Just a technical note (machine side) to effectively use cutter comp on something like that, you'll need 3D Tool Comp G41.2/G42.2 and TCP G45.5 (IIRC) .. and a post that supports that..

Was the IP reporting from Southern California?

I've done a bunch of parts like that. Usually with full radius keyway cutters or lollipop tools... depending on how much negative it goes. You have to turn off gouge checking.

Those walls' CAD looks problematic. I don't see how smooth motion is even going to be possible. You're going to have to make some decisions...

Sequence numbers at Tool Changes is handled in the Control Definition. Sequence numbers at each op would be handled in ptlchg0. Change the pbld$, n$, to pbld$, *n$, Been a while since I've edited a Mastercam/IHS post. That should get you close.

My 1st look would be WCS/T/C plane conditions in the Operation. After that if it checks out I'd make sure that absolute/incremental switch in Misc. Reals/Int. is correct since backplate is good to go from what you say. Next, I'd change the depth of the feature to absolute and re-pick a point on the feature and see if it matches the number I'm getting in my code. After that... I'd have to give it some more thought.

YOU win the Internet for the day. Where all the Celeron's at? Were you looking over my shoulder sometime around 1994?

I trust Backplot MUCH more than verify.

FWIW 99.9% of the time I use incremental. Because if I ever have to change a plane depth or change workholding stack height (which changes my simulation stuff) or something like that,, I want to be able to change it and regen everything without having to reset depths. I 100% of the time will miss something somewhere... guaranteed. Incremental helps me miss less.

I miss that one too from time to time. I don't use the transform very often so I forget what settings. Thank goodness for regen.

I had 32GB in my 2nd to last rig with a Zeon ( so maybe 7 years ago...) anyway, I had a span of 4-5 months where all I was getting were these massive projects that were large complicated assemblies (yes the fasteners threads were removed) and it was bring my system to it's knees. I bumped up to 64GB and it began performing as I would expect. Now, with price of RAM, to go from 64 to 128... really not a big deal. My current rig is doing OK with 64, though I'm working in a project that is 350MB+ (only 10mb is CAD) that is slowing things down a little so I don't run simulation simultaneously on this project. I may bump it up. I'm almost done though so we shall see. Long story short, it's not always an ego thing @medaq... sometimes it's a "I've gotta be prepared for whatever comes through the door thing." When I go out to a customer's site and I'm waiting forever for their file to load so we can do a simulation... it's a BAD look. When you own your own company as Ron does, he's routinely in front of customers... it's his livelihood. Time is money and impressions are a key to success.

Spiral method is cool, however it has it's ummmmmmm, tradeoffs too. With spiral, you tend to have leftovers on the 1st and last passes unless you have/can extend the surfaces... at least on the majority of surfaces I have tried the strategy on. I'm sure theere are many feature topologies that it looks awesome on, just not many I have used it on. Like Ron and the other guys said, check your Control Def to make sure that you have the appropriate settings for arc creation "on" or enabled.

You want to use the transform rotate function. Transform itself is move from A to B, rotate which is go CW/CCW around a point, and then you have mirror which is opposite about something... basically. On the 2nd page you set the rotational values, plane to rotate about and point to rotate about.

Thank you sir may I have another! Wonder what it would run the benchmark in? Though it being a Xenon, not as fast as one would think/hope.

No document. What is different? I may know.