Sounds like your G1 spec is with no lens and the other includes a lens. They are essentially the same size cameras.
I already veered the thread off-topic from peters review.
Sounds like your G1 spec is with no lens and the other includes a lens. They are essentially the same size cameras.
I already veered the thread off-topic from peters review.
Both came from the Panasonic site - but I take your word for it.
An odd mistake to make - but once made it will stick as all the reviews will simply copy the 'official' specs.
I agree that a one man band, a photo journalist and everyone wanting a dual function but relatively competent camera can like the GHx though.
My question here; is eoshd.com known to be a quality review site, or is it perhaps known to be a so-so place? Or is it the same as with still photography that you need to read several reviews to get nearly the whole picture?
After reading a few reviews and comparisons of the video capabilities of different candidates, my impression is that the results go in all different directions. I have a feeling that it's partly a question of what people are looking for. Review parameters for video don't seem to be nearly as well defined as those for photo, maybe because most still camera review sites are looking towards dpreview when they set their standards. Although the reviews at dpreview are often more or less irrelevant to my photography, they have made a very elaborate routine, and apart from real life photography (disregarding the snapshots that accompany their reviews), it's rather complete. I haven't found anything nearly as good when it comes to video.
i'm with the latter, read more.
agreed, there isn't a benchmark that everyone on the web uses to test video.
the gh1 video wise is crippled because of its "AVCHDlite" codec and not sensor size in my opinion. There have been many canon 5d 7d users who switch to the gh because of that "filmic" look....i'm guessing not super clean that it almost appears fake look. It also doesn't suffer from overheating, recording limit, and does a better job at avoiding moire. It comes down to a matter of personal taste really. In the first zacuto clip you can see the amazement from the industry pro's as to what this little sucker can do (unhacked too). Of course it has its weaknesses like DR (the gh1 chart was all unfocused, crocked and looked liked it was thrown in last minute..ha) and FPN.
The gh2 has a better codec that now includes B frames though still not true pro AVCHD (maybe the AF100 has that) and still suffers from shadow artifacts from what i have seen but it does have hdmi out which can possibly give the option to record onto a separate recorder whatever uncompressed codec you like. That is slowly being nutted out and fine tuned by users. They finally addressed the FPN in low light, high iso, and underexposed areas too which was what ever gh1 owner was hoping for. Though there is a strange single line horizontal blue band that sometimes appears in very high iso and changes position based on frame rate but being able to capture decent image at 3200iso, its a minor flaw. The cleaner image in low light high iso is enough reason for me to upgrade, ill still be keeping my gh1 though=)
Something interesting is the gh1 iso's is alot brighter than the gh2 if you compare the two side by side.
I have a friend that works in tv and he hates it when they decide to shoot with the 5d, could be workflow or just dslr in general, i should ask him why =)
Last edited by MRfanny; 18th January 2011 at 18:35.
>he gh1 video wise is crippled because of its "AVCHDlite" codec and not sensor size in my opinion.
It is AVCHD and not lite (lite is for 720p in other cameras). It is crippled by:
a) low bit rate
b) no B-Frames
c) GH2 uses better encoder
There are certainly not many of all the pages DPR presents with each review that are of interest to me. Checking noise levels at very good light and at small 100% crops are for example pretty much meaningless. Fining outhow the camera handles in real life is also not easy.
The Zacuto clips; the bathroom scene, maybe from part 3?, showed very well that the GH1 had the worst DR of all the tested cameras. The big jump was between film and digital though.
From what I have seen (untrained eyes though) this far the differences between the GH1 and the GH2 aren't that big. I'll have to go back re-check some samples. The high ISO noise speaks for the GH2 though as I really dislike banding.
I've come to the conclusion I would benefit from something in between my 5D11 and Panny LX3. This would be particlarly for travel photography and wilderness hiking, when the 5D and lenses are getting a bit heavy (well camera is same weight but feels more to my ailing discs!). LX3 capable of surprising quality on latrge prints and 24mm and F2 is great, but limited useable ISO range and 60mm top end excludes a lot of possibilities
GH2 seems to fit the bill - no APC camera offers appreciable wieght bulk saving over 5D. But I'm struggling with lens selection. I don't want another system, just coverage form 24/28 to 200 ideally, with reasonable close focussing. In the UK, only sold with 14-42 or 14-140; latter seems a bit bulky when discretiuon is the key but there seems no alternative 2 lens option as the 14-42 does not seem to do sensor justice, so can't see any alternative - would be nice to have 20mm pancake for when zoomn is just not discrete enough.
