35mm film vs 5DII - Low light performance

<p>Karim, this is the shot with NR off you requested.</p><div></div>

<p>Rishi,<br>

this is the shot processed in LR3 you requested. It was processed by a colleague to offer the most detail - it includes backlighting and other adjustments.</p>

<p> </p><div></div>

<p>As we discuss how to make the 5DII look better, a word of caution when someone considers shooting hundreds of pictures digitally and plans to sort through and edit (via RAW processors and then PS to) this body of work to match the look/detail of film. This is many many hours of work.</p>

<p>Just remember I spent zero time in on PS modifications on film. B&W development was 30 minutes (both rolls) and color negative zero (drooped off and picked up). Scanning was unattended other than to reload the cartridges with no adjustments either.</p>

 

<blockquote>

<p ><a name="00YECL"></a><a href="../photodb/user?user_id=3995956">Mauro Franic</a> <a href="../member-status-icons"></a>, Feb 15, 2011; 08:51 p.m.</p>

</blockquote>

 

<blockquote>

<p>"Similarly using a faster shutter speed with the flash means that the background areas, which would normally be underexposed on film, are instead black, so we can't see the lower tonal range."<br>

Zack, flash shots had a slower shutter speed (limited by the synch speed). You may clearly see the tones of the brick wall in the back on the Portra shots. they are just out of focus bcs of the large aperture.</p>

</blockquote>

<p>Okay, so a slower shutter speed. Your speeds aren't listed, so I made an assumption based on the fact that the backgrounds are MUCH darker in the film shots. The exposure still relies on the flash though, which means that, for whatever the reason, the foreground and background are exposed in (essentially) two different ways, while your digital shots are exposed entirely by ambient light. Theoretically, if you rated each shot at the same EI/ISO and used the same camera settings, we would see two similar exposures. Since you did not do this, we can't give a fair comparison. The Portra shot for instance, is HEAVILY reliant on flash, while the 3200 TMAX one is not.</p>

<p>That brings up another question though ... if the 5DII could take a well-exposed photo at 3200 ISO and f/1.8, why the heck didn't you just use the same settings with your film camera? I could understand if you didn't want to change lenses at a rock show, but since both lenses support that aperture there's no reason that the flash even needed to be on. By shooting at f/4 and using the flash, you're assuring that the film images will be sharper - not only because of the flash itself, but because the 50 1.4 is sharper at f/4 than the 1.2 is at f/1.8.</p>

<p>Again, I applaud your efforts, but it's not apples to apples. It's an apple to a hybrid that is very much like an apple.</p>

 

 

<blockquote>

<p>The Portra and TMAX were set at 1,600.</p>

</blockquote>

<p>Wow. Generally, though, the proper philosophy is to overexpose negative film, correct? Typically I set my Portra 160NC to ISO 100 or 80 in my Canon EOS-3.</p>

<p>Also, that LR3 processed shot looks terrible. Doesn't look like masked sharpening was used...</p>

<p>While I love your work in general, Mauro, I feel it's a bit unfair to compare the Portra w/ flash vs. the 5DII with ambient lighting. With the 5DII, you're amplifying pixels (on his face) w/ low signal:noise, whereas you probably had a higher signal:noise (on his face) on the Portra shot b/c of the flash. Naturally there'd be more detail on his face with the higher signal:noise.</p>

<p>So... that doesn't seem fair... does it?</p>

<p>"if the 5DII could take a well-exposed photo at 3200 ISO and f/1.8, why the heck didn't you just use the same settings with your film camera?"</p>

<p>Zack, we experimented with different settings on 3 different cameras to cover the event. We were not trying to replicate the shots for exact comparison.</p>

<p>Rishi, you can isolate each shot and evaluate the results independently.</p>

<p>If you want to match the lighting, judging from the guitars, the TMAX EI 3200 and the 5DII EI 3200 shots were shot with similar lighting conditions. The TMAX EI 1600 and the Portra EI 1600 were similar together as well. </p>

<p>Next time I can remote trigger all the cameras at the same time for a few shots. I did not anticipate so much detailed interest.</p>

<p> </p>

<blockquote>

<p>Next time I can remote trigger all the cameras at the same time for a few shots. I did not anticipate so much detailed interest.</p>

</blockquote>

<p>LOL, quit being so modest. Your film vs. digital threads are <strong>always</strong> epic. You should always anticipate detailed interest :)</p>

<p>In my opinion the 5DII did excellent for low light action at ISO 3200 (independently of the results on film). There is no need to put it down.</p>

<p>Film did fantastic as well (independently of whether the 5DII could have captured the same shots similarly in terms of detail or not).</p>

<p>The key observations for me are:<br>

- Film provides super sharp and detailed results at high EI. Especially TMAX although its DR is compacted during push processing. <br>

- TMZ (not posted here) offers wider DR and tonality than TMY-pushed but produces larger grain.<br>

- Portra's DR is wicked wide for a fast film.</p>

<p>LOL Rishi.</p>

<p>Just out of curiosity, Mauro, is there any reason you didn't push the Portra 400?</p>

<p>Sorry I missed answering. I did push Portra but not in a controlled environment. I had it developed at Target with special instructions to push process it - develop it as 1600 film. Go figured what actually happen.</p>

