ISO100|11-22@11mm|1/320s|f8,0|0,0EV
The title gives the dilemma I posed to myself. Recently I cleared out a large chunk of my photographic gear. I have realised that I mostly photograph people, either in reportage style, pretending to be a PJ, or a stage perfomance, pretending to be.... whatever. For this, I need reasonable coverage from wide to normal angle of view and a nice bright mid-tele (zoom?). For the second purpose, I am still thinking about to get an 50-200 (again) or bite the bullet and buy a 35-100 - but this is another story. For the first purpose, the 11-22mm f2,8-3,5 is more than good: fast, sharp, perfect coverage (for me) and I shall be greatly surprised if the new 12-60 f2,8-4 beated it in the overlapping region (if there is such a lens anyway). The 11-22 is the lens I have kept of all I had so far for its excellent quality. But using it on my E-1 always makes me feel that it cries for more resolution. An example is the following image:
ISO100|11-22@11mm|1/320s|f8,0|0,0EV
Although I do not think of myself as a victim of "megapixel illness" (an E-1 user for more than one and a half year now), I think that a little more than 5 megapixels would be nice. I know that one can print presentation-grade A3 images from ISO1600 E-1 files. Sure. But you are almost quadrupling the pixels in that case and this can be noticable, even is subsouciously. This is the obvious point, illustrated by the image above. Cropping is the other obvious and analogous point.
ISO1600|50-200@200mm|1/250s|f4,0|0,0EV
A perhaps not so obvious point is that if you shoot using high sensitivity with noisy cameras such as those of the E-System, you will definitely want to apply some sort of noise reduction. As there is no free lunch, this will cost you sharpness and detail, even if you use zero luminance smoothing. One thing is to use the best lens available. In a case where you can get away with the über-sharp 50-200 or the 50 macro (no to mention the 35-100 or the 150), there is no point to try kit lenses. They are fine in well lit scenes, but they simply don't deliver the sharpness that you can lose from and still get acceptable images. The image above is an example of a high ISO image that appeals to me and others a lot (save for the Mickey Mouse lighting, which is debatable). The key to the success here is accurate exposure, stability, sharp lens AND a subject with relative few details (and yes, there still remains some noise that could be dealt with - note to self: fix this). But clearly, if the camera can deliver sharper images, you retain more detail. Because of this I consider sharpness as critical for high quality low-light photography. The image below was taken with the 11-22 and I think it would greatly benefit from more detail:
ISO1600|11-22@11mm|1/10s|f11,0|0,0EV
We all "know" that the new bodies will have live-view among other features. I do not need LV and we already feel the pressure of the coming E-410 as E-400 prices go down steeply. More importantly, a body only version is available right now. The new small lenses are relatively (read: compared to the competition's offerings) nice, but are no match for the "Pro", not to mention the "Top-pro" line. This suggested that I leave my snobbish attitude behind and consider dropping the E-1 in favour the E-400. If end resolution is what I expect, and handling of the new baby is nice, I can live without splash-proofness and a magnesium body (for a while, at least). To find out if this is feasible, I went to a local shop called MegaPixel, where the guys were kind enough to let me play with the E-400. I took along my friend's 14-54, 50-200 to sidestep testing lenses and also his E-1 as a reference (mine is on loan to another friend).
Please note that this is not a test and not a review. It is only an illustrated log of my contemlations for others with similar inclinations. I know it has several flaws and I shall point them out due course! In spite of this, if you find any mistakes, inaccuracies or have suggestions, any feedback to the address given on the bottom is appreciated!
So how shall I compare the two bodies? One is (almost) 5 megapixel, the other is 10. One is a professional-grade body, the other is an OM-style smallish one. I have no intention to perform a feature by feature comparison, as that will be done by dpreview, among others. What I care about is the end result. Can the E-400 give me identical or better results than my E-1 at wide angle (which means 14mm on the 14-54 as I do not have the 11-22 at hand) and is it possible to use the 50-200 fully extended on it? The main point is probably handling and to see what happened to noise. Is it lower or higher than what we have seen on the E-1? Initial samples were quite encouraging, but is it feasible to pack twice the resultion on a phisically identical chip and improve noise at the same time? There is nothing to discuss if noise is lower in identical conditions. But if it is higher, can we say that there is a net benefit from the higher MP?
