You think that particular design can withstand that? I'd have to disagree.
The nacelle pylons of a
Constitution-class ship are three metres thick, support gigantic nacelles that account for over ten percent of the ship's volume each, and yet can withstand several thousand
g while bearing the weight of something that must have a mass of some 30,000 tons
It's common sense. You don't put something big and heavy atop something really light and thin, even in a microgravity environment. There's bound to be some structural deformation.
...You have
seen nacelles, right?

All we're establishing here is that starships
cannot possibly function without structural integrity fields. Which is kind of my argument given the kind of forces we see them routinely dealing with.
A Galaxy-class starship can't, however. Without a SIF, a Galaxy-class will experience structural damage with accelerations of only 3g. So we've got a discrepancy there in our figures.
So you're agreeing that the structural integrity field does increase the structural strength of a ship by hundreds of times? Cool. There's no discrepancy there.
See above. While accelerations of 1000g are not unheard of at full impulse, it is by no means something that's supposed to be routine.
The
TNG Technical Manual specifies that "required performance" of the impulse engines is an acceleration of 10km/sec². "Full impulse" is normally given as being 0.25
c and if a ship is going to dawdle along pulling only a sluggardly 3
g at maximum then getting to full impulse from full stop is going to take it about a month.
A SIF almost has to max itself out to keep a Galaxy-ship's hull from buckling at such times, but thankfully it isn't something that it has to do for very long.
Says who?
No one actually said that, certainly not me, but most Federation starships aren't homogenous from stem to stern, they're comprised of different sections with varying amounts of mass. So some sections may fail before others even if they are all subjected to the same forces simultaneously. A SIF is there to ensure a big ship's spaceframe isn't distorted by using forcefields to tighten and flex parts of the hull where needed.
Because a SIF has limits. Just like deflector shields, It isn't some infinitely-powered device that can protect a ship indefinitely.
Nobody's saying otherwise. Something not being
infinitely powerful does not mean it's
pointless. Shields are going to collapse anyway, so do you not bother to raise them? Of course not.
Well, antimatter containment isn't that remarkably robust, because there's been times it nearly failed on the Enterprise and completely failed on doomed non-hero ships. And I would argue that artificial gravity is tied into a ship's life-support system and will always be among the last systems to fail even in the event of total power loss.
When fully fuelled a
Galaxy-class starship holds
450 tons of antimatter. Not to mention over 200 photon torpedo warheads. To carry around that much antimatter when less than a gram coming into contact with regular matter would destroy the entire ship means that the containment systems must be, in fact,
remarkably robust, because the ship doesn't explode when it gets a bump. Think about what it would be like carrying around 450 tons of concentrated nitroglycerin with all the rocking and jostling starships do, then think about how antimatter packs about eleven billion times more blam that. Yeah. It's
robust.
Robust does not mean
impervious.
Because "a few g" can be enough to kill some people. 3g is sort of the maximum safe acceleration for everyone, although admittedly there are those that can easily withstand much more. And while the IDF does work in tandem with the SIF in regards to keeping a ship together, its primary purpose is to protect the crew.
But that's not what we're talking about. We're talking about why ships should struggle with handling a few
g when they routinely handle hundreds or thousands.