TMax, scanning and grain

[leif e]

Until recently, I haven´t shot much b&w for several years. After rediscovering the RF cameras, I have started all over again - mostly using my favourite of old; the T-Max 400. I don´t develop films myself anymore, though. I give them to a pro´s lab with very good reputation. Still; when I scan these negatives, I´m often disappointed at the graininess of the scans and resulting prints. Especially I miss the smooth, beautiful skin tones. I use a Nikon LS 5000 scannner.
Any suggestions? Should I convert to another brand of film? They say C-41 process negatives are better for scanning. True?
Another question. I sometimes used the ISO 50 (approx.) Ilford Pan F in the old days - due to sharpness and almost no grain. Does anyone have any experience with bringing these negatives into the digital world?

leif e

 

[kaiyen]

Well, many folks say that the latest TXT emulsion is actually finer grain than TMY, but I have never done the comparison. Regardless, there are way too many variables here to know what's going on for you. Perhaps they are using a developer not suited to your needs?

C41 films do scan well basically right off the bat, but I _still_ believe that you can get the results you want with conventional B&W films if you have control and use the right tools and techniques. I will say this as often as I can (and have said it twice today now .

First, find out what developer your lab is using. Perhaps we can do something there. Then, consider using a different film. Then, consider using C41 film if you have to (I personally don't like it other than convenience).

My comments about getting desired results with conventional film apply to Pan F as well. Expose correctly, develop properly, and then scan well. You'll get what you want.

allan

 

[sf]

PanF scans well, but still has grain.

The most grainless film I have used, in terms of scannability, is Neopan 400. Or 100, I have not used the 100, but assume it is at least as good as the 400. The 400 is essentially grainless according to my scanner (Konica Minolta Dimage Scan Multi Pro). I am scanning 35mm at 4800 dpi and MF at 3200dpi.

I have heard that scanning the chromogenics is nice - but have never used one.

I just double-checked one of my Neopan 400 scans - and it is really really smooth film.

Here is attached a 3200dpi scan of a 645 frame of neopan 400, and a 100% zoom crop to show the grain. On 35mm, you'll have more grain, of course, but no more per unit area.

Note that the Multi pro really brings out the grain in anything. Even Velvia.

 

[telenous]

Leif, I just developed and scanned a roll of TMax 400 in the lab I take my film. I absolutely loathed the grain. Could be like kaiyen says, and proper development with the right chemicals will deliver the goods with conventional B&W. Having said that, I also developed an Ilford XP2 which I shot concurrently with the TMax - it was lush, with beautiful sheen, great blacks and appealing structure of grain.

 

[kaiyen]

Remember that XP2 doesn't even have grain - they are dye clouds. The "grain" that folks perceive is actually noise from underexposure more than anything else.

The finest grain films I know of that retain that quality while scanning almost without qualification are TMX and Delta 100.

allan

 

[Ukko Heikkinen]

Hello

My choice: XP2, 200 ASA.

Ukko Heikkinen

 

[Nikon Bob]

leif e

I think allan has a point in that the results from B&W film depend greatly on how they were developed. Scanning on a high rez film scanner will really bring out any grain there is. I have had variable luck with the only local lab that does B&W and I don't develope my own B&W film so I switched to using Kodak C41 B&W film. If you use C41 B&W film and do not want to see dye clumps try shooting it at 200 asa. It scans well in my Minolta 5400 as colour neg film and finnished in PS Elements 2.0 with the last step being conversion to greyscale. I imagine Ilford XP2 is similar judging by Ukko's comment.

Nikon Bob

 

[Gabriel M.A.]

Hmm, actually, I've noticed that XP2 has some "grain", even more so than Kodak's CN400.

All of this, however, is due to the fact that these emulsions "behave" differently when you scan them. A lot of this grain is actually noise that comes from the scanner. Some scanners are better at not showing this than others.

Kodak's done a nice thing and "optimized" some of their emulsions for scannability: CN400 and HD400. I've also noticed that Ultra Color 400 exhibits virtually no "grain" when scanned.

