Satory Лучше вряд ли, класс не тот, но факт, что при увеличении количества пукселей шумы на уровне Е-3 - это радует.
я вот посмотрел кадры по последней ссылке и не вижу, чтобы матрица у Г1 была лучше, чем в Е3. посмотрите в левый верхний угол снимков, где изображено белое здание. у Г1 и Л10 оно явно "пересвечено" чтоли по сравнению с Е3, деталей не хватает...
Acid_Reptile Джипег есть джипег, каждая камера выдает его по разному в зависимости от своего интеллекта. Да и в целом оно так и должно быть - преимущество Е-3 - чуть шире ДД, чуть приятнее картинка, более тонкая обработка шумов. Класс сказывается
Acid_Reptile elevar Если экспозамер не справился и цвета выбиты, то это вина экспозамера камеры, а не матрицы. Конечно, имея матрицу с меньшим динамическим диапазоном, труднее избежать пересветов, но данный отдельно взятый кадр явно не показатель. Очевидно, при использовании той-же матрицы в E30 - будет другой процессор, другой замер и другие результаты. Лучше или хуже - увидим.
Aperturer А разве у Г1 не матричный замер? :huh: По остальным частям снимка видно(где тени и штора), что картинка у всех камер практически идентичная, соответственно можно сделать вывод, что у Г1 ДД не такой уж и широкий, на уровне Л10. Никакое сравнение с Е3 быть не может.
Acid_Reptile А по мне так картинка Г1 заметно светлее чем у Е3 :) Кроме того присмотритесь - у Е3 деталей в тенях меньше :) Так что дело тут мне кажется всетаки в немного разно экспозамере и не более. А вот, что шумодав кушает детали у Г1 - это да, правда и зерно шума при этом не такое навязчивое.
Acid_Reptile: Да, заметная глазу потеря деталей здания на заднем плане за окном при одинаковом, примерно, переднем плане. Констатируя этот факт делаю вывод, что ДД у Г1 меньше и картинка Е3 приятнее.
Т.е. Вы думаете что цифро-аналоговые преобразования происходят в самой матрице? :D Да и разговор был про экспозамер. Процессор не подбирает премя экспонирования и не влияет таким образом на пересветы?
Зато я про него и говорил, только вы удалили из цитаты. Другой процессор - другие результаты работы экспозамера. В данном случае конечно процессор не в вакууме, а со своим фирмваре и АЦП (который как раз очень многое значит для деталей в светах и тенях). Так что с той-же матрицей результаты будут теми-же, только если вся связка полностью перекочует из камеры в камеру, но вроде предпосылок для этого пока не видно.
Появился первый владелец адаптера м4/3 на 4/3. Поскольку впечатлениями он делится на сайте, на котором требуется регистрация, привожу его текст здесь (в конце - ссылка на тестовый снимок с Зуйкой 50/2.0 ;) "I got one of the 4/3 adapters off of eBay (seller i-qju from Japan). (By the way, if you want one and you're in the US, be sure to check the live.com cashback; currently you can save 25%.) I ordered the adapter on the 7th, got a shipping notice with the tracking number on the 9th, and received the adapter on the 12th. The adapter initially didn't strike me as overly sturdy. The two mounts look like they're both steel, but I can't tell whether there's any supporting steel between them. I feel comfortable hanging the ZD 50mm on it, and the Leicasonic 25mm also seems solidly supported even without holding the lens. The camera is already getting a pretty front-heavy with the 25mm though. I even tried my recently-acquired 7-14mm. While objectively it seemed sturdy, I didn't feel at all comfortable with such a heavy weight hanging off the front. From a practical standpoint, you have to hold the lens anyway in that case, since otherwise the whole thing is completely unbalanced. When you put on one of the 4/3 lenses that doesn't support AF, the camera displays a message telling you to switch to MF. The only lens that I have which will AF is the Leicasonic. For that, it shows a message to switch to AFS if you happen to have the camera on AFC. For regular 4/3 lenses, the magnified MF assist works just like with the kit lens. The only issue is that you realize how helpful the IS in the kit lens is in terms of stabilizing the view while focusing. That said, with my eye to the camera, the jitter doesn't cause any problems when focusing the ZD 50mm. I tried the 50-200mm at 100mm and it's still not too bad, but at 150mm and up, I was struggling. Especially at 10x, things are just shaking around so much. I did manage at 200mm though, so it's not totally impossible. Of course, some people will be steadier than me, and if you're using a tripod or monopod, it will be much better. Out of curiosity, I also tried turning off MF assist. At 200mm, I actually did well. I doubt the viewfinder can really show critical focus unmagnified, but you can definitely tell unfocused, so "too close, too far, to the middle" seems to work pretty well. Anyway, here's hoping that Olympus manages to cram in-body stabilization