Kodak, Ilford, Efke Film Shootout Redux - A More Controlled Experiment

[chuckroast]

I while back, I did a quick and dirty comparison of 400TX and HP5+, but the test conditions were not well controlled. So I decided to be more rigorous about it.

There is a dead fallen tree in a forest preserve near where I live that just lights up with high angle light in the mid-morning to early afternoon. Not only do the dead branches have a lot of detail, so too does the forest behind it.

I tripod mounted a "Baby" Speed Graphic with a 180mm f/5.5 Tele-Arton lens there on two different days.

The first day. I shot old (11/1969) Tri-X and and (2018) HP5+ sheet film the first day. Which was semistand processed for 25 min in D-23 1+9 with 0.5g/l NaOH added to it.

The second day, I took the same two films with me but added Efke PL100M, all of which were semistand processed for 25 min in D-76 1+3.

Although taken on different days, the lighting conditions were essentially the same with bright high sun and clear blue cloudless sky above.

I wanted to see how much difference the developers and films made as regards to sharpness and grain, noting that I was working with out of date/unavailable films in these tests. (I happen to have a fairly large stash of out of date HP5+, PL100M, and Tri-X in 2x3 in my freezer.)

I made 8x10 silver prints from these negatives and roughly matched them for exposure and contrast.

Preliminary findings, subject to revision as I inspect more carefully:

• All things being equal, HP5+ is somewhat less contrasty in this controlled experiment than Tri-X. However, it is possible to get nearly identical print results from the two negatives using split-VC printing and exposure management under the enlarger.
  
  
• The Efke PL100M produced a somewhat thinner negative than the Tri-X and HP5+ did in D-76 semistand development. However, the negative was properly exposed and - again - split VC printing and enlarger exposure management brought the resulting print in line with the other. It's not shocking this negative was different since the emulsion is rated at 1/2 the daylight speed of the other two films. Probably some tuning of the development time might be in order. I care about this, because even though Efke is long out of business, I have tons fo this film in 2x3 in my freezer. Moroever, Adox CHS100II may be a close replacement, so what I figure out or the Efke should translate, at least to get going.
  
  
• I would have expected the ultra-dilute D-23 with lye to produce much sharper outcomes than the D-76 1+3. It did not. The observed sharpness of the final prints is almost indistinguishable between the two developers.
  
  
• Even though this is very old Tri-X out out date as of 11/1969 (Kodak has not made 2x3 sheet film in ages), it seems visibly slightly sharper with less "crunchiness" than HP5+. This is consistent with my less rigorous testing of the films with more recent 35mm emulsions.
  
  
• The prints are still just a tad damp and I need to go look in detail, but the Efke looks like it may have been the sharpest of the lot. This is consistent with my prior use of this film is which has very high acutance out of the box.
  
  
• I did not notice any objectionable grain, but I wouldn't expect to at 8x10.
  
  
• At some point, I'd like to drill into this a bit more and compare, say, Tri-X and Efke against these findings but using HC-110 1:128 and Pyrocat-HDC 1.5:1:200, again using semistand development.

Questions, comments, corrections, complaints, condemnations etc. welcome ...

 

[john_s]

Interesting comparison. The one thing that stands out for me is that the 50+ year old Tri-X did so well!

 

[chuckroast]

Interesting comparison. The one thing that stands out for me is that the 50+ year old Tri-X did so well!

I have generally gotten very good results with old films using semistand development.

The one exception was old Plus-X that had terrible bromide drag when I did this.

I even got good results from Super XX expired in 1961, although the film surface itself was compromised mechanically having been stored all those years. The emulsion stuck together and some of it chipped off. Still, the film was otherwise photographically sound.

 

[pentaxuser]

chuckroast, is it possible to show us the negs and or prints ?

Thanks

pentaxuser

 

[chuckroast]

chuckroast, is it possible to show us the negs and or prints ?

Thanks

pentaxuser

I will try to scan some of the prints when they are flattened to share here.

 

[Lachlan Young]

Health warning:

This whole 'test' is so fundamentally flawed for various (rather obvious) reasons to the point that it's essentially irrelevant to anyone other than the tester. Nothing wrong with playing around with various expired materials if that's what makes you happy, but nobody should place any meaningful value whatsoever on the conclusions.

