Short answer

There is no Rolex Milgauss 10190. The real Milgauss references are 6541 (1956), 1019 (1960 to 1988), 116400 and 116400GV (2007 onward), plus the Z-Blue dial version of the GV. So when a listing is tagged "Milgauss 10190," the number is either a seller's own catalog code or a typo. Ask which genuine reference the replica is copied from before you look at anything else, because a vintage 1019 clone and a modern 116400GV clone fail in different places.

Why the Milgauss is harder to fake than a Submariner

Three parts of this watch are difficult to copy on a budget.

• The soft iron Faraday cage around the movement. Genuine Milgauss watches resist magnetic fields up to 1,000 gauss. Most replicas skip the inner cage, which is why their caseback sits flatter on the wrist and why the watch often weighs a gram or two less.
• The caliber 3131. It has a non-magnetic escapement and a paramagnetic alloy balance spring, not a plain 3130. Replicas usually run an A2836 or a clone 3131 that only looks correct from the back.
• The green sapphire crystal on the 116400GV. The color is in the glass itself, not applied on top. Cheap builds use clear sapphire with a green tint film, and the rim of the glass gives it away in direct light.

What I check first on a replica Milgauss

Crystal and dial edge

Tilt the watch toward a window and look at the rim of the crystal, not the dial. On a real GV the green stays consistent from every angle. On a coated copy the tint goes darker at the edges and can wear off. If the seller's photos were shot in soft studio light only, ask for a shot in sunlight.

Lightning bolt seconds hand

The bolt should be sharp, flat, orange, and parallel to the dial. Weak copies show a rounded bolt, a wrong orange (too red or too peach), or a hand that wobbles as it sweeps. Compare it to press photos of the real watch side by side, not from memory.

Dial printing and lume

Check the thickness of the ROLEX and MILGAUSS text, the shape of the coronet, and the spacing of the minute track. Lume should be even with no bubbling or dust under the glass. On Z-Blue dials the blue needs to be an electric blue, not navy, and it should shift as you rotate the watch.

Case, crown, and caseback

The modern case is 40mm with a domed bezel and no crown guards. Because of the Faraday cage, genuine casebacks sit slightly proud, so a thin flat caseback is a clue the shield is missing. Look at engraving depth and font, and feel whether the caseback threads on smoothly.

Movement

The seconds hand should sweep at 28,800 beats per hour, smooth and without stutter. It should hack when you pull the crown, and the date should flip cleanly if the reference has one. Ask for a timegrapher photo showing rate and amplitude rather than trusting a spec sheet.

Bracelet and clasp

Oyster bracelet, solid end links that sit flush against the lugs, no rattle at the clasp, and an Easylink extension that actually works. Stamped clasp codes should match the case reference quoted by the seller.

Price tiers and what they buy

Budget builds under roughly $150 get the silhouette right and little else: coated green glass, a flat caseback, and a generic movement. Mid tier builds around $250 to $450 usually fix the dial print, the hand shape, and the bracelet feel, but still skip the magnetic shielding. Top tier builds cost more and are the only ones worth comparing to a genuine watch, with a clone 3131, a proper green sapphire on GV versions, and correct caseback geometry.

Vintage 1019 versus modern 116400GV

The 1019 (1960 to 1988) has no lightning bolt hand, no green glass, a slimmer case, and a brushed bezel. A replica of it lives or dies on dial print and how the patina is faked. Most listings using the number 10190 are pointing at a modern 116400GV, so confirm which watch you are paying for in writing before money changes hands.

Bottom line

Treat 10190 as a flag on the listing, not on the watch. Pin down the real reference, then judge the crystal, the bolt hand, the caseback thickness, and the movement sweep. If a seller cannot name the reference they copied, they cannot tell you what they got right either.