New TMax P3200

[Mackinaw]

I forgot to ask, did you process at the given time for EI 800 or EI 3,200?

Thanks!

I processed for E.I. 800 times.

Jim B.

 

[Mackinaw]

So, I'm not expert in all this, but why would you choose to pull the more expensive 3200 film down to 800 instead of just pushing the 400 up one stop?

From the Kodak data sheet, the true speed of P3200 is E.I. 800 -1000.

Jim B.

 

[Sirius Glass]

So, I'm not expert in all this, but why would you choose to pull the more expensive 3200 film down to 800 instead of just pushing the 400 up one stop?

From the Kodak data sheet, the true speed of P3200 is E.I. 800 -1000.

Jim B.

Which begs the question, why spend the money for P3200 if one is going to shoot it at 800?? 800 is well within the range of Kodak Tri-X 400 and Ilford HP5+ both of which are cheaper on either side of the Atlantic Ocean? The reason for D3200 and P3200 to exist is to have higher film speeds.

 

[MattKing]

Which begs the question, why spend the money for P3200 if one is going to shoot it at 800?? 800 is well within the range of Kodak Tri-X 400 and Ilford HP5+ both of which are cheaper on either side of the Atlantic Ocean? The reason for D3200 and P3200 to exist is to have higher film speeds.

Perhaps because with an ISO 800 - 1000 speed film, you will get better shadow detail than when you under-expose a 400 ISO film by a stop.
The D3200 and P3200 films don't have higher speeds than their ISO ratings. They are used in low light situations because their mid-tone and highlight response responds better to increased development and the contrast increase that arises from it.

 

[jim appleyard]

So, I'm not expert in all this, but why would you choose to pull the more expensive 3200 film down to 800 instead of just pushing the 400 up one stop?

What Matt King said. When you push film, contrast goes up. Sometimes this is desirable, sometimes not. If I were to shoot a stage show, or a wedding in a dark church, I would probably go with a 3200 film @800 rather than take the risk of blowing out my highlights.

 

[fjpod]

What Matt King said. When you push film, contrast goes up. Sometimes this is desirable, sometimes not. If I were to shoot a stage show, or a wedding in a dark church, I would probably go with a 3200 film @800 rather than take the risk of blowing out my highlights.

I see,... but Kodak doesn't even consider a push of TMY from 400 to 800 a push at all. Developing times are listed the same.

 

[MultiFormat Shooter]

I processed for E.I. 800 times.

Jim B.

Thanks!

 

[MattKing]

I see,... but Kodak doesn't even consider a push of TMY from 400 to 800 a push at all. Developing times are listed the same.

"Push" means an increase in development. It doesn't mean a combination of decrease in exposure plus an increase in development (although somehow that misuse of terms has become more and more common).
If you increase development (a "push" development) the quality of rendering will increase in certain ways, and decrease in other ways.
In particular, increased development decreases the quality of the highlight rendition.
It is Kodak's opinion that when you under-expose TMY by one stop and develop normally, the reduction of shadow and near-shadow quality that results is less than the total loss of quality in the shadows and near shadows and the highlights when development is increased.
I agree with that opinion.
Every time you make decisions regarding exposure and development you are balancing several factors, because very decision will improve the quality of some factors while reducing the quality of other factors. The balance is more difficult to maintain when you work outside the standard recommendations.

 

[crypt47]

Which begs the question, why spend the money for P3200 if one is going to shoot it at 800?? 800 is well within the range of Kodak Tri-X 400 and Ilford HP5+ both of which are cheaper on either side of the Atlantic Ocean? The reason for D3200 and P3200 to exist is to have higher film speeds.

Speaking about P3200 at EI 800. Yesterday I printed a photo in a darkroom. What striked me immediately is so called 'silver greys'. Middle greys, leafs and grass, on a sunny day (not full contrast, slightly hazy day) all are much lighter then with HP5+ (not an expert with this film). Also I don't know how to achive silver greys with TMAX400, it's a contrasty film even at EI 200. You can print grey lighter but fallback to blacks is different. Here I've got blacks, shadow detail on black cloth and light middle tones. What I've got reminds me ADOX Silvermax 100. Tonalty touched me to the bottom of my heart, but sharpness didn't. Reversing Kodak own slogan: world's UNsharpest. Is it really a T-grain film? Scanning also is worst thing to do with this type of asymmetric big grain.

