Gallery Update: Panasonic Lumix DC-G9

Always nice to have a gallery to see what cameras are capable of. But in this case not so easy, because most of the more challenging shots (low light, higher ISO’s, indoor, sports) are images processed by the reviewer without the out-of-camera files to compare . So be careful before passing judgement on them. You are looking at the reviewers preferences. They might not be yours. Just try to process some RAW’s yourself if you can.

5 months agopermalink

Out of curiosity I used some tools you can find on the internet that allow you to trace back settings of image files. For some basket ball files it looked like this:
Edited with Photoshop CC 2014 on Mac
White Balance : custom (not chosen by camera)
Tint : +48 shifted to purple
Settings exp., contrast, highlight, shados, whites, blacks all edited to taste
Region : Lights and Darks edited
Profile : Vivid
Sharpening : 48
NR Luminance : 25 (very modest)
Smoothness : 25 (very modest)
Now, this is meaningless if you don’t use Photoshop/Lightroom, but anyone who does knows that the output file with these settings will not look like any out-of-camera jpeg. Better? Well, that’s personal taste again. But certainly noisier and more purple. So if you are interested, do your own research.

5 months agopermalink

yeah, the Vivid profile says it all ;-)

5 months agopermalink

soundknight21

Nice catch Robemo.
They should have before and after edits

5 months agopermalink

Most people would agree that on full frame ISO6400 is about the point where the images becomes a bit ratty.

The way ISO and sensor size works, one can expect roughly equally noisy images at ISO2600-ish on APSC, ISO1600 on MFT, and ISO800 on 1".

With more than half of the sample photos above ISO1600, I'd venture you are pushing the G9 harder than you push other, larger format cameras in comparable galleries. It's no wonder so many of the images seem of rather poor quality, you are treating the gear as you would a full frame camera and cranking the ISO two stops to compensate for the smaller sensor. To my way of thinking that puts the G9 in a worse light (pun intended) than fair. In terms of "this is the job, here's what it does" ok, I get it, but anyone investing in the MFT system would I hope understand that two stops more available light is required to compete against FF on noise and would tailor their use scenarios accordingly to stay under ISO1600 as much as possible.

5 months ago*permalink

zakaria

I wonder why micro four thirds cameras are so expensive while the sensor is small and DR is poor.

5 months agopermalink

Nomoreheroes

I wonder why Pentax are so expensive with their lack of AF.
So there are different priorities

5 months agopermalink

I wonder why the Nikon D5600 is so cheap while it has great AF and great DR.

Am I doing this right?

5 months agopermalink

@zakaria I wonder why micro four thirds cameras are so expensive while the sensor is small and DR is poor.

How many licks does it take to get to the center of a tootsie troll . The world may never know...

5 months agopermalink

kristian2000

It's not the size of sensor that matters, it's what you do with it.

5 months agopermalink

@kristian2000 some need to compensate with hardware bitching for lack of skills.

5 months agopermalink

utphoto

Dynamic range isn't that bad with the current m4/3 sensors. They match the DR of a number of Canon APS and FF sensors.

4 months agopermalink

That bad huh?

... Cheap shot but couldn't resist XD

4 months agopermalink

Interesting to see how many keyboard warriors are in these forums and comment sections that don't have any gear, don't produce any photographs, don't have any positive contribution, just silly messages. Yet the DPR-staff just sits there doing nothing. Wow, must be a fun job ...

5 months agopermalink

AstroStan

Malevolent trolling infests social media and nobody has figured out a way to simultaneously censor while allowing free exchange and to do so cheaply (automated) while generating profit (ad clicks).

But there are forums that have very low infestation and a commonality is no ads, moderation and/or participant qualification. Related to DPR - I find it interesting the B&H equipment review forums have virtually no trolling. There are many low value comments (e.g. many can't wait to hit the keyboard immediately after opening their package) but many posts are good useful experiences with real particulars (not useless opinions). B&H forums do not sell ad clicks and probably do maintenance to remove trolls (who then migrate here) and there is verification that the poster bought the equipment.

5 months ago*permalink

AstroStan

P.S. It is obvious that DPR encourages trolling in many ways, not the least being the selection of news items. They have learned which topics generate the most heat and seek them out.