It also needs a title change to make explicit that these are all (very) expired films.

P.S. if you know about the effective real-world shadow speed differences between 400TX and HP5+, all of the confusion about HP5+ supposedly being 'less contrasty' or 'grainier' than pre-2003 400TX makes a whole lot of sense - you're overexposing it. Not a lot to it. Ilford had a commercial interest in getting HP5 to deliver ISO 640 in Microphen.

 

[chuckroast]

OK, some scans.

But first a few notes and caveats:

• Scans are of a small central section of 8x10 workbook prints on an Epson V800.
  
  
• The only post scan processing was to adjust contrast and white/black point as best I could to match the actual look of each print. There was no sharpening or other fiddling.
  
  
• There is no guarantee that you will see what I see because monitors may not match and monitors never properly match reflective print media.
  
  
• Tri-X and HP5+ were rated at EI 250, PL100M at EI 160.
  
  
• Development for both developers used was 3 min prewash, 2 min continuous agitation, 15 sec agitation at around 13 mins, and total development time of 25 min. Processing was in a Kodak 1/2 gal rubber tank with the film suspended vertically from a single pinch clip (Kodak Hanger #6).
  
  
• I did these tests to check my film and way of working. I have hundreds of sheets of 2x3 stashed in a freezer and I want to make sure I get everything I want out them, since they are a non-renewable resource. If this doesn't interest you, do feel free to ignore - it won't hurt my feelings

tl;dr

The actual prints show differences:

• PL100M sharpest followed by Tri-X then HP5+
• The difference in sharpness between Tri-X and HP5+ was more pronounced with D-76 1+3. D-23 1+9+lye showed both to be very sharp.
• Tri-X showed most natural contrast, followed by PL100M, then HP5+
• The Tri-X and PL100M contrast differences were more subtle.
• The HP5+ contrast differences were more obvious
• With very close inspection, the PL100M had the least grain, and the Tri-X and HP5+ were comparable but in no case was it objectionable or visible in the 8x10 work prints.

Tri-X D-23 1+9 0.5g/l NaOH

HP5+ D-23 1+9 0.5g/l NaOH

Tri-X D-76 1+3

HP5+ D-76 1+3

PL100M D-76 1+3

 

[Harry Callahan]

Health warning:

This whole 'test' is so fundamentally flawed for various (rather obvious) reasons to the point that it's essentially irrelevant to anyone other than the tester. Nothing wrong with playing around with various expired materials if that's what makes you happy, but nobody should place any meaningful value whatsoever on the conclusions.

It also needs a title change to make explicit that these are all (very) expired films.
...

+1

 

[Craig]

Ilford had a commercial interest in getting HP5 to deliver ISO 640 in Microphen.

Why? What was special about ISO 640?

 

[Graham06]

I love tests like this. Thanks for taking the time to post. I was expecting to not see differences that mattered to me, but that is not the case: I like only one photo: the Tri-X in D23. It's the only one with natural looking tones. For all the others the highlights of the big logs are too flat and ugly. The tri-x in d76 is too contrasty.

I wonder if you would be able to make prints I like from the other negatives ( not asking, just wondering) e.g. if the hp5 in d23 were just a little bit contrastier and a bit darker, would the ugly flatness go away)

I wonder if the semi-stand development is helping. I understand that semi-stand has a 'compensating' effect which means shouldered highlights (haven't really looked into it) but perhaps it gives you something you can actually print

I think it is worth trying to do the best you can with each film you have, and with each negative you have.

My votes for what to do next:
try make the tri-x in d23 better ( don't have any suggestions)
Rate the hp5 a bit higher speed to avoid the ugly flat highlights
try an exposure developed with more agitation
try make better prints of the negatives you have ( perhaps the feel of the real prints is quite different from the scans, and of course these are just crops, but I kinda like it as a photograph in it own right)

I might try edit the scans you posted to see if I can make edits that would make me like them more.

I wonder what opinions actual printing experts have.

 

[Lachlan Young]

Why? What was special about ISO 640?