I tried to standardise the comparison as much as possible, but - as you shall see - there were some problems, which I think are minor. To achieve comparability, I have used the same lenses (14-54 and 50-200), tried to shoot the same subjects as close as I could in a hurry and without a tripod, and shot RAW, developing them in Studio 1.5.0 using the same settings (everything set to "0"), using the "High function" mode. And the first, perhaps the greatest caveat comes here: we know that a high-iso E-1 image is best developed in the "Advanced High Speed" mode to avoid artefacts . As far as I know, the "Advanced" modes are available only for the E-1, but the simple "High speed" is available for all. In Studio version 1.5.0 however, the "High speed" mode failed to read the E-400 files. What I am left with is therefore a suboptimal conversion. Note however that the difference is mostly that the "High speed" mode does not introduce the zig-zag patter the "High funcion" does, yielding marginally unsharper, but perhaps better de-noisable images. Obtaining a 8 bit tiff, I used Photoshop to crop and resize the images, which were finally sharpened using PhotoKit Sharpener (I have checked the effect of going for 16bit, but could not see one. Note: no post-processing was done in Photoshop). Noise reduction - if any - was done using the Noiseware Photoshop plugin. I would like to thank an anonymous graphic artist friend who made all this possible!
My real comparison of the two cameras would be to produce a 40*30cm print. Short of time, I could not do that, but to achive such a size, both files should be upsampled, the E-1 one a little more than tripled, the E-400 increased by a half. Comparing the two at the E-400 native resolution yields the same subjective result with less hassle. For this, I upsampled the E-1 images using Photoshop's bicubic resize command in one shot. There can be arguments about the validity of this method, but it is fast, simple and suggested by at least one stock agency Alamy. (I am not saying that this is the best agency, but I find their practice reasonable. They do not have megapixel requirements, but look at only a file that can give 40*30cm print @ 300dpi. If they find it good, you are accepted. E-1 files are OK. ;))
On comparison shots, the E-1 image is always on top and if you hover your mouse on the image, the E-400 version appears (the last one is the only one exception). Note that I shot the images hand-held, having to go away for a camera change. The end result is images which are not perfectly aligned. I kept this flaw to show a sense of realism and not to lose resolution.
Warning, I repeat: all the 100% crops from the E-1 are upsampled to the resolution of the E-400! (using PS's bicubic method)
Before the illustrated comparison, I look at ergonomics because it does affect image quality. And I have to state the obvious: the two cameras are really different! Although this is trivial, many people say that the E-400 is not good, because it does not have a big enough grip, a magnesium body, etc. I am quite astonished to hear this (it is almost as good an argument as saying that the E-3 or whatever can not be $2500 or more since you can get a D200 for that price), as we are comparing apples to oranges. The E-1 is ment to be a professional-grade body, that is: capable of taking of several thousands of images, withstand the elements, operated in gloves, etc. So there is no point to ask the question if the E-400 meets these requirement: clearly, it does not. It has only one command wheel, and pushing the modifier button for compensation is impossible in winter gloves. But if you are shooting without gloves (or you do not intend to take this camera to the cold) and not shooting extra quick events, pushing that extra button makes one no harm. The E-400 has a smaller and - so they say - less bright viewfinder, but this has not bothered me. I wear glasses, so I could not see the edges of a big finder anyway. I can not reliably focus the E-1 manually either, so I do not care for a bit of a reduction in the image. One could continue for long. If one wants a jack of all trade camera, one has to look at the rumored "G" enthusiast's series, not the "S"mall one. And those wanting a pro-spec camera should look clearly at the pro line or an alternative.
A valid comparison is that among the entry-level cameras of different makers. This is something I shall not do here either, but based on my very limited knowledge, the E-400 is one of the best, if not the best. Besided full manual control you have spot metering, DOF preview, to name just the most interesting features. Considering the positioning the E-400, its feature set is impressive, yet free from unnecessary baggage. Other than this, the controls are fine, positive, etc, although definitely not as universally usable as on a pro camera. I shall not digress.
If size matters however - and it often does if you want to live with your camera - the E-400 delivers. It is unobtrusive, and although its mirror is louder than that of the E-1, it is not a great problem. Note: I am no fan of photographing people candid, because my experience is that they do not like this. Being known to the subject, my aim is to photograph so that I do not disturb her or him. In that case, size is more important than sound, I think. Of course, only up to a limit. (When I went to photograph modern dance with my E-1 and 50 f2, the mirror was way softer than the Canon ones. I thought: cool! But then the 50mm lost focus and went all the way twice, emitting a strong, loud whirring sound. In the meantime, the Canon USM motors were working inaudible. After that I became much humbler.)
