majid

I was so disgusted with my defective M8 I returned it and got a R-D1 and a 5D for essentially the same price (after rebates). With the 5D's high-ISO noise advantage over the M8, you don't really have the traditional rangefinder edge for low-light photography.

<p>

Image quality is perfectly fine in daylight, but terrible at ISO 1250 and higher.

<p>

For more info, consult my <a href="http://www.majid.info/mylos/weblog/2006/11/23-1.html">review of sorts</a>. Of course, I may have been spectacularly unlucky with my copy, but the poor QA is inexcusable.

Don't get a camera bag. They are not very versatile and are a magnet for thieves. A messenger bag will fit your camera and lens with ease, plus a few sundries like books or magazines, and be much more discreet to boot.

If you are using a light tent, the white background is throwing off your exposure. You could use an incident light meter, but it's probably easier to dial in an exposure compensation. With a little trial and error (and bracketing), you should soon know how much is required.

The 12mm is definitely rectilinear, not a fisheye. I have <a href="http://www.photo.net/photodb/folder?folder_id=323427">a few samples</a> on a full-frame Leica film camera. I ended up selling mine because it was simply too wide, but then again, 35mm is my idea of extreme wide-angle...<p>

 

The lens does not have rangefinder coupling, so you would need to use the engraved distance scale to guesstimate the focus. With the depth of field it typically has, that's usually a non-issue. With the M8's 1.3x multiplier, it's still a 15mm lens, and any tilt to the lens will lead to massively out of whack perspective, be sure you understand what you are signing up to if you buy the U-W Heliar.

The Magnuson-Moss act does not allow sellers of consumer goods to void a warranty simply for using compatible third-party parts, if they are unrelated to the problem. That said, if the damage is indeed due to the batteries, they are off the hook.

Why don't you start with the free <a href="http://www.photo.net/learn/">Learning</a> section on Photo.net before you spend money on dead trees? It's pretty well written, and definitely more witty than any book out there, and thus more memorable.

<p>

The Internet is usually a better resource than most books, which have to be camera-independent, and are usually obsolete by the time they are printed. One example is <a href="http://photonotes.org/articles/eos-flash/">NK Guy's Canon Flash tutorial</a>, which is far superior to anything available in book form. Online articles usually have less useless filler material added, unlike books that need to be bulked up for marketing reasons to make them more impressive and sellable.

<p>

Also check out Apple's <a href="http://manuals.info.apple.com/en/Aperture_Photography_Fundamentals.pdf">digital photo fundametals manual</a> (PDF) for Aperture, which is extremely lucidly written and a good intro to topics like RAW conversion that will let you get the most out of your camera.

<p>Where books are useful is in more general topics like composition or criticism, and many of these are found in the visual arts section of your bookstore rather than photography instruction. If you want a very technical book on all aspects of photography, I highly recommend the <a href="http://www.majid.info/mylos/weblog/2002/03/26-2.html">Ilford manual of photography</a>, which is a book that's now reached its third century of printing! It won't help you take better pictures, though.

<p>

Consider also attending a photo class at your local taxpayer-funded community college. Many pros offer one-day or weekend seminars that can be inexpensive (check out ads in the like of American Photo or the excellent Digital Photo Pro magazine). Personalized human feedback will always be preferable to abstract instructions in a book. I would suggest the Photo.net galleries, but they seem to have devolved into adolescent popularity contests with a premium to the conventional.

If he has the 16-35 L, I am assuming the 18-55 does not get much use any more... You could get a lens like the 50mm f/1.4 or the 85mm f/1.8, but someone who paid $1400 for a L series lens is probably picky about his lenses. He may be planning to get the 50mm f/1.2 or one of the four 70-200mm L zoom, so your purchase might be wasted.

 

The flash is certainly a good option, but what I would recommend is you get a color calibrator for his PC monitor if he doesn't have one already. Monitor usually have color casts, and if you are editing photos on an uncalibrated monitor, you will correct for the cast and end up with over-corrected images in the other direction. A calibrator is a little gizmo that looks like a hockey puck dangling from a cord, you put it in front of the monitor, and it goes through the colors and measures them so you can accurately compensate for any cast.

 

The decent monitor calibrators (avoid the Pantone Huey!) cost between $150 and $250 depending on how sophisticated the software you want. The good models are the Monaco OPTIX XR, the GretagMacbeth EyeOne display and the ColorVision Spyder2. Each of these is available in an entry-level or a pro version, the hardware is the same, the only difference is in how much control the software gives you. Unless your husband is a control freak, the base package should be sufficient.

This site has a <a href="http://www.photo.net/equipment/canon/">pretty good tutorial</a> on the Canon system. I would suggest you start there.

<p>

Image stabilized lenses are great for handholding without camera shake, but they don't make toddlers run around slower, and are not much help in freezing their motion. For that, you need a fast lens. The 50mm f/1.8 (or the slightly better f/1.4) make fine portrait lenses, as does the 85mm f/1.8, which Phil Greenspun explicitly recommends as good for families with kids (I am assuming the stellar 85mm f/1.2L and 135mm f/2L are out of your budget, but if they weren't, my recommendation would be the 35mm f/1.4L, which is near permanently mounted on my Rebel XT).

