Certificate (verdict 87). The 20 mm rim figure for mugs is Canada’s rule, not the FDA’s (verdict 56).

Last, the date. One cost table is marked accurate as of October 2023 (verdict 53). The ink certificates behind section 03 were announced in 2018 and 2020.

Sources: Winnerjet · Transfer Superstars · Roland DG ink certification, 2020 · Brother DTG ink certification, 2018 · Verdict 50 · Verdict 53

A rating we had to take back

Our own register made the same mistake. Verdict 84 rated “confirmed” the statement that curing takes 1.8 to 2.4 kWh per 100 garments and adds 11 to 19 cents a shirt at $0.12 per kWh, because the site it names says exactly that. At the page’s own $0.12, the first figure comes to a fraction of a cent a shirt, and the second is 38 to 88 times larger. We ran it as Verdict of the week in the Pulse of 6 October and caught the arithmetic a day later, while writing this issue. On 7 October the rating moved to Figure unverified, and the page keeps the original reasoning under the revision. Checking that a source says something is not checking that it adds up. Section 02 shows the arithmetic.

Sources: PackPrintMatrix · Verdict 84

Dates

Where this issue is thin

Section 04 rests on four verdicts, two of them rated unverified because the checks read only part of the text, and on primary documents we could read in full. We wrote what those documents support and stopped.

Sections 02 and 05 rest on sellers’ own pages. Where an independent measurement exists, such as the dental perceptibility study in section 05, we used it. For wash life, cure time and embroidery speed there mostly isn’t one, so the comparison between pages is the finding, not a figure.

Section 06 reads some legislation directly where no rated verdict exists yet, namely the formaldehyde entries and the GBL consultation. Those passages cite the regulations themselves and no rating.

Section 01 stands on the makers’ own releases and spec pages and on the organiser’s release of 26 September, whose figures are preliminary; most of what it reports has no verdict yet, so it quotes the pages and rates nothing. Impressions Expo Dallas runs from 21 to 23 October, after this issue. What is announced there goes into Pulse once it can be checked against the makers’ own pages; this issue does not preview it. The market and factory-floor section is also the issue’s gap in material. As of 7 October the register holds eleven verdicts on market and factory-floor claims published since 1 September, and three of them are web-rated. The same two sections in the first issue stood on fifteen.

There is no Voices section this time. All seven community verdicts in the register date from 26 to 28 August and belong to the first issue. None has been added since 1 September, and a section built on forum threads nobody has checked would be filler.

Sources: Verdict register · Impressions (Dallas exhibitors, 5 Oct 2026) · PRINTING United release, 26 Sep 2026 · First issue’s overview