It's good and nice to have an APS-C sensor in a mirrorless body with full manual video controls. Powerful new image processor, crying out to be hacked :)
But GH2 clearly wins on resolution and the cleanness of the image with less aliasing.
If you want a larger sensor and 1080/60p the 5N is a pretty good choice though!
thanks for the test. Based on what I've seen so far, and taking only video mode into account, the hacked GH2 destroys the 5N in resolution, sharpness, Latitude, aliasing, moire, audio control, color, user interface and surprisingly, even in low light, with sexy sharp grain:) NEX5N wins in fast wide angle lenses(because of sensor size) and has 108060p. It would be interesting to compare the GH2 720p 60fps to sony 1080 60fps. Were you serious when you said "soft is cinematic" ? :) I think what you meant was soft can be cinematic :)
very interesting stuff andrew. thanks for the review!
i am getting mine soon too, so i can compare the gh13, gh2, 550d and some other cams lying around. i hope it does not show strobing tendencies in wide shots like my gh2 compared to my 550d in 24p...
i love that they opted for OSS on the 50mm prime at a fair price, something m4/3 users would love to have at this aperture - speaking of primes, don't you find aliasing and moiré is less pronounced with primes than the "kitlens" ?
did you make any tests with a fader ND attached outside? i feel it helps the overall analogue look quite a bit...
@dts -aliasing and moire has absolutely nothing to do with aperture, primes or zooms. -a fader ND will in no way on this planet give any footage a more analogue look, what it might give u is the ability to use wider apertures and have shallower depth of field.
what do you mean by strobing tendencies in wide shots? horizontal banding?
- i found that he kitlens for my gh1 got me more aliasing than old analogue prime lenses, but then again, they were not as sharp. in general i hope decreasing digital sharpness inside 5n and using not the sharpest lenses might help a bit... - well i prefer the 70's look and for me the fader nd does help and not only in using lenses wide open - i mean the strobing with horizontal pans, that has been discussed a lot, and seems to be divided into two groups, one is saying "thats 24p and learn how to shoot" and the others who compare with other 24p cams say "something strange is going on - perhaps oversharpening etc."
Will a hacked GH2 improve the highlight fall off and limited tonal gradation as depicted in the review? Surprised that the Sony beats the 7d for stills quality.
@brianluce a hacked GH2 will not improve high light roll-off, perhaps bringing out the native sensor grain will aid with dithering a bit. Please don't say things like "limited tonal gradation" (i'll get an ear-bleed:)), it doesn't mean anything on its own, too vague, do you mean latitude, dynamic range, color space, 8bit vs 10 bit? Also, in my opinion, the review does not depict a limited tonal gradation.
Does 5N suffer from overheating like NEX5 did? Or does it have any artificial time limits like A77 that shuts down after 13-30 minutes depending on ambient temperature (along with other alphas that follow a similar pattern)?
I had bought a second GH2 6 months ago and sold it thinking that A77, EX7 or EX5N will outperform the GH23. Time goes on and now I think I'm wrong ... All sonys appears to have more moire, less resolution and poor dynamic range ... The only thing that can let me buying the 5N could eventualy be the HDMI live at 1080-60P and plug a nano recorder ... which seems to be impossible at all ...
Less resolution?... hmm... A daylight comparison might probably reveal the GH2's better performance in resolution, but the lowlight scenes Andrew has posted do not show any clear advantage over Sony. They're all pretty much soft. At least to my eyes.
@Stefanos Please don't say things like "limited tonal gradation" ................
That's precisely the term EOSHD, the guy who wrote the review used. If I'm citing the findings of the review, I thought it best to use the language he used as to allay confusion. Personally I've never heard the term either until the review, but I'm not a video engineer.
I think some are just too analytical...its down to the look your after,ergonomics and uses. APS-C sensors give a different look and feel to M4/3rds...in many cases I prefer the "look" that that the sony brings.
The GH2 has great video qualities...but lets be blunt here...its a poor stills camera....Its not great with skintones either....and the battery life pisses me off.
On one end you have guys who want to see the farthest reaches in super fine super high bitrate resolved detail...on the other you have the same guys complaining about halo's or oversharpening or videolook and finding ways to reduce it for that "film look".
