Tmax 400 vs G10

Discussion in 'Film and Processing' started by mjferron, Feb 23, 2009.

  1. I was challenged to do my own test and so I did. On the left is the 100% crop from a jpeg shot with my Canon G10 at ISO 400. I was pleased to see 400 on that little camera was not all that ugly compared to ISO 80. Sure a bit more noise and a slight loss of detail as is expected. On the right The new Tmax 400 developed in D76 1-1 for 10 minutes shot with a Nikon N80 and 50 1.8 at F8. Both cameras were tripod mounted and the self timers were used to assure sharp images. I also used mid range apertures to provide a decent depth of field to help balance any possible focusing errors. It's tough to keep image size exact. One camera you are framing on a 4/3 screen and the other a 2/3 viewfinder but managed to keep them pretty close. Both images were give the benefit of levels adjustments and sharpening. BTW the film was scanned on a K/M scan dual 4 @ 3200 dpi. I know there will be a bunch of folks telling me this is flawed or an Imacon scan would have blown the doors off my scanner but when you enlarge the images up to 200% it's quite clear there is really not much left to extract on the film image and the results are pretty much as I expected. Judge as you will. Critics please post your own honest comparisons.
  2. Enlarging a scanned image to 200% doesn't tell you if there is any more detail to extract from the film. Slap the film under a microscope to find out how detail is actually on it.
  3. Detail level looks pretty similar. You might want to run noise reduction on the film scan before any sharpening; it really exaggerates the grain a lot otherwise. The camera does it too, after all.
  4. Neil I'll send you the digital file and you can look at that under a microscope. Real world dude. Real world. As I said elsewhere scanned 35mm Tmax is worth about 10mp tops. Do keep in mind the advertisement I pinned to the garage wall filled about 15-20 percent of the viewfinder/screen. Then a 100% crop was taken from that. This is a tiny crop of an image that would look great as an 8x10 from either camera.
  5. Janne the detail is not far off. The film though is not going to show 5x the detail of the digital file as claimed elswhere no matter how you treat it.
  6. "the self timers were used to assure sharp images"
    Hello Michael: Did you have the mirror locked up on the Nikon?
  7. Hello Lous the N80 doesn't have a mirror lock up but I can assure that camera's mirror is the lightest hitting thing around. Plus my shutter speeds were up around 125. Also I used my best tripod with an expensive Novoflex head that was sitting on concrete.
  8. Not a "try", just a question. You will get many. And I think you mean to say your shutter speeds were ONLY around 1/125. That's not very fast at all.
  9. And the lenses are different.
  10. Yeah but on that camera using the scenario above the mirror is not a factor. Plus mirror vibrations are mostly a problem at 1/30 and under. It's plain to see that the grain is becoming the limiting factor on the film image. You hit that point where grain starts blurring detail and it's over. I don't have an ax to grind as I will continue to shoot both digital and B&W film. Maybe someone else can show an example with the 35mm film showing a slight advantage. That's fine. Just don't show me a pixilated digital image that's obviously been handicapped.
  11. Larry I would have to give the 50 1.8 the edge here unless someone is sure the g10's is as good. Both lenses were shot at mid aps hoping to keep things equalized.
  12. It is just that a camera with a non interchangeable lens is mated for the perfect... I am not disputing just pointing out another factor. Developing film scanning it the choice of developer... all factors....
    I say it just can't be done properly..
  13. Larry you are right and I'm sure the results would vary a bit on another test. (I'm done though) My whole point here is to show folks 35mm film is not going to blow away good digital when it comes to detail. Those days are gone. Hey this was just a point and shoot. The comparison would looks worse if I had an A900 to shoot with.
  14. Film is my friend. I am not going to get into this... Have fun enjoy see your stuff in the years to follow. To each his/her own but since I am shooting mostly MF these days I want you to know that Casual shooting is getting better with Digital and some day a few of those new photographers will move up...
    You are also my friend.
