Working Pro with M8

[keithwms]

Btw, my photo teacher told me to go get a 8x10 camera if I was so worried about image quality, but he wasnt being snide about it, he thinks I should be shooting higher quality imagery.

Haha, true. But if the gear impedes your shooting style then the gain in resolution and tilt & shift function is meaningless. So I would say get an RF-coupled 4x5 😉 You can easily pull 50+ meaningful megapixels off a 4x5 sheet!

Of course, how many megapixels is sufficient for big enlargements is a topic that has been hashed and rehashed and served with a side of bacon on the net. There is no answer for everybody. I won't even bother to tell you what I think. Try it for yourself!

I will offer one comment on megapixels and the importance thereof. I am sure you know this already... but anyway, the actual resolution goes as the square root of the mp count. So a sensor that has double the resolution of a 6mp camera is... a 24mp camera. In other words, the difference in actual resolution between 10 and 12 or 16 is pretty small. What does matter in terms of enlargeable detail (ultimately signal:noise) is the pixel size. That's my main beef with the m8, the ~1.3x crop factor, which gives the camera kinda soso noise performance compared to its dslr colleagues in the same price range. This matters because when you enlarge the image, you enlarge the noise along with it. If you denoise, you trade resolution to do so. Take a look at what a d3 can do at ISO 3200 or 6400 and you'll see what I am getting at.

 

[keithwms]

Yes, yes. But, is there somewhere on the web that someone has objectively compared M8 image quality to 5D image quality?

Maybe this helps...

http://www.dpreview.com/reviews/leicam8/

It's pretty much a head-to-head comparison of the two. The 5D does trump the m8 in terms of resolution. Noise performance is kinda similar up to a point. I am not saying that this is necessarily important though (see above), I am just pointing you to the most neutral comparison that I have seen between the m8 and it's dslr competitors. I think the m8 is an amazing piece, by the way, just not for me. The price point is way too high for me and I almost never use 35mm any more.

 

[hofrench@mac.co]

There are long and interesting threads here on printing with the M8 and on file quality from the camera:
http://www.l-camera-forum.com/leica-forum/leica-m8-forum/9022-30-x-40-inch-m8-prints.html

http://www.l-camera-forum.com/leica-forum/leica-m8-forum/34540-big-enlargements-m8.html

 

[jlw]

Of course, how many megapixels is sufficient for big enlargements is a topic that has been hashed and rehashed and served with a side of bacon on the net. There is no answer for everybody.

But the original question wasn't about printing in general, but specifically about the ability to produce two-page magazine spreads.

Graphic arts professionals (I'm one of 'em) have a fairly blunt mathematical way of looking at this; it depends on the final printed size of the reproduction and on the linescreen, the spacing of the "rosette" ink patterns that produce a color image in offset printing.

At typical newsstand-magazine page sizes, you'll seldom see a double-page spread that's more than about 17 inches wide, and a typical linescreen for a consumer magazine might be 150 lines per inch. That makes the entire spread about 2,500 rosettes wide.

You want each of those rosettes to represent more than one image pixel; that provides some "oversampling" that helps avoid the appearance of moiré patterns. The minimum factor that will do this is 1.414 (square root of 2) but most engravers would rather to bump it up to 1.5 to provide a little wiggle room.

2,500 rosettes x 1.5 = 3,750, so most magazine printers would like to see a file at least 3,750 pixels wide to print that 17-inch-wide spread (and actually they'd be happier with a bit more.)

The specs on DPReview say the M8's sensor yields a width of 3,970 active pixels -- so you'd be just on the hairy edge of what a magazine printer would consider the acceptable minimum, and that's assuming you don't need to crop the image.



Of course, if it were an absolutely once-in-a-lifetime shot they'd use it regardless of pixel count, and we know it probably would look fine. We also know that daily and weekly media routinely use lower-pixel-count images at large sizes (although probably at coarser linescreens) with no problems.

But if the production team has a choice of two otherwise-equally-good shots to use for a spread, and one of them has a comfortably high pixel count vs. a barely adequate pixel count for the other, the one with more pixels is probably going to win the tie-breaker.

(Now you know why those stock agencies are pushing for images with more pixels -- it can be the difference between making and not making a sale. Or to look at it another way, the higher the pixel count, the more ways an image can be used; so, more sales opportunities and higher value.)

The numbers start going up quickly when you begin looking at more demanding print jobs, such as high-end catalogs -- these may have larger page formats, glossier papers, and effective linescreens of 200lpi or even more. The people who shoot highly-detailed subjects, such as food, for these publications are likely the ones still using the 8x10 view cameras your instructor mentioned.

 

[TJV]

I'm sorry but this discussion is rediculous. No one is questioning Mikhael Subotzky of Magnum Photo's about his image quality. He uses a 10D.
The M8 isn't perfect but you'll like it, and if you like it you'll use it to make more pictures. End of story.

 

[MJones]

Great Explanation

Great Explanation

But the original question wasn't about printing in general, but specifically about the ability to produce two-page magazine spreads.