I didn't test Delta 3200 the same way, but it's sharper that's for sure, but tonalty is totaly different I guess.

 

[Lachlan Young]

Is it really a T-grain film? Scanning also is worst thing to do with this type of asymmetric big grain.

Yes, it is T-grain. Sounds like you are talking from the perspective of never having actually seen a properly first-rate scan. What did you process in & what are the terms of your comparison with D3200?

 

[RattyMouse]

Why is this film not available in 120 size? Anyone know?

 

[Anon Ymous]

Why is this film not available in 120 size? Anyone know?

Never was.

 

[Lachlan Young]

Why is this film not available in 120 size? Anyone know?

It was claimed that the 120 backing paper did not offer the necessary degree of protection from atmospheric radiation (or something like that) to such a highly sensitised film compared to a 135 canister. Then Ilford made a slightly faster film in 135 & 120 without problems - so it's now questionable as to what limitations there are other than market demand. TMZ in sheets would be something else though (and probably too much of an engineering effort to be viable).

 

[Anon Ymous]

It was available in 100ft 70mm rolls though.

 

[RattyMouse]

It was claimed that the 120 backing paper did not offer the necessary degree of protection from atmospheric radiation (or something like that) to such a highly sensitised film compared to a 135 canister. Then Ilford made a slightly faster film in 135 & 120 without problems - so it's now questionable as to what limitations there are other than market demand. TMZ in sheets would be something else though (and probably too much of an engineering effort to be viable).

Yes, Delta 3200 offers no issues in 120 size. I've never seen fogging of any kind from this film due to poor paper protection.

My 35mm lenses were so fast that I never saw much need for high speed film in this size. My medium format gear however, never gets faster than f/3.5 and is mostly f/4.5 so high speed film is often called for.

 

[crypt47]

Yes, it is T-grain. Sounds like you are talking from the perspective of never having actually seen a properly first-rate scan. What did you process in & what are the terms of your comparison with D3200?

Please, define 'a properly first-rate scan'.

 

[Lachlan Young]

Please, define 'a properly first-rate scan'.

Not an Epson or similar low-end flatbed or an oversharpened minilab scan.

 

[crypt47]

Not an Epson or similar low-end flatbed or an oversharpened minilab scan.

Well, definition by exclusion is not a difinition actually, because it assume too many gradation but not one. No I wasn't talking about flatbed. I think scan is acceptable if grain is less then pixel otherwise multiple pixel produce an aliased image of grain, which in case of subject in discussion is even worth cause too much is aliased. Optical printing without discretization differs drastically. Althout print less sharp then pushed delta 3200.

 

[Adrian Bacon]

Well, definition by exclusion is not a difinition actually, because it assume too many gradation but not one. No I wasn't talking about flatbed. I think scan is acceptable if grain is less then pixel otherwise multiple pixel produce an aliased image of grain, which in case of subject in discussion is even worth cause too much is aliased. Optical printing without discretization differs drastically. Althout print less sharp then pushed delta 3200.

Ermmm... a scan that has grain smaller than the optical Nyquist response of the scanner is not a good scan.

You actually want it the other way around, meaning at least 2 resolved pixels per the smallest grain you want to resolve for the source scan. More pixels are better. From there you scale the image down to either your working resolution, or just scale down straight to your output resolution. Doing that results in a scan that requires very little if any sharpening once at the destination resolution and any good downscaling algorithm will apply a low pass filter to smooth out frequencies that are above what the destination resolution can represent without aliasing. This has the net effect of smoothing out the grain while still retaining maximum detail and sharpness.

A particularly bad scan is one where the scanner’s optical Nyquist frequency is at or near the grain size of the bulk of grain comprising your image. If that is the case, then you need to scale down that scan down to 1/4 resolution of the source scan as your working resolution. Depending on your scanner, this may result in a pretty small scan. It may be better to either get a higher resolution scanner, or use a different emulsion.

All scanners have a fixed sensor/optical path resolution and have this exact same challenge no matter what emulsion you are digitizing. A good scanner has 4 times the native sensor resolution as what the scanner kens can render due to diffraction.