  15. I'm not disputing your results or the quality of one medium over the other. Your results are certainly valid for the equipment and technique you used. The point I am making is that the playing field is not entirely level and to assign any global relationships beyond your personal set up would not be accurate. Not grinding any axes here, I shoot both as well.
    " the mirror is not a factor. Plus mirror vibrations are mostly a problem at 1/30 and under"
    This may hold true for normal magnification general photography but not high resolution imaging of high contrast targets. It doesn't take much to lose maximum edge definition. Combined with the slightly smaller size of the film image, variables can multiply quickly. I won't even go to scanning variables.
    You've make a good test that has given you great information about your equipment, imaging chain and working technique. Use that information to your advantage. Making larger, more global attributions, would not be completely accurate.
  16. Yes Louis there are many variables but I'm not sure any would make the film image a lot better than we see here. Maybe a little better. I'm afraid Mauro stated on the other thread that 35mm film has 5x the actual resolution of his Canon D40. My whole point is this just isn't so.
  17. What's up with that scan? It looks like the capture was smeared.
  18. Nah that's just the end of the resolution Dave. Hit the wall so to speak. Again this is a very tiny crop. you would not see the difference in an 8x10 image. If it were smeared the grain would be smeared. The grain is sharp. This test pretty much sums up what I've seen in the past in other such comparisons.
  19. Your outcome is EXACTLY what I would have expected and what I see in my scans on a Nikon ED9000 with 35mm film. Your test is valid. Last year I compared 120film in a 645 Mamiya with the output of my 5D - same ISO, same focal length, etc. and the result was pretty much the same - the 5D file had more detail (albeit not as obvious as your comparison here). I love film and will not give it up, but we have to realize that digital has come a long way and has surpassed 35mm film quite awhile ago.
  20. Thank you Juergen. I like film because I like the looks of the highlights better and i can print in the darkroom. .
  21. Digital is better for low light work as it is inherently faster. Try doing a comparison with microfilm and/or a larger film format like 4x5; 400 speed for film is very fast for film.
  22. Michael, I have a scan Dual IV. I've never had a scan look so smeared, with TMX, FP4, HP5, Tech Pan, Fortepan 400....none of them. I'm not questioning your comparison, it's just that I've seen better from the Scan Dual so I'm not sure what's up. As I've been meaning to try a roll of the new TMY, I do a scan of something and compare it to my 40D or K20D.
  23. Michael, look at your film with a microscope, if it looks like your scan, throw away all the chemicals and start over. If it looks good, service the scanner.
    At 3200 dpi (this is only 65 line pairs per mm) you shouldn't see much grain and have well excess resolution in the film to fill every pixel of the scan with information.
  24. Mauro did your even read the part about this being a tiny little crop?? The chems are fresh. This what 35mm 400 B%W film looks like @100%. Other folks know this to be true as well. I might suggest you return your 40D as there is something wrong with it looking so pixelated.
    Dave the grain is not smeared. If the image was smeared the grain would be smeared.
  25. See this a full sized image image on Tmax 400. Looks great but if you take a tiny corner it's not going to look so good. Do you understand how this works?
  26. Ok here it is.
  27. I thought I left this conversation. Just leave it alone and shoot what you want to do what you want.. the time we spend fighting here could be spent better.
  28. Agree with Larry. Let's just enjoy the shots.
  29. Michael, can you post the unsharpened film image crop for us please?
  30. Michael, looking at the unsharpened scan can pinpoint the problem. At 100% please. Straight from the scanner with no processing.
  31. Dave this shot has no levels adjustment or sharpening.
  32. Thanks Michael. It looks a tad better and I can read the numbers but still not what I expected.
    Do you have a way to look at the film on a microscope and verify if the bottleneck is the scanner or the film?
  33. Actually both images look more natural in the second photo. Sorry Mauro I don't own a microscope. The shot is sharp and it's in line with another test I did in the past. Maybe a better scanner would improve things a bit but I don't think the improvements will be earth shattering.
