TO DO langfristig

[CsAngel-github]

Not for the prototype, but for the future platform:

Core elements:

1. How can customers create their own modules?
2. How or who reviews the new modules?
3. How can customers edit their setups? As soon as you upload them, they are permanently available on the platform, and even if you change just one letter, the entire thing has to be saved as a new setup. This will quickly become very confusing.
4. We need a way to create the module SVG. Automate files - synchronize with Inventor
5. Customize the Wildcard in a dedicated editor
6. If you buy a lot of one Module the price shall decrease per piece - skaling effekt

Nice to have:

• The setup browser should be able to search for categories / words / dates / popular.
• You can overlay two cubes, but then you won't see the second one, which can often lead to an incorrect BOM.
• Modules should be able to be deleted by right-clicking or pressing the "Delete" key.
• Modules schould be alble to be copy pasted
• The ability to select multiple cubes and move them simultaneously; otherwise, you have to move each component individually if you make a mistake by moving a single square.
• create new module feature:
  - it should be possible do select the drawed shapes and move them around
  - the frame itself is very small. its hard to understand that this tiny canva is the place you schould draw. it would be great if you could draw in a zoomed in canvas

[beniroquai]

• orientations vs. optical axis will be tricky to map if we don't have a standard
• how to deal with non-optical axis components?
• How to import optical components and wrap them in inserts?
• select properties in parts like lenses?
• economy vs professional versions?
• well-defined api per module - how? branches? repos?
• https://github.com/kmdouglass/cherry

[beniroquai]

• It will be useful that we can still select the cube module which is behind the Frame body. Once the frame is placed in the viewer, cant get access to the cubes behind it
• Group function of different cubes together once it is built correctly
• 2.5D representation
• Fluorescence spectrum viewer FPBase

[beniroquai]

• variations of location of principle not working
• we should make the curvature of the lens w.r.t. its focal lengths and the lens adaptive to the
• when exporting the diagram as a svg, we also want to export the rays
• the lenses' focal lengths value seems to be flipped (i.e. negative values form a positive lens and vice versa) What'S the reason?
• beamsplitter and mirror on first load are wrong w.r.t. its orientation
• bottom part of optikit is blacnk

From @CsAngel-github :

Dringend benötigtes feature: auch bei Strahlquellen die keine Laser sind, sollte es die möglichkeit geben nur einen Stahl anzeigen zu lassen weil die anderen strahlen schnell sehr verwirrend werden, auch wenn es physikalisch korrekt ist

zu überabeitende Module:

[ ] Torch: Strahlquelle um 90° Verdreht
[ ] fiber combiner for galvo system: hass two light sources, but the blue one is at the wrong field and in the wrong direction
keine wellenlängen oder cutoff wellenlängen einstellbar
[ ] single fiber cupler: has no light sourcs
[ ] LED Matrix: has no light source
[ ] IR bases autofocus: has no light source, no funktion
[ ] Mirror:

strahl wird auch dann reflektiert und nicht geblockt wenn die Spiegelrückseite in den Strahlengang gedreht wird

[ ] kin. Mirror 45°:

verhält sich noch wie 90°

strahl wird auch dann reflektiert und nicht geblockt wenn die Spiegelrückseite in den Strahlengang gedreht wird

[ ] XY-galvo mirror setup has no funktion jet

[ ] Bandpass: blockiert keine Strahlen falscher Wellenlänge
[ ] USB Camera with tube lens - Strahlen werden von ihr nicht blockiert
[ ] aquarium: doenst block rays

Ist es einfach die Module zu überarbeiten? könnte ich das auch erledigen? wenn ja, wie? :D wenn ich das übernehmen soll, wäre es mega wenn du für eines der Module ein kurzes video tutorial machen könntest, den rest kann ich mir dann gut erarbeiten.

LG Christine

[beniroquai]

Things I was working on:

GitHub Issue 1 (Programmer A): Configurator UX, Editing, Persistence

Title: OptiKit Configurator: editing UX + setup persistence + canvas usability

• Open a setup loaded from DB/GitHub, edit, and save back under the same name (overwrite remote version)

• Setup browser: search + filter + sort

  • search by keyword/category
  • filter by date
  • sort by popularity

• Delete modules

  • right click context action "Delete"
  • Delete key support

• Clipboard actions on modules/cubes

  • Ctrl+C copy
  • Ctrl+V paste
  • Ctrl+X cut
  • keep relative placement when pasting

• Multi-select + move

  • box select / shift-click add/remove
  • drag move all selected together
  • optional: arrow-key nudge

• Group/ungroup cubes

  • group selection into one movable unit
  • ungroup back to individual cubes

• Selection behind Frame body

  • still able to select/move cubes that are visually behind the Frame
  • fix hit testing / z-order / lock-frame toggle

• Playground viewport/layout bug

  • bottom of playground is clipped/cut off

• SVG export

  • export diagram as SVG
  • include rays in export when rays are enabled in UI

• Ray recompute trigger wiring (UI side)

  • any parameter change should trigger recompute when rays mode is enabled (debounced)

GitHub Issue 2 (Programmer B): Raytracing + Optical Model Correctness + Physical Rendering

Title: OptiKit Raytracer: optical element correctness + physical icon rendering

• Principal plane location variation is wrong

  • fix principle plane placement logic so ray results match expected model

• Lens focal length sign appears flipped

  • confirm sign convention and fix (positive convex vs negative concave)
  • align UI labels + stored data + raytracer math

• Mirror + beamsplitter wrong orientation on first load

  • fix default orientation on load
  • ensure rotation/orientation persists without needing to open parameter panel

• Torch source ray orientation is rotated by 90 degrees

  • fix coordinate transform / source frame

• Add fluorescent sample element for dichroic testing

  • converts wavelength (example: blue in -> green out)
  • define conversion rules + intensity behavior

• Light source completeness

  • single-mode fiber laser: point source + divergence parameter
  • LED matrix: ensure it emits rays (currently not a light source)

• Mirror mount behavior

  • Kinematic mirror mount 45 behaves like 90: fix
  • XY-galvo mirror should behave like 45 kinematic mirror (define correct reflection + axes)

• Spectral filtering correctness

  • bandpass filter must block rays outside passband (wavelength dependent)

• Aquarium interaction

  • should block rays for half its volume (define geometry boundary + absorption/refraction rule)

• Replace SVG-only icons with model-derived rendering

  • compute appearance from the same physical model used for ray tracing
  • examples: mirror drawn as 45 degree line, lens curvature/diameter/position shown
  • add toggle: computed physical icons vs hand-drawn icons
  • applies to both library view and canvas view

• Ray export correctness

  • exported rays must match current computed rays (including after parameter changes)

[beniroquai]

Small image you mean they will be as shown there as separate unit? Maybe a 2d block illustration how they are stacked together will be also helpful

Regarding the fluorescence marker configurator: your idea is good, but you’re providing too much information for dyes that are measured using the same filter combination, it's overestimated.
DAPI and Hoechst use identical wavelengths in practice; the same applies to GFP, Alexa Fluor 488, and FITC. And I don’t think we even have that many slightly different lasers or LEDs.