Did Kodak color neg lose sharpness when it was reformulated to improve scanning?

[interneg]

Pere, if you spent less time opining about materials you have clearly never used, optically printed or scanned on high end machines, everyone would be better off. Not one of the photographers I work for who used VPS III in the early 1990's and subsequently used the various Portras (and who either exclusively optically printed at the time or were having optical prints made for them) is of the opinion that VPS III complies with your frankly bizarre claims. From my own experience working with their negatives, I would agree with their assessment.

Visual inspection of materials via microscope is largely worthless for anything other than basic diagnosis of processing problems. This has been the case for decades. What matters is the photographic behaviour of the materials, not the often easily fooled human eyeball.

Finally, if you actually put the work in to find that Kodak Gold datasheet, you'd discover that some of the faster Gold films have MTF responses closer to those of the Portra films - and that Portra has a drastic granularity advantage. The truth of this matter may not be dramatic, simplistic or comforting to your deeply held beliefs or agendas, but rather it is deeply complex in terms of aspects of photographic systems design unfamiliar to you - and which is not really very interesting to most people.

[Pere Casals]

Visual inspection of materials via microscope is largely worthless...

Yes, worthlessfor practical photography, and sharpness can be the least important factor in a great image. But inspecting the color clouds tells a lot about a color film, the day you try it you will know it, if you are able to understand what you see.

Pere, if you spent less time opining about materials you have clearly never used...

interneg, you don't know that, in the late 1980s I was using a pentax program a, that still I use often. This kind of arguments speaks about you, not about me.

Finally, if you actually put the work in to find that Kodak Gold datasheet, you'd discover that some of the faster Gold films have MTF responses closer to those of the Portra films - and that Portra has a drastic granularity advantage. The truth of this matter may not be dramatic, simplistic or comforting to your deeply held beliefs or agendas, but rather it is deeply complex in terms of aspects of photographic systems design unfamiliar to you - and which is not really very interesting to most people.

I don't propose comparing Gold to Portra. I propose to compare:

> Vericolor III VPS to Portra (Pro films)

> Pre (say) 1996 cosumer films vs say (2000+) consumer films. This is Kodacolor VR 100, VR-G 100 and pre 1996 Gold, compared to (say) any post 2000 consumer film.


________


For around 14 years CN consumer film sported 100lp/mm at 30% MTF, this is since T-Grain introduction until digital minilab era. In that moment preformance dropped to 50-75 lp/mm.

If I understand well you admit that this performace drop happened in the 1990s but you say this was not related to adaptation to hybrid processing, and that the "reengineered to deliver ... improved scanning performance" (datasheets telling it) did not had an impact in the resolving power.


Me, I see clearly that "reengineered to deliver ... improved scanning performance" ended in larger overlaping clouds to avoid aliasing in the scanning, and in that performance loss.

So from my side debate is finished, in the future well informed readers will judge our opinions.

[Oren Grad]

> Compare professional Kodak Vericolor III VPS (discontinued 1997) vs Portra 160/Ektar 100 (2019)

Attached MTF curves are for Vericolor III (1997) and Portra 160 (2016).

EDIT: Added Portra 160 NC and 160 VC (2003). 160 NC and VC were merged in 2011; the Portra 160 MTF in the 2011 data sheet that Pere linked upthread is identical to the 2016 one I've attached here.

[Oren Grad]

Two more: Portra 160 NC and VC (2006). In the series of data sheets I've referenced for these curves, 2006 is when the "reengineered to deliver significantly finer grain at all speeds for improved scanning performance and greater enlargement capability" wording first appears. The 2003 data sheet says "An emulsion overcoat and scanner friendly design make PORTRA Films the perfect choice for photographers and labs who scan negatives."

I have curves from 2008 and 2009 as well, but those are identical to the 2006 curves.

[interneg]

Cheers Oren, I've generally found the VC Portra a hair less sharp than the NC and the current 160 sharper than either. The big problem with the VPS chart is the lack of individual RGB plots, which of course makes absolute comparison trickier - and one of the more interesting aspects is the changes in the 30 & 50% MTF response of the red sensitive emulsion over the years which mirrors changes in the cinema films, especially 5203 - possibly because it gives a tonal/ colour behaviour that more closely mirrors the look of Ektachrome etc. Ektar does this particularly strongly.

[Pere Casals]

Attached MTF curves are for Vericolor III (1997) and Portra 160 (2016).

Oren, if one sees those curves one may think that both films are equally sharp, but they are very different graphs, splitting the curve in 3 ([1] suposedly) monochromatic tests delivers a nice graph. It has to be noted that the Vericolor graph is limited by the red sensitive layer performance, that is the worse one.

Now supose that we have a RGB MTF graph for Vericolor in what the Red curve is as good a the full spectrum MTF graph.


Say we have a lens with chromatic aberration but you want a nice graph to illustrate a datasheet with meaningless technical information. Then we make R,G and B MTF graphs for the lens, it looks that we deliver better information, but we simply plot a nicer graph like if it was a super APO.






[1] (suposedly) monochromatic tests because those Portra tests were probably made with very monochromatic (or narrow bad) sources rather than with color spectral bands, as wavelength of the peak sensitivity in the layer only takes information of that layer, while a color band would take information from the other layers from the channel crosstalk.


Portra NC sported 65Lp/mm at 30%MTF, if one one sees the 3 RGB monochromatic tests then it looks it's much better.

The big problem with the VPS chart is the lack of individual RGB plots, which of course makes absolute comparison trickier

yes, trickier...

and one of the more interesting aspects is the changes in the 30 & 50% MTF response of the red sensitive emulsion over the years which mirrors changes in the cinema films, especially 5203

I guess this is an interesting factor to analyze.

[Bill Burk]

I thought they did something to the base, the tooth that reduced Newton’s rings or helped retouching was interfering with scanners leading to what seemed to be grain, while you could tell under a microscope there was no grain.

[Drew Wiley]

Short answer: HECK NO. These are their sharpest films relative to ASA speed ever. I print these all the time. The very subtle tooth has a secondary effect of somewhat reducing rings (I still use AN glass on both sides of the carrier). I can't address specific scanning issues or any intermediate degredation that might hypothetically be involved. But when printing direct optical (optimized), the superior characteristics of the newer products are apparent. There is zero secondary texture seen even in big enlargements. This is the case regardless of whether I use a narrow-band RGB colorhead or a conventional YMC subtractive one. Color paper sees dye clouds differently than the human eye looking through a loupe or even microscope. If you do that, also use a medium blue filter to null out the orange mask. Ever notice that RA4 paper has a blue coating over the emulsion?

[interneg]

Drew - nothing to disagree with there!

Bill - that's largely my understanding as to the major modification specifically for scanning

[Pere Casals]

largely my understanding

IMHO to understand that it is necessary to understand well how film behaves depending on contrast in the test, if you remember I was pointing that to you not long ago:

> https://www.largeformatphotography.i...=1#post1507794

> https://www.largeformatphotography.i...=1#post1507818





About the 1999s cloud growth, I'll provide the evidence, "smoking gun" class. I need a while.


For the moment let me advance that if a cheapo V700 equals an Scanmate 11000 drum scanner with Portra (https://www.largeformatphotography.i...=1#post1509776 ) this is because Portra is not as sharp as 1980s kodacolor VR 100.