  34. I think I would do better with a Nikkormat FTN and a 55/3.5 Micro Nikkor P with the mirror locked up. You might also have done better by simply printing the negative and then scanning it with a good flatbed. Very nice work can be done with either a G10 or a film camera with b&w film. All of the tests we see here are interpretations of what film can do after the film is scanned and the scan is then viewed on a monitor. A more carefully done test with the negative printed in the traditional way would show different results.
  35. I love film of all kinds and my Nikon F6 is my favorite camera. That being said, if I want vibrant color and unlimited depth-of-field, I have not seen a scanned 35mm slide yet that will out-resolve my G10. I really think it's that good.
    Now, the G10's optical finder is only 80%, it's much slower, doesn't meter as well, and isn't weatherproofed, but it DOES fit into an outer pocket pretty easily and is usually with me, whereas the F6 is at home waiting to go out on photo jaunts.
    So, they are totally different pieces of equipment. My opinion is that the perfect "do everything" camera has not been invented yet. The flip side (and a positive), is that it gives me more excuses to buy numerous cameras that each have a different strength.
    Enjoy them all! Don't get wrapped up in the minutae and shoot shoot shoot. By the way, I happen to think that the G10 at 400ISO looks like my Tri-X when the digital shot is converted with care. The Tri-X, of course, does this effortlessly, but I have to spend a bit of time in the dark with some chemistry to appreciate it.
    By the way, the motion picture guys use these differences to help tell their story. So, if you're in the city, embrace film and grain, or shoot fast with some sensor noise! If you're in the country, pump up the vibrancy, using whatever method is appropriate for the capture. Point is, use the best tool...
  36. Thanks for the unsharpened version. I've been meaning to try TMY2 for a bit. I guess I'll have to put it through the paces and compare.
  37. Michael, this is the best 40x microscope you can get for the money for the light table:
    Just $27.
  38. I can't comment on TMY-2, but I have printed pictures in my darkroom taken with TMX that outresolve what my Nikon 9000 can see, which will outresolve a K/M scan dual 4. They weren't test shots but real pictures I took out in the field. Real World.
  39. I find the same with Acros....which is about the same as TMX. With it, fine details like grass and leaves hold together that don't in my 40D. That said, there was more detail in print from my old Nikon 9000 (which I never should have sold!!!!) then there was/is in my Scan Dual IV. Visible in 12x18 and 14x21 print.
  40. Agree, there is at least twice the data on my TMX under the microscope than my Coolscan captures.
    Dave, never asked you why you sold the 9000?
  41. I was going to look at the Sony A900 as a replacement for MF. I had purchased the Epson V700 with a 4x5 fluid mount holder. I figured if I wanted 35mm scanned, I'd use my old Scan Dual IV.....and for 4x5, the Epson. I (wrongly) figured that for the odd time I used MF, I'd just use the Epson. In the end, after selling unfortunately, I decided I liked the 9000 scans of 35mm and MF much better. The funds from selling the 9000 went to my Bessa R2a, 35mm Voigtlander f1.7 Ultron, and towards the Pentax K20D.
    I'm not sure I'm gonna purchase another scanner for now though. The Epson gives me very good 16x20 from MF....which really is the largest I ever printed from MF on a normal basis. Any larger and I always made sure I had the original on 4x5 film. 4x5 on the Epson is far better than MF on the Nikon.
  42. Sounds like a sad story (sniff sniff). I can hear the violin...
  43. From my test experience there ar the following problems with this test:
    1. It is not a digital vs. film comparison, but a digital vs. digital comparison. A scanner is a kind of digital camera. This scanner (digital camera) here has a rather limited resolution of only 3200 ppi, which means significantly less than 70 Lp/mm resolving power. Furthermore most of these amateur scanners can't fully achieve the resolution values of their official data sheets. So in real world the resolution is probably slightly under 3200 ppi (less than 60 Lp/mm). But TMY-2 is able to achieve resolution of more than 100 Lp/mm (I have no problem to get that with my 50 Nikkor primes).
    2. 1/125 shutter speed even on a tripod is not fast enough to achieve the best resolution, especially if you don't use MLU. Faster shutter speeds and MLU give 20-30% higher resolution. I have tested it.