Te basis for a real comparison to me was to see whether the E-400 can be handled comparably to the E-1 and see its image quality. Apart from the limitations inherent in a small form factor, the E-400 is nice to handle with bigger lenses (I forsee some shake-related problems with the kit lenses, especiall the tele extended to 150mm, where it starts from f5.6. Luckily this is not my problem.). Oddly enough, key to this is the lack of a grip. While small grips are often difficult to grab, here you do not think that it is there, so you adapt. Note that my impression is based on having long, bony fingers - chunky hand might find the real estate a bit small. In the range provided by the 14-54, there had no problem whatsoever with using the E-400, and I expect that the situation is the same with the 11-22 too. These lenses are big enough to give stability through weight and size and something to hold on, yet they are light enough no to be difficult to keep steady. My question therefore reduced to see whether it can support a large lens, or rather: can I hold the lens steady almost by itself. This time the only available big lens around was the 50-200, so I took that one with me. It is no substitute for trying the 35-100, but given a little more than half the weight and the worse handling when extended, it might be a good approximation.
There is unfortunately no quick answer to this question, and maybe no final verdict either. One thing is for sure (altough, again, obvious): a grip does help in stabilizing the lens. I anticipate however that this difference is often compensated by the body, except for stressed situations. Again, this is a difference between pro (or pro-like) and casual use. Nevertheless, one should almost statistically test this difference to be sure. You might notice that below I use a portrait for comparison, an obvious candidate for comparing resolution. Yet, I provide no 100% crop. This is because I messed up and took only one comparable shot with the E-400 and that turned out unrealistically blurry - probably the effect of camera shake - while the E-1 shot is nice and sharp. I might even go back and repeat the exercise to see more clearly.
Update: I have received questions from current E-500 owners asking whether some functions, such as ISO and AF pont selection for example, are mapped to the four-way buttons on the back of the camera. I was sad to discover that this is not the case. Although this can be fixed with a firmware update, there seems to be no space left in the viewfinder to show the selected ISO setting - a real shame considering that there is plenty of ways to show it within the current viewfinder. This is expecially bothersome to me, as I often find myself shooting in manual mode, setting largest aperture and a "safe" sutter speed, adjusting ISO according to availability of light.
The first, and rather stupid-looking question is how apparent resolution changes by doubling the number of pixels in the cameras. The answer is easy: a lot.
ISO800|14-54@54mm|1/1250s|f3,5|-0,3EV
Basing the decision solely on resolution, the answer is a no-brainer: twice the pixel is twice the pixel. There are at least two other aspects apparent on the images, however. One is difference of colour, a white balance or colour interpretation issue, and the other is noise.
The E1 has an outer white balance sensor and many would swear that this is one of the secrets of the (some would argue) unique look of the E-1 images. Such sensor suggests something professional, although contemporary top of the line cameras do not have one (the Canon 1D used to have one, if I remember well, but it is gone now). On the above image, we have seen that the E-1 produced a slightly warmer image under the same conditions. Is this a rule or an exception? After several trials, I could not decide.
ISO400|50-200@200mm|1/640s|f3,5|0,0EV
The above portrait was taken outside the shop under (an overcast) daylight. Even to my untrained eyes, a cast is evident on the E-1 image (or on the E-400? who knows?). A fellow pointed me out a while ago that some Olympus cameras have a magenta cast. This one is however not magenta but blue. Thinking about the situation, it is in the shade under daylight, where light is bluest, in general, so the E-1 image might be actually "truer". In any case, the two ways of setting WB is different. Another examle below is the rendition of red. It is important to me as one of the "problems" I had with the E-500 is that its red was markedly harsher than that of the E-1, and I did not like that. This example does not show a great difference to me, just a slight. And yes, the E-1 red is paler again, but not much paler.
ISO400|50-200@200mm|1/640s|f3,5|0,0EV
ISO800|14-54@14mm|1/800s|f2,8|-1,0EV
The above image, shot from within the shop, shows the opposite of what we have seen on the portrait: here the E-1 image has more yellow (less blue). It might be premature to draw such a conclusion, but it appears that the E-1 tries to reproduce actual light conditions (blue outside, yellow inside), whereas the E-400 attempts to correct this by counteracting the actual lighting. It is up to Oly engineers to comment on this, I guess. And again, I hope my conclusion does not come from the novelty of the E-400 and some weird behaviour of Studio.
I babbled a lot about noise and resolution in the introduction. Now I take an image to see them both. Note that this is not a sharpness test and focus was set in the wrong way - on the statue on the top left. Next time I'll be more cautious. Also note that the image was taken in a way to preserve both highlight and shadow detail. The end result would clearly be different after post-processing.
