project settings for final HD export with 4K footage

I’m doing a project where all of the footage is 4K. The final product will be HD.  I’ll use the extra resolution to reframe, zoom in and what not.
My question is, should the project be in 4K and then exported in HD or should I make an HD project from the start, using the extra res on the 4K footage.

Is there one option better then the other in terms of the final result. My system can handle the 4k project and I'm using proxy files for the edit anyway.

Thanks in advance for the feedback.

I think I've found the information I was looking for in this video.
Don't want to waste anyone's time, but if you have more insight on the subject it will be most certainly welcome.
Thank you

I edit my 4k footage in a HD timeline/project and have no quality issues.
I did test to edit in 4k to export to HD but it didn't make a difference.
And editing in HD resolution gives you instant feedback how far you can push your 4K material.
So it works very well for me.

If you want to zoom and pan, edit in 1080 with your 4K footage. I've been doing it daily for TV broadcast for years now and it works great.  The clips you want to zoom in to, set Spacial Conform to None.  100% scale will be the full 4K resolution.  You can even enlarge up a bit safely.  50% will shrink it down to 1080 size.  You'll have good latitude to play with.

That works a treat. That is exactly what I was looking for. The best part is that, as soon as you change the spatial conform to none, FCP immediately "shows" you how far you can push.

Thank you so much

Ok workflow but I did tests that showed there is no image quality difference whether you set fit to none, and then zooming out, or leave fit to fill and zoom in over 100%.

Basically a 4k image is double the size of HD. So if you leave it at fit to fill you could zoom in up 200%.
When using fit to none you could zoom out to 50%.
My experiences is that this relationship is not really 1 to 1 meaning you can only push a 4k image so far, and it depends in large parts what camera you use.
My rule is that I can zoom in to about max 160-170% (fit to fill setting) with my C300 mkII footage. and even then I have to sometimes add a very subtle sharpening to the image so it's not too soft.

I do like to fit to fill option better (and it produces the exact same result) because I don't have to add another adjustment to my workflow. Also I start with seeing the whole content of my shot, not a zoomed in version.

I can't prove it but I'm pretty certain that the fit to fill vs fit to none is just mathematically a different setup and not a different image pipeline.

Best way to find out is to test the different scenarios and see what works for you.

I never said there was a quality difference. Setting Special Conform to None makes it easier to deal with zoom/pan adjustments for many of us. And everyone starts with it set to Fit, as that is the default. It's not a different image pipeline, no one said it was, it's an easier workflow for many of us.

Hey Ben, I didn’t want to imply that you did. I just wanted to point out that both approaches create the same result as I thought that maybe wasn’t clear. Apologies if that came across differently.

Thank you both for your input. Not only it answer my question practically but also provided more insight in this matter, I now have a good baseline of how much to push with either method.
I’ll see what feels more organic in the workflow but since I won’t have a lot of time for testing I’ll probably go with none for spatial conform to insure the best quality, in the off chance it is effectively better.
Will definitely test this further in the future.

Thanks again