impactx icon indicating copy to clipboard operation
impactx copied to clipboard

Output of transfer maps

Open jonasbjorklundsvensson opened this issue 6 months ago • 4 comments

Hi,

I'm very interested in starting to use Impactx, but some features that I routinely use are currently missing. I thought I'd let you know you have a "customer" lined up when these are implemented :) I'm very bad at coding, and am not confident that I can contribute this functionality myself, certainly not in a programmatically good way.

One feature I use a lot is transfer maps. Being able to get the first- ("R") and second-order ("T", perhaps also third-order, "U") maps for all coordinates for an arbitrary accelerator section/collection of components is paramount for a lot of my work.

I've been looking through the documentation, but I can't seem to find such functionality (might be user error on my part). It is required to optimize, for example, imaging conditions, bunch compression, CSR cancellation schemes, ...

Best Jonas

jonasbjorklundsvensson avatar May 08 '25 08:05 jonasbjorklundsvensson

Hi @jonasbjorklundsvensson,

Thanks a lot for the report! Yes, this feature (diagnostics) is definitely something we had planned to add, too.

We are currently a bit low in staff since I am on parental leave for another month, but @cemitch99 and I will sync once I am back and update here how and when we add this. Expect an update on this in mid/end of summer.

Best, Axel

ax3l avatar May 23 '25 19:05 ax3l

Hi @ax3l,

Thanks for the reply! I'll keep my eyes peeled for updates, then :)

Enjoy the remainder of your parental leave! Best, Jonas

jonasbjorklundsvensson avatar May 26 '25 06:05 jonasbjorklundsvensson

Hi again @ax3l,

Just one thing I realized I should mention in this context (in case you haven't already considered it): a related feature that is very handy to have for visualizing things is to be able to get the transfer map elements as a function of s, and not just from start to end. It's useful for visualizing the impact of, for example, sextupoles on higher-order dispersion, similarly to how one can look at the impact of quads on the Twiss functions.

Best regards, Jonas

jonasbjorklundsvensson avatar Jun 16 '25 12:06 jonasbjorklundsvensson

@cemitch99 and I discussed the following near-term roadmap:

  • we will add support for the linear map output first,
  • then add support for higher-order maps (a bit more work, as a follow-up PR)

ax3l avatar Jun 28 '25 20:06 ax3l