Testing and evaluating CatLabs "X Film 320 Pro (2022 version)

[aparat]

{Moderator's Note: The Posts here have been copied from this earlier thread:

CatLABS X FILM 320 Pro now available in 35mm and 120​

Which can be found here: https://www.photrio.com/forum/threads/catlabs-x-film-320-pro-now-available-in-35mm-and-120.194245/
For that reason, you may find them quite familiar, and aparat will see a definite jump in his post count.
In addition to asparat's tests and comments, people are encouraged to add their results with or tests of the film}

I am slowly trying to return to photography after a long hiatus due to a long-term illness/disability. It turns out I missed out on a lot of new products over the past three-four years. I was intrigued by the Catlabs Pro 320 announcement. I bought a roll of 35mm and 120. The film arrived very quickly directly from Catlabs.

The film makes a great first impression. It's packaged very nicely, and I really like the artwork. Both 35mm and 120 are very easy to work with. The base is very clear, with B+F of only around 0.08. The film does not curl very much at all and dries very, very flat. I have not noticed any emulsion imperfections. The edge markings show the Catlabs branding. Clearly, a film made specifically for them.

Following the manufacturer's recommendation, I exposed the film at EI200. I ran a typical film test, using a recently calibrated and certified sensitometer, giving the 35mm variant an exposure of 3.68 Millilux Seconds, which is appropriate for ISO 200 film in my particular device. For example, Ilford HP5+ exposed at 3.38 gets essentially very close to box speed processed in D76, as does 400TX. Expired Neopan 400 gets about a 1/2 stop less speed.

Disclaimer first. I do not claim that my results are representative of the film's performance. It's possible that I made mistakes and/or that I got a bad batch of the emulsion. Please, keep that in mind.

In my test, the film ended up being severely underexposed. I processed four strips for 4, 5:40, 8, and 12 minutes in D76 1+1 at 20C, using a rotary processor with constant agitation. The resulting curves show a very long toe and rather steep G and CI values For example, flare-corrected CI ranges from 0.6 to 0.9. According to my calculations, effective film speed is around ISO 25.

Because my results are somewhat unexpected, I am hesitant to share a more detailed analysis. I should have bought more than one roll of 35mm and re-run the test. I have taken some pictures with my medium format camera and got decently exposed frames, mostly showing similar characteristics as the 35mm variant, namely, underexposed and overdeveloped. Unfortunately, I am still unable to make prints, so I cannot really tell much beyond basic exposure and a rough idea of tonality.

 

[aparat]

I digitized one abstract frame with my phone camera just to give a rough sense of tonality.

 

[aparat]

I finished running my tests. I have no reason not to like this film. In fact, I find it very easy (and pleasant) to work with. It has exceptionally low B+F density (for a 35 mm film) and it dries super flat. It also seems to resist drying marks rather nicely. The grain is fine and tight. I would imagine this film would scan really easily. I discovered no emulsion flaws, whatsoever. One interesting bit is the fact that on all three 35 mm cassettes I bought, the DX code was covered with a piece of black electrical tape. Finally, as there's been some speculation as to the film's origin, in my experience, the film does not resemble any of the current 35 mm films by Harman (including Kentmere), Foma, or Kodak. The film base, is similar to that of the Rollei Retro 80s, but I make no claim that the film is made by the same manufacturer as the Rollei Retro 80s.

I encourage you guys to try to find flaws in my methodology, especially if they would help account for the results. I do not claim to be an expert sensitometrist!

I have tried to run the tests using as rigorous a process as I could within my modest home setup. My results DO NOT include any real-world photography, so they are probably completely useless to most photographers. I simply exposed the film using a calibrated and certified sensitometer at 3.68 Log Millilux Seconds, processed four film strips for 4, 5:40, 8, and 12 minutes in D76 1+1 at 20C using continuous agitation in a rotary processor. I then computed all the usual parameters, such as Gamma, G, CI, film speed, fractional gradient, etc., and I am reporting my findings below. Please, refer to my previous post in this thread for more details.