Either a Goseen Digipro F for a good, easy to use incident and reflective meter, or a Pentax digital spotmeter if he already has a basic meter. Sekonics are nice on paper but too many functions crammed into a single unit means you have to fiddle with a lot of buttons to get anything done, and that's not good trait in a tool.

Oh, and for Ken Ng: the Cayenne is a monstrosity, like every other SUV, and Porsche could perfectly well have produced something like the <a href="http://www.teslamotors.com/">Tesla</a> if they had been truly forward-looking.

Marc, The JPEG is for people who can't read DNGs, but I have also posted a link to the actual DNGs themselves: <a href="http://www.majid.info/mylos/weblog/2006/11/L1020007.DNG">DNG 1</a>, <a href="http://www.majid.info/mylos/weblog/2006/11/L1020017.DNG">DNG 2</a>. I was able to get the same results as you, but only by using curves - no amount of switching WB settings in DCR was able to do it. The results are still unacceptable because you end up with weird green highlights everywhere (the wooden armoire in the background is all brown in real life, as are the candlesticks, so I wouldn't count that as a recovery.

<p>

That said, the IR sensitivity is a known issue with a work-around, so there is no point in rehashing it, or even testing without the IR filter, except to demonstrate that the IR filter is an absolute must-have for indoor shooting.

<p>

My main point with this example was that the problems indoors are not limited to black synthetics, but apply across the entire gamut. And also to provide criticism of the camera from someone who actually owned one, even if only for a few days, rather than someone from the peanut gallery indulging in schadenfreude. Also to vent some of my frustration, of course. If there had been any other M8 bodies available for an exchange, I might not have been in such a hurry to ask for a refund.

<p>

I saw your <a href="http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00Iv2D">other post</a>, and the noise levels do seem fine. It is quite possible my M8 was a lemon in more ways than one. Given the, ahem, spotty quality control exhibited so far, there could be a wide variation in performance from one individual sensor to another.

<p>

I might even have a look again at the M8, but not before another year has elapsed, the teething problems have been resolved, and the Phil Askey review is published, as I have zero confidence in mindless boosters like Michael Reichmann (as if he of the "The Canon D30 offers better quality than Velvia" fame deserved any to begin with). In the meantime, my MP still works as fine as ever, even if E-6 processing is getting tougher to find.

Alex, I held off on buying a R-D1 for 2 years, waiting for the M8. I bought a M8 last Saturday, returned it on Wednesday (it was defective, beyond the known IR issue) and bought a R-D1 instead. For more details, see <a href="http://www.majid.info/mylos/weblog/2006/11/23-1.html">my article</a>, with samples. Perhaps I was just spectacularly unlucky, but when combined with the IR fiasco, I ended up simply disgusted with Leica.

Michael "what IR problem?" Reichmann's opinions carry little weight. I put up original DNGs up for inspection. Perhaps your tolerance for noise is higher than mine. Perhaps my M8 was a particularly defective copy (but I saw the same noise levels on a Leica rep's demo M8, so I doubt it).

 

Keep in mind that with digital, noise is most visible in shadows, unlike film where grain is seen in midtones highlights. If you shoot scenes where there is enough light to make the entire frame moderately lit, it won't be as visible as with really dark scenes shot at EV 1 where by necessity most of the frame is dark with very visible noise patterns.

The thin IR filter was Kodak's design decision, not Leica's. The decision to release the camera with either inadequate testing, or inadequate documentation of the problem, was all Leica's. I didn't mind the IR filter issue so much, even though I never use UV filters on my lenses because they degrade image quality.

<p>

For whatever it's worth, I bought a M8, used it, realized it was defective (an entire column of dead pixels, the kind of glaring flaw that should have caught by the flimsiest of QA processes), that the ISO 1250 noise performance was abysmal in the shadows, that the RF patch was vertically misaligned by about 2%, and that the IR/magenta issue was not limited to black synthetic fabrics but made every single indoor picture I took last Sunday irrecoverably magenta-ified.

<p>

The M8 went out, was replaced with a R-D1.

<p>

For more details, read my <a href="http://www.majid.info/mylos/weblog/2006/11/23-1.html">write-up</a>.

I need ISO 1250 because I shoot often in available darkness, in places where you need to go down to 1/15 or 1/30 even at f/1.4 and ISO 1600 film. Even then, most of the scene is in shadows (where noise is more readily apparent). The only way you can do this is with a tripod or a rangefinder. Others may choose not to work in such extreme (non)-lighting.

 

My Canon DSLR system simply can't deal with those situations, which represent close to half my shooting, and I had already been impressed by the R-D1's low-light performance, despite using the same noisy Sony CCD used in the Nikon D70. I decided to pass on it and wait for what Leica would bring out, but the end result was underwhelming.

 

In any case, I am not suffering, since I ditched my M8 for a R-D1. Any "suffering" would be very relative, in any case, we pampered westerners barely understand the concept any more, even on Thanksgiving.

