THe first torch he had was oxygen - which was running out. That was when he thought he might have to change plans for the blue surface-decorated weight.
He normally uses oxy-acetylene.
The original plan was to make the two weights, the clear crystal close-pack and the blue surface decorated.
I think they've found that it's a practical and economical method to do two weights at one time - it certainly was for the purposes of this demonstration because he wasn't using the glass from the pot, but his own crystal and things which were kept in a second annealling oven - perhaps with the temperature set for his needs there on the day.
When the second torch was produced, he was able to continue with the original plan.
Wuff has very kindly written to me privately, to point out my error in thinking that interesting bits of the canes might have been left behind in the bit of the clear crystal glass which was chopped off.

The actual canes had been "squeezed" into the dome - the outer ring of white-surrounded canes were slightly longer than the others - it would only have been a few squished bits of them.