Having read a few Google entries it seems that tubular bullets might be fairly viable and something that's been overlooked? 

On paper they would solve the problem of poor dispersion because the 'pusher' shotgun style wad,  is behind the bullet.  Plus it has full contact with the rifling. 

Then you get the extreme B.C. of a flechette without the many drawbacks. 

Any thoughts?  with regards to a long range DMR application. 

It could be as simple as a hole drilled in a normal bullet,  or conversely a completely hollow,  thin walled projo. 

also what about disk-shaped bullets (like a CD ROM).... would they need to spin for stability? 

www.dtic.mil/ndia/2004solic/flat.ppt

this apparently was a U.S. experiment... I can't open it though. 

People talked about the bullets having a poor 'mach range'.... not sure what that means

The British Army has used tubular projectiles for decades: the 30mm L15A1 DSRR - Discarding Sabot Reduced Range (formerly known as the RRTR - Reduced Range Training Round) for the Rarden AFV gun.  It was designed to match the L14A2 APDS trajectory in the early part of its trajectory, out to around 1,500m. At that point an aerodynamic shockwave effect chokes the flow of air, making the projectile unstable, and it falls to the ground at about 2,000m. It also has an annular tracer good for about 1,500m. 

The above pic, from BOCN, shows the projectile still attached to the base of the sabot.

I think this stability problem is what users meant when they talked about the 'mach range' problems...

Given that we'd want to avoid shockwave problems until 600m,  I wonder if a 762-level gun would be the minimum 'scale' required.... 

Do you think it could be brought down to a 556-level scale? 

And would it be better to have a small hole in a regular bullet,  or a full-on ring projectile

I have no idea of the technicalities involved.

Probably similar to aerodynamics to jet intakes on supersonic aircraft (except there they are trying to get the airflow subsonic for fuel burning.

Heaps of stuff on this but the maths/arithmetic gets out of control pretty quickly also the manufacture - the inside of the duct is one of the more expensive parts of the aircraft with variable geometry to handle the speed range

On a completely random note Tony,  I was told by a Swede marksman that they managed to get MOA at 900m with their 762 SLAP.  Is that reasonably impressive? 

Also another Swede posted extremely detailed info on his SLAP performance here 

https://www.sniperforums.com/threads/408-cheytac.30/

he was basically singing its praises,  saying that for him it shoots better than regular ball. 

Their setup looks simplistic,  it's a very short-for-caliber bullet in a cup-shaped sabot. 

To me, the thread you linked is not so much about dispersion, but about the significant advantages of a flat trajectory and a short time of flight at medium and long range. Errors in sight setting, holdoff (moving target) as well as crosswind have much less negative effect on the mean point of impact.  

The Swedish intention to outweigh a somewhat larger dispersion (sabot) by having the mean point of impact more often on the target (flatter trajectory, shorter time of flight)  seems to have worked for them. If they really achieved the dispersion improvements indicated -I am a little sceptical in this regard- the better for them.    

We already know the increased hit probability you get from sabot projos. 

My point was he was suggesting impressive core accuracy- MOA or under.  

This conflicts with other sources which said it could be up to 33% less accurate than normal 762. 

It also maybe suggests that bullets,  not flechettes,  are the way to go with sabots.  

Under some circumstances, yes, APDS / SLAP works better than APFSDS, in others it's the opposite.

Things must also be weighed regarding accuracy and costs - 33% worse accuracy may be worth it in some roles but not in others.

Last but not least, this debate belongs to another thread - this one is about tubular bullets, not saboted ones.