I have no interest in beating you up.

You can do as you say, but I found, with 4147 Plus-X and 4164 Tri-X that you
had to slide exposures so far up the curve to get acceptable (for me) shadow
contrast, that I was using EI of 20 and 80, respectively, for those films.

With TMax film, I can use an EI of 50 and 200. If I test those films, they
are as it says on the box (100 and 400), but I get better prints if Zone V
is at about 0.9 density when developed to normal contrast and to get that
requires a little more exposure. If I used those old films (as I did in the
past), the exposure times got ridiculous.

   The toe and shoulder are areas that have lower contrast
than the rest of the curve. The area in-between is called
the straight line portion even though its not really
straight in many films. The contrast of both shoulder and
toe change gradually, the toe becomming greater as it
approaches the straight line portion and the shoulder slowly
loosing contrast as it is approached. For many modern films
the range between the end of the toe and beginning of the
shoulder is more than ten stops but it also depends on the
developer, especially the shoulder end. Some developers do
not produce as high a maximum density as others.
   There is a minimum exposure to get the shadows up far
enough on the toe to get good shadow detail. Many decades
ago a researcher from Kodak Research Labs named Loyd A.
Jones (that's the correct spelling) conducted extensive
experiments over a period of decades to determine the
optimum exposure of the negative for good tonal rendition on
the print. What he discovered was that the deepest shadow
that was to have any detail had to be at a point on the toe
where the gamma (contrast of a point) was about 1/3rd of the
contrast of the straight line section of the curve. Greater
exposure made little difference up to an increase of perhaps
ten stops but even a small reduction resulted in poor shadow
detail and a noticable reduction in print quality. The
judgement of print quality was done by double-blind testing
of prints made from actual scenes. I no longer remember the
number of observers but it was considerable and there were
hundreds of scenes of various sorts. From this data Jones
developed a speed measureing system. This was used
internally at Kodak for many years. Jones wanted to find the
_minimum_ exposure for good tone rendition because
relatively thin negatives have an advantage in looking
sharper and having somewhat less grain than denser ones.
This was more true of the thick emulsion films of the time
(1920s through 1940s) than it is now, but its still
considered good practice to make negatives on the thin side
provided the tone rendition is good. In 1943 the Jones/Kodak
system was adopted by the American Standards Association
(ASA) as its speed system. Unfortunately, the ASA decided to
add a one stop fudge factor to the speeds. This resulted in
excessively dense negatives. Probably the reason for the
fudge factor was an attempt to make sure that casual
photographers got a printable image even if it was not of
optimum quality.
   In about 1958 the ASA adopted a different system. The
main reason was the difficulty of measuring speeds by the
Jones method. It required measuring two curves
simultaneously. The system which replaced it was a
modification of the second version of the German DIN system
(there was an older DIN system which was different). The DIN
system based its speeds on a certain minimum density above
gross fog and the base density. The ASA found, after
surveying nearly all the pictorial films on the market at
the time, that if a multiplier of 1.4 was added to the speed
needed to get a density of log 0.1 above fog and base
density the point would be, in virtually all cases,
identical with the one found by the Jones minimum gradient
system and would be a lot easier to measure consistently.
This new system was adopted and at the same time the fudge
factor was dropped. As a result the speeds of all films
jumped by a full stop!  The current speed standard
promulgated by the ISO and, in the USA, by NIST, is a
ammended version of the 1958 standard. The main difference
is that no standard developer is specified. The measurement
can be made with any developer the film manufacturer wants
as long as that developer is stated with the results.
   In fact, the new system still incorporates Jones' idea
of making exposures so that the darkest shadow to have
detail is still on a part of the toe having sufficient
contrast to print well. For many films many photographers
find that a small increase in exposure gives them better
shadow detail. This is probably due to variations in
exposure measurement, flare, and processing than in the film
or speed method.
    Note that the speed measurement is made at what
ammounts to a fixed, standard, contrast. This contrast is
about right for contact printing and diffusion enlarging. If
some other contrast value is desired and a different
developing time is used to obtain it the _effective_ speed
of the film will change. The standard specifies a range of
exposure at the film surface which will translate to a
specified range of densities on the developed film.
Development during testing is adjusted to obtain this range
of densities. Of course, if development is varied the range
of densities will also vary, this is, by definition, a
change in contrast. Generally a lowering of contrast will
reqire more exposure and raising of the contrast will
require less. The ISO standard contains a chart for
calculating the variation in effective speed (sometimes
called exposure index) which results from this variation.
   A note on the shape of film curves. Kodak's published
curves appear to be genuine and not approximations. They can
be compared and some idea of what the variations will do to
the reproduced gray scale can be seen. Some film and
developer combinations result in a slight hump in the mid
gray range and some films, notably ISO-350 Tri-X, have a
delibrately very long toe, meaning the gamma increases
gradually all throughout the curve until reaching extremely
high values of density (out of the usable range). If the
same scene is photographed on films of different curves and
the shadow and highlight points are chosen to be the same,
the effect will be that long toe films will tend to
reproduce the mid-grays darker and the humped films
reproduce them lighter, than in the original scene. A
stright line film will reproduce them linearly. All three
types have their places. The long toe film will tend toward
dramatic highlights where the humped film will tend toward
luminous gray values.
   Of course, the same sort of variations occur in printing
paper but there the effect is the reverse of those from the
film. A paper with a long toe will tend to reproduce mid
grays brighter than one with a staight line.