Lens capability on the M8

[mikemyers]

I'm wondering if anyone can provide a link to someone who's done a lot of testing of different brands of lenses on the M8.

I know that to get the best resolution, one needs to shoot in DNG format, not JPG. However, do the Leica, Zeiss, and Voigtlander lenses all provide "more" capability than the camera is capable of recording, or do you "lose" something if you don't use the very best lenses?

Years ago, with film, the limitation to "sharpness" seemed to be the lens, and had little to do with the camera body. Now that we're in digital, the M8 is limited by the 10-meg image sensor. If you try to capture more detail than the sensor can record, it's wasted effort.

From what I've read, all three lenses (Leica, Zeiss, Voigtlander) work splendidly on the M8, and I'm not sure if it's possible to "see" the difference between them. When it comes to a lens that I'm not going to use very often (such as an ultra-wide), the question is do I want to spend $4000 on the lens (which I could never afford) or can I get the same image using one of the lenses that sells for $400?

From what I think I've read so far, on the M8 I'm not going to be able to "see" the difference between lenses. They're all better than the capability of the camera to record detail.

Any opinions or advice on this?

If the Leica lens is one of those from 35 years ago, how would it compare with today's lenses?

 

[jaapv]

Each and every lens has its own "fingerprint" or "feel" which works out in considerable differences in the end result. It is one of the joys of this system to discover all the details. And don't worry, the M8 -and many other digital cameras are far more "sharp" and able to render subtle differences than film. Don't fall in the megapixel trap. It is a marketing ploy. Actually the latest sensors outresolve any lens by a fair margin. 10 Mp on a 1.33 sensor it quite close to the sweet spot in resolution.
For tests see Sean Reid's site: http://reidreviews.com/reidreviews/

Not free, but well worth the small contribution.

 

[Ben Z]

I'm wondering if anyone can provide a link to someone who's done a lot of testing of different brands of lenses on the M8.

I think lens test reviews are very entertaining. The free ones are well worth it. The ones you have to pay for...well, to each his own. Unless the tester has an optical bench and provides hard data, it all boils down to a subjective opinion, and I consider mine to be as good as anyone's. Of course you must understand that I don't subscribe to the [internet-forum-based] notion that being a pro photographer or even a good photographer has any relation to one's ability to judge the sharpness of a lens. After all, TTBOMK the guys who design the lenses are neither.

To answer your question, I use a multitude of different brand and age lenses on my M8. They do perhaps have their individual signitures and subtleties, but most of my photography at least doesn't bring them out in a glaring way. Shoot most Leica-mount lenses a few stops down from maximum, focused out past 5 ft, and not pointed at a bright light, and they all perform quite well.

Don't fall in the megapixel trap. It is a marketing ploy. Actually the latest sensors outresolve any lens by a fair margin. 10 Mp on a 1.33 sensor it quite close to the sweet spot in resolution.

But you're not cancelling your pre-order for the M9 are you? 😀

 

[mikemyers]

I know about the megapixel game, with all the manufacturers trying to claim more and more megs as a sales gimmick. There are many other things at least that important in getting a "good" image.

However, when I shoot 60 people as a group, maybe 40 or 50 feet away, I'd like to be able to see some detail in their faces. With my old 4 megapixel D2h, this was impossible. With my D2x and then D3, things are better, but still not as good as I'd like. I suppose if I were to get a D3x, maybe it would be slightly better, but that's more money than I can spend.

My impression so far is that the M8 does a great job of providing good image quality. I tried to take this group shot I mentioned above, but at the time my only wide-angle lens (35mm) didn't even come close to getting in everyone, so I used my Nikon.

I'll probably never get what I want - ideally, when I take a group shot like that, I'd like to be able to view one person in the finished photo and see them as a sharp, high-definition image, maybe to where I can even read some of the print were they to be holding a newspaper when I took the shot. I guess that's out for many more years, when we're routinely taking 100-megapixel photos at a "normal" setting.

Anyway, back to the M8. I read years ago, in a comparison of film vs. digital, that digital wouldn't be able to match the resolution of film until images were created by 20-megapixel cameras, with all the other factors equally as good. The M8, with the 1.33 crop factor, is approaching that level.

My own personal experience is that I'm producing images now with the D3 and M8 that are at least as good as what I used to get from my F4 film camera (using the same lenses as on the D3). I'm very pleased with the way things are going. For much of what I'm doing, I don't even shoot at the "large" image setting on the D3, as the smaller images have all the resolution I need.

Still, I'm always wondering how things would work out, were I to try to get a really, really good image of something, with enough resolution and sharpness that it would look great when printed in a very large size.

Every chain is no stronger than the weakest link - so in the above case, would the limitation be the capabilities of the sensor, or the lens? That's what I was getting at when I first asked this question.

 

[ferider]

10 Mp on a 1.33 sensor it quite close to the sweet spot in resolution.

The M8 sensor has about 150 pixel per mm, on either side. To resolve a line you need at least two pixels. To out-perform the sensor the lens needs to exceed around 75 lp/mm resolution. Many modern and even some 50 year old lenses are capable of that, some (like the ZM Biogon 25/2.8) exceed it by far, at least theoretically, on a tripod, etc. etc.

