I'm sure its capacity isn't limitless, but I have never been able to run this developer to even near-exhaustion. I always seem to pull some bonehead stunt like pouring the used #2 into the #1 bottle, knocking one of them over, etc. before I extend the stuff too far. But I get a lot of mileage out of it before the screwup, typically.
you are quite right that, when designing or choosing to use a developer, there are a number of interrelated variables such as film speed, granularity, resolution, capacity, acutance, etc. However, all of these are based on the idea of the 'holy grail' of increased film speed, virtually no grain, exceptional acutance, blisteringly sharpness and unending capacity.
My main point was that, if you throw away this unachievable wish list and accept that you can't have it all, then any given developer can be the 'perfect' developer if it delivers the results that you desire. Thornton's two-bath developer gives me (subject to my own preferences) sufficient sharpness, acutance, fine grain and capacity that I get the prints that I desire. It provides a very consistent and reliable workflow that lets me concentrate on finding new images (knowing that I will get good results technically) and then knowing I have all the information that I need on the negative to realise my interpretation of the scene in the final print.
The downside of a two-bath is generally stated as a loss of speed. Using iso 400 film this is no big deal as iso 200 is more than enough for me. However, when one undertakes rigorous tests to determine personal EI in relation to personal metering and development techniques, I have found that every developer that I have tested resulted in an approximate halving of the box speed if I want to achieve good shadow detail. My personal experience is that the biggest variables with developer choice are in regard to grain, sharpness and acutance and, for my taste, the most important of these is achieving minimal grain. For others, accepting some loss of detail in deep shadows is not a problem and these people generally report that they use the box speed.
Going back to the OP, the Thornton developer processed as I indicated earlier in this thread will give good and repeatable results. If, after a number of films something is missing, then the OP has a strong foundation to post for further advice. For example, by explaining current technique and then asking how to achieve more speed, more/less sharpness, more/less grain, more/less acutance, etc. In my experience this is the best way for someone to rapidly advance their technique and, in doing so, swiftly move towards creating the basis for predictable results thereby enabling one to concentrate on the most important thing - making images and enjoying their photography.
I agree, my only concern was with the word "perfect" and people getting off on the wrong path. Sadly there are still some that constantly seek a better developer at the expense of actually taking pictures.
A rock pile ceases to be a rock pile the moment a single man contemplates it, bearing within him the image of a cathedral.
~Antoine de Saint-Exupery
if one of the op's priority's is speed the i suggest this modification ( to which i have tested and used )
replace the 6.5g/L metol with;
i've used this ( it's basically from crawley's fx4 ) with 80g sod. sulfite/L and without the 1g Kbr/L with a B bath of sod. metaborate. ?f i was using sod. carbonate ( say 12g - 20g /L ) as the B bath, i may consider using the Kbr. ?'m used to having to print through some fog, knowing i got every ounce of speed but sometimes it can get a bit much ( the fog ) and i add some in. i was using this combination for when i had to shoot hand held 120. for tri x you can easily ei800. i was shooting gp3 shanghai 100 at 200. it was fine grained ( typical of 80g sod sulfite/L ), sharp and a nice 'S' curve look. i simply don't use this often enough because i mostly work with tripod relegating speed as the factor which i compromise on.