I do see definite grain patterns when scanning my old Kodak crap-MAX film (400 or 800) negatives, which is a little different. ::sigh:: It's too complicated to explain in this "tell me in a nutshell" world. I wish I could be better at explaining things like Allan is.

 

[dadsm3]

Agree with Gabe. In my limited experience I've found that scanners emphasize grain....and the problem is worse with 400 films, especially if slightly underexposed or if the dev time is pushed. Also, scanning as a b&w neg at 16 bits instead of 8 increases the grain. I've tried various grain reduction techniques, but couldn't really reduce it very much. I've found, however, that grainey negs print just fine at 4x6...but not 8x10.
Surprisingly, wet prints of these negs look just great, even at 8x10.

 

[varjag]

C41 films like XP2, contrary to classic BW films, tend to become increasingly less 'grainy' with overexposure due to the nature of dye pigment accumulation. Exposing 400 ISO C41 at EI 200 is a sound idea.

 

[kaiyen]

So some lessons from the last series of posts:

-scanners increase grain. This is because they act like a condenser enlarger (intense, focused light) with conventional film and because they get noisy in areas of thin dye clouds with color film. You need to act accordingly to address this issue (again, you can get what you want, but you gotta do what you need to do to get there).

-with c41 film, overexpose. latitude is so great that _starting_ at 200 is not an unwise choice. I shoot xp2 at 320 or 400, though - if I wanted a 200 speed film I'd have shot a 200 speed film (but a conventional one, of course). But I use the c41 stuff only for specific things and for convenience, not as my main stock.

-grain in conventional films is the result of one big, uncontrollable thing and 3 smaller, controllable things. The first is the film's grain structure itself. As I said before, TXT has bigger grain than Pan F Plus. There is no avoiding that. However, from one shot and/or roll of TXT to another, you increase grain via:

1 - overexposure (so get your EI right and meter correctly!)
2 - too much time in developer
3 - too high a temperature in developer

The last two points are tricky. Yes, part of it is dialing the process in - exactly 8 minutes and 15 seconds, for instance, and not 8 or 9 minutes. But you can't decrease time in the soup without increasing temperature, and vice versa.

This illustrates the need to know what your lab is doing, or to develop yourself.

allan

 

[titrisol]

1. Turn ICE off in your scanner

2. If you are sending it off to be developed elsewhere use always the same lab and sacrifice a roll. Expose one scene bracketing -2,-1,0,+1+2, then a blank frame, repeat for another escene til the end of the roll.
Have that processed and scan without ICE. Check which of the scans give you the best grain and tonality on your scanner and adjust the film speed that way.
Tis is due to the fact that most labs run standardized cycles (say 5,7,9 minutes or so) and thus negs maybe too dark or too light. Scanning exagerates grain anyways.
Normally scanners prefer "thin" negatives and your lab maybe over/under processing so you can adjust the filmspeed to account for it.

To check if your negs are too dense place them on top of a book and see if you can see the book through teh darkest part of your negatives. If you can not they have been over developed.

PS1. There was a thread about getting good results with Tmax 400 CLICK HERE
PS2. Try the same sacrificial roll with Delta 400 and compare results.

 

[titrisol]

These negatives can be scanned, specially if you use a good scanning software such as VueScan, but be aware of 2 things.
1. Scanners exagerate grain, specially if ICE is turned on
2. Over exposure/development is bad with scanners

Another question. I sometimes used the ISO 50 (approx.) Ilford Pan F in the old days - due to sharpness and almost no grain. Does anyone have any experience with bringing these negatives into the digital world?

leif e

 

[kaiyen]

Scanners don't just exaggerate grain with ICE turned on while using conventional film. Everything goes wonky. Don't use ICE with silver-based film. I always forget to mention this, so I'm glad Pablo caught that one

allan

 

[titrisol]

I disagree on the temperature part.
After processing for years with variable wtaer temperature (summer around 75F, winter around 68F) there is no significant difference if you adjust times properly. Most pro-labs will process film at 75F/25C or higher to shorten times in their pull-machines.
I remember when Xtol came out a lot of people where otraged that most of their dev times were at this temperatures, but it was not randomly chosen.