As promised earlier in this thread, I ran my film test twice, just to be sure I didn't botch the first attempt completely. The results are very much the same, e.g., the "speed point" differs by 0.01 Log Exposure and the CI by 0.02 between the two runs (comparing 8 min. in D76 1+1 at 20C with continuous agitation), so within the margin of error. Compared to expired Fujifilm Neopan 400 and expired 400TX, the Catlabs 320 Pro, when processed to approximately the same CI (or G), using the ISO standard, as described in Davis (1993), Eggleston (1984), and elsewhere) appears about 3 stops slower. In essence, when exposed and developed according to the manufacturer's film box recommendations, the film has the property of being significantly underexposed and overdeveloped, or "pushed."

I computed film speed by two different general methods: (1) the ISO standard method, using absolute log exposure values, and (2) the BTZS method, using relative log exposure values. I obtained speed point densities by means of the commonly used 0.1 over B+F method, as well as the fractional gradient (Nelson and Simonds, 1955), and approximate CI (Phil Davis, 1993) methods, all yielding consistent results. Admittedly, there is some, small, gain in speed when developed for the manufacturer's recommended time, but then you're getting the CI of 0.72-0.74, which some photographers will love, but others will hate. . Perhaps this is the "street film" look that Catlabs refer to on their website. I can totally see this film being very popular because of that look!

The results are somewhat troubling to me. The Catlabs 320 Pro does appear significantly slower than advertised. However, different manufacturers have different methodologies of computing their "box speed," so I don't really worry about the actual ISO 320 value per se, as most photographers run their own tests anyway. What's more significant, though, is the fact that the Catlabs Pro 320 appears to be much slower than (expired) 400TX and (expired) Neopan 400. It's a pity I did not have any fresh 400-speed film to compare the 320 Pro against.

Still, I want to take the Catlabs advertised film speed and development time in good faith, so it's possible, maybe even likely, that there are flaws in my methodology. If I were to try to poke holes in my testing process. I could easily find at least two possible sources of variability (or error). One, Catlabs 320 Pro has a much different reciprocity failure property than 400TX and Neopan 400. The sensitometer exposure was 1.068 seconds. Not ideal, but that's the only option I had (I was aiming for an ISO 200 exposure). Two, Catlabs 320 Pro has a very different spectral response, which resulted in a significant dip in sensitivity, particularly, around the peak of 530 nm. I am sure I am missing something. I welcome your comments and criticism.

 

[aparat]

I wonder why this was the case? Have you taken the tape off to see what the code underneath reveals assuming there is a code there?

pentaxuser

I do not know why, but here's a picture

 

[aparat]

Thank you for your time and effort in testing this. So I get that this has an ei of 25-50? and maybe partially blind at 530?

I guess the data seem to be pointing that way, yes.

 

[aparat]

This is interesting and useful. Thanks for doing the tests and sharing your findings.

It would also be useful to compare the real-world results that you get with i) manufacturer's recommendations vs ii) your experimentally determined EI + developing time. If you plan to do this comparison, please do share your results.

Thanks again.

Yes, it would be very useful to do a photographic test. However, I don't have a film scanner, and using a phone to digitize film is not ideal. I need to wait a bit longer until I am able to make silver gelatin prints. Then, I will do such a test.

 

[aparat]

Found this review of Catlabs 320 Pro.

Thank you for posting the link to the review! It's nice to see people enjoying the film.

As I mentioned, I have no reason not to like this film. I think it has an interesting "punchy" kind of look, when exposed and developed according to the instructions on the box. It's also great to see that one's personal EI works out very well for them. That's what it's all about.

Testing film in a controlled environment is kind of different from finding one's preferred parameters with actual, real-world photography. There has been some discussion in the film sensitometry literature over the years emphasizing the role of subjective negative and print assessment, rather than analysis purely by the numbers, so keep that in mind, please.

I am attaching the results of my analysis as a PDF so you guys can see the them more clearly. Please, feel free to pick holes in my analysis. I want to learn to do a better job in the future!

