mp equivalence to 6cm square

[bruce_erickson1]

What is the mega pixel equivalent of a 6cm square slide using Kodak

E100 film? The next time someone thinks my mf camera is "quaint" or

says I should enter the 21st century, I want to be able to say that

each shot is x mp, or something like that. (When I scan at maximum

resolution I often get over 100 mb .tiff files, but this probably is

not an accurate assesment of the equivalency.)

[kbreak]

actually your answer was pretty accurate.<BR>you can choose to scan your MF at one Mb file size or one million Mbs, it's your choice.<BR><BR>100Mb is probably the commercial average, or in the ball park.<BR>

[grinder]

I too shoot 6x6 but I really could care less what the mp equiv. is. I understand why you want to know but the truth is the next thing out of there mouth will be But my D** looks beter than you **MP scan. No crap of course it does the film scan is a second gen. image there digi file is a first gen. This aurgument grows old, I shoot film I appreciate digitals points I still chose film. I have no reason to use digital I am not a newpaper or press photog I do not need images now I enjoy chemistry. If anything I am working my way backwards I miss my 1970ish sears ricoh. I have a eos rebel and use manual mode 90% of the time and now I shoot my autocord more than anything. If I ever buy a bigger enlarger I may go buy a 4x5 field camera. My style is slow I have no need to take 400 pictures this afternoon. Bruce if you enjoy film be proud stand up for your choice comparing MP is like comparing MHZ it is only a tiny part of the overall picture (Pun intended) I love computers I am excited that digital is getting better I just enjoy film is that so wrong.

Tell them "I do not agree with what you have to say, but I'll defend to the death your right to say it." voltaire

[hakon_soreide]

Someone might have the exact answer for you, but some general remarks might be in order: at some dpi it might be argued that you just keep adding definition to the film grain rather than getting new details out of the film. To the extent that film grain is a wanted part of the creative work, you could go rather high in dpi before you stop getting anything more out of the scan.<p>

A scan is also a true RGB rendition of each point, whereas a digital camera with very few exceptions uses Bayer interpolation, meaning that the resolution is just for luminosity, and colour information is interpolated, reducing the actual effective resolution when compared to a CCD getting all three channels of colour information for each pixel.<p>

I've detected details in Velvia slides when viewed through a loupe that, if they were to be distinguishable in a scan, would require at least 8000dpi.<p>

Using 8000dpi as a general guideline for where you might stop adding detail if resolution is increased, a 6x6 (which are usually around 54x54mm, right?) would make 17.000 pixels by 17.000 pixels, or 289 megapixels.<p>

Wonder how long until digital cameras can do that...<p>

Hakon Soreide<br>

Bergen, Norway<br>

<a href="http://www.hakonsoreide.com">www.hakonsoreide.com</a>

[kelly_flanigan1]

Most of the MF films shot with 620 and 120 have been by amateurs; ie the Brownie format. Most of the films are decent; most the film ever shot was with single element lenses. A camera and film combo shoots an image. Just mentioning the film type is not enough info. A Kodak Duoflex image with a tripod can hold more info than a blured Hasselblad image. This question is many decades old. Kodak was looking a film for storing digital info in the 1970's. One might as well ask how much an acre of land is worth; without mentioning the location.

[grinder]

Good point kelly

[hakon_soreide]

pixel count is of course just one aspect of this. Regardless of how many pixels you make a photo into to discern between details or to refine the grain, will it look better?<p>

Of course, digital vs. film comparisons usually compare digital vs. digital in that they compare digital photos to digital scans. And usually to scans made on low-end scanners. Film usually has a lot more potential than it's possible to glean from it.

<p>

Digital, however, seems to be more a case of getting what you see with no extra fuss.<p>

Wonder when I will go digital...<p>

Hakon Soreide<br>

Bergen, Norway<br>

<a href="http://www.hakonsoreide.com">www.hakonsoreide.com</a>

[jnorman2]

the common wisdom at this point is that about 8mp equals the quality of 35mm film. the testing i have seen has demonstrated pretty clearly that at around 12mp, you are equal to, or exceeding, the resolution of MF scanned film. at 16mp, you are at least equal to the best MF film scans, but not quite yet equal to 4x5. so far, nothing quite smatches LF. the "MF" digital backs, like those on the hassy H1, push the limit and are astounding in terms of resolution and sharpness. of course they come with a ridiculously high price. however, there is a certain look and feel of film prints that digital has not yet accomplished, even at similar resolutions. if you view, side by side, prints made from high quality digital and those made from film, there is still an apparent qualitative difference that is hard to deny. if your goal is to produce large fine prints, film still seems to be the way to go. if you are shooting for publication, or for almost any other kind of application, digital has become the de facto standard already. fashion, sports, news, product, wedding - all digital now. landscape, fine art - still film-based. oddly, my own genre, architecture, has also gone largely digital, though at the highest levels (canon 1DSmkII, nikon D2x), and even in this field, where i have spent the past 20 years shooting a 4x5, LF is essentially dead. while i have owned a number of MF systems, my next camera will likely be a nikon D2x - i am almost appalled at saying that out loud, as i have always been a die-hard manual camera fanatic - i love intricate machines, and have revered the wonderful engineering and high quality of my nikon F2s, leicas, rolleiflexes, hasselblads, and of course my cambo, for years. i never even liked electronic shutters, and i f***ing hate autofocus and motordrives. i alwys liked camreas that didnt need a battery at all. i am wary of trying to figure out the complexities of a computer-based camera like the D2x, but i have come to accept that this is where the future of photography lies.

