Armour is expensive, tiring to wear, can make it harder to breathe, hear and see, and can be uncomfortable. If it didn't work - if things went through it like it wasn't there - why bother? The short answer is that movies that show armour as useless are unrealistic.

However, the disadvantages mean that there are compromises.

It's important for armour to work against common threats. One of the main threats on the Medieval battlefield was arrows and crossbow bolts (and not just in Europe - the same was the case across much of the world). How much armour is needed to stop arrows from bows with draw weights of well over 100lb? Bolts from crossbows that were just as, or even more, powerful? Simplifying somewhat [1], plate armour of 2-3mm thickness will stop all arrows, and crossbow bolts from all but the most powerful crossbows (Williams, 2003). Unsurprisingly, this is a common thickness for breastplates (Goll, 2013). Breastplates intended to be worn over mail were usually thinner, often about 1.5mm - this, together with the underlying layer of mail, would be enough to stop the arrows/bolts.

Not all armour was plate. Some armours were made of small plates, overlapping, and either joined to each other by lacing (lamellar armour), or riveted inside cloth (brigandine), or laced or stapled to the outside of a cloth foundation (scale). The individual plates in these armours were thin, often about 0.8mm thick. However, they overlapped, and often by enough so that any incoming arrow/bolt would need to pierce 4 layers of metal. As with plate, this would be enough to stop arrows/bolts.

Mail was another common type of armour. There have been very few good modern scientific tests of the effectiveness of Medieval mail. However, from literature and law, we know that a sandwich of textile armour, mail, and more textile armour was effective. For example, Baha ad-Din, in his Life of Saladin, noted the effectiveness of this sandwich:

every foot-soldier wore a vest of thick felt and a coat of mail so dense and strong that our arrows made no impression on them. ... I saw some with from one to ten arrows sticking in them, and still advancing at their ordinary pace without leaving the ranks.

and European chroniclers wrote similarly:

a hauberk and a linen tunic padded many times and difficult to penetrate, ingeniously worked with a needle and consequently called in the vernacular a Pourpoint.

(as quoted in Heath (2019)). The Saracens would also wear a similar textile-mail-textile sandwich, sometimes made as a single garment - an armoured kaftan called a khazagand. The Ordinance of St. Maximin de Tréves (1473) prescribed similar armour for some soldiers, with a mail shirt under a ten layer jack being an acceptable alternative to a brigandine. The additional textile armour worn on top of the mail suggests that the mail itself (and the textile layers underneath) were not by themselves sufficient.

All of these alternatives - plate, lamellar, brigandine, and textile-mail-textile - are heavy. To armour the entire body, arms and legs as well as the torso, to this degree would result in impractically heavy armour. Two other factors matters: arrow wounds to the torso are much more likely to be fatal than arrow wounds to the arms or legs, and weight carried on the arms and legs is more detrimental than weight on the torso. Together, these meant that the best armour was the torso armour and helmet, and arms and legs were more lightly armoured. For example, plate armour on the arms and legs was often 0.8-1mm thick (Goll, 2013). Sometimes, arms and legs were unarmoured.

Since the torso armour was the minimum that would reliably stop high power arrows/bolts, this means that the arms and legs were vulnerable to such attacks. Thus, we can ask what the function of arm and leg armour was. First, it could still be effective against arrows and bolts. Plate armour on arms and legs was curved, and could deflect arrows/bolts that struck it at an angle. Arrows and bolts lost energy through drag, and were less likely to penetrate at long range. Second, this thinner armour will still be effective against swords. Contrary to what is shown in some movies, swords are not very effective against armour - a sword typically needs to make a bigger hole in armour than an arrow/bolt to be effective, and a sword blow/thrust typically has less energy than an arrow/bolt from a powerful bow/crossbow.

Plate armour of 2-3mm thickness could be effective against early guns:

• https://i.pinimg.com/564x/26/aa/f7/26aaf7b5085e23c4932f574d21980bef.jpg

However, as muzzle energies climbed (notably, with the introduction of the heavy musket), such armours became vulnerable. As guns that could effectively pierce such armour became more common, that armour was no longer worth the weight and the expense, and was either (a) abandoned, or (b) thickened, which happened with cavalry breastplates which climbed in thickness to 6mm or even more.

If they were in fact effective, how did the soldiers die?

As noted above, while the most protective parts of an armour would stop almost all threats, arms and legs were usually less well protected. Armours typically had gaps. Many armours left the insides and backs of legs exposed, buttocks exposed. Armpits, insides of elbows, the palms of the hands, and the face were exposed by many armours. Swords could be effective against such gaps, and tactics such as half-swording were used to drive the sword point into gaps. The dagger (often used in combination with wrestling) was often a very effective anti-armour weapon. While swords were not very effective at going through armour (as opposed to going around armour, as just described), other weapons such as halberds and other heavy polearms could strike with much more energy, and should have been effective against the thinner parts of an armour.

Footnote:

[1] The resistance of plate armour to penetration depends on the hardness and toughness of the metal, as well as the thickness. The hardness is affected by carbon content and heat treatment - steel armour (with more carbon than iron) is more effective than iron armour, even if unhardened by heat treatment, and with good heat treatment is even more effective. Toughness can be reduced by excess slag inclusions in the iron/steel (and can also be reduced by bad heat treatment which can make the armour brittle). The thickness values given assume iron or low carbon steel, but of good quality (i.e., without excess slag).

References:

Beha ed-Din, The Life of Saladin, Palestine Pilgrims' Text Society, 1897. https://archive.org/details/libraryofpalesti13paleuoft/page/n4

Matthias Goll, Iron Documents: Interdisciplinary studies on the technology of late medieval European plate armour production between 1350 and 1500, PhD thesis, University of Heidelberg, 2013.

Ian Heath, Armies and Enemies of the Crusades, 2nd edition, Lulu, 2019.

Alan Williams, The Knight and the Blast Furnace, Brill, 2003.