There are a few things I need to mention first.
1. The film is underexposed, so the curves have a very unusual shape. Modeling such curves is a challenge. There is some extrapolation involved, which, any statistician will tell you, is a big no-no . The extrapolated bits are marked with black dotted lines.
3. I considered giving the film more exposure, but that would involve significant contribution from reciprocity, which would make the analysis a lot less reliable, without knowing the actual reciprocity failure values from the manufacturer.
4. All of the numbers computed here are done entirely by algorithm, so they are approximate. It is particularly true of Gamma where finding the "straight portion of the curve" is typically subject to interpretation by the photographer. Here, Gamma is computed by finding the straight line statistically (the green dotted line). Fractional gradient is somewhat similar in that regard. Other parameters are also subject to some variability, but should, nevertheless, get you in the ballpark.
5. On page 3, I overlaid the curve from expired Fujifilm Neopan 400, exposed the same way as the Catlabs 320 Pro and processed in the same tank for 8 minutes. You can see much higher B+F density due to the type of film base itself and increased fog due to the the film aging.
6. The "ISO curve" and triangle are marked in pink on p. 3
7. There is a lot more detail that's not printed in the PDF, so please free to ask questions.
8. The plots on pages 8-10 are similar to those you get in the BTZS-type of anlalysis and are available in the application called WinPlotter. It's an old piece of software, but it has some really interesting functionality.

 

[aparat]

In my haste, I forgot to include a plot comparing the curves of Catlabs 320 Pro and Kodak 400TX (purple). I gave Tri-X the same amount of exposure as the Catlabs 320 Pro (3.68 Log Millilux Seconds) and developed in the same tank for 8 minutes. Clearly, Tri-X did not receive enough development, as the contrast Index (CI) is only about 0.45. Even so, it presents a well-formed curve, with a fairly short "toe" and a long, monotonically increasing "straight line" portion of the curve. This would translate into a negative with a wide, smooth tonal range; one that should print easily on Grade 2 paper through a condenser enlarger (though I prefer to develop Tri-X to the CI of about 0.55). The speed point is about 3 stops "in front of" (or faster) than that of the Catlabs film of G=0.62 (the pink "ISO" curve and triangle). The Tri-X curve also shows much higher B+F density, which is to be expected as the film has a very different base and is a few years expired.

The manufactures of both films claim their respective products to be conventional, panchromatic, black and white films, with the "box speed" of around ISO 320-400. Both manufactures list D76 1+1 as a recommended developer. Catlabs lists 10 minutes at 20C (so that should translate to around 8:00 minutes in a rotary processor), whereas Kodak lists 9:45 minutes for 400TX in a rotary processor (again, my test sample did not receive enough development). Therefore, one would expect the curves to look very similar, except, of course, for the difference in base fog density and some other minor idiosyncrasies. As I mentioned earlier in the thread, the results suggest that, compared to Tri-X, the 320 Pro would have a much more "punchy" look, with significantly less shadow detail and a somewhat narrow, steep tonal range. I am sure it's a look that a lot of people love. It would require a softer paper grade to print well. A good scanner or DLSR, on the other hand, should be able to get some of the shadow detail back, with endless Photoshop editing possibilities.

Finally, while looking at each film through a magnifier, the Catlabs film appears to have somewhat finer grain than 400TX? Maybe? But that is just "by eye," so please take it with a grain of salt (pun not intended). My eyes are not what they used to be.

I am looking forward to seeing other people's tests of the Catlab 320 Pro and possible comparisons with other films.

 

[aparat]

Hold on there for just one moment......
Sorry to shatter the pillars of the earth here, but that is just erroneous.

Just taken directly from your website: "D76 1+1 10 min"
Typically, in a rotary processor, you develop for about 15-20% less time, to start, and then adjust to taste. Nothing carved in stone, but a reasonable starting point.
I developed this particular strip of Tri-X for 8 minutes, along with a strip of Catlabs 320 Pro. What is erroneous about that?

 

[aparat]

Indeed, especially considering that sample of one was developed using erroneous assumptions resulting in wrong processing times.

Please, show me where the error is. I processed not one, but four strips of the film in times ranging from 4 to 12 minutes. That is how film testing is done. I chose to compare to Tri-X at 8 minutes because it was the closest time to your own recommendation so I could develop both in the same tank, to avoid adding a source of variability to the analysis.

 

[aparat]

As someone above said unironically, not to nitpick this analysis, but there are an infinite number of uncontrolled and unaccounted for variables in your testing protocol

I am not saying its wrong, just that it might be correct for anyone in any other situation other than yours .
It would seem that the better alternative to ending up with years expired Trix that was never used and then sacrificed at the alter of an experiment that did not even involve the taking of one photo is to go out and take more pictures, on fresh film. If you buy fresh film, you support the industry that makes it. If you dont support it by buying more fresh film it will not stay around for much longer.