I had a Nex 5 and it never really bothered me in realworld shooting with moire/alaising...and videos looked just as sharp as my GH2 footage on my 50" and matched well.
Biggest gripe with sony is lack of high quality AFFORDABLE lenses...999 dollars for a 24mm f1.8 zeiss or otherwise is a bit of a stretch in my book.
As for the video..hard to compare when its using two different lenses (If I read right in the past the samyang outperforms the canon L FD so this might explain the corner sharpness difference)
Skin tones, the smooth gradation of tones from one subtle shade to another, overall colour, stills and battery life are a strong point on the 5N. I feel the GH2 has a better scaler from 18MP to 1080p, but it has a bias on resolving clean detail whilst the 5N's scaler has the bias toward resolving the colour space better.
@Brianluce that phrase isn't in the review? Gradation is the smooth transition from one colour or shade to another. Tonal range is the number of steps in between.
@zigizigi no overheating so far on the NEX 5N or A55-like limitations, that was due to the movement of the SteadyShot Inside system causing a heat built up, according to Sony. They have taken the system out of the A77 and the mirrorless cameras never had it, instead relying on optical stabilisation in the lens.
The GH2 definitely has more detail, low light or not!
What I mean by 'soft is is cinematic' is that you don't want a hard digital look to cinema with a lot of electronic sharpening and jagged aliasing, a smoother and softer look to detail is beneficial, can be a bit harsh and contrasty on the GH2 sometimes. That's why I like the older optics.
One thing, how "bad" is the aliasing? I had a 7D but sold it because of the extreme levels of aliasing and moire. I've never had any problems with GH2 with it. How is the Sony in this regard? I understand there's more, but could you possibly post a test? A Canon can't shoot a brick wall or water, but I'm ok with some (for example the FS-100 showed some moire and aliasing in Phillip Bloom's test, but it was more than manageable).
Also, is it true that the EU version is only 25/50p?
Aliasing on the 5N in 1080p is similar to your 7D in 720p mode, but with far less moire, certainly no rainbows! In depth tests are coming, only had the 5N since Thursday.
In resolution the 5N is in between the 7D and GH2, smack in the middle I'd say.
Indeed, but don't analyse this area too hard, it comes down to the bigger things in the end as @Mimirsan said, APS-C sensor and 1080/60p. Certainly the 5N at 18mm wide angle at F2.8 gives you a shallower depth of field than the GH2 at 14mm F2.8 wide angle. It's a different look. 60p great for slow mo.
Also I am getting more info in the shadows and the 5N footage holds up better to grading, with more dynamic and tonal range.
@EOSHD: Absolutely! There was however a reason I felt more or less nothing but hate when I sold my Canon... There were so many parts that were rendered unusable because of bad aliasing or moire... But again, it will be fun to see how this camera handles! I love the picture from it and with that low of a price, it might still be worth it... Basically, as long as the aliasing is manageable I'm ok with it, as the moire seems to at least not be a problem (and that I think is a much larger issue on a Canon).
EDIT: After looking at your footage, the aliasing seems to be manageable. Again, the biggest problem I feel with Canons comes from the moire, which I see very little off in these tests!
Andrew, the feature I'm most interested in is live peaking; how fast is it, how's shooting with it? From what I've seen so far latitude looks worse than hacked GH2, you say 5N has an advantage in dynamic range?
For now (early days) in my opinion it definitely has a dynamic range advantage.
It surprised me also how much cleaner the shadow detail is on the 5N at 24Mbit compared to the GH2 44Mbit and AQ 4 (Chris Brandin's setting). Look in the bottom left hand corner of these test shots... I have adjusted levels to extenuate visibility of what's going on the lows, this is not directly off the camera. I exposed to just about protect the very bright highlights on the book in the sun, base ISO. Cinema Gamma mode on GH2 and Portrait mode on 5N.
GH2 left, 5N right, click to enlarge
Come to think of it, what exactly IS going on in the lows on the GH2!? What causes the heavy macro blocking - a lack of data from the sensor? Maybe @cbrandin can explain?