    3. To fully exploit the potential of modern films and sensors you need absolutely exact focusing. Depth of field does not help you at all! What exact focusing means can be seen under a microscope. Only slight focusing errors can result in dramatically different resolution values. I've got tests where my resolution figures varies from 60 to 120 Lp/mm! Because of differences in focusing. And all shot with f8! Focusing errors are the most underestimatet problem in all these tests made by people with no test experience. We have no really exact focusing systems, neither AF nor MF. You have to use focus bracketing for exact test results.
    4. I've done lots of tests with Kodaks new T-Max 400. Incredible stuff. Especially concerning resolution and fineness of grain. I achieved more than 100 Lp/mm resolution with my 50 prime lenses. Object contrast was moderate 4 stops (about 1:16). The lines of the 100 Lp/mm structure are so clear, that you immediately see that the resolution is even higher (100 Lp/mm was the finest structure on the chart). A digital sensor must have 35 MP to achive a resolution of 100 Lp/mm.
    For 100 LP/mm you need a very good drum scanner. I use my Rodenstock Apo-Rodagon for enlarging in my darkroom. No problem at all with this optical jewel to print 160 Lp/mm onto paper, I have tested this with Rollei ATP negatives (outstanding film).
    Friends of mine have achieved similar results. Zeiss has achieved 120 Lp/mm with TMY 1. Look at their website to the publication "camera lens news" No 17,19, 20, 24. And the article about MTF in their last camera lens news of Dr. Nasse, chief optic designer at Zeiss. He proved that TMX is able of resolution values of more than 160 Lp/mm (see the diagramm on page 11). That confirms former tests at Zeiss made by K. Mueller (published in the camera lens news as well).
  44. In my resolution test of the new TMAX 400 I measured 220 lines per mm with the microscope. That is 40MP of actual resolution. Also the tones are nothing short of fantastic.
    Grain is not only the smallest I've seen for a 400 film but also very attractive.
    Tones (shades of gray) are superb.
    The G10 only tested at 10MP by DPreview.
    I would try Xtol 1:1 at 75F for 7 miinutes Richard. And then look at it directly under a microscope.
    I will not post samples to avoid yet another upset to a thread.
  45. Michael, please let us know what you see when you look at your film under the microscope.
    Or send me the film and I'll scan it for you.
  46. I second what Francois posted.
    These "digital vs film" tests and debates can never really prove anything definatively.
    The camera bodies are different. The lenses are different. The sensors are different. The scanner used for the films are different. The software used to scan is different.
    There are way too many variables and there is no way of comparing the two without having to consider those factors.
  47. Francois and Craig I also agree there are far too many variables that could change the outcome in another test. In practical, real world situations either camera above will make a wonderful 8x10 or 8x12 which is most often as big as I personally print anyway.
  48. Michael, I think that sums it up very well. Most people want more and more resolution, yet when I actually see what these people print, it often goes no further than 8x10, or 10x15. At most, prints from the overwhelming number of these resolution seekers never reach even 16x24. As most would consider a 240dpi print at 16x24 to be more than adequately detailed, that is reached by todays good gear. In fact, considering most prints never go beyond 12x18, any current DSLR in the 15mp range, any fine grained 35mm film well scanned, any MF or LF camera will all produce adequate results.
    For me, the main reason I shoot MF and LF is because many of the prints I produce are 16x24, 24x30, 30x40.....and for that, current DSLRs run out of steam in the larger sizes.
  49. Dave I've never liked pushing 35mm too far. No matter how much detail you may have. Over 10x15 and that's the territory of larger formats. Subject matter comes in to play as well but in general if you buy a quart of paint for the ceiling that really needed a gallon then things are going to be spread pretty thin.
  50. With B&W 35mm film I try avoid print sizes where grain is visible but not obvious. I like to either avoid it or make a statement with it. I'd say in general the 16x20 to 20x30 print range from TMX100/XTOL 35mm falls in this visible but not ovious region. This is just my personal take on it.

Share This Page