If you shoot large format, that of course works, if developed for a higher contrast. If not though then you'll have to either do digital negatives, a positive process sheet film (ie, enlarge negative to make bigger negative), or an interpositive -> large negative process. I've tried the digital negative stuff and just couldn't get something that looked natural and correct. Maybe I wasn't patient enough with the curve manipulation, but either way, it's not really something that holds my interest. The positive process film requires some chemicals I don't have (dichromate bleach etc), so that's also out for me. I do have large format camera and some nice negatives, but 4x5 is pretty small. So, I did the whole interpositive -> negative process
I have a DSLR scanning setup, so I mounted my Mamiya 645 with a macro lens on it and basically "scanned" the negatives I have using it. I used FP4+ film, shot at 50 ISO, and pulled about 1 stop using my GVK1 developer (something finer grained would be ideal, but don't have anything on hand). The resulting positive film was pretty nice, excluding some dust spots on the negatives I scanned. The positives are quite low contrast as that was my aim, but basically it captured all of the detail the negative had to give. I then put this 6x4.5 film into my enlarger and enlarged onto Arista Ortho Litho 8x10 film (super cheap, high contrast by design, completely clear base, and can be handled under safelight). The speeds were quite a bit slower than typical paper printing speeds. I used Liquidol as the developer, 60ml to 1L of water seems to be about right. Develop for 1m30s - 3m depending on the negative being produced and contrast aims. Extend the time for lower shadow contrast, increase agitation for longer scale highlights. It's almost impossible to develop by inspection unless the subject is very contrasty, it'll basically just look all black under safelight when developed to the proper aim The resulting negatives take some practice to really judge the quality of, just print a few of the same subject, and try a test print of each negative produced. Sometimes it can be quite weird, like you need more contrast yet highlights clipped to white. I'm unsure of any technique to perfect this quickly other than practice.
Alternatively, if you have some slide film, it can be the easiest to start with. Simply enlarge the slide film onto sheet film. For Ortho Litho film, keep in mind it'll give an ortho color rendition, and also keep in mind that slide film is practically very high contrast when enlarging it like this.
One interesting theory for how to get reversal results without reversal chemicals is that you could in theory print a sheet film positive, then bleach it all the way back using ferricyanide or copper sulfate. Then take another sheet of film out and expose it to room lights, fully fog it but don't solarize it. Finally, set out two trays one with an alkali, and one with a developing agent (optionally kept in weak acid). Dip the positive in the developing agent, hang until it's no longer dripping (maybe drying it even!?), then dip the fogged sheet in the alkali, and also hang until it's no longer dripped but ensure it's still wet. Put the positive film onto a sheet of glass. Then, get another sheet of glass and put the fogged sheet onto it (it should stick with surface tension, if not then use tape). Finally, and with speed, sandwich the two pieces of glass together, aligning the sheets as best as possible, but DO NOT move or realign the sheet after pressing them together. After 30s, development is practically complete. Unsandwich the sheets. The top sheet will be a negative, the bottom sheet will be a positive. Fix the the negative sheet. To preserve the positive used for "pressing", develop in a normal developer afterwards to completion, no fixing should be required. I've tried this before, but with paper and the process does work, but there's definitely problems with high contrast and uneven wetting. I'd love to hear if anyone figures out a way to make this process work better. It basically works via local developer exhaustion. Where the positive develops, the negative won't, thus creating an inversion of density