Diluting developer is another factor of graininess, generally the more diluted the developer the larger the grain.

So some lessons from the last series of posts:
---snip---
-grain in conventional films is the result of one big, uncontrollable thing and 3 smaller, controllable things. The first is the film's grain structure itself. As I said before, TXT has bigger grain than Pan F Plus. There is no avoiding that. However, from one shot and/or roll of TXT to another, you increase grain via:

1 - overexposure (so get your EI right and meter correctly!)
2 - too much time in developer
3 - too high a temperature in developer

The last two points are tricky. Yes, part of it is dialing the process in - exactly 8 minutes and 15 seconds, for instance, and not 8 or 9 minutes. But you can't decrease time in the soup without increasing temperature, and vice versa.

This illustrates the need to know what your lab is doing, or to develop yourself.

allan

 

[Nikon Bob]

I don't think Kodak Max 400 is crap film. This example was scanned originally at 5400 on a Minolta 5400 before being finished and down sized in PS Elements 2.0 for web use. It aws scanned with ICE on. The shot was taken with a Nikon S2 and 50/1.4 lens. I do not think it is too grainy/showing dye clumps.

Nikon Bob

 

[Buze]

I think the thinest grain on ISO 400 traditional is the Delta 400, hands down. TMax is also pretty good. The "heaviest" grain I get is usualy with HP5+

To preserve sharpness while loosing grain/noise when scanning, always scan 1) at the highest resolution you can 2) disabling USM and all the other gizmos the scanner software tries.
Then import the TIFF in photoshop and possibly apply an anti-noise filter, before resizing it to a more manageable size.

For example I scan at 4800dpi on the Epson 4990, with everything disabled. I use the "Auto exposure" button but then inmediately widen the spectrum of it's tone mapping to encompass the whole of the tone range, and apply a "linear" contrast curve.

Then in photoshop I add a "Levels" and "Curves" adjustment layer to get the tones I want, I do a small amount of "Noise Ninja" to remove any scanner noise/some of the heavy grain. I then rescale the image using "Bicubic Sharper" or "Bicubuc Smoother" if it's a portrait for example.
At that point you can possibly sharpen the result, or apply the USM using the "enhances microcontrast" technic.

Finaly I save as a JPEG for printing. On a 6x6 neg I get a JPEG that is roughtly 25 megapixels (5000x5000) that I can resize to whatever print size I want.

When you neg is under/overexposed the scanner software "autoexposure" amplifies the noise a LOT on a pixel basis to try to reach an "good looking" contrasty exposure. By scanning at a much higher resolution than necessary, you "drown" part of that problem in the extra surface, and if you do a bit of cleanup and use a "pro" resampling software to do the rescaling (read, photoshop) later you will get the smoother results.

 

[amateriat]

From what I can tell:

- The best scans from conventional b/w film developed somewhat on the "thin" side.

- Chromogenic b/w works best overexposed (i.e. XP2 @ IE 200-320, in my experience)

- Scanners differ in terms of light source: Nikons use LEDs, the equivalent of a condenser enlarger; Minoltas (except for the second version of the 5400) use a fluorescent tube, which is more the equivalent of a diffusion enlarger.

- VueScan helps a lot when scanning conventional b/w (doesn't hurt with chromogenics, either).


- Barrett

 

[kaiyen]

I would say the Minolta's tubes are more like a condenser enlarger with a very slightly cloudy piece of plastic over the bulb. The difference in grain intensity is minor.

allan

 

[arthury]

I, personally, enjoy Ilford PanF+ and loved the TechPan but it's no longer available in the market. I also have good experiences with Fuji Acros in terms grainlessness and tones. If you have software control over grain (Nikon Scan4), Tri-X is also fine.

Make sure you turn ICE off when you're scanning the true B&W negs. Chromogenic is fine with ICE.

I have a few posted in this other thread ...
http://www.rangefinderforum.com/forums/showthread.php?t=22612