Again, this is source resolution, not working resolution or output resolution. Many people make the mistake of looking at a scan at source resolution and rendering judgement on the quality of the scan. Source resolution should always be as high as possible, working resolution should be 4x output resolution, output resolution should be 300 pixels per inch of display, so an 8x10 is 2400x3000 pixels as the output, so working resolution would be 4800x6000, and the source scan should ideally be 9600x12000 pixels or more.

 

[Lachlan Young]

Well, definition by exclusion is not a difinition actually, because it assume too many gradation but not one. No I wasn't talking about flatbed. I think scan is acceptable if grain is less then pixel otherwise multiple pixel produce an aliased image of grain, which in case of subject in discussion is even worth cause too much is aliased. Optical printing without discretization differs drastically. Althout print less sharp then pushed delta 3200.

But you're not willing to divulge the name of the scanner? Or what you processed in?

The contrast grade required for a good darkroom print can also impact dramatically on the perceived sharpness of an image - and at that point we start having to think about subject contrast, CI developed to & the curve shape of the film & it's impact thereof on the highlights. At its simplest, the harder the grade you print on, the sharper you may perceive the image to be.

For what it's worth, I've never had any complaints about the perceived sharpness of either film within the context of what they were designed to do. And that's involved darkroom printing (generally DeVere & Rodagon or Componon-S) & scanning (generally high end CCD/ Imacon). If anything, D3200 is perhaps a tick softer than TMZ.

 

[Shoom]

I have two rolls with a 2015 expiration date that have been kept frozen (not refrigerated), do you think that they'd be worth using? Maybe expose them at EI 1000 or 1250?

I picked up a 2015 exp date roll a few months back and shot it with no ill effects. I think it had been stored at room temperature before I bought it as well so you should be fine. I've heard P3200 fogs regardless of temperature, anyway.
I either shot it at 1600 or 3200 (didn't seem to write it down on this one...) but I still got pictures. They were a little thin, so I think the wisdom of developing N+1 for your EI applies to P3200 as well as Delta 3200. It also might have been due to indoor lighting.
Especially considering that the stuff is back in production, along with the short shelf-life, I don't see any reason to hoard them

 

[MultiFormat Shooter]

I picked up a 2015 exp date roll a few months back and shot it with no ill effects....

Thanks, I was just trying to determine if I'd be able to get anything useful off of it. Now that it is back in-production, I'll just buy as I use it.

 

[crypt47]

But you're not willing to divulge the name of the scanner? Or what you processed in?

The thing is I wasn't going to discuss my technics. I wanted to share my first impression that's all. Also you were the one who started commented without any technical criteria, using something undefined like 'first-rate'. That was total lack of sense to me. I just brought your attention to it.

 

[crypt47]

Thanks you Adrian for such a bit reply, but I leave the theory of frequences to you. Based on my experience:

You actually want it the other way around, meaning at least 2 resolved pixels per the smallest grain

Considering black grain this will give you 4 (not 2, cause there are 2 dimentions) grey pixels. Considering you multiplay resolution by 2 this will give you 16 avaregly greyed pixels. The edges always be a smoothed aproximation. So unless you have a resolution that results in high edge/center ratio you will always have mudy edge reproducion. The asymetric cluster's shape of P3200 grain in particula will give you more edges and more mud then delta 3200.

If grain more or less equial to pixel size then to one black dot on a print you'll get one black pixel. My test confirms two things: it's sharper that way and there is no much difference in which film to scan after some threshold. Both HP5, Delta 3200 and TMAX P3200 all exihibits this threshold in case of my scanner. ADOX CMS 20 which has smallest grain gives sharpest (and overly the best) scans which contradicts your statement:

a scan that has grain smaller than the optical Nyquist response of the scanner is not a good scan.

Another matter:

working resolution should be 4x output resolution

Yep, say it to people who downscale their film scans to 0.5 mpx and talk about sharpness all over the internet. Resulting resolution does matter but the lower resolution you made (with sharpening) the bigger impression of sharpness you can achive. Especially with P3200 which doesn't resolve fine details anyway.

so working resolution would be 4800x6000, and the source scan should ideally be 9600x12000 pixels or more.

And this I guess would be important to a man above who can't define 'a first-rate scan'.

