When tanks were used primarily for infantry support, there was no such thing as too many machineguns. Little machinegun cupolas or full on turrets were quite useful. However, as tanks evolved, the bristle of machineguns boiled down to just one in a ball mount in the front hull. Even this feature came under scrutiny, however, as this British document shows.
"As far as is known no useful progress has been made on this and as has been stated on innumerable amount of occasions before, unless the Ordnance agrees to change the shape of the front glacis plate, they are unlikely to attain a satisfactory solution. It is extremely unfortunate that no honest attempt has been made to tackle this design in the past nine months, since the short ranges of Tunisia have shown that the machine gun will come into its own and this bow machine gun with no sight is merely a means of wasting valuable ammunition. Quite recently it was reported from Tunisia that a tank had expended twice its normal complement of S.A.A. Obviously it cannot be wasted hose-piping it from an unsighted gun.
In the absence of any sight, the following alternative solution is offered. As the bow machine gun stands today, it is the weakest point of the tank. That is to say, the strongest part of the front glacis plate of the M.4 will withstand a striking velocity of 2400 f.s. from 3" A.P.C. M.62 ammunition. The bow gun represents only 1300 f.s. S.V. Officers returned from Africa are of the opinion that this bow machine gun, whether provided with a sight or no, is a waste of ammunition, since it cannot be commanded. They would much prefer twin machine guns in the co-axial mount. We continue to press the Armoured Force for this. Meanwhile, it is recommended that the whole question should be reconsidered by D.R.A.C. with the view to agreeing to the elimination of this bow machine gun in tanks as long as no sight is forthcoming."
-British Army Staff (AFV) Situation Report as on 18th July, 1943
Sadly for the officers who gave the recommendation, the situation did not change. The machinegun remained, and no sight was added. Not until April 5th, 1945, were trials held to see if a sight helped out after all. A stock Cromwell, as well as one with a 1.9x No.39 telescopic sight were used for this test. Firing was performed against 3x3 foot targets. Guns were fired continuously for 10 seconds. Unfortunately it is not mentioned how many rounds were fired.
When stationary, the results for gunnery without a sight were pretty poor. Even with 1 tracer in 3 instead of the usual 5, the chances to hit at any range were low. A linked telescopic sight helped considerably.
The next step of the trial was firing on the move. The Cromwells started at 500 yards, driving continuously to the target at 10 mph. The guns were fired from 400 to 300 yards, then a 100 yard pause was taken, and then fired again from 200 to 100 yards. An estimated 60-80 rounds were fired, but it was difficult to estimate because the BESAs were jamming constantly, even with carefully inspected belts and ammunition.
The addition of a sight still led to an improvement, but not as great as when the tank is stationary.
A more interesting trial followed: a showdown between the modified Cromwell and a Sherman V.
Curiously, the fire of the Cromwell at 300-400 yards was less effective. The trials record explains it as: "when firing without a sight, the gunners tend either to fail to find the right position for the gun, or else find the correct lay and retain it for the remainder of the 10 seconds allowed, the number of hits obtained follows no statistical distribution. This is more evident when firing at the halt as the movement of the tank is continually altering the position of the gun relative to the gunner, so taht he is unable to learn the right position. ... A curious phenomenon appears in trial iv, where the number of hits without a sight at 400-300 yds. exceeds the number obtained with a sight (tanks moving). It is considered that this is due to the gunners realising that when firing without at sight their only chance of hitting the target was to fire continuously. When using a sight, they fired in short bursts when they thought they were on target. Since, on the move, the chance of any round hitting is small at this range, whether a sight is used or not, when firing from a unstabilized gun, the number of hits is largely a function of the number of rounds fired."
Tank Armament Research Report No.27, The Value of a Sight when Shooting with a Bow Machine Gun