GH4 vs GH3 file comparison by Quesabesde
Quesabesde (translation here) is the first site giving you full size GH4 RAW files to download. And it proves that the GH4 isn’t only a killer video camera, but also a better stills camera! Enjoy the pixel peeping and tell us what you think about those images!
Some more Gh4 image samples and also some image samples with the new Leica 15mm lens can be downloaded at the German Dkamera site (Click here).
North American preorders, shipping date and price:
GH4 camera at Amazon (Click here), Adorama (Click here), BHphoto (Click here), and Panasonic (Click here). Price: $1,698. In Canada at Vistek (Click here).
GH4 camera with interface unit at Amazon (Click here), Adorama (Click here) and BHphoto (Click here). Price: $3,298
Interface unit only at Amazon (Click here), Adorama (Click here), BHphoto (Click here) and Panasonic (Click here). Price: $1,998
Europen preorders, shipping date and price:
Germany: Wexcameras, Marcotec-shop.de.
UK at Wexphotographic, CameraWorld, UKdigital, CVP, TipTop.
Holland at Fotohanskeuzekamp.
Belgium at Fotokonijnenberg.
Norway at Fotovideo.
Sweden at Cyberphoto, Skandinavianphoto.
Finland at Telefoto, Topshot, Verkkoauppa.
France at Photocineshop.
Spain: Fotoboom.
72 Responses
Photos should be looked from a distance twice their size. But pixel peeping can be good for a chat at the pub.
That ISO 200 looks like sh_it, but 6400 rather good. Not worth the price at all, but it may be interesting for video shooters, if it does RAW video.
The ISO 6400 sample has more detail on the GH4 but the noise is far worse. To be honest, I expected more of an improvement since it has been confirmed it has an all new sensor. At least video has been more impressive so far than stills.
The noise is fixable in post. Detail isn’t… I think it’s a massive improvement here.
But reducing noise in most will also remove detail…..
The comparison with the GH3 would be 10x more meaningful if instead they have provided unprocessed RAWs.
Sure, the lettering on the GH4 is a lot sharper (and readable too) but I’m much more interested in how textures are reproduced and that new JPEG sure looks like a backfiller. I wonder how much detail is available on the RAW.
Yawn more flawed tests on useless parameters, to sell more overpriced toys.
Here’s a great look at the GH4’s high ISO video: http://vimeo.com/90885582
Here’s a look at the cine profiles shadow and color treatment. http://vimeo.com/90598002
There is more sharpening and noise in the GH4 Raws, that’s all. You could apply the sharpening to the GH3 RAWs, too. So basically they seem to be identical in quality. No major difference, not step forward in stills quality
exactly, the sensor output seems to be the same, just different processing.
If it’s Panasonic, then it might only be the processing.
The tired recital of ignorance to Panasonic sensor tech.
Not only the fine details at high ISO are clearly better on GH4, but also chroma noise (still showing up through the NR) is distinctly different.
unfortunately these particular RAW-files are not available for download, otherwise we could definitely have a look at them both with some alternative raw-converter and some hex-editing.
Thanks for posting o.luv.
i tried the iso5000 raw example from their site without any processing and it still looks pretty noisy:
the jpeg looks better of course, so i still think it is just more advanced NR and post processing.
even ACR opens the hex-edited raws easily, just make it think it is a GH3:
with some NR and sharpening this is how the iso5000 raw looks in ACR.
below again the crop from the camera-jpeg, so no sensor-magic going on at all.
Here’s something for picklespeepers 🙂 http://newcameranews.com/2014/04/02/royal-society-of-pixel-peepers-proposes-truth-in-exif-act/
Long overdue!
The pixel peeper allegations always seem to flow from those using smaller sensors towards those with larger sensors. With your own chosen format , obviously the “right ” choice lol. So the mFT mafia target the larger sensor larger MP count cameras { despite them having sometimes considerably lower pixel density than we do} for the pixel peeping accusations meanwhile those using smaller formats { Sony 1″ etc } are ripe for insults aboiut image quality , which is far closer to mFT than mFT is to FF.
Yet , there are people here with thousands of dollars worth of mFT lenses and cameras .Including those buying the pricey and often large fast primes from the likes of Voightlander . When apparently to the “artists ” { few of whom will ever take part in a camera forum let alone a brand specific one} are not concerned with , shadow noise, DR, DOF ,MP count , high ISO .
mFT has exactly one significant advantage over DSLR users , namely size/weight of the gear. Take that out of the equation and you can easily get “better” image quality and very often for less money. I accept that eventually smaller sensors will get to be good enough that anything larger will be for very specific needs .Sadly this is not yet here I am actually quite disappointed in the image quality from my E-M1even as a predominately low ISO shooter , I would have liked the E-M1 to bring better shadow noise , a genuine 50ISO base , and yes I would have liked a higher MP count lol. Despite the complaints there are few downsides to high MP counts. At low ISO you gain detail and at high ISO you have more flexibility { a high ISOD800 file downsized with a little NR which it can take better due to the detail advantage is not that far from a D4 }
Brilliant, yet sadly true! 🙂
Awesome site. Onion News for Photogs.