[Ed_Ingold]

It's not easy to answer this question objectively. At 4000 dpi, using a Nikon LS-8000, I get about 72 MP from a nominal 6x6 image, which is a good number for your purpose (silly banter with a non-believer). In TIFF format, that's about 225 Mb for an 8-bit/channel file. I use both a DSLR (Nikon D1x) and an Hasselblad with film - most often Fuji Reala.

 

For a more serious comparison, as for someone deciding which medium to use, the answer is more subjective. Not all pixels are the same. Those from a digital sensor have much more acutance than similar resolution by scanning film. Scanned, high-resolution images tend to have more acutance than optically enlarged images.

 

Based on my experience with a D1x, including images I've composed by combining several D1x shots, I estimate that a 16 to 22 Mp sensor would contain about the same useful information as a 6x6 film image.

 

To get the most out of an Hasselblad, you must use a fine-grained film, a tripod and good technique. Otherwise, it's hard to tell the difference. In an 8x10 enlargement, prints from my D1x appear nearly as sharp as from my Hasselblad. At 11x14 the Hasselblad is significantly better, etc.

 

Sharpness aside, film tends to have smoother gradients than digital, and a natural, dynamic compression at the heel and shoulder of the exposure/density curve.

[kelly_flanigan1]

which one(s) are digital or film?<BR><BR><img src="http://i4.photobucket.com/albums/y148/ektar/150mmRonar/Ronar150mmFull.jpg"><BR><BR><img src="http://i4.photobucket.com/albums/y148/ektar/150mmRonar/Ronar150mmWWVCLOCK.jpg"><BR><BR><img src="http://i4.photobucket.com/albums/y148/ektar/150mmRonar/Ronar150mmStones.jpg"><BR><img src="http://i4.photobucket.com/albums/y148/ektar/hockey/tripods-547.jpg"><BR><BR><img src="http://i4.photobucket.com/albums/y148/ektar/NZT%20dish/ntzdish640.jpg"><BR><BR><img src="http://i4.photobucket.com/albums/y148/ektar/NZT%20dish/ntzdishBase640.jpg"><BR><Br><BR><img src="http://i4.photobucket.com/albums/y148/ektar/biker/tripods-498.jpg">

[jan_brittenson]

30MP, or a 100MB 8-bit file sounds about right from a well-made CT even though I don't shoot 6x6. But 6x7 to me is about 40MP, and my 16MP 1Ds2 produces images with detail content somewhere between what I get from scanned 35mm and 645. Assuming a fine-grained film that scans well (like E100G/RAPF) of course.

[troyammons]

Depending on a lot of factors, like film, lens, scanner, between 20-30mp. I figure conservatively that clean film like E100G is worth about 5mp per square inch.

 

If you are shooting microfilm in a Rollei with MLU probably a good bit more.

[chrisaitken]

"What is the mega pixel equivalent of a 6cm square slide using Kodak E100 film"

 

Well, the simple answer is multiplication. Take the x-axis pixels of your 100 Mb image, and multily by the y-axis pixels. Thats your pixel count. Divide by 1'000'000 to get the MPixel count.

 

If I scan 35mm at 2400 ppi, I get 2400 pixels by 3600 pixels - about 8.6 Megapixels. This is in agreement with JNorman.

 

Note this is a simplified assumption, to be used against those number numpties (I've got more MPix, ergo mine is better).

[gaius1]

I get a 7000x7000 pixel image from 56x56mm Rollei negs, which I crop and print at 20x30" at 254 DPI - its uninterpolated resolution is 233 DPI, so that's not bad, a poster sized image that can be examined from 6 inches away at full quality. With a Lanczos interpolation, it would go much, much larger.

[david_holland1]

To add my two cents worth at a deep discount price, the question as I understand it is how do you equate a 6x6 with the commonly bandied MP counts for digital cameras (i.e. 6mp, 8mp, or 16mp). TIFF files are usually something less than 3 bytes per pixel, so the file size for a 6mp camera when converted to TIFF at the same resolution will be about 16mb or so, depending on the scene, and the 16mp will be about 45mb.

