Originally posted by gripen
Well, then I quess the Spitfire had really excellent inlet design because in every case in the flight tested critical altitude seems to be at least 2k higher than given in the power charts. Another possibility is that flying speed used for power curves is very low.
gripen
Keep in mind that there were many extarnal modifications, ie. the Spit I you mentioned was tested w/o the additional armor plates, windscreen, which came with a 9mph speed decrease (IIRC), and as such a reduction in rammed FTH. I don`t know if the Spit had any special inlet design, to me it looks like a simply hole on the fusalage, but I would hardly think it would have anything to do with it. Anyway, if you have any info on the details of the "really excellent inlet design", please share.
Looking on the Bf 109 G-2 and G-6, for example, the additional drag of the G-6 slowed it down by -25 km/h at FTH compared to the G-2, and interacted with the decrease of FTH of 400m (1300 ft). You may notice the exactly similiar decrease in case of MkV W.3134, with and w/o snowguards fitted (370.5mph/20.1k and 365/18.8k). The loss of FTH decreases power available at altitude, which decreases speed, which again decreases the power and so on until it balances out.
So, IMHO it`s simple a case that the Merlin III (and possibly the others) curves were done using later Spits with more extra drag from new equipment, such as filters, armored windscreen etc, and comparing their FTH with the earlier, cleaner ones makes little sense. Looking at MW`s rammed power chart for Merlin III, the FTH is at around 17.5k-18k ft or so. This agrees very well with the measured FTH of Spit I (Merlin III) of N.3171, at 18.9k ft. I dont know what equipment the plane exactly had, the difference is not great, and easily explainable by extra drag from newer equipment or difference with indidvidual planes. As an example for the latter, the well known JL 165, the same plane in different test was found the have a FTH of 14 400 and 13 800. Same plane, just different tests!
In any case, the point is that MW`s chart shows dynamic powers in max level speed flight, and this should be taken into account when compared to, say, a static chart. Some might come to false conclusions about high altitude performance while being busy comparing apples and oranges.

Voodoo : The acceleation figures were
calculated by greg shaw - he posted in this thread, ask him about the details of the calcs - , based on available max SL speeds at given powers and weight. Thus they are based only indirectly on German testings, but should reflect the relative merits very well. After all, it`s not hard to see that 2000 HP will accelerate a 3362 kg plane better than 1840 HP at 3859 kg heavy one, especially if the latter needs more power to obtain the same speed.