One may measure a lenses' resolution at a particular point on its
covered field, when focused at a particular distance, with a particular
type of light using a particular type of media.

Taking for granted that focus errors (or varying focus "strategies"),
collimation error, movement, sample variation and a few other factors
aren't influencing the result.

I've done a lot of lens testing and I can tell you:
=> It is not simple!

I have a couple of Zeiss lenses with not-so-good bokeh. They luckily
tend to fill roles where bokeh isn't especially important.

<BIG SNIP of a rant which be be the opposite in about two weeks time>

Yes (though not immediately noticeable).

Or better build quality?

Yes (though not immediately noticeable).

 And, if such lenses do exist, what do

More than the Takumars, but not necessarily outrageous sums.

Do you *need* "better" lenses than your Takumars?
Why do you keep agonizing over this subject?

I know what lenses I like, I know what lenses I'll get soon and I know
what lenses I dream about and might never afford...

It is possible, just a little complex.
http://www.luminous-landscape.com/essays/bokeh.shtml
http://www.flarg.com/bokeh.html

Those links mostly talk about ways to quantify the degree of donut ring
versus soft edge which is indeed not a simple thing to measure. A
standard test might look something like the luminous-landscape link's
pinhole graphs done at various distances and apertures. Then graph the
results on something like MTF charts. Rockwell's test was not that
thorough but he did a nice job of picking at least one typical
meaningful OOF scene and comparing various Nikon 50mm lens options. I'm
surprised how little difference there is! It would have been more
interesting to throw in some really cheap zooms to show that there
really can be a difference.

Other factors like roundness of aperture blade are straightforward.
Degree of cat eye effect could presumably be measured and designed for.

I'm really only interested in Nikon options but I googled around & saw
some Leica noctor 50mm f/1 samples that looked super creamy soft to me,
even in the most harsh shooting conditions like bright lights at night
but it had a dizzying cat's eye effect so I guess some people think
that's too freaky looking, plus it has severe vignetting and isn't
particularly sharp wide open. So there's always a trade-off. The Nikkor
35mm f/1.4 has some really awful donut ring bokeh wide open plus the
cat's eye thing so it really does look freaky but it's capable of
shooting that wide and it's supposed to be razor sharp stopped down just
a bit.

It seems most decent Nikkors don't have too much of a donut ring problem
except under the most difficult conditions, the OOF circles are evenly
illuminated but their edges are crisp lines so that can produce somewhat
odd looking out of focus backgrounds. If I add a teleconverter and or
closeup lens I can see things go downhill pretty quickly and the donuts
raise their ugly heads. My new 105 micro seems better than anything else
I've tried which makes sense given it's not a zoom and not cheap. It
holds up to the TC & closeup lens better than the fantabulous 70-200
zoom. My new manual focus 45/2.8 P has nice round circles at any
aperture, it seems a bit sharper than the 70-200 and with
similar/typical Nikkor bokeh circles.

Here's a comparison I did of $300 vs $1700 Nikkor zooms at around 100mm:
<http://www.edgehill.net/1/?SC=go.php&DIR=Misc/photography/bokeh/compare&PG=1&PIC=1>
but as I recall the cheap zoom did OK at around 50mm and it's possible
to get nice bokeh out of it at 200mm. But you can clearly see a
difference in this test. This could be charted & graphed.

I *would* like something for an f-mount that could do creamy soft-edged
OOF circles. Perhaps an 85mm f/1.4 or something like that or a $2,000
28mm f/1.4 (ha ha). I would hope that one doesn't have harsh bokeh! I
just got an old 28mm f/2 AI-S and we'll see how that does, sample photos
all seemed quite reasonable & it was only $200.

I'm still not clear if I can get a screw mount adapter without
additional corrective glass in it to use old 3rd party lenses but it
doesn't appear to be an option. There's always lensbabies though I'm not
sure how often I'd want a total blurry mess, it might be fun occasionally.

My Sigma 12-24 makes hideous donuts but I have to really try to get
anything out of focus at 12mm so that's not a big problem. It's nice &
rectalinear though and full frame if that ever becomes useful. A
17-35/2.8 would be nice, probably typical stern looking Nikkor bokeh
though and too similar to the 12-24 in range for me to fork out for that
one now.

Empiracle data doesn't exist in a vaccuum. There must be a mind to
evaluate, interpret and/or use it to some end, a subjective mind. In
either case, Paul Furman's post and links have probably said everything
I was about to say on this point (pardon the pun). Whether you use a
set of numbers, a diagram/graph (is it sagital vs. tangential MTF
curves that indicate bokeh and/or spherical aberration or both, my
memory fails me on this) or a photo as the means to carry "empiracle
data", bokeh (at least one aspect of it) can be easily seen in the
light distribution pattern within a blur circle or line patterns (ie.
gaussian blur soft edged evenly illuminated discs, de-doubled lines,
etc.). You canhave all the empiracle datain the world but a lens's
bokeh is either to your taste/purpose(s) or it isn't.