into their micro-4/3 cameras... AF with the 25mm makes me realize just how much work must have gone into the dedicated micro-4/3 lenses. The 25mm is both significantly slower and significantly noisier than AF with the kit lens. When it focuses, it zips back and forth two or three times before finally picking its spot. Probably the kit lens does the same thing, but it's much quieter and less noticeable. I'd guess AF time is around 3/4 to 1 second. I also quickly got the idea to turn off pre-AF, since otherwise the lens noisily dances around on its own occasionally in a somewhat disconcerting manner. Any hopes that existing 4/3 lenses will ever work as nicely as the micro-4/3 lenses were completely stomped by this initial experience. Even if they AF, they're likely to be considerably worse for photographing anything that moves, and the experience of using them is nowhere near as pleasant. The 25mm is the main lens I use indoors on the E-1, usually at ISO 400. That gives me a shutter speed of around 1/60 to 1/80 at an aperture of f/1.4 to f/1.6. My main subject is Daniel, our 4-year-old, as he bounces around the house. I'm lucky to get 30% of the pictures usably focused, and that's when he's in an especially placid mood. Even when he's holding relatively still, the E-1 is just not an AF speed demon. It dances around a bit as it tries to home in on focus at f/1.4, and sometimes it's just off enough that the picture isn't quite acceptably focused. And I lose time in focus-recompose, which causes further misses as Daniel jumps around. (I do lean on the continuous drive, so at least some misfocused pictures happen on subsequent pictures after he moves. On the other hand, sometimes the focus is off initially, and he moves into a position that results in better focus.) To compare, I set up the G1 in the same way (ISO 400, program mode, continuous drive). (Probably with the G1 it would make more sense to use shutter priority at around 1/100 and let the aperture and auto-ISO take care of getting the right exposure. I didn't think about that until later though.) I like the face-detecting AF when photographing Daniel, since I don't have to do the focus-recompose dance constantly. Based on the sluggish response when first I first tried the 25mm, I wasn't expecting much. I filled up the memory card as he was playing in the living room, about 400 pictures. When I checked them, I found a whopping 65% keepers. I was just totally blown away by how much better it was than the E-1. Many of the ones that had to be tossed were suffering from motion blur too, so it's not like the G1 really missed the focus. It was a jaw-dropping performance. I'm completely sold on the ability of CDAF to critically focus a fast lens, and absolutely drooling at the upcoming dedicated 20mm f/1.7. If they eventually make another fast micro-4/3 prime in the 50-100mm range, I'll be in photographic heaven. So the overall story on adapted 4/3 lenses is a mixed one. Certainly the lenses are useable, and for mostly static subjects, not at all problematic. The smaller lenses fit naturally with the camera and don't seem at all out-of-place, though obviously you lose maximal compactness. On the other hand, the only lens that I have which autofocuses is very noisy, and not nearly as responsive as the kit lens is. Using regular 4/3 lenses on a micro-4/3 body probably won't ever be as pleasurable as using them on a normal 4/3 body. Still, I was very impressed by the G1's ability to make use of the 25mm with a moving target. I can hardly believe that the AF really performed so spectacularly. In bodes very well for the future if we can get more dedicated fast micro-4/3 lenses. Update on Thursday, November 13, 2008 at 04:13PM by David Long I posted one test image from this morning using the adpater. Tripod, ISO 100, ZD 50mm @ f/5.6, self timer, focus via 10x live view, etc. Basically the most strenuous resolution test that I can give the sensor. Only in-camera JPEG though (sorry, I run Linux and don't have raw support installed yet). I'm impressed with the amount of detail, though since I'm used to the E-1, almost anything is a big step up in that regard :-). There's a little bit of moire here and there, so presumably Panasonic has been fairly agressive in toning down the AA filter. http://davidlong.org/sal/album.php?dir=pho...bum034/01g1test
Не знаю как это устроено у панасоников, но во многих cmos сенсорах зеркалок преобразование происходит непосредственно на кристалле матрицы.
Очевидно, что "Leicasonic" это не "оликовский блинчик", а панасониковский 25мм F1.4 с надписью лейка.
И Leica 25 f/1.4 и ZD 25 f/2.8 давно уже выпущены :) Автор статьи пробовал на камере первый из них и он фокусируется медленнее чем китовые стёкла и издаёт больше шума при этом. Надо бы опубликовать перевод тут...