Hi Nigel, welcome to the forum.
The 14-140 + 20mm is a good two lens kit. All purpose lens plus fast and discreet pancake (at a good focal length to be useful in many situations).
You are also right that there is very very little difference between the GH1 and the GH2. They have partially cleaned up the banding noise. It's still very much there but it's better now. Also they improved their internal pre-procesing noise reduction routines by a tiny tiny bit - but at the cost of some global and micro contrast. When this is combined with the new 160 ISO base it seems to be better at higher ISOs but it's really just a combination of the new base value and the improved NR. Now combine those two things with smaller pixels (finer noise grain per screen scaled inch) and you have people thinking that it's a great improvement - ummm, after they scale both GH1 & GH2 images to 1000 pixels for the web that is. It is better but very very slightly when one evens the playing field via 100% and adjusts ISO to base compensation. As far as still IQ goes that's it. That's all there is.
Video Mbps in the GH2 has been almost doubled (to 24Mbps) over the native GH1. But with the hack the GH1 can achieve 75Mbps peaks and sustained average rates of 50Mbps. Additionally the GH1 with hack can set GOP to 20 or 10 frames, producing 3 key-frames per second at 60p or 50% more in 1080/24p(I). The codec's 75Mbps peak bitrate allows it to encode a large amount of image detail into each key-frame yet compression efficiency remains high enough to limit average bitrates to 50Mbps. You'll need a good class 10 card for this but that blows away the GH2 in almost every situation. The situation improves when using the MJPEG encoder with such setting and if you're doing green screen there's no beating these bitrates - well, within the M4/3 world anyway... The shorter GOP frame structure in combination with the 75Mbps peaks also significantly improves the codec's ability to track moving objects without producing mud! Nice aye?! The GH2, AF-100 and the Native GH1 can only dream of such things.
>"[the gh2's] Resolution is simply amazing, makes still frame video grabs look like the 60D's stills..."
This is BS even if exaggerated.
I have to disagree about the video quality of the GH2 compared to the Canons.
At slashcam.de they see it as a reference in DSLR video quality even in lowlight compared to 60d and Nikon D7000: http://www.slashcam.de/artikel/Test/...erflieger.html
I have written about my personal experience 2 days ago.
Last edited by lcubed; 19th January 2011 at 22:41.
Show me the bitrates and keyframe intervals, tell me the sensor size, post some samples if you know what you're doing, and that's all we really need from these guys. All this on and on endless drivel and speculative opinion about what they think is nothing short of retarded.
Last edited by Tesselator; 20th January 2011 at 02:01.
Tesselator, do mean by "becoming an instant expert" the slashcam site? This is a very trusted site and is testing camcorders for many years now. If you mean me, I have to say that I am not an expert, but also not an novice. I started filmmaking (documentaries) 13 years ago with a Sony VX1000 and have made my experiences during the years.
Here's a 100% crop from the part of my house which typically shows the worst lateral CA which a lens has to give. This crop is from the left edge of the frame. I processed both 9mm f/4 files in Raw Developer, which doesn't apply any corrections. On the left is the Oly mZD 9-18, on the right the Pana 7-14, both at 9mm and f/4:
Here is the same area with the file processed in Lightroom 3 (which applies CA corrections automatically for Pana lenses shot on Pana bodies):
RAW files for download:
Oly 9mm f4.RW2
Pan 9mm f4.RW2
Naw, I didn't have anyone or any site in particular in mind when I made that remark. But the reviews I read on every single commercial site from "professional" reviewers almost always offer an "opinion" which is wrong at lest as often as it is correct. To me there is no such thing as a "trusted site". People are way too unpredictable and influenced by all sorts of absurd things to be trusted with delivering unbiased and absolute fact. Especially sites that have advertising, get review products from manufacturers for free, or who receive any kind of funding from OEMs or VARs. If they can show what they're saying directly then there's a chance it might be true but still I don't completely trust such and almost no review sites do this satisfactorily anyway - German ones included.
There's only one intelligent way to go about learning of some gear's abilities, and that is to read the specs and the manual and then try it for yourself. Generally speaking, that's about it.
Luckily for me most Japanese stores will allow you to play all you want with their demo models! It used to be that way in the USA as well - I think it still mat be too...
This one only - I processed so the CA would stand out clearly. I brought it in at 16mpx instead of 12 so it's bigger than normal. It the same image as above - but processed.