Yes, it is. See above and Paul Furman's post too. I don't know why you
are so hooked on numbers (perhaps overly influence by resolution
tests?) but whatever floats your boat :-)

Actually with Leica's current set of lenses its the other way round.
Their reformulated Asph (aspherical) lenses sacrifice smoothness of
bokeh for sharpness wide open. The 35mm f/2 Summicron 4th/last
generation and the 35mm f/1.4 Summilux M mount lenses verses their ASPH
counterparts are often cited examples of this new(ish) Leica trend.
Before this (1990's? till present) Leica lens's bokeh was a big factor
in the "Leica look" (still is to some extent) - how do you know they
didn't or don't design for good bokeh which is balanced against but not
at the expense of other lens attributes? Do you really know their
optical trade offs?

I can argue neither for nor against this point as I don't see how
cementing elements, other than some kind of general haze-like
non-spherical aberration soft focus effect would create any kind of
aberration, slight or otherwise. It would be like putting a slightly
dirty window inside the lens wouldn't it? - I don't see how it would
bend light rays differently to cause the slight undercorrected
spherical aberration that would engender "good bokeh".

Ironically? some early Nikon rangefinder lenses had really pleasing
bokeh (numbers, graphs and empiracle data aside).

Shall I say "crappy" or less well corrected lenses like the Holgas tend
to introduce their own set of very stylized/"artistic" lens aberrations
to the image to give it a certain soft focus (spherical aberration?)
look. Scan the web, if you want, for Holga, Diana and or Len Baby shots
and you'll catch my drift. The lens defects give a very moody look to
landscape shots.

Reportage includes portraiture as part of its umbrella. Not every shot
is a group shot taken with a 17-45 lens at f/16. Portraiture is not
limited to the K-Mart or upscale portrait studios.

If your needs are satisfied by that particular game plan then more
power to you :-). Just trying to offer you a way to keep your Pentax
lens gear and enter the digital DSLR arena with minimal financial pain.

Neither do I.But as you probably know, not all Pentax's AF lenses past
or present are plastic. Especially the current Limited edition (as well
as some current FA*, A*, A, M, K, and screw mount lenses of course)
with many built to high optical/mexhanical standards.

 What (digital) lens out there is going to

See above.

No reason, other than you wanting to make Pentax rich allover again
;-).

Good idea. A paint can tied as a weight to the tripod or some downward
pressure applied to the camera on the tripod can add "image
stabilisation" to the ricketiest of pods.

I've done the same thing with a Leicaflex when I owned it. Good trick.

Only for those shots where a tripod (due to topography (rocky/uneven
terrain, tight spaces, etc.) or some other reason/encumbrance) would be
impractical and flash (of course, with most landscapes) is not an
option.

Film helps too ;-)

Well then you know your needs and your current equipment, it seems, for
the most part, satisfies those needs.

Tony would beg to differ with you, he tends to think that Pentax lenses
have more (barrel?) distortion.

multicoating that is

35mm format lenses (all with excellent SMC coatings) will work on the
Pentax DSLRs too.

(We don't know, because it is difficult or

Pop may test them, but my standards are much higher than their ratings
it would seem...

Leica. Zeiss. Current Pentax lenses such as the 85/1.4 FA*, possibly
the 77mm f/1.8 Limited, 50/1.4 FA (probably, other than me,
"thebokehking" of lenses ;-)) and my 20-35/4 FA when focused close and
used wide open has beautiful bokeh too - there are probably other but I
don't own every current or near current Pentax lens, just wish I did
;-). Some of the new 4/3 Olympus lenses have "Leica worthy" (if not
Pentax worthy bokeh.

 Or better build quality?  And, if such lenses do exist, what do

A lot :-) ;-)

You seem to know what you want to do and how to achieve it with the
gear you want. But if you ever did want to go the DSLR route, don't
discount Pentax's current DSLR efforts. They are very competitive (more
so in some areas such as price (under $1000 for the pro body and about
$350 for one of the amature bodies) vs. the usefulness (perhaps not to
you but to others) of the features you get for those low prices (IE.
anti-shake with any lens built into two of the bodies and water
proofing built into the pro model). And as mentioned, there are many
Pentax lenses that, though not as highly well built (othe than the
Limiteds and perghaps some FA* lenses) are no small shakes (pardon the
pun) optically bokehwise, anti-flare coatings wise, sharpness-wise,
etc.).