 

Since a 6x6 is more accurately 56mm square, or 2.2 inches (not quite the so-called 2 1/4 square), if it is scanned at 2000 pixels per inch it will produce 4400 by 4400 pixels, or 19.36mp and around 55 megabytes. If you weren't tuned into this difference, you might want to reread the posts which go back and forth between the two terms without noting the difference.

 

Scanned at 4,000 pixels per inch, you get 4 times the megapixels (about 77) and 4 times the file size (say 225 megabytes).

 

Since 4,000 seems fairly common these days, you should be able to stand your ground fairly easily with the answer that you are shooting a 77 megapixel camera.

 

The point, of course, and detailed above, is that pixels and megabytes aren't everything. If you ignore grain, color depth, second generation duplication, the vagaries of lens resolution on different capture aspects, plus the other factors mentioned above and the variety of others mentioned in the unbelievably tiresome expanse of digital vs. film posts, pixel count is only important when you enlarge enough to cause the image to suffer from pixelation (not to say enough that pixelation becomes visable to the naked eye). So it all boils down to whether you are going to enlarge enough to cause pixelation, assuming that you are comparing a film scan to a digital capture. Although that isn't your precise quesiton, unless you assume that you are going to scan the film, you're really comparing apples with avocados. All of the above posts, taken together, in my opinion make just that point.

[bruce_erickson1]

Thank you all for you replies, I learned a great deal, I think. If the issue ever comes up again, I will say 80. I too like chemistry, but am stuck with scanning with a sprintscan 120 for the time being. Right now I make small prints 8X8 now with a Kodak 8500, so I suppose I would do just as well with a digital camera. But J'refuse. Besides I like my camera too much, very simple all mechanical except for the meter -- a tlr Rolleiflex. Perhaps when film finally disappears or becomes too expensive (unlikely in my opinion) in 5 or 10 years, the price of a 22mp digital camera will be as affordable as a new Hassy V-system with normal lens was 5 or 10 years ago. I have to admit that before I got the Rolleiflex I was very tempted to go digital, but the Rolleiflex definitely reiforced my love of film. (I may still get that Rollei digi-miny or whatever they call it though!) Actually, my next step is to set up a darkroom.

[demetri_p.]

Depends on the film/lens combination. Assuming no scanning; just optical enlargements: test for 35mm show line pairs per mm resolution in the high 50s range for the best lenses and a sharp film and 40 or so for mediocre lenses.

 

We can expect a medium format lens/film combo resolving an excellent 40 lpmm to have a pixel reolution of (56*40*2)*(56*40*2)=20Mpix or a mediocre folder stopped down with maybe 25 lpmm giving a resolution of (56*25*2)*(56*25*2)=7.8 Mpix.

 

Res tests on digicams done on dpreview and imaging-resource show about 1800 lines height for 6 mpix models and about 2000 lines height for 8Mpix models (with the best lenses). This corresponds to (1800*1)*(1800*1.5)=4.8Mpix resolution for a 6Megapixel 3:2 sensor and (2000*1)*(2000*1.5)=6Mpix resolution for an 8 megapixel 3:2 sensor. The ratio of pixels on the chip to resolution is about 6/4.8=1.25 and 8/6=1.33.

 

Using the conversion above for final resolution vs. sensor pixels we get the MF camera with 40 lpmm giving 20*1.3=26 Megapixel digital sensor equivalent Mpixels and our little old folder with possible 30 lpmm yielding around 7.8*1.3=10 Megapixel digital equivalent.

 

The short and fast answer thus is between about 10 and 25 megapixels but someone asking a question like that will likely not care about all the caveats and conditions for that to be true. Like that this is for ideal optical enlargements. Like that optical enlargements are not usually ideal. Like, if you use a scanner, this is highly dependent on how good a scan you can get which is a difficult art in its own right. Like, the aesthetics of film grain which is emphasized by most scanning and so on...

[mattb1]

In 5 years you will not want to buy a 22mp back as it will seem as good as a 3mp does today. Most people forget that we are talking basically about computer technology, not so much about the traditional camera. 18 months is about the amount of time for a doubling of speed/memory, so 5 years will be about 6.6x the abilities that we have today. So, we may be approaching the capability of 6x6 film at that time. For us with somewhat more limited budgets I think we can expect to afford 3 to 4x what is available today, maybe replacing 645s.

 

35mm format will hit the wall at around 22mps, because even the best lens will not support more. Hopefully then we will see innovations such as Foveon, or other advances beyond Bayer. Better photo sites that remove the necessity of moire filters. And higher density photo sites, today less than 50% of the chip surface is photo sites. 48 bit color, LAB, and so on. All of these can only enable us to make better images.