Thanks for the great link. I found this one further down that was funny too: http://newcameranews.com/2013/12/08/daughter-iphone-takes-better-photos-father-nikon-d800/
Okay, that site is hilarious. And tragically accurate on many of the articles. Makes me feel better for having looked at the comments hoping (in vain) to see intelligent commentary.
After the required EXIF tag one, this was my favorite:
http://newcameranews.com/2014/02/17/royal-society-of-pixel-peepers-mulling-1000-crop/
Pretty good results overall. Seem better than the E-M1…
http://photobyandrei.wordpress.com/
I do the same comparison as well , with my GX1 & GX7 . Only at high ISO , the different can be significant spotted .
At low ISO , the GX7 deliver more brightness ( might due to light gathering capabilities as reported ) than GX1 .
Overall it did improve but I expect more . When will Pana start to use Fuji’s sensor tech in it’s camera ?
I hope that will neve happen…xtrans sensor is a joke…water colour effects highlight blowout, inflated iso ….and so on…
The e-m5 inflates its ISO by almost a full stop, Olympus was and is pretty well known for lets call it “optimistic ” ISO claims. The main problem with Fuji is the poor RAW support in the most common editing software. Though I am happy with mFT for my needs download some Fuji high ISO raw files and use Photon Ninja the results are amazing near FF quality.
What a joke “GH4 RAW photo files available for pixel peepers!” and with what should we edit them, no program supports GH4 RAW yet…
sure, just hex-edit them and make them GH3 raws
I personally do not know how to do that, can you develop few picks and upload them here?
i found the images pretty uninteresting, most of them are shot above iso800 and need some processing to really look good. i would have welcome some raws at base iso, but so far there are none.
yep o.luv’s correct, just edit the file into gh3 and it works. that’s how I did it.
I see only different sampling kernel / filter with more sharpening.
Most people do not realize with m4/3 you shoot at 2 stops lower ISO than FF. If you PP dpreview Nikon DF or D4s you will realize there is no advantage to FF today
What drugs have you taken?
What do you mean? It is true. If you shoot the same image with 4/3 and FF with the same dof you stop down 2 stops on your ISO. maybe your on drugs or maybe you don’t understand photography
It is silly because you don´t get shallower dof when you use lower iso values – you have to close/open the aperture in order to influence dof.
1/250 @ F4.0 ISO 3200 on FF is the same as
1/250 @F2.0 ISO 800 on m4/3
Same image same DOF 2 stops lower on ISO. That is what I said.
And two stops bigger aperture – iso has nothing to do with dof. The two stop wider aperture should give you 4x faster shutter speed – but the two stop lower iso slows shutter speed so that it´s the same as above.
I never said ISO has anything to do with DOF. You can raise your shutter speed or lower your ISO. Why would you raise your shutter speed?
That rather conveniently assumes you need the exact same DOF and shutter speed. Other than in low light situations where you need extensive DOF this is irrelevant nonsense . In decent light {natural or man made} you can simply shoot with a slower shutter speed, that is still adequate for the job in hand. Or in low light you can either use tripods/IS as long as the subject is not moving there are FF lenses with IS covering from 16mm to 1200mm { 1800 in crop mode}. So for your plan to work you would have to be shooting in a very narrow field indeed.
Plus many forget that though mFT drifts closer to the P&S everything in focus DOF , 35mm film {FF digital} has quite extensive DOF { hence the fondness for MF }other than at close distances using very fast lenses wide open
A little hard to decode but essential you’re right. To get the same DOF from FF you need to stop down and compensate by ISO.
You girl…….do not compute
So to sum the comments up…This essentially same sensor is fantastic on the em1 but crap on the gh4?
I thought that the sensor on the E-M1 wasn’t as good as the E-M5 in low light conditions and the E-M5 sensor is more likely the one revised from the GH3 for the GH4, since there are no Phase Detection pixels.
E-M1’s (Panasonic) sensor has more noise on longer exposures, as compared to the E-M5’s (or GH3’s) Sony sensor. Otherwise IQ is very similar.
GH4 has again Panasonic’s sensor. Because it was silly for a flagship to have a competitor’s sensor.