Yes, i wish I had fresh film but I didn't. Sorry about that. Expired film is going to raise B+F and cause some speed loss, mostly. I can send you my Tri-X to see how much it differs from a fresh roll. My bet is, not enough to make a statistically significant difference.

I asked specifically for people to tell me what was wrong with my analysis. I would love for you to provide your own analysis, to compare to mine.

And yes, I did include a photo, a bad one, but it was there.

 

[aparat]

Read between the lines. The statement of what can’t or won’t be shared is completely obvious and has be restated consistently and repeatedly. The basis of the 320 or the 200 speed rating is to be taken as an act of faith as a starting point.

The 25 rating is based in an experimental analysis and has been disputed. The 80 is conjecture. All worth considering yet none so definitive that the others can be excluded.

Buy and try is how the film is being marketed. Who are we to try to change that?

It’s the continued arguments about not getting the answer and not liking the style of answer given that has made this an embarrassing mess… more for Photrio than for CatLabs IMO.

Yes, I even disputed my own results after I got them back the first time. This is why, I bought more film and ran the test again. The results were nearly identical, well within the margin of error. I actually, ran the test the third time. I would gladly continue testing and sharing my results but the OP made it clear that they are not taking this thread seriously, so I will no longer post in this thread. It's a waste of time.

If I had to give you my conclusion, I would say that the Catlabs film is technically slower than the advertised ISO 320 and, not even close to being capable of reaching ISO 1600 (Catlabs recommends "pushing" the film to ISO 1600 on their website). Even if you don't put faith in the actual ISO number my analysis produced, you can see that the film is about three stops slower than 400TX under experimental conditions. In the "real-world," one should expect similar results, subject, of course, to the usual variability inherent in film (and hybrid) photography.

As to why the film turns out slower than advertised, I would hazard an (educated) guess that the film, in addition to being inherently slower than mainstream medium-speed films (e.g., 400TX, HP5+), is not quite panchromatic, and has a significant dip(s) in sensitivity along its spectral response curve. I started testing its spectral response, but I can see that it is a waste of effort so I will not conclude the test. The OP does not care, and it's absolutely their prerogative. I have no intention of telling them how to run their business. Bringing a film to market (whether new or re-badged) is a huge undertaking. Catlabs deserve sincere congratulations on succeeding. I am seriously impressed that they were able to offer this film in 135 and 120, and at competitive prices!

Nick

 

[MattKing]

Posts here have been moved/copied here from the following thread:

CatLABS X FILM 320 Pro now available in 35mm and 120​

Which can be found here: https://www.photrio.com/forum/threads/catlabs-x-film-320-pro-now-available-in-35mm-and-120.194245/

 

[mshchem]

Yes, I even disputed my own results after I got them back the first time. This is why, I bought more film and ran the test again. The results were nearly identical, well within the margin of error. I actually, ran the test the third time. I would gladly continue testing and sharing my results but the OP made it clear that they are not taking this thread seriously, so I will no longer post in this thread. It's a waste of time.

If I had to give you my conclusion, I would say that the Catlabs film is technically slower than the advertised ISO 320 and, not even close to being capable of reaching ISO 1600 (Catlabs recommends "pushing" the film to ISO 1600 on their website). Even if you don't put faith in the actual ISO number my analysis produced, you can see that the film is about three stops slower than 400TX under experimental conditions. In the "real-world," one should expect similar results, subject, of course, to the usual variability inherent in film (and hybrid) photography.

As to why the film turns out slower than advertised, I would hazard an (educated) guess that the film, in addition to being inherently slower than mainstream medium-speed films (e.g., 400TX, HP5+), is not quite panchromatic, and has a significant dip(s) in sensitivity along its spectral response curve. I started testing its spectral response, but I can see that it is a waste of effort so I will not conclude the test. The OP does not care, and it's absolutely their prerogative. I have no intention of telling them how to run their business. Bringing a film to market (whether new or re-badged) is a huge undertaking. Catlabs deserve sincere congratulations on succeeding. I am seriously impressed that they were able to offer this film in 135 and 120, and at competitive prices!

Nick

Are you going to be doing any further testing?

 

[warden]

Posts here have been moved/copied here from the following thread:

CatLABS X FILM 320 Pro now available in 35mm and 120​

Which can be found here: https://www.photrio.com/forum/threads/catlabs-x-film-320-pro-now-available-in-35mm-and-120.194245/

Thanks Matt.

