Comparing films by their specs
Michael K. Davis
http://www.smu.edu/~rmonagha/mf/filmsort.html
Bob Monaghan is hosting a spreadsheet I put together that allows you
(users of MS-Excel 5.0 or higher, PC or Mac) to sort several print and
slide films by a composite score that incorporates your preferences
(expressed as a percentage) for each of the following performance specs.
Grain RMS
Resolution at 1.6:1
Resolution at 1000:1
Sharpness
Bob has converted the default values to a static table in the html
document above, but a link to the spreadsheet itself is also available
there. Thanks again Bob!
Giving equal weight to Grain, Resolution and Sharpness, the spreadsheet
confirms that print films have better marks than slide films in the higher
ISO speeds. It also shows that Velvia still has a lot to offer over
Provia 100 F. Grain isn't everything.
Downloading the spreadsheet, you can manipulate the weighting percentages
as you see fit. Personally, I'm willing to forfeit fine grain before I'll
suffer a loss in sharpness, for example. So for my tastes, I don't give
these two criteria the same weight. I'm also more concerned about low
contrast resolution than high. As Bill Tuthill pointed out recently,
"Have you noticed that blue sky grain is more a function of low-contrast
res than RMS?" Those large expanses of blue sky is where grain is most
easily evaluated in scenics.
Thanks to Bill Tuthill for compiling the manufacturers' data at his page:
Mike Davis
--
Brian Walsh
Your table looks like a fun exercise. One minor point: You took the
manufacturers at their word, and that could send your ranking system
down the tubes.
For example, you list Fuji's NPS, Reala, and Superia 100 as tied by the
default ranking system. I can believe the sharpness and resolution
figures are identical for the three films, but Reala and NPS are quite
different in grain. Fuji lists both as RMS 4.0. Of course, that decimal
implies significance, but there must be something about RMS testing, at
least as Fuji reports the results, that makes "4.0" mean something other
than "more than 3.95 and less than 4.05." If grain size matters, you
can't sort it out with Fuji's numbers.
Brian Walsh
BTW, I still enjoy your focal length conversion and depth of field
spreadsheets.
Thomas Cooley
Kodak Royal Gold 25 was removed from the market,
but what became of Kodak Royal Gold 100 color print film in your
spreadsheet?
"Michael K. Davis" <zil...@primenet.com> wrote in message
news:8lvjd2$t3q$1...@nnrp02.primenet.com...
dj
"Brian Walsh" <mc...@dcn.davis.ca.us> wrote in message
news:39839FB3...@dcn.davis.ca.us...
> Fuji lists both as RMS 4.0. Of course, that decimal
> implies significance....
FWIW, I just checked the Fuji data sheets and they don't report the numbers as
4.0 etc. They just report 4 (ie one decimal place). Seems silly given the
information of interest is in the second decimal place. Yet I see it often
reported that Fuji reports the numbers as 4.0...is there somewhere other than
the Fuji UK site that holds their tech data?
Also, FUJI doesn't list the magnification when measuring RMS of slide films.
So it's anyone's guess how the numbers compare. More sloppiness to protect the
innocent I guess. Is there somewhere else Fuji lists this information since I
note the spreadsheet says to divide by 2.5?
Kodak PGI system can't be compared with the RMS system so who knows how they
should be listed. Still, Kodak's new measurements at least has the resolution
to be meaningful and does seem to accurately reflect the films that I've tried.
Cheers,
Byron
Bill Tuthill
> For my curiosity,
> Kodak Royal Gold 25 was removed from the market,
To my knowledge Kodak never released resolution or MTF numbers for
any RA emulsion. If I had to guess, I would say Royal 100 resolves
63 - 140 lp/mm and its sharpness falls below 100% at 19 lp/mm.
That would place it among the Fuji 100 films, but I believe Reala
has higher resolution than Fuji claims. I haven't tried NPS II.
Thanks to Michael and Bob Monaghan for the interesting spreadsheet!
Michael K. Davis
Brian Walsh <mc...@dcn.davis.ca.us> wrote:
: Hi,
: Your table looks like a fun exercise. One minor point: You took the
: manufacturers at their word, and that could send your ranking system
: down the tubes.
Garbage in, garbage out, right?! We'll have to eat from the
manufacturers' trash cans until we can find a better source.
: For example, you list Fuji's NPS, Reala, and Superia 100 as tied by the
: default ranking system. I can believe the sharpness and resolution :
figures are identical for the three films, but Reala and NPS are quite :
different in grain. Fuji lists both as RMS 4.0. Of course, that decimal :
implies significance, but there must be something about RMS testing, at :
least as Fuji reports the results, that makes "4.0" mean something other :
than "more than 3.95 and less than 4.05." If grain size matters, you :
can't sort it out with Fuji's numbers.
I've always been annoyed by the scale of RMS-Granularity figures - with
5.0 being twice the grain of 4.0, for example, how is that so many films
have whole-number RMS figures? An incredible coincidence? At least the
new Kodak Print Grain Index has a finer scale.