5 months ago*permalink

mxx

In the internet world a click by a troll is still better than no click at all.

5 months agopermalink

Rich Evans

If you think these images look as good as a APS-C you must be smoking something and i want that. These images have the look and feel of Micro 43: noisy, low dynamic range and poor color reproduction.

BUT, if you can't see it and all looks the same. More power to you, since you can spend less money on less powerful cameras. Oh wait, you can't cos this costs as much as a full frame camera.

5 months agopermalink

Thorgrem

And yet more than 80% of your Frontpage posts are MFT related... A system you have no interest in whatsoever.

5 months agopermalink

gianstam

Thorgrem: I think there is the word "obsession"

5 months agopermalink

soundknight21

It's a balance of many things, yes a 4$k camera body with an am equivalent lens will look better but that's already twice the price.
A good photographer can get great and useable or even excellent images from most crop sensor systems.
Plus its WAY lighter to carry as a kit, has lots of pleasant menus and features and video options. Can't say that last point about Nikon and Canon systems.

5 months agopermalink

Loga

Thorgrem: even if you are right, that does not change the quality of these pictures.

1 month agopermalink

Imager of

When I zoom in 800% I’m noticing quite a bit of noise. Is this normal?

5 months agopermalink

Rich Evans

You don't need to zoom to 800% and normal view you can see the shadows are populated by specs. Its a normal behaviour from a small sensor to have this high amount of noise.

5 months agopermalink

Conrad567

LOL. :/

5 months agopermalink

It gets better at 2000% ...

5 months agopermalink

fillkay

Seems to be much adieu about nothing...goodbye.

5 months agopermalink

Michael Ma

I own many Panasonic cameras. One thing they don't do well is white balance and color science. You can say it's their look, but it's not a good one. Luckily there is raw. Standard profile is the worst. You should just not be shooting people with it as it accentuates people's blemishes. If it's anything like the GH5, for shooting people, I recommend the Portrait profile with -3 saturation and +2 or +3 Hue. For video, you have to pay close attention to white balance as there is very little room for wb correction.

5 months agopermalink

https://www.dpreview.com/sample-galleries/0107841691/panasonic-g9-sample-gallery/3575116859

Judging camera output on jpegs is never wise. I used several DSLR's from different brands (also top models) and jpeg output always needed thorough tweeking to get for me 'acceptable' output. And that's personal taste of course. And even then I didn't like all results all the time and often had to process RAW's to get it how I want it.
I don't see why it should be different with any camera. Just look at the endless debates in these forums about 'Nikon color versus Canon color' ....

5 months agopermalink

lotzi

Maybe it is just me but some of the settings used in these pictures do not seem to be the ones that would maximize the image quality.

Picture of an alligator, likely static, in bright sunlight
35 mm, 1/1000 sec, f/2.8, ISO 200, -1 EV

Why use f/2.8 in bright sunlight? The lens is wide open, showing problems at the corners, which would probably disappear at f/5.6. And you wont get any kind of bokeh effects anyhow, not in this setting and distance. Alligators usually move very slowly, there is no way that you needed 1/1000 here.

https://www.dpreview.com/sample-galleries/0107841691/panasonic-g9-sample-gallery/4422857139

Static picture of a dimly lit room, no movement:
15 mm, 1/100 sec, f/2, ISO 12800

Why 1/100 on a static scene, on a camera that has 6 1/2 steps of image stabilization? Wouldn't this be the point to do: 1/5 sec, f/2.8, ISO 200?

Am I missing something?

5 months agopermalink

1. To demonstrate ISO 12800.
2. I suspect the main concern there was "OMG ALLIGATOR!!!"

5 months agopermalink

lotzi

Richard,

1. Well, you have the test scene for that. I think that the samples are more useful in the sense of "if I would buy this camera, what kind of pictures I would be able to take in a situation like this?".
2. ... ok, maybe living in Florida had kind of made me immune to it... there are like, everywhere.

5 months agopermalink

Carey Rose

IIRC, the first photo was taken from a moving vehicle powered by a huge fan as part of a bayou tour.