Of course there is more than resolution to a lens signature ...

Cheers,

Roland.

 

[JNewell]

Well, I've owned a lot of Leica lenses, old and new, and some third party lenses, over the last 35 years, and I think you can see differences, but not all newer lenses are better than older, and not all Leica lenses are better than third party. I will also say that broadly speaking I think you get what you pay for as long as you're willing to incorporate a very steep cost:benefit ratio. Example: the 25mm/2.8 ZM may be a little less adept than the 24mm/2.8 ASPH, but it costs, what, half as much? Those last few percentage points are hugely expensive; whether they're worth it or not is an individual call.

With the caveats that there is no substitute for your own tests, that every individual example may be different from whatever example someone else used or tested, and that digital and film may be different, you can look at:

1. Reidreviews.com - (subscription; I have no connection) Sean Reid tests all brands of RF lenses and gives very down to earth assessments with examples. Perfect? No. Exhaustive? No, but close. Worth the price of admission? IMO, YMMV, very definitely.

2. Puts, in his various loci. There is a free pdf on the LUSA website called something like "Leica Lenses: Their Soul and Secrets." It's a slightly older, shorter version of his "Leica Lens Compendium." And, he's got reviews (of his idiosyncratic sort) on his website.

That would get you started...I wish there were a Nikonlinks.com equivalent for Leica but I guess the market is not broad enough to engender the creation of the underlying content...

 

[MCTuomey]

in the end you use the lens and only you can determine its fitness for your purpose. that said, if you're pushing the very limits of enlargement with your files and want as much detail preserved as possible, then you must have the very best. consensus is probably leica, at the very margin of performance distinctions.

but i've got to say that you ought to consider another format for your large group portrait work. resolution, the more the better - that's your friend in your endeavor.

 

[JNewell]

I hate to talk fuzzy-speak, but beyond the MTF numbers, beyond all the other quantitative stuff, when the engineers hit it right, there is an intangible quality that is hard to describe but very real. I have been mostly a Leica and Nikon user over the years. A few Leica lenses hit that level. A few(er) Nikons do too.* Read Sean Reid's writeup, and look at the sample pictures, for the 24/3.4. There is mojo there, even when posted on the web. One of the most amazing lenses I've ever used is the Leica 100/2.8 AME R lens, but most R lenses are not as good as that one.

*Example: I owned and used a Tamron 17-50/2.8 v1. On a dare and because Amazon had a temporarily price anomaly, I bought a Nikon 17-55/2.8. I did the usual static tests - brick walls, bookcases, etc. Inconclusive. I turned the lenses on 3D subjects - totally different results. The Nikon was clearly outstanding where the Tamron was merely very good.

 

[mikemyers]

Thanks for all the information - I've now got a lot of new sources to check out.

I don't think I'm being clear in what I'm asking though. Let me give an example.

Let's say I've got an Excel spreadsheet for collecting data, and it displays information to two decimal places.

Let's say I've got access to two sources of data to input into my spreadsheet, one of which is providing data also to two decimal places, and one of which is much more expensive, and provides data to six decimal places.

No matter which source I use, when I input the data into my spreadsheet, which is only going to show up to two decimal places, that's the limit. Providing better data than that is a waste, as it won't be seen.



That's what I was asking here.

If the lenses are "better" than the M8, it doesn't matter which lens I use, as the limitation is the M8 itself.

On the other hand, if the M8 is "better" than the lenses, using a better lens will make a noticeable difference.



It's really a very simple question, but I don't know how to get the answer. For now, It really comes down to what's good enough for me......

(I've got the same situation with the D3 - I've got both Nikon and Sigma lenses, and in terms of resolution, the D3 seems to be the limiting factor, not my lenses.)

 

[MCTuomey]

mike, if it was that simple a question, it would have been answered long ago. it is not a "weakest link" question that can be answered by perusing mtf charts or comparing lp/mm data. or relying on other people to tell you.

you have to do it empirically, yourself, on subjects you shoot, on output you favor. or find a proxy that will correspond to your findings. Which will require you to do testing yourself anyway.

 

[JNewell]

You truly can't reduce this to numbers. If you wanted to, I guess that MTF results would be what you'd use, but a lens with high MTF numbers can still produce very hum-drum images, and vice versa. Reading other people's subjective opinions, looking at sample images (even if only on the web - better than nothing) and ultimately trying them yourself is the only way to really get a firm grip on what a lens can do.

 

[jaapv]

The M8 sensor has about 150 pixel per mm, on either side. To resolve a line you need at least two pixels. To out-perform the sensor the lens needs to exceed around 75 lp/mm resolution. Many modern and even some 50 year old lenses are capable of that, some (like the ZM Biogon 25/2.8) exceed it by far, at least theoretically, on a tripod, etc. etc.

Of course there is more than resolution to a lens signature ...

Cheers,

Roland.

Indeed there is more, Erwin Puts has a nice chapter on it in his lens book. It turns out film resolves about 40 Lp/mm once you analyze the data of the makers and that most lenses about match it. The real question is: resolution at which contrast?