In answer to your first question: Neither is correct. The laserdisc is too bright and the DVD too dark. I have no proof of this, and in truth, when it comes to timing film, it was probably all over the map even from the beginning, but I'm willing to bet that the intended exposure is somewhere in the middle, and that's where I put it. When a cinematographer exposes, it's generally for both the brightest and darkest areas of the frame, and you can just sort of tell where the falloff curve was intended to be, unless a deliberately stylized look is planned (think Fincher silver-retention Fight Club look). There's no question that Star Wars was intended to be grittier than not, but still the DVD just has that overly crunched curve to it, where the laserdisc and CERTAINLY the VHS's have that clearly washed-out, overly bright midground. Also remember that things like this "gamma" balancing can only be approached satisfactorally if you have a calibrated reference monitor and signal path. That is no small undertaking. And of course, the results will appear radically different on consumer televisions, 99.9% of which people leave in the horrifically overbright, aliasing-ridden, oversharp and color imbalanced factory-preset modes. My consumer televisions have all been calibrated by ISF technicians, but still I check the factory modes to see what the rest of the world would see, and they're night and day. Although I do use the consumer mode to check things like garbage mattes, because on a properly calibrated monitor, you can't see them, but the overbright consumer modes show them right away. (In truth, I simulate consumer television response curve through an adjustment layer in the computer - I don't actually have to output to a television).

On the starfields: There is a blend between CG elements and original elements in the starfield. The original starfield has in some places been replaced, and in some places been enhanced. The original starfield, as a result of Lowry clean-up, had most of it's subtler stars eaten and the bigger ones lost their photographic bloom, so it can be hard to spot. But the crawl needed to be replaced, so it was impractical to use the original field for that section of the opening. But past the pan-down, it's mostly original. It's original in the next shot with the Star Destroyer approaching, too, as I recall. I have had to roto and re-balance every single starfield in every shot of the film, because no matter what the source, the original balance between background and foreground elements has long since been lost. Again, it's something you can sort of "see".

I had done digital grading for quite a few years before I took on Legacy, but I have learned more since I started than I would've thought possible. In many ways, ANH is like a grading final exam. So many worst-case scenarios. And without separate elements to recomposite, it means a lot of rotoscoping. In fact, it's mostly rotoscoping. And this would be true even if you had a pristine 4K scan - it'd just be easier to see where to roto.

_Mike