 

[removedacct1]

Yes, I even disputed my own results after I got them back the first time. This is why, I bought more film and ran the test again. The results were nearly identical, well within the margin of error. I actually, ran the test the third time. I would gladly continue testing and sharing my results but the OP made it clear that they are not taking this thread seriously, so I will no longer post in this thread. It's a waste of time.

If I had to give you my conclusion, I would say that the Catlabs film is technically slower than the advertised ISO 320 and, not even close to being capable of reaching ISO 1600 (Catlabs recommends "pushing" the film to ISO 1600 on their website). Even if you don't put faith in the actual ISO number my analysis produced, you can see that the film is about three stops slower than 400TX under experimental conditions. In the "real-world," one should expect similar results, subject, of course, to the usual variability inherent in film (and hybrid) photography.

As to why the film turns out slower than advertised, I would hazard an (educated) guess that the film, in addition to being inherently slower than mainstream medium-speed films (e.g., 400TX, HP5+), is not quite panchromatic, and has a significant dip(s) in sensitivity along its spectral response curve. I started testing its spectral response, but I can see that it is a waste of effort so I will not conclude the test. The OP does not care, and it's absolutely their prerogative. I have no intention of telling them how to run their business. Bringing a film to market (whether new or re-badged) is a huge undertaking. Catlabs deserve sincere congratulations on succeeding. I am seriously impressed that they were able to offer this film in 135 and 120, and at competitive prices!

Nick

Congratulate them for producing a 25 ASA film and marketing it as 320 ASA? Not sure why that deserves kudos, but ok.

I used the CatLabs X 80 for awhile, and found it was actually about 12 ASA in practice so I didn't bother with it again. It had terrible reciprocity traits.

 

[aparat]

Are you going to be doing any further testing?

Yeah. I'll finish what I've started and try to get a few extra rolls soon. At the very least, I'd like to test the film with exposures closer to its actual speed. That should get us a more realistic-looking curve family and development times. Then maybe I can give tone reproduction analysis a shot.

 

[MattKing]

Yeah. I'll finish what I've started and try to get a few extra rolls soon. At the very least, I'd like to test the film with exposures closer to its actual speed. That should get us a more realistic-looking curve family and development times. Then maybe I can give tone reproduction analysis a shot.

If your initial conclusions can be applied by some real world users, it will be interesting to see what shows up.

 

[aparat]

If your initial conclusions can be applied by some real world users, it will be interesting to see what shows up.

I agree completely. It would be fantastic to see photographs. I know I am guilty of not posting photos.

 

[Disconnekt]

Just by skimming through the other thread, the demeanor of Catlabs/person running the account on here, any slight interest I had with the company/their "new" film has pretty much gone out the window. :/

 

[bernard_L]

I asked specifically for people to tell me what was wrong with my analysis.

Your testing is IMO near-perfect experimental (scientific) procedure, at least by serious hobbyist standards. You don't need to apologize when Catlabs just produces vague arguments:
"I am not saying its wrong, just that it might be correct for anyone in any other situation other than yours" (not even sure I understand the meaning of that, even after reading three times)
Or shifts the discussion to another level:
"If you buy fresh film, you support the industry that makes it."

 

[AgX]

Diskonnekt, Bernard, please leave such comments at that other thread.
Matt's idea was to put up a thread thread here devoted solely to testing this film, either by sensitometric testing or by photographs .
The latter ideally taken together with common film at same ISO setting, same subject, and the manufacturer advised developer and time, e.g. D76.

 

[faberryman]

Diskonnekt, Bernard, please leave such comments at that other thread.
Matt's idea was to put up a thread thread here devoted solely to testing this film, either by sensitometric testing or by photographs .
The latter ideally taken together with common film at same ISO setting, same subject, and the manufacturer advised developer and time, e.g. D76.

After all, CatLABS can't afford to do this so we have to rely on one another. That's what forums are for. Providing product documentation to manufacturers. Thanks aparat for starting things off.

 

[pentaxuser]

AgX makes a very good point. This thread was very sensibly established by Matt to separate our comments on what some of us find to be lack of information from CatLABS and a general discussion from simply testing the film

We really need to separate these 2 sides.

pentaxuser

 

[MattKing]

CatLabs is welcome to add more results to this thread if they wish - of testing or of use.