Graphic arts professionals (I'm one of 'em) have a fairly blunt mathematical way of looking at this; it depends on the final printed size of the reproduction and on the linescreen, the spacing of the "rosette" ink patterns that produce a color image in offset printing.

At typical newsstand-magazine page sizes, you'll seldom see a double-page spread that's more than about 17 inches wide, and a typical linescreen for a consumer magazine might be 150 lines per inch. That makes the entire spread about 2,500 rosettes wide.

You want each of those rosettes to represent more than one image pixel; that provides some "oversampling" that helps avoid the appearance of moiré patterns. The minimum factor that will do this is 1.414 (square root of 2) but most engravers would rather to bump it up to 1.5 to provide a little wiggle room.

2,500 rosettes x 1.5 = 3,750, so most magazine printers would like to see a file at least 3,750 pixels wide to print that 17-inch-wide spread (and actually they'd be happier with a bit more.)

The specs on DPReview say the M8's sensor yields a width of 3,970 active pixels -- so you'd be just on the hairy edge of what a magazine printer would consider the acceptable minimum, and that's assuming you don't need to crop the image.



Of course, if it were an absolutely once-in-a-lifetime shot they'd use it regardless of pixel count, and we know it probably would look fine. We also know that daily and weekly media routinely use lower-pixel-count images at large sizes (although probably at coarser linescreens) with no problems.

But if the production team has a choice of two otherwise-equally-good shots to use for a spread, and one of them has a comfortably high pixel count vs. a barely adequate pixel count for the other, the one with more pixels is probably going to win the tie-breaker.

(Now you know why those stock agencies are pushing for images with more pixels -- it can be the difference between making and not making a sale. Or to look at it another way, the higher the pixel count, the more ways an image can be used; so, more sales opportunities and higher value.)

The numbers start going up quickly when you begin looking at more demanding print jobs, such as high-end catalogs -- these may have larger page formats, glossier papers, and effective linescreens of 200lpi or even more. The people who shoot highly-detailed subjects, such as food, for these publications are likely the ones still using the 8x10 view cameras your instructor mentioned.

My graphic design backgorund is in different areas in interpretive signage where resolution is not as demanding, but this is a great explanation where the M8 file size would fall into a full spread for your typical magazine.

 

[CameronDavidson]

My primary RAW developer is Capture One or RAW Developer. I use ACR 4.2 with the M8 for some files, if I am not getting the look I want from Capture One or RD.

With Sharpening turned off, I have up-rezzed an M8 file with Adobe Camera Raw 4.2 and the file looks very good. Actually, much better than I expected. I then sharpen the file using Capture Sharpening in the PhotoKit Sharpening plug-in by Pixel Genius.

My stock images are with several of the large agencies and I've have not problem with the Leica M8 files being accepted into their libraries.

Hope this helps.

 

[Joe Mondello]

That's interesting. Could you post some links to information that indicates an M8 has better resolution than a 5D? I would really like to see that.

I was stating my personal opinion based both on anecdotal evidence I'd seen on the web PLUS my own comparisons of images shot with all 3 cameras (NOT testing).

I think the comparison link posted above that compares the M8 at 200% to the 5D at 177% is interesting but fundamentally flawed.

I made a similar comparison once between my D200 and M8 and came up with astoundingly similar results favoring the M8 . . . BUT I know that the differences are really not THAT pronounced. So I think there is something wrong with his technique.

 

[tomasis]

avotius, if you take better pictures with RF, take this by all means. It is not about technical perfomance but handling ability. Don't waste your time on wondering which is better camera. Just decide which type suits you. DSLR or DRF. The end.

My shooting improved a lot when I did sell everything of DSLR, lucky me 🙂

 

[HAnkg]

But the original question wasn't about printing in general, but specifically about the ability to produce two-page magazine spreads.

Graphic arts professionals (I'm one of 'em) have a fairly blunt mathematical way of looking at this; it depends on the final printed size of the reproduction and on the linescreen, the spacing of the "rosette" ink patterns that produce a color image in offset printing.

Rules of thumb like this or doubling the line screen to get the reqiured resolution are not that useful. If you are shooting herring bone jackets and textiles with geometric patterns you will need much higher resolution then 10MP and the printer will need to output images at much higher then 300DPI to avoid moires -if the shot is a portrait with a softer look you could get equal quality from a 6MP file upsampled.

The 10MP M8 properly processed can make CMYK 11x17 spreads equal to anything out there as everything is going to get crushed into the anemic CMYK color space and pushed through that halftone screen. The biggest problem is the lack of flexibility in cropping. If the art director decides to crop away 70% of the image 10MP isn't going to cut it.

When you are looking for an image for an ad campaign you might need portrait and landscape versions for 1 page and 2 page spread ads as well as larger sizes for point of purchase displays and outdoor advertising. From the agencies point of view bigger files provide more flexibility in cropping and future applications and as they know MF backs are available up to 39MP and 35mm up to 22MP all other things being equal why not have the bigger files.