Using data found at Bill Tuthill's site again,
I used this single set of values in a sentance Bill wrote:
RMS 4.0 ~= PGI 28, RMS 4.5 ~= PGI 41, and RMS 5.0 ~= PGI 53, but the
equivalence is not linear and there seems to be a lot of slop in the
reporting of RMS numbers.
to come up with a formula that pretty closely plots the relationship
between the three pairs of values. It definitely isn't linear. I am
reluctant to publish it because I haven't tested it using Kodak's PGI
Grain Ruler (or any other method), but I have a strong hunch it's
accurate. You wont' find it elsewhere, so use it at your own risk!
To translate Color Print Film RMS Granularity to Kodak's Print Grain
Index:
Kodak PGI = 0.5335 * (RMS**2.8669)
To translate Kodak's Print Grain Index to Color Print Film RMS
Granularity:
Color Print Film RMS Granularity = (PGI / 0.5335)**(1/2.8669)
Here's a table of equivalent values already calculated. Remember, this is
Print Film (not slide) RMS-Granularity vs. Kodak PGI:
PGI RMS
12 3.0
14 3.1
16 3.3
18 3.4
20 3.5
22 3.7
24 3.8
26 3.9
28 4.0
30 4.1
32 4.2
34 4.3
36 4.3
38 4.4
40 4.5
42 4.6
44 4.7
46 4.7
48 4.8
50 4.9
52 4.9
54 5.0
56 5.1
58 5.1
60 5.2
62 5.3
64 5.3
66 5.4
68 5.4
70 5.5
72 5.5
74 5.6
76 5.6
78 5.7
80 5.7
82 5.8
84 5.8
86 5.9
88 5.9
90 6.0
92 6.0
94 6.1
96 6.1
98 6.2
100 6.2
: Brian Walsh
: BTW, I still enjoy your focal length conversion and depth of field
: spreadsheets.
That's great. Thanks!
Mike
: "Michael K. Davis" wrote:
: >
: > Have a look at
--
Michael K. Davis
: For my curiosity,
: Kodak Royal Gold 25 was removed from the market,
: spreadsheet?
I don't have all four specs for that film (beginning with Grain RMS) so, I
couldn't include it. Almost all the Kodak color print films are missing.
I just posted an article describing how I have worked up a somewhat
dubious formula for converting PGI to RMS, so I might ought to convert all
the Kodak PGI figures in Bill Tuthill's data to RMS and give you a simple
sorted table of RMS values for all the color print films.
OK, I couldn't resist... here it is (about thirty minutes later...):
4x Magnifiction Print Grain Index values for Kodak Color Print Films were
converted to RMS-Granularity using the formula: RMS = (PGI / 0.5335)^(1 /
2.8669)
RMS-Granularity values for films other than Kodak Color Print Films were
converted to PGI using the formula: PGI = 0.5335 * (RMS^2.8669)
Slide Film RMS values were converted to Print Film RMS values by dividing
by 2.5
Slide Films are indicated with an * by the name.
Brand Film Name RMS PGI
Fuji *Provia F 100 3.20 15
Agfa Portrait 160 3.50 19
Kodak *Kodachrome 25 3.60 21
Fuji *Velvia 50 3.60 21
Kodak Royal Gold 25 3.83 25
Kodak Supra 100 3.93 27
Kodak Ektapress PJ100 3.98 28
Kodak Royal Gold 100 3.98 28
Fuji *Astia 100 4.00 28
Kodak *Ektachrome 100s 4.00 28
Kodak *Ektachrome 100sw 4.00 28
Agfa HDC 100+ 4.00 28
Konica Impresa 50 4.00 28
Kodak *Kodachrome 64 4.00 28
Fuji *MultiSpeed @ 100 4.00 28
Fuji New Superia 100 4.00 28
Fuji New Superia 200 4.00 28
Fuji New Superia 400 4.00 28
Fuji NPH 400 4.00 28
Fuji NPS 160 4.00 28
Agfa OptimaII 100 4.00 28
Fuji *Provia 100 4.00 28
Fuji Reala 100 4.00 28
Agfa *RSX 100p 4.00 28
Agfa *RSX 50 4.00 28
Fuji Superia 100 4.00 28
Fuji Superia 200 4.00 28
Fuji Superia 400 4.00 28
Kodak Portra 160 NC 4.08 30
Kodak Portra 160 VC 4.22 33
Agfa Ultra 50 4.30 35
Kodak old Pro 100 4.35 36
Kodak Supra 400 4.35 36
Kodak old Pro 400 MC 4.39 37
Kodak *Ektachrome 100 4.40 37
Kodak *Ektachrome 100+ 4.40 37
Kodak *Ektachrome 100vs 4.40 37
Fuji *MultiSpeed @ 200 4.40 37
Kodak Ektapr Multspeed 4.47 39
Kodak Royal 400 Select 4.47 39
Kodak Vericolor 3 160 4.47 39
Agfa HDC 200+ 4.50 40
Agfa HDC 400+ 4.50 40
Agfa OptimaII 200 4.50 40
Agfa OptimaII 400 4.50 40
Kodak Ektapress PJ400 4.55 41
Kodak old RoyalGold 400 4.55 41
Kodak Portra 400 NC 4.55 41
Kodak Royal Gold 200 4.55 41
Kodak old Pro 400 4.59 42
Kodak Portra 400 VC 4.62 43
Kodak Gold 100 4.70 45
Kodak Gold 200 4.77 47
Agfa *CTprecisa100 4.80 48
Agfa *CTprecisa200 4.80 48
Kodak *Ektachrome 200 4.80 48
