So, I used a roll of Copex to do a sharpness comparison between two lenses yesterday. Something went wrong with my advance mechanism after the main bit was over, but I wasn't sure how screwed up everything was, so I decided to try out HC-110 instead of Caffenol LC+C. I messed with the Massive Dev recipe a bit - used a 1:90 dilution at 24c (I have no way to cool down my solution to 20c) for 12 minutes, with one inversion per minute. The results were surprisingly good!
- No edge effect
- Slightly lower grain
- A bit denser
- Shadows definitely required more of a boost, but were totally acceptable after said boost.
I'm guessing Caffenol is still the way to go, and that my process just needs work, but if I had to choose between the two based on my current experience, I'd go with HC-110. Here are some relevant shots:
The first two are Caffenol devs, both on grey days. Not really fair, since my HC-110 roll was in sunny weather, but what can you do. Note the heavy edge effect on the first shot. I had better luck with the roll that the second shot came from, but it was still noticeable in some frames.
One from the HC-110 roll:
Now for some 100% crops. Note that my setup is
really not ideal, especially since the NEX-7 I use imparts its own noise, even at ISO 100. These are all equivalent 4154dpi (5888px on the long edge - forgive me if I have the math wrong). No sharpening, etc.
I also decided to go a bit nuts on the scanning with a frame from the HC-110 roll, using an extension tube to stitch together an 8114dpi shot (11500px on the long edge). Again, not scientific at all, since I didn't even shoot at the same shutter speed. It's also extremely difficult to get everything level and in focus at that level of magnification. Still, even though I'm sure I could do a better job with more patience and better gear, I feel that this is the first time I've really been able to "see what this film can do" in terms of resolution. Also no sharpening here.
You really do get an improvement in detail with the higher-res scan, too. On the left is the single-shot, 4154dpi scan uprezed to match the larger one:
Now, in the end, we come back to the question of whether these films actually come anywhere near the megapixel counts they claim. The answer, obviously, is a big 'no.' A qualified one, though. Looking at this image next to one from an A7RIV, with both simulating a 6-foot print size, the A7RIV's is clearly more detailed. The Copex shot, however, is more subjectively pleasant to look at, given its more 'organic' quality. Still, if I had the money and all I cared about were making huge prints, that 61mp digital camera would be the obvious choice as far as convenience goes.
Having said that, this setup cost me less than $100 total. And I think Copex makes a
great normal-use high-res film, considering you can shoot it 2.5x faster than CMS 20. The only real problem with it is that light-piping is a genuine concern, which makes it tricky for travel if you want to get it hand-inspected at security.