I'm writing to say thanks to those that posted to this topic and to leave additional information for those who find themselves here in the future.
I was having the same issue--vertical stretch commensurate with a degraded right triangle waveform controlling V-Lin. This is the only place on the internet I could find actual detail around the cause of the non-uniform stretch. It's fairly obvious once you've seen the waveforms, but this is the only place it's explained and which there is a reference image of the actual problem, on-screen.
I've just completed the teardown of my beloved 1084s-d and I can confirm the same exact failure. It appears that the glue that was put in place to minimize component movement/drop damage was excessively applied at C306. It also appears to be the only place where the glue is directly in contact with the capacitor leads. After removing the glue with a pick, I found that the C306 capacitor leads were corroded fully through. The glue must be highly corrosive in direct contact with the leads. Fortunately, most of the capacitors on the board which are glued in place (except for C306) are so close to the pcb, the glue is only in contact with their plastic casings--not the leads. They appear to be safe.
I would note for future reference that the schematic supplied, while incredibly helpful, has C306 incorrectly labeled as C506. Floobydust correctly stated "C506" in his message, as it is labeled that way on the schematic--however, the board is stenciled C306--which is in line with the numbering for components in that physical location. I would figure that C306 is in fact the correct label, given how the numbering of components is laid out. I suppose we've discovered a typo in the original documentation--ironic given that it's the part that failed.
I also discovered during the teardown that C116, one of the two large 100uF capacitors on the isolated side of the transformer--and filtering a 320Vpp signal--was leaking gel from the top (though I seemed to have no other issue with the monitor). Every other capacitor on the board looks pristine. The same is true for my Amiga 500 rev. 5 and the power supply it came with. I suppose the Daewoo (in this monitor) and the Nichicon capacitors of the late 80s are just solidly made components. I can't say so much for the adhesives of the time.
Due to other demands, I've yet to replace the capacitor (planning on doing the entire board), but will report back what I expect to be success when I'm finished.
In the meantime, if you find yourself here troubleshooting the 1084s-d, I would expect C506 (or C306 depending on what you're looking at) will need to be replaced regardless of other issues you are having if that glue is present and in contact with the capacitor leads--come to think of it, anywhere that glue is in contact with capacitor leads (or other metallic portions of components), remove the glue and replace the component.
Cheers,
Cargo