That should NOT be cyan!
Maybe our adjectives are out of sync but for nearly a thousand dollars I call this "massive" amounts of CA. :-) My $200 Rokkor 16mm has much less than this and it's 40 years old. ;-)
OK then, we disagree. Compared to all the ultrawides I have used, the Pana 7-14 does not have above average CA, so by any criteria, I wouldn't call it "massive" (which to me implies much worse than average).
As as a practical matter using my default workflow, I was never troubled by color fringing of any kind with this lens. What body were you using? RAW or JPEG?
GH1, Firmware 3.2, RAW, ACR demosaicing.
How many other UWA lenses have you used?
It does neither of us any good to argue this issue. We both have hands-on experience with the lens and have formed our own opinion of it. For the others reading the thread, I just wanted to go on record as strongly disagreeing with your description of a massive CA problem with this lens and put up a couple of carefully controlled comparison crops with the corresponding RAW files for download. Having accomplished that, I'm done here .
I have like, 50 lenses between 20mm and 35mm. None of them have as much CA as the Lumix 7-14 and none of them cost over $250. All but three are manual focus - most of them are primes but there's lots of zooms too. I've have only 4 UWA lenses though (below 20mm). None of them cost over $500 and only one has CA like the Lumix. Three are MF lenses.
It's also cool that we have different opinions. That's the human condition. Sharing such opinions is fun for me. That you didn't see and couldn't show the CA from this lens might also have something to do with the difference in conditions you and I used it under too. It's all good and thanks again for sharing your experiences!
I am with Amin on this one. In the situations I use my 7-14 I am very happy with the lens and do not see problems with CA in my copy.
I might later, yeah... But the GH and G firmware can't correct it when it's in the OOF areas like this. Which is why I tried to get the potentially troubled areas in the OOF portion of the scene. You're right tho. The camera's firmware does a lot of shifting of the color plates for that lens and images that are all sharp and in focus are typically well corrected.
The one UWA I mentioned that also has a CA problem is the exact same deal except I have to correct for it in ACR by hand. One setting fits all images and almost all of it's images need it. The difference between it and the Lumix is only that the GH1 does it for the Lumix automatically without user intervention. Handy.
At 7mm at f4.0 with a subject distance of 3 feet the you should be OK from 1.42 feet to infinity hyperfocal 2.7 feet
At 14mm at f4.0 hyperfocal is 10.8 feet.
@ Tesselator: how about posting real images? By this term 'real', I mean an image that you took to be used for a purpose, focussed on something, properly exposed, etc. I think I could get the 12–24/2.8 to show this kind of CA if I duplicated those setups, but that would give no indication of how the lens performs in 'battle conditions'.
And you wrote:
That's all fine, but as a philosophy prof. once said to me, "Opinions are like *bottoms*; everyone has one—we want to know why this opinion is more valuable than that one". If the goal is fun, no problem, but sometimes it's hard to divine the intention from how you write (again, as an ex-philosopher, to me you can sound like a logician, with the weight that this entails, but you might be intending rhetoric, or polemic).It's also cool that we have different opinions. That's the human condition. Sharing such opinions is fun for me.
Then again, you might be able to see things that others can't, too. If this is the case, evidence please.
Best wishes, KL
As for experience with wide angles, I've used over a hundred with angles of view greater than 90 degrees.
As for your example of the Rokkor 16mm; it's dead easy designing a fisheye with low chromatic aberration in comparison with rectilinear lenses of the same focal length. Have a look at the concurrent 20mm/2.8 Rokkor. You in particular criticize reviewers for comparing apples and oranges, so you leave yourself open in this regard.
The 7-14 Panasonic is an outstanding lens, even at the price.
Tesselator showed two types of CA. The lens cap crop he showed crop was greater than 100% and processed to maximally show fringing. I can assure anyone who shoots on a Panasonic body using Lightroom 3 that lateral CA is just not a significant issue with this lens in practice. I have tested quite a few lenses by taking identical pictures of the front of my house, and the region I showed in the crops above is where lateral CA rears its head from the great majority of them.
The other crops Tesselator showed are clearly not lateral CA, because the fringing is of a single color completely surrounding blown highlights. While this could be related to a sensor/microlens issue, it's probably longitudinal CA. If I look for this type of cyan discoloration surrounding blown highlights, I can see it in selected images from lots of lenses, and the Pana 7-14 is IMO not especially poor in this aspect.
In any case, the 7-14 performs admirably with respect to lateral chromatic aberration.