The claim was that you could get 2/3 stop shadow speed boost from a correctly balanced PQ developer at a normal contrast index. When you take into account what the core (35mm) HP5/TX market was in the late 70s, a real world effective boost of about a stop of useful speed without 'pushing' was seen as potentially commercially advantageous. The upshot is that under most circumstances, HP5+ is a bit faster than 400TX (with pretty closely matching curve shapes in conventional developers until about 1.2 above base + fog), but 400TX is designed to look like what people think Tri-X should look like when exposed at 400. Overexpose 400TX and it suddenly starts to look more like what people think HP5+ exposed at 400 looks like. As has been said elsewhere on here recently, an awful lot of what people think are inherent differences between some films have more to do with their inability to control for real-world exposure differences rather than the actual differences that exist.

 

[chuckroast]

I love tests like this. Thanks for taking the time to post. I was expecting to not see differences that mattered to me, but that is not the case: I like only one photo: the Tri-X in D23. It's the only one with natural looking tones. For all the others the highlights of the big logs are too flat and ugly. The tri-x in d76 is too contrasty.

I agree with this assessment. Generally speaking, I have found MQ type developers develop a kind of harsh contrast with semistand extended development. The hyper-dilute D-23 gives very sharp results while holding contrast in bounds.

Just a reminder though. This is very old Tri-X. It is not a predictor of how modern Tri-X would do. I'd have to cut some 4x5 to size to test it since 2x3 is no longer made.


Also, the PL100M would probably do well in the dilute D-23.

I wonder if you would be able to make prints I like from the other negatives ( not asking, just wondering) e.g. if the hp5 in d23 were just a little bit contrastier and a bit darker, would the ugly flatness go away)

There is a fair bit of room to improve this. I wanted the inherent contrast of each combination to show up so I visually matched them for comparison's sake and did no real manipulation beyond that. Some split-VC tweaking could do a lot to improve any of this, as would selective burning in the highlight areas which are washed out in some of the prints but likely have detail in the negative.

I wonder if the semi-stand development is helping. I understand that semi-stand has a 'compensating' effect which means shouldered highlights (haven't really looked into it) but perhaps it gives you something you can actually print

I think it is worth trying to do the best you can with each film you have, and with each negative you have.

My votes for what to do next:
try make the tri-x in d23 better ( don't have any suggestions)
Rate the hp5 a bit higher speed to avoid the ugly flat highlights
try an exposure developed with more agitation
try make better prints of the negatives you have ( perhaps the feel of the real prints is quite different from the scans, and of course these are just crops, but I kinda like it as a photograph in it own right)

I might try edit the scans you posted to see if I can make edits that would make me like them more.

I wonder what opinions actual printing experts have.

All good ideas. What this whole exercise confirmed for me is that D-76 isn't a great developer for semistand/EMA and I will stick to Pyrocat-HD[C] and D-23.

 

[snusmumriken]

@chuckroast: Could you just remind me what the point of adding NaOH to 1+9 D-23 is, please?

 

[chuckroast]

To restore alkalinity lost when you dilute the stock that much.

Without it, D-23 1+9 is a very low contrast/low activity developer essentially useless in normal application.

 

[snusmumriken]

To restore alkalinity lost when you dilute the stock that much.

Without it, D-23 1+9 is a very low contrast/low activity developer essentially useless in normal application.

Thanks, got it. So then, what is the reason to choose NaOH for that purpose rather than Kodalk, or even borax?

 

[chuckroast]

Thanks, got it. So then, what is the reason to choose NaOH for that purpose rather than Kodalk, or even borax?

That particular concoction was recommended by someone (@Raghu Kuvempunagar) long ago and I liked the results.

In principle, you can use other alkalais to fix developer alkalinity it's just a matter of how much to use. NaOH is extremely alkaline and very little is needed to restore proper development action.


At least, that is my understanding. If someone has a better or corrected explanation, do chime in.

 

[Milpool]

To restore alkalinity lost when you dilute the stock that much.

Without it, D-23 1+9 is a very low contrast/low activity developer essentially useless in normal application.

It’s a good question snusmumriken asks. Hydroxide is fairly extreme but since this is a dilute metol/low sulfite developer in theory the higher the alkalinity the greater the edge effects if this is a goal (although this is somewhat of a simplification). The story would be different for Phenidone(s). All other things being equal fog levels should also increase with increasing alkalinity.

It’s worth noting D-23 diluted 1+9 is not as extreme as it might seem. If you used carbonate you’d have something similar to Beutler. Lower pH values using metaborate or even borax could be avenues to explore to see if tone reproduction changes.