<p>In the end, it needed no adjustments after scanning and I can tell you what confidence what EI it was shot at but that's all. Based on this, they either correctly pushed it two stops or Portra is so generous that can eat two stops for lunch.</p>

<p>"Jeff, this are real life results from live shooting. When controlled testing is posted, the answer is that they are not real life, and viceversa"<br>

This doesn't really excuse anything. Either you should make controlled photos, or not present the results for analysis/discussion.<br>

Personally, I found the posted example to be too small to discern anything. Also, it would be helpful to know the scanning method.</p>

<p>Scott,<br /> <br /> There is no need for excuses and this is not a scientific test. It is just sharing results. If you find no value in this analysis/discussion you can ignore it.<br /> <br /> If you are interested though, you can follow the link and see the shots at 100%.<br /> <br /> To your question, the scans were made with a Coolscan 9000 with all adjustments off (other than auto f&e and ICE for color of course).</p>

<p> </p>

<p>Scott, you may also look at the results like this if you prefer. Are you able to appreciate the observations about film I made now?</p><div></div>

<p>Mauro,<br>

Sorry, I didn't see the link. Interesting, I almost prefer the Tmax @3200 to 1600.<br>

Scott</p>

<p>Mauro, great comparison!<br>

I'd like to know how you developed the B&W film and what did you scan it with?</p>

 

<p>Sorry but with different lighting, different lens and different ISO in most cases for the 5DII I have to ask if this is a valid comparison. I am by no means championing any particular product but lets have a bit more scientific methodology.</p>

<p>No problem Scott. </p>

<p>The TMAX 3200 looks smoother and more natural (though with a tad more visible grain do to the pick up in PS brightness). The highlights are better controlled too.</p>

<p>The TMAX 1600 has a gritty look though that makes me my favorite.</p>

<p>Tony, this is not a scientific comparison. Light changes by the second in a concert.<br>

As we discussed in this thread, shots 1 and 4 seem to have caught very similar lighting. Shots 2 and 3 were also similar with each other.</p>

<p>I actually have one more question, Mauro: how does the grain in Portra 400 behave if you overexpose? If I had the luxury to indulge, I'd like to see comparisons with Ektar 100 and Portra 400 @ 100. Ektar will probably hold more detail but Portra might have less grain... maybe.</p>

<p>Next time you shoot Portra 400 for comparisons, I hope you don't push because it's more valuable knowing the native latitude of a film. That way you know if it's safe to change ISO at will on the same roll.</p>

<p>Pete,</p>

<p>TMAX was developed in Xtol 1:1 at 78F for 8 minutes with 8 agitations every 30 seconds (5 inversions and 3 vertical long taps).<br>

<br /> All film was scanned with a Coolscan 9000 with all adjustments turned off using the Nikon Scan software (with the exception of ICE for color).<br>

<br /> No PS processing was done on the film with the exception of adding brightness to the TMAX 400 in PS.</p>

<p>"how does the grain in Portra 400 behave if you overexpose?"<br>

Karim, it does superb at EI 100. Grain will still be larger than Ektar. Dynamic range will be materially wider than Ektar as well.</p>

<p>It is a very noble film.</p>

<p>Here is a test similar to what you ask:<br>

http://figitalrevolution.com/2010/11/17/kodak-new-portra-400-review-part-2-skin-tones/<br>

http://figitalrevolution.com/2010/11/18/kodak-new-portra-400-review-part-3-usable-ei-range/</p>

 

<p>Related to that,</p>

<p>TMAX 400 at EI 3200 and TMAX 3200 at EI 3200 are very different:</p>

<p>TMAX 400 at EI 3200: Sharper with smaller grain at the expense of high contrast and narrow er dynamic range.<br /> TMAX 3200at EI 3200: Very sharp and detailed too with much larger grain. Midtones and dynamic range are more controlled.</p>

<p>Both look super nice and very different.</p>

I do have lots of past experience shooting TMax 400 and 3200 for a newspaper. I shot it, developed and printed it all with TMax developer. In practical usage the 400 was just fine for newspaper work and if I used flash I used it with 400. We used 3200 for low light situations where we could not use flash and we did not use it unless we had to because it was grainy and the contrast, as I remember, was not great in low light particularly so it did not look great in the paper. I tried it for sports but did not like it. I pushed 400 as a preference to 800 or 1600 if I had to but mostly got by with good lenses. My 5d 3200, IMO, is a hell of a lot better in color than the TMax 3200 in B&W at least in my experience. I just printed a usable 5D 3200 13x19 without flash print of a group and although colors are not as intense as an ISO 400 print they are acceptable and with a little noise reduction in processing the picture is quite good. I never went that big with Tmax 3200. I have to say that newspaper work is highly varied in all sorts of lighting conditions. So while the test is nice there is more to real world usage than just shooting a guitar player. It is hard to tell anything anyway with the low resolution on PN. Comparing flash to non-flash pictures truly distorts the results, IMO. Look, I have made some pretty stunning pictures with TMax 400 with darkroom developing. I used it at weddings and made my own prints just for the B&W I have never had much use for TMax 3200 and used a lot of it several years ago. If I wanted to get seriously back into B&W I would use TMax 100 and would not scan but print with an enlarger. I love the film. I cannot seem to get as good a B&W results with digital, particularly with deep blacks. As a former R&D director Mauro, I would call it a comparison as I don't think it qualifies as valid test. It, however, has shown its usefulness to some of the posters.