ISO200|14-54@54mm|1/1250s|f5,6|0,0EV
The images taken with the two cameras show almost no difference, except for the shift due to camera change. The difference is apparent however, once we look at details. The difference introduced by the increase in resolution is dramatic. The E-400 image can show details such as wooden stripes on the blinds, details of the rooftop, and so on, which are rendered as a homogeneous block by the E-1. There is visibly more noise, however.
ISO200|14-54@54mm|1/1250s|f5,6|0,0EV
Running the files through Noiseware with 0% luminance and 100% colour filter changes the scene quite a bit: noise is gone from the E-400 image. There is a slight decrease of resolution, but the advantage of the E-400 is there. Hey, I was always on the opinion that Olympus should bundle Noiseware with its cameras. My opinion is unchanged.
ISO200|14-54@54mm|1/1250s|f5,6|0,0EV
So far we have looked at the midtone detail, but how about shadows, most prone to noise? There the difference between the two images is more marked. The E-1 has little noise and because there is only little fine detail, it does not suffer from low resolution that much. The black of the E-400 on the other hand is quite horrible. Filtering does eliminate chroma noise (which is trivial given the settings), but grain has even increased! Such dark areas are however quite small and can be made almost black during post processing if the grain is bothersome.
ISO200|14-54@54mm|1/1250s|f5,6|0,0EV
ISO200|14-54@54mm|1/1250s|f5,6|0,0EV
If you have not noted, I have to say that the above images were shot at ISO200! If you are still alive from the shock, you might be wondering - how about iso1600? That is exactly I am looking at with the image below. This is the same brick wall shot we have seen at the beginning, but pushed digitally a stop further, to almost ISO1600 (on the E-1, I found this equivalent to shooting ISO1600 up front). Here I am looking at shadow detail only, as I do not believe in "shooting without light" low light photography. When I use ISO1600, the main subject is always well-lit, well-exposed and separated from the background.
ISO800|14-54@54mm|1/1250s|f3,5|-0,3EV|+1,0EV digitally
We see two already familiar phenomena here: different colour and higher noise of the E-400. I think that filtering through Noiseware does a good job here, too. If you liked the E-1, you'll like the E-400 too.
ISO800|14-54@54mm|1/1250s|f3,5|-0,3EV|+1,0EV digitally
Can we correct for the colour difference in post-processing, increasing saturation? Yes, we can. E-1 tone is back, at least in such a simple color patch.
ISO800|14-54@54mm|1/1250s|f3,5|-0,3EV|+1,0EV digitally
Update: I have stated before that this is a lens test, so I am not going to make it one. Nevertheless, all the Zuiko Digital lenses I have tried so far seem to be up to the task of serving the 10MP sensor with sufficient amount of detail (this means the 14-54, the 50-200, the 11-22 - the 50mm is surely able to deliver). I have not only looked at 100% crops, but compared an A4 print from the E-1 and the E-400. From normal viewing distance, the difference does not jump at you. It is to be found - as you would expect - mainly in the fine details, but those little details add quite a bit to the perception of the image. You might imagine the difference as the effect using some fast lenses wide open as opposed to stopped down to its optimal performance.
Update: As I have sold my 50mm f2.0, I had to look at a near-alternative. In the past days, I have been playing with the OM 50mm f1.8. This seems to be a not too bad lens to use on the E-1 (resolution-wise at least), so I was interested in how it performs on the E-400. Firstly: its size fits the E-400 perfectly. It gives the OM feeling just fine. Secondly: here I could feel the effect of a smaller and dimmer viewfinder. Focussing the lens was just bit more difficult than it was on the E-1, but this did make a real difference: the number of misfocussed images was considerably higher and focussing itself took more effort (I used the lens @ f2.8, the maximum usable aperture suggested by Olympus). But this is not everything. While this lens produces sharp images on the E-1 at f2.8, it is apparent that the 10MP of the E-400 is too demanding at given apertures. Moral: anyone considering using legacy lenses on the new generation of Olympus bodies should seriously look at their performance, as they are most likely to show their weaknesses clearly.
As we have seen, the image quality of the E-400 is pretty comparable to that of the E-1, having all the ups and downs of the higher resolution. They are obviously different beasts, but I should say that the E-400 beats the E-1 overall. There was one phenomenon, which made me worry about the E-400. A blue cast appeared on the left hand side of the E-400 images taken with high sensitivity. An example is given below with the usual comparison to the E-1.
ISO1600|50-200@200mm|1/1250s|f3,5|0,0EV
The image is small, but I guess you see my point. This is a phenomenon that resembles the bright patch that appears on images generated by a CCD during long exposures at the point where the data is read (upper left corner in the case of the E-1). This is a visually similar, but by no means identical problem, as the exposure was quite short. Krisztián says that this might be a reflection on the CCD. I do not now the details of the sensor used in the E-400, but I really doubt this is a result of design. I shall try to mention this problem to Olympus Hungary and perhaps a response will come. In any case, I keep you posted.