 

[Adrian Bacon]

Thanks you Adrian for such a bit reply, but I leave the theory of frequences to you. Based on my experience:


Considering black grain this will give you 4 (not 2, cause there are 2 dimentions) grey pixels. Considering you multiplay resolution by 2 this will give you 16 avaregly greyed pixels. The edges always be a smoothed aproximation. So unless you have a resolution that results in high edge/center ratio you will always have mudy edge reproducion. The asymetric cluster's shape of P3200 grain in particula will give you more edges and more mud then delta 3200.

If grain more or less equial to pixel size then to one black dot on a print you'll get one black pixel. My test confirms two things: it's sharper that way and there is no much difference in which film to scan after some threshold. Both HP5, Delta 3200 and TMAX P3200 all exihibits this threshold in case of my scanner. ADOX CMS 20 which has smallest grain gives sharpest (and overly the best) scans which contradicts your statement:



Another matter:


Yep, say it to people who downscale their film scans to 0.5 mpx and talk about sharpness all over the internet. Resulting resolution does matter but the lower resolution you made (with sharpening) the bigger impression of sharpness you can achive. Especially with P3200 which doesn't resolve fine details anyway.



And this I guess would be important to a man above who can't define 'a first-rate scan'.

They are used to seeing images that have a contrast response of 100% all the way up to the Nyquist limit and therefore confuse sharpness with resolution. Resolution and sharpness are related, but they are not the same thing. Sharpness is the contrast difference along edges. If you scale down so much that your edge is two pixels, then yes, after you jack the contrast up between those two pixels, it will look very sharp and clean. It however only has the bare minimum resolution to represent that there is even an edge there, and only if that edge happened to fall exactly in the right spot so that two pixels could represent it, otherwise it would be aliased, and not be sharp at all. If you want true sharpness, you must have enough sampling frequency/resolution to show the contrast response fall off of the medium you are digitizing. You have to have more sampling resolution than what the medium you're digitizing contains to do that. The *minimum* is 2x, significantly more is better. I'll take high resolution with loads of subtle gradation and detail with a contrast response that rolls off as spatial frequency increases over a small overly sharpened image any day, but that's just me. The bulk of the internet prefers 100% contrast response up to the nyquist limit of the image because that's all most of them have ever seen.

So with that being said, our definition of a good scan does not match. There is nothing wrong with that, it just is what it is.

In your experience, how do you deal with a grain that is the same size as your pixels but happens to be right on the boundary of two different pixels? That is the definition of grain aliasing and exactly the reason why you want *at least* two pixels per grain, again, significantly more is better.

In the case of ADOX CMS 20, the frequency of that films 100% contrast response is significantly higher than what pretty much every scanner can sample at, which means that you're only digitizing it at the limit of your scanner and not actually digitizing it at a resolution that does the film justice, so how exactly is that a good scan? It might be good for you as in it's the maximum resolution your scanner can deliver, or maybe it's all the resolution you want out of the scan, but technically, no, for that film, it's not a great scan because you're not extracting all of it's available resolution.

Every system that samples and digitizes analog sources has to deal with frequencies that are higher than its sample rate. In audio land, as a best practice a high pass filter is applied before the ADC to attenuate those frequencies as much as possible before they hit the converter because it reduces audible artifacts and aliasing, and the converter sample rate (or source resolution) is as high as possible. Good CD quality audio is not digitized at CD sample rates, it is digitized at significantly higher sample rates and scaled down after some processing. Every good sound engineer knows this and does this as a best practice. On a good sound reproduction system, you can in fact hear the difference between a CD that was digitized at CD quality and one that was digitized at a significantly higher sample rate and scaled down after some processing. This is especially true where the source being recorded has a lot of frequency information and a wide stereo image like a live performance. The stereo imaging is better, the higher frequencies sound cleaner, despite the fact that it's been scaled down to CD quality, pretty much everything about it sounds better relative to a CD that was digitized at CD quality. It is the same concept for film. Digitize at the maximum resolution of your sensor, low pass filter it either optically, or after the fact digitally to reduce visual artifacts and aliasing, then scale down to an intermediate working resolution to do the bulk of your work, then scale down to your destination resolution. It looks better once at the destination resolution.

Is doing that more work? yes. Is it worth it? It depends on the image.