 

I wanted to believe just as much as anyone, although the use of a Kodak sensor should have rung alarm bells (in my experience, every single product Kodak makes is utter dreck).

 

Leica's rush to release the camera and their handling of the IR issue was terribly bungled, but the root problem is Kodak's use of a BS7 borosilicate glass from Kyocera that lets in too much (10%) IR in the 1000 to 1200 nanometer range (oddly, it blocks IR fine in the 700 to 1000nm range). The reason why I returned my camera in disgust is the poor quality control evidenced by the dead pixels.

 

Part of it certainly has to do with the unrealistically raised expectations from early reviewers. The M8 does not feel entirely like a M, no matter what over-enthusiastic reviewers may have said. The shutter noise is incredibly disconcerting, for instance. I used to complain about the rewind knob on the R-D1 and how it is an affectation, but the shutter click and manual rewind on it is much less objectionable than the M8's loud motorized whine.

 

If you are in the market for a M8, I would strongly advise you to wait for a local Leica Demo Day so you can hold the camera in your hands and take low-key sample shots at ISO 1250. Once you receive your camera, check the rangefinder alignment, shoot pictures at low and high ISO and check them for defective rows or columns.

Just to be clear, I'm glad you are happy with your M8 and that it meets your needs. When I do available light photography, I use my MP with either a 50mm Summilux ASPH or a Noctilux at full aperture, loaded with Fuji Neopan 1600, Fuji Natura S 1600 or Fuji Provia 400 pushed 2 stops, because I need every last drop of light.

 

Unfortunately, the M8 does not meet my needs, and handheld low-light photography is the one place a digital rangefinder is invaluable (for daylight shots, the Canon 35mm f/1.4L is plenty sharp).

Just for laughs, the M8 at ISO 2500.

All three conversions were done in PS CS2 with the same noise reduction settings, BTW.<div></div>

And a Canon Rebel XT at ISO 1600<div></div>

For comparison purposes, a 500x500 crop from a R-D1 at ISO 1600<div></div>

ISO 640 hardly qualifies as high ISO. I'm talking of ISO 1250, which is quite a bit worse than ISO 1600 on the R-D1. ISO 2500 is a complete joke.

<p>

Here is an example at ISO 1250: <a href="http://www.majid.info/mylos/weblog/2006/11/L1020007.jpg">L1020007.jpg</a>. Look at the vertical line about 3/4 of the way to the right. Also, note the horizontal streaking that emanates from the lights.

<p>

Now it's quite possible my sample was defective in more ways than simply bad pixels, but the demo M8 the Leica rep had at the Calumet photo 9th anniversary in SF had the same noise levels.<div></div>

I bought an M8 last Saturday. Lousy ISO performance, vertical misalignment of the rangefinder, low-light photos with a massive magenta cast (and not just on black synthetics). To top it all off, an entire column of dead pixels. I could live with the IR filters, but certainly not with noise levels worse than a $400 Fuji F30, massive smearing and blooming of highlights across shadows, and spotty quality control.

<p>

I returned it yesterday and got a R-D1 instead. So, yes, as far as I am concerned, too little too late. I wrote up my <a href="http://www.majid.info/mylos/weblog/2006/11/23-1.html">impressions</a>.

<p>

If you shoot only in full daylight, this is a fine camera with excellent detail, that lives up to the lenses. I didn't buy a Summilux and Noctilux to shoot in broad daylight, however. Come to think of it, the low magnification is a give-away: this is a camera designed for wide-angle junkies, thus not fast lens/available light shooters.

<p>

I am amazed by the amount of cognitive dissonance it must take for the see-no-evil camp to sustain their denial.

There's an additional lens that makes the viewfinder slightly less bright, but reduces the likelihood of the rangefinder patch flaring out of visibility (that can easily happen if your eye is not perfectly aligned). I paid $275 to have my M6TTL upgraded in June 2004.

 

If you seldom experience the rangefinder patch disappearing, don't bother. If you do, it might be a worthwhile upgrade.

The 35mm frame line is barely visible as it bumps aganst the edges of the 0.85 field of view, so, no, the magnifier is counter-indicated with 35mm lenses. I have no problem focusing my Noctilux with the 0.85 finder (or my old 0.72 for that matter). The 90mm would benefit (barely), but it's not required.

 

The magnifier is really meant to be used with the 0.58 and the 0.72 finders, it is overkill for the 0.85

I would suggest you read this tutorial:

<p>

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

Your assumption that modularity means lower cost is far from proven.

 

Look at the Digital Modul R, which has a nearly identical sensor to the M8 and presumably the same electronics package, and provides that modularity to Leica R users.

 

The DMR by itself costs $6000, so a modular M would cost at least $9000 with body. In addition, the result would be a bulky monstrosity like the DMR, an antithesis of the lightweight compact M spirit.

 

Making a modular M that is designed to be replaced at the factory rather than by the user would probably cost less, but the point still is that building this kind of modularity has real costs in terms both financial and how design compromises would degrade the functionality of the camera.

 

Leica is a small company with limited engineering resources. Ask yourself why Canon or Nikon don't offer this service to EOS 1Ds or D1X users, whose cameras are significantly more expensive than the M8?