BUT: in the meantime, it would be very useful if any of you E-400 owners could replicate the situation. This might be a flaw of a single unit and thus not even worth mentioning. (Mentioning such a problem can be thought as controversial. But given that we are talking about an commrecially available camera, I do not feel bad about mentioning it.)
Update: a handful of E-400 users have took the trouble and replicated my experiment and they all experienced the blue/magenta cast on the left of the images. Moreover, one of them is getting his third E-400 body and still obtains the same result. Conclusion: the problem belongs to the camera either by design or manufacturing. Either we shall see a mass-recall or this little beauty will not be usable at any reasonably high sensitivity setting. I have to say this is very sad...
The E-400 is a wonderful little camera. To me it is the digital OM many were waiting for. The mirror is smaller, the number of buttons is larger, but that is the way digital is today and probably shall be in the foreseeable future. The smaller view is really not an issue for me, given the accurate and reasonably fast auto focus - the end result is almost always an image that is sharper that used to be with analog. The E-400 is a camera of its own class and a perfect second camera in my opinion. Critical points, such as overall image quality and noise in particular is very good and can be handled well, respectively. Given the limitations of size, this tool appeals to me very much. A fellow E-1 owner remarked that he can not see any change in image quality after more than 3 years of development, set aside resolution. I think that this is perhaps the greatest compliment and E-1 owner can make to the E-400. If one needs the smallest SLR or the highest resolution Olympus right now, and does not go into the swamp or the desert, the E-400 is the right and a very good choice, given that it provides not only one, but both of these features.
Shall I trade my E-1 for it, then? Not sure. There are obvious differences, which come from the form factor: some functions do not have dedicated buttons, and even using clever engineering, this can not be overcome and thus be a hidrance (you have to press a modifier button, as there is only one command wheel and you can not select iso blindly, to note the two most important "missing" features). Another factor is stability. If I was to use the wider zooms, such as the 14-54, the 11-22 or (God forbids!), the 7-14 on it, I would buy it instantly (again: second body!). I am not sure about it use as a single body, though. Mounting a larger lens helps greatly in stabilizing the camera, but the lack of grip can be a problem in very critical applications. Note however, that these are quite rare and appear only in special settings of special lenses, such as the 50-200@300mm f3,5 or the 35-100@100mm f2. Read: with razor-thin depth of field only. In every other case, it is just fine, I think. [Pity I need my camera to perform in exactly such circumstances. Result: no decision yet for me.]
Update: As you might have already seen, the blue-patch-on-the-left problem seems to be unavoidable with the E-400 as of now (19/2/2007). This means that the E-400 is a definite no go for me. Sad; very sad.
Update: After almost exactly a year, I could not resist any more: I have bought the E-400 hardly used for third of its original price (and even paid more than I should have, I think). What can I say? I think I was the first to know about the weaknesses of this little camera. I know it has a relatively noisy CCD, it has no IS, the old AF module, etc. But after a year I know one other thing: the new MOS sensors from Panasonic show their own peculiar behaviour. iso 100 seems to be nonexistent on the E-3, being equivalent to iso 200 in terms of noise. Apart from being (of course) relative noisier at low iso in general, it also shows quite heavy deep-shadow noise. It is very compeitive at iso-s between 400 and 1000, but after that point, performance is less good than that of the competitors'. And a big disappointment: banding above iso 1250. It is not random and not rare: it's there and hard to get rid of. No matter what popular wisdom says, I see high iso performance as important, so I am disappointed. I know other brands have it and I know they have either inferior lenses or much higher prices. Still, unless pushed hard, I am not prepared to buy the E-3 right now. Maybe later, I will convince myself. Besides, I felt that even if I buy it, I shall need a low-iso machine. Not for action, but as a poor man's studio cam. And here I have it! So far, the E-400 shows no worse image quality than my E-1 with double the resolution and perhaps half the weight and the size. And it gives silky smooth rendition at iso100. Not the tool for the dark. But for well lit scenes, you can get perhaps the best A3 images using four-thirds at present.
Tell me what you think. Check out our review of the Olympus Top-pro lenses. Take a look at my photos.
Please note that although this site is not a commercial venture, there are costs of running it. To contribute, consdier visiting sites that you like (and only those) through our advertisments, or get Firefox through our link. This also helps deciding to continue with reviews, etc. or not. Thanks in advance!