I suppose they could have caught up to what Sony was doing and without the extra Phase Detection pixels, they have more room for good image quality.
The weird thing is that FujiFilm claims 86,000 Phase Detection pixels on their (supposedly Sony-made) sensor and it doesn’t really seem to have a problem, other than what raw development applications are having with the color filter array.
Gee whiz I try to never shoot with asas above 800. if I need more light then I use lights. works really well that way. ASA6400? Maybe for the Blair Witch Project IV. YeeeOuch!
As expected, stills quality is just a bit better. Video quality, OTOH, has taken a huge leap forward. The competition is basically the BM 4K camera and then other gear that costs at least 5X as much.
What, better?! This sensor is a crap, much worse than the first Four Thirds ones (L1, l10, etc.)!
Even a vacuum cleaner from Canon makes better pictures!
http://englishrussia.com/2009/02/02/canon-vac/
I have both L1 and GX1. I can say that the GX1 is a lot better than L1.
So, I think GH4 is better than L1 for sure.
But it’s clearly not better than the Canon vacuum cleaner. 😛
*sarcasm*
They sure know how to vacuum the money.
What? The DMC-L1 wasn’t nearly the first Four-Thirds camera body.
Agree, but no Olympus body could have weak image quality, no matter how old it is (just ask Olympus fanboys) so I’ve forced to instance a Panasonic body. 😀
Just don’t compare the DMC-L1 to the E-520, which used the same sensor but the AA filter of the E-520 was brutally stronger, smushing much of the detail away. 😀
Not an Olympic fan Boi, are we?
Looks to me like the GH3 was not really focused properly. The
lettering around the sides is considerably fuzzier than the center of the label, as if it were barely focused on the closest part of the label. The GH4 seems to be focused somewhere between the nearest and furthest part of the label. Almost as if someone wanted
to prove GH4 is better, no matter what it takes. The GH4 is obviously fantastic for video, but I wish people would stop insisting it is better for everything. It’s not. From what I have seen it shows more noise.
How about some comparisons to the EM-1?
Or not happening in a million years?
The E-M1 has a current gen Panasonic sensor so what do you expect to be different other than Olympus using a crap video codec? Here in the UK the GH4 has the exact same launch price as the E-M1 , with it you get a great 5 axis IS system { if that is important to you} and typical mFT image quality { sub par compared to the best APS and FF models } with average video , the GH4 will no doubt have equally average image quality but at least it comes with excellent video .
Or are you one of those Olympus nfans who through magic rose tinted specs sees multi stop improvements in Olympus cameras lol. Remember all those claims about how superior the image quality of the E-M1 was right up until it was shown to have a Panasonic sensor lol
It has a sensor designed by Olympus and built by Panasonic. That is not the same as being designed and built by Panasonic.
As for the video, although it does not have all the different frame rates the video quality is better than the A7 and A7r and the GH3. The GH3 also has better video quality than the A7 and A7r.
lol the old Olympus design routine they have zero experience in sensor design or manufacture { hell they used the bog standard Panasonic 12mp for years } their video is downright poor for a £1300 body only camera.
The video quality of the E-M1 is nowhere near that of the GH3 to anyone not swallowing the Olympus happy pills. I love my E-M1 for all its qualities but as I am also a GH3 owner video is certainly not one of them . Only a delusional fanboy would even claim that the E-M1 video is as good as the GH3 { i cant speak for the Sony FF body’s , but even the humble RX10 has better video than the E-M1
I guess it a good video camera…I don’t know anything about that…but whatever you are trying to sell me here as a stills photographer….this is TOTAL BS. Better? Thank you…but, no…I will stick with my Olympus bodies.
With the top Panasonic’s you get pretty much the exact same still image quality , with typically far superior video quality so I do not see the downside .
I want to believe that the GH4 is a good leap forward for stills, but even small leaps are fine. I’d love to see better image quality at ISO 3200 and usable photos from ISO 4000~ISO 6400.
I don’t believe that tests like these are going to convince me either way, since I’m not in control.
Byebye GH4, hello Sony 4K A7s.
http://www.sonyalpharumors.com/sr5-sony-a7s-has-12-megapixel-full-frame-sensor/
Wow. The FE-mount seems to be flourishing! *3* bodies and whooping (drum rolls) *2* native lenses!!
4 FF lenses
20+ cropped lenses
also infinitely adaptable since it is mirrorless
1 very good quality lens
1 good quality lens
2 poor quality zoom lenses for the price.
Looks more like better JPEG processing to me, but having said that also looks very useable for ISO 6400.
Load of nonsense.
Lol tomorrow Sony will bring out a new camera which will be so much better then this very very expensieve and very very small sensor camera.