5 months agopermalink

lotzi

Carey: Oh, I see. That also explains a bunch of other pictures with 1/1000 + wide open on apparently static scenes. Maybe adding a single line of explanation of circumstance and photographer intent would greatly increase the usefulness of the sample galleries - I mean text like:
-catching an alligator from a flying air boat
-trying out ISO 25600
-trying out handheld stabilization at 2 sec

5 months agopermalink

Carey Rose

@lotzi I agree. Dan's out of the office today, but I'll point him to your comment when I see him next.

5 months agopermalink

Even with noise reduction from Dpr, I find these jpegs pretty noisy full size ( the pic of the resto at 4000 iso verges on the unacceptable ) Now downsized by half ( 2600 ) and they all look good . For me it's enough , for printing huge I don't know ...

5 months agopermalink

That's why it would have been nice if they had done a real comparison with the cameras they mentioned in the review. It's hard to judge this without knowing the conditions. Doesn't look that great at 100% indeed.

5 months agopermalink

The more I look at the gallery the more I get the impression the reviewer was maybe challenging the camera deliberately to show us how it performs in low light/higher ISO. He is certainly not going for a 'flattering' approach if you see how harsh most high ISO images are processed without OOC jpegs to compare. And this particular image was therefore probably deliberately taken with unnecesary high ISO setting. The shot certainly didn't need 1/250. Half or even less would have been fine which would have resulted in much lower ISO's and better quality image. But even when you process this RAW image a bit more subtle, you get more pleasing (subjective of course) results. So again a pitty it's not a real world comparison of the cameras mentioned. Would have been nice.

5 months agopermalink

I think you have to just take it face value: if you like what you see in the gallery, great! If you don't, there are plenty of other larger-format cameras, fortunately many of them considerably less expensive, to choose from. Simpler to just move on to a bigger detection surface than forever be messing around with smoothing optimizations on intrinsically noisier data, or having to continually rely on stabilization to get the images you desire.

5 months agopermalink

@ R Murdey
I do agree ...yet switching to another format , in my case I've been looking at FF for quite some time, is a costly endeavour that requires a bit of thinking, because 4/3 is still quite good enough in most cases , so better identify thoroughly which scenarios require FF and which don't , before parting with your hard earned euros ( or dollars etc..). Now when I see an Oly 45 f1.2 at 1299 euros and then a good used Canon 85 1.8 at 300 on my favorite second hand site, that definitely helps making a decision ....

5 months ago*permalink

"Now when I see an Oly 45 f1.2 at 1299 euros and then a good used Canon 85 1.8 at 300 on my favorite second hand site, that definitely helps making a decision"

^^this

MFT is fine for what it is. Diminishing returns start to kick in though once you try to compensate for the smaller sensor by buying fast lenses. Past a certain point it's just not worth it.

5 months agopermalink

Happy to see some pics with the Panaleica 15 mm 1.7 , which seems to be a fiiiine lens
The 12-60 2.8-4 on the other hand ? meh

5 months agopermalink

And too expensive for this quality ...

5 months agopermalink

For every one of you can't can't afford it there's 10 others that can.

5 months agopermalink

Eric Hensel

That's not what he said, red.

5 months agopermalink

noisephotographer

It would be much better if one could also see the out-of-camera jpgs of the edited raw files.

5 months agopermalink

It's useful to download some RAW files and process them to your own taste, because Dan processed his shots to his taste. Especially the basket ball shots looked quite yellowish (skin tones) on my calibrated screen and since there were no OOC jpeg's to compare I tried my own processing. The yellow disappeared. It's a matter of taste of course.

5 months agopermalink

rfsIII

Maybe Panasonic did the right thing taking the sensor stabilization out of the GH5s because to me all these photos look a little bit blurry—as though the sensor were moving.
And perhaps the lack of an ISO lower than 200 is also holding the camera back.
On the other hand I'm glad Dan risked having seizures from the flickering viewfinder to capture the basketball shots. They're an interesting test.

5 months agopermalink

noisephotographer

It would be nice to see the out-of-camera jpg of https://1.img-dpreview.com/files/p/TS5184x3888~sample_galleries/0107841691/3955563524.jpg

5 months agopermalink

Very nice photos and colors but the IQ is mediocre at best.