Also to note is that we do not know how the lens performs in terms of LaCA as the camera corrects for that before we ever get to see the images. It could be fantastic and it could be the worst in the world. I guess we would need to put it on a M4/3 body that has no such correction. Has anyone done this and also tested for LaCA yet?
I do like the lens! I'm not trying to unwarrantedly knock it. And if you need AF on M4/3 in that focal range I think it's the best option there is. I dunno the Oly so well - is it better or worse do you think?
I have the micro 4/3 version of the Olympus 9-18mm zoom and before it I had the regular 4/3 version. I have seen the Panasonic 7-14 in real life for five minutes only.
I'm happy with the micro 9-18. It is half a notch "worse" than the regular 9-18. Judging from images I have seen the Panny 7-14 is clearly better than my lens. But the 9.18 takes standard filters and is "good enough" for my use of a super wide angle lens.
As I have followed the Olympus SLR forums out there for years and the micro forums since Nov 2008 I feel pretty safe saying the a general vote between the available Super WA ooms would end like this:
Olympus 7-14 (reg 4/3)wins, quite close followed by the Panny 7-14 and then there is a small jump to the Olympus regular 4/3 9-18 close and finally followed by the micro Zuiko 9-18. That would include every sort of aberration, different levels of bragging rights and further Internet opinions one can find. I'm pretty sure about the order here but one should remember the difference from one lens to the next one is small indeed.
Adobe built the capability into ACR very early on. Then about 2 years ago the DNG standard was changed to be able to capture this information. Aperture took a longer time to support. Some other RAW development programs have said, we aren't going to make the corrections.
So, it is possible to see uncorrected files if you use the right RAW developer. Here is more info:
Here is the 7-14mm lens on the Oly camera that does not correct for CA
That doesn't look "terrible". Not great but not terrible anyway. Too bad that's DPR tho... They've been known to fudge, omit, and cater somewhat in the past. I'll assume these don't fall into hat category tho.
My oldest G1 is from November 2008 and I followed all those discussions about the in-camera corrections, and about the very big DNG files we got in the start, and... Then there is always this small possibility they changed something between the G1 and the GH2. In this case I don't think anything changed and thinking about it it is hard to see how the raw data itself should be corrected on the fly.
Well, there's actually quite a lot of "preprocessing" that goes into a RAW. A RAW file is not just a sensor dump like some think. Or so I've been told by those who say they know. They were convincing, I believe them.
There may very well be more on the topic. Do you have any sources or links?
If the RAW files were sensor dumps, all files from the same sensor, regardless of camera brand, should look the same and open with the same RAW converter settings. But they are not the same, so some processing is obviously going on. This is logical; all camera manufacturers want to give us what looks like "imaging according to them", even in RAW. Giving us pure sensor dumps would give us a situation where the lens manufacturer, the software houses and Sony decided how the output looks, at least for those using Sony sensors. And the software programmers wouldn't even need to talk with Nikon to know the "inner secrets" of their cameras. Sony would know what they needed to know.
No, the camera guys need their own secret formula, so that we can have that yellow Nikon look and the Olympus blue skies
>On the correction issue I don't know for certain but I believe this is a preprocessing in-camera operation
It is in camera for JPEGs but not for Raw.
I think I "know" Panasonic and Olympus uses different colors on the bayer overlay. That way everything is changed also if the raw files are "sensor dumps". Or one may say the sensors aren't the same.
Another thing which may make some (not you) users think things are different is that most raw converters apply the in-camera JPG settings (fully or partly) when opening the file in the converter.
So, to me there isn't any obvious processing going on. I know, now that i have written that somebody will tell me...
But I'm willing to learn. Do you have any other arguments for pre-raw processing I would like to learn about it.
Don't forget the Canon orange, or red...
>Do you have any other arguments for pre-raw processing I would like to learn about it.
In few cases the Raws maybe tweaked for noise. At least it is rumored.
There was a rumor about Canon sensors and raw noise reduction. I couldn't find any reliable information telling me anything but this being about on-sensor hardware noise reduction circuit. I don't think about that as software processed raw files.
Then we have Sony and their FF cameras where there is some on-board noise reduction going on (not that one can see it in their high ISO images...). Again, as I understand it it is about on-board noise reduction circuits.
I'm also not at all sure about how there could be any software based noise reduction made but sometimes my brain isn't that flexible. Maybe it can be done, or maybe it is done. Maybe somebody knows about this?