 

I'd also love to see a modular system for the computer components similar to what we have for MF gear today. That is having the ability to plug-and-play CPUs, image chips, memory, system bus, and so on all in the same MF body. You could select the sensor with the photo site density you want, and then pick a CPU that would fulfill your needs for processing in camera. Much like you can select the lens, view finders, focusing screens, film backs, and such today. I think that kind of approach could breath some life back into the MF market.

 

Personally I cannot wait for digital to overtake film, but we are at least 5 years or more before it replaces MF. And I hope we are beyond Bayer by that time.

 

My personal view is about 50mps for 645, 90 for 6x7 and 110 for 6x9. But, I'm motivated by detail in the image. So, I would agree that 77 or 80mps would be a fair gauge of what your 6x6 can do. However, I agree that it takes great technique to achieve that degree of detail.

[bruce_erickson1]

In 5 years a 22mp back will seem like a 3mp now? Wow. Pre WW2 Leicas and other brands of camera made back then are just as useful today as then; one just needs the film and talent. Can one expect any digital camera made today to last 75 years and still be as useful? The upgrading in digital is endless. (And of course there is the need to get the latest upgrade in the imaging software, not to mention the computers and os software.) Digital almost seems like a step backward, but it sure has been and continues to be the motherload for the industry -- vastly more lucrative than film cameras. They are quite happy to see the demise of film.

[mawz]

Honestly, it's doubtful that resolution will climb too much higher. The reason: Physics. There's only so small you can make the sensor sites before you run into S/N issues. This is the reason for the poor high-ISO performance in high-megapixel P&S digitals. Not enough photons hitting each site. And as the sensor sites get smaller, you lose dynamic range.

 

So they will go up, but not all that much more. The next step is someone coming out with a Foveon-like sensor at a reasonable resolution.

[bruce_erickson1]

Adam, your answer gives me hope. But, what is this Foveon sensor I keep hearing about. Can you or someone give a simple explanation of it? How rugged is it? How long has it been in development? Is it perhaps one of those inventions that are superior but will never replace the current digital technology because it isn't superior enough? (Recall the MS-Intel computers vs. Mac, everyone said Mac's were superior.)

[simonpg]

There is lots of debate about the true mathematical calculation of megapixel equivalents from a scan.

 

But I think the basic calculation is: dpi scan resolution x vertical x dpi scan resolution x horizontal. So if you use a 4800 dpi scanner, it would be: 4800 (dpi)x2.25(inches)x4800x2.25 = 116,121,600 = 116MB file. If that is the same a megapixel, I'm not 100% sure. But it tells you the data size of what you're producing.

 

However different file formats produce different file sizes due to data compression. This is where I'm at a loss. My scans of 6x6 negs at 4800dpi scan resolution do produce open files of that sort of size. I have has 4800 dpi scans of 6x12cm trannies produce massive open files of something like 250MB.

 

Print resolution is different, but say to produce a full 300dpi print say 40x40 inches, you'd need a data file of 40x300x40x300 = 144,000,000 = 144MB.

[bruce_erickson1]

Simon, what scanner are you using that produces 4800 dpi scans?

[simonpg]

Done for me on Canon and Epson's latest flatbed scanners with full sets of 135 film and 120 roll film masks (6x4.5, 6x6, 6x7, 6x9, 6x12) and 4x5. I recall they were both released early this year in 4800x4800 resolution. Amateur Photographer (UK) did a detailed review a few months ago - highly regarded with the Canon winning the review by a bee's d..k (only in the area of smoother tonal gradation)!

 

My printer uses Epsons big colour plan type printer to do huge prints to a beautiful standard off these scans and Epson upgraded the flatbed scanner there to the new 4800dpi version.

 

What grabbed my attention was a sensational print of 24inches x 36inches from a Leica M shot.

[simonpg]

One thing I feel I should add is that there is no "common wisdom that 8MP eaquals 35mm film". While I don't shoot digi; one whing I am sure of is that equalling resolution quality of the best film (likely to be 50ISO trannie) is:

 

1. not simply about quantity of pixels - yes a large quantity is necessary, but there is more to it than simply quantity;

2. involves the "quality" of the imaging sensor technology - some 8MP sensors produce images to a higher quality than some 11MP sensors;

3. importantly involves the "depth/range" characteristic of the sensor's imaging.

 

Both quantity and quality of the pixels are vital to the final product.

 

If there is any "popular opinion" it may be that Canon's latest full frame DSLR (IDsMkII) is the first to fully challenge the imaging quality of the best trannie film available; but that is limited to 135 format.

 

What I have learned about digi imaging is that it involves very complex components and that somewhere along the line the debate is worthless unless the extreme end of the best quality imaging makes a visible difference to the user.