Since bores are not "straight", (in fact, they are not even "curved', they have "kinks" in them), you have no choice but to true two points and then some way get a third point running true with those thwo points. From a machinist standpoint, the only way to do this is to single point bore that "third point", thus establishing three points that run dead true with the machines bearings, (ie, "true").
Exactly. One picks two points on the basis that a straight line between them is the best approximation of the bore for purposes of maximizing accuracy, then single point bores to make the chamber coaxial with the straight line between those two points. The problem is in the assumption, and it is an assumption, that only two pionts can be picked and in the assumption that a point near the muzzle and a point near the throat are "always" the best representation of the true bore axis of the barrel and will result in maximum acuracy chambers and crowns.
As I understand it, the two points of critical interface for the bullet with the bore are where it enters it at the throat and where it leaves it at the muzzle. So, what two points best represent the bore axis of interest at the chamber throat? What two points best represent the axis of the bore at the muzzle? Given that the bore is kinked and wanders around randomly inside the barrel, what are the chances it is always the same two points, one near the muzzle, one near the throat? In my opinion, while the chance is not zero, it is so close to zero that the difference doesn't matter.
On the other hand, it may not matter when one gets to the range and starts shooting.
The other consideration is that one can really choose 4 points in the barrel, two near the throat, two near the muzzle. The two near the throat can be used to align for chambering, the two near the throat can be used to align for cutting the crown. That is in essence what Gordy does, which most of you already know.
The only down side to Gordy's approach is the need to index the barrel to point the bullets exit path as close to in the vertical plane as is practical, which is so easy to do even a novice like myself was able to do it on my very first barrel (which was a Savage with barrel nut - a tenon with shoulder would have been even easier). Other than that, it's really no harder to do.
By establishing the muzzle end true, and then establishing a point in the bore's area that will represent the first thing that the bullet "sees" as it leaves the case neck, then single point boring the chamber so it runs dead true with these two points, you will have no problem establishing a chamber that is "true" with the barrel's ID. That is, true with the originol two indicated points, heck, one inch further from those two points may show as much as .001 runnout. There is nothing you can do about this, because the bore is not even straight with it's own self.
I respectfully disagree with your conclusion there is nothing that can be done to compensate for the fact that 1" farther from the throat may show 0.001" (or more) runout. What can be done to compensate for that is to use two points, one in the throat, one an inch, or more, beyond, to establish the bore segment that is aligned with the spindle axis. Then, when the drilling and boring is done, the third point is created that in fact exactly aligns the chamber with the bore segment that begins at the throat and proceeds an inch or more beyond it which is exactly what Gordy's method was designed to do.
That is the difference between how most do it and Gordy and those that follow his approach do it. That said, I don't know if the difference matters or not. Gordy thinks it does, those who don't use his approach think it doesn't. I rather suspect there is no <light bulb moment> that will cause folks to change their opinion. Some folks will do it one way, some will do it the other.
Rifles done using either approach have shot world records. One approach may increase the odds of a "hummer", or not. I certainly don't have the money or time to set up a statistically significant experiment which would require doing about 32 barrels with each approach (64 total) and then having the shootout using a rail gun in the Houston Wearhouse (which is sadly no more) to take pretty much all the human element out of it as the definitive test. That's not going to happen. But if it did, and we got a difinitive answer, there would be the next topic to keep the forum discussions going.
May your reamers not chatter, your actions be true, and your groups no bigger than caliber sized bullet holes. One can always hope ...
Fitch