Agfa *RSX 200p 4.80 48
Kodak Max Zoom 800 4.80 48
Kodak Gold Max 400 4.84 49
Kodak Portra 800 4.87 50
Kodak Supra 800 4.87 50
Kodak Ektapress PJ800 4.97 53
Fuji NHG II 800 5.00 54
Fuji Superia 800 5.00 54
Kodak Ektapress 1600 5.10 57
Kodak Gold Max 800 5.10 57
Kodak old Pro 1000 5.10 57
Kodak Royal Gold 1000 5.10 57
Fuji *MultiSpeed @ 400 5.20 60
Fuji *Provia 400 6.00 91
Kodak *Kodachrome 200 6.40 109
Konica 3200 @ 1600 9.00 290
Fuji HG 1600 10.00 393
Fuji *Provia 1600 10.00 393
Please don't allow this to be any more than a conversation piece. My
conversion formulae are based on only three known pairs of PGI vs. RMS
values.
Enjoy!
Mike
: "Michael K. Davis" <zil...@primenet.com> wrote in message
: news:8lvjd2$t3q$1...@nnrp02.primenet.com...
Bill Tuthill
>
> Downloading the spreadsheet, you can manipulate the weighting percentages
> as you see fit. Personally, I'm willing to forfeit fine grain before I'll
> suffer a loss in sharpness, for example. So for my tastes, I don't give
> these two criteria the same weight. I'm also more concerned about low
> contrast resolution than high. As Bill Tuthill pointed out recently,
> "Have you noticed that blue sky grain is more a function of low-contrast
> res than RMS?" Those large expanses of blue sky is where grain is most
> easily evaluated in scenics.
However, Jonathan Sachs' very interesting comparison of Velvia and ProviaF,
www.dl-c.com/Velvia%20vs%20Provia%20100F/Velvia%20vs%20Provia%20100F.html
definitely shows that ProviaF has finer blue sky grain than Velvia, even
after doing a Normalize (or StretchHSV) to regularize color (don't do an
AutoLevels, it wipes out the blue!).
This is contrary to my assertion, as Velvia resolves 80 versus 60 lp/mm
for ProviaF in low-contrast areas.
In terms of color balance, Velvia seems a bit too cool (blue) and ProviaF
seems way too yellow.
Joe Manthey
granularity in 1990 in order to overcome several limitations of the RMS
granularity metric. As you've more or less concluded, RMS methodology
is a logarithmic scale whereas the PGI scale is linear. Consequently, a
change of 4 PGI is a just-noticeable-difference no matter what the
absolute values are, but the effect of a change of 1 RMS value will
depend on the absolute numbers, i.e., going from 4 to 5 RMS is a bigger
granularity change than going from 5 to 6 RMS. It is also standard
practice to round RMS values to the nearest whole number, which -- as
your table shows -- makes granularity distinctions among films more
difficult to assess.
What's less obvious is that the RMS method measures granularity at a
specific point on the characteristic curve which for most films is in an
overexposure region. The PGI method measures over a weighted range of
exposures that go from under to overexposure. The importance of this is
that granularity generally decreases as the film is overexposed, so the
RMS value probably doesn't correspond to the average image. This is
also why an attempt to mathematically equate RMS and PGI values will be
error-prone.
Another issue is that RMS values are strictly valid only for 12X
enlargements, whereas PGI numbers are usually quoted for more than one
magnification, allowing the user to determine the pertinent comparisons.
There are additional techno-geek reasons for preferring PGI methodology
for color negative films, but these are probably the most important
ones. It's an attempt to provide useful and improved information.
Joe
Sent via Deja.com http://www.deja.com/
Before you buy.
Michael K. Davis
Brian Walsh <mc...@dcn.davis.ca.us> wrote:
: BTW, I still enjoy your focal length conversion and depth of field
: spreadsheets.
You probably have v2 of my DoF spreadsheet. It has evolved a bit
since then. You can find v5 at
http://home.online.no/~gjon/mdofcal5.xls
It's still in the MS-Excel 5+ format, for PC or Mac
My thanks to Jon Grepstad for hosting it! Take note that he no longer
maintains his pages at
If you have any bookmarks to the great links he has, you might want to
update them to point to:
Mike Davis
--