5 months agopermalink

Thorgrem

Perfect fit for 99% of us it seems. Or do you have any photos in your gallery that suggest you are with the 1% who needs better IQ?

5 months agopermalink

Make that 99.999%, all phone users included.

5 months agopermalink

photomedium

JACS
Agreed! Count me among the 0.001%...and based on Thorgrem gallery I agree with him as well. Except maybe he could be even better off with just a phone.

I never make such gallery-show requests to people, because they are silly and they indicate that I have already lost the argument, but I must admit it cleared things up here.

5 months ago*permalink

Thorgrem

Always nice to attract a troll (and off course without a photo gallery). But your comments on different subjects say everything. To much trolls at DPR, it's time DPR does something about the obvious trolls.

5 months agopermalink

You can always self-ban yourself, you know...

5 months ago*permalink

@J A C S

Very nice post you got there and good effort but the value to the community is mediocre at best.

5 months agopermalink

brendon1000

Next thorgram will ask anyone reviewing movies to he an accomplished director themselves before they can dare comment on a movie :p

5 months agopermalink

Old Cameras

Thorgrem
I strongly agree with your comment regarding trolls.
Whole conversations degrade to silly brand partisanship and straight up insults - just read the comments about Nikon's rumored Z-mount. It would be some work to moderate but immediately deleting off topic comments intended only to bait others would be a great thing. More civility is badly needed.

5 months agopermalink

Whatever you feelings on this, "perfect for 99% of us" is just silly. The results come in about midway between an APSC and 1", just as you'd expect, plus some characteristically Panasonic color rendering in artificial lighting which many people would find lurid but which is presumably fixable with sufficient effort.

Teh G9 will appeal to people who value compactness/speed over image quality, or those who mostly operate within the format's strengths: daylight, telephoto, macro, and wide DOF shooting generally. Wanting better image quality than what is on display in the gallery does not make you part of the 1%, far from it. It doesn't make you a troll either.

5 months ago*permalink

photomedium

Nicely put Richard Murdey!

5 months agopermalink

Looks as good as an APSC sensor. It is strange actually. Performance of APSC sensors has stayed nearly the same since Nex7 (OK, some improvement but not huge) but somehow m43 sensors have been improved considerably. I think Sony has put too much effort to make its APSC sensors faster (although Sony also produces m43 sensors).
I am talking as proud owner of Nex7, A6300 (sold last week), EM1 II and EOS M5. EM1 II's low and high ISO performance is so close to A6300 that I sold A6300.

5 months agopermalink

It doesn't matter. Majority of modern cameras are capable enough for majority of shooting scenarios. If you cannot get compelling results with one format, just switching to another will rarely help.

5 months agopermalink

Correct.

5 months agopermalink

tinternaut

Things have improved since the Nex 7, but that doesn’t mean you’re wrong. Once you are able under expose by a stop, with no penalty in post when you push the shadows in post (based on equivalent ISOs), then what you have is more than sufficient for most scenarios. The Nex 7 was there.

The issue with Micro Four Thirds, for me, is actually the base ISO of 200. I took a lot of photos over Christmas, comparing my Olympus kit with my Ricoh GR. In the difficult winter light, having a (actual) base ISO of 100 proved to be a big bonus for the GR.

5 months agopermalink

@Otto: I have been using APS-C for years and I am considerung to "upgrade" to FF. Following your reasoning.. wont that give me more possibilities?

5 months ago*permalink

DingieM

I also think the ability to have lower hase ISO is important. Even low ISO modes of my e-m5 (100) resulted in photo's where shadow's could be recovered much better.

5 months agopermalink

tinternaut

CaPi, In perfect light, the differences are marginal. However, a larger format will always deliver an expanded shooting envelope if we assume equivelent sensor technology is in play.

5 months ago*permalink

SylvainP

I dropped aps-c and went m43 and FF for that reason. In some scenario a FF is really useful but for me it's only a very small portion of the time.

I won't complain about improvement on small sensor since it mean smaller lighter lenses for me. I waited for m43 to be usable in iso 1600 and 3200 to move most of my investment in this system and the last generation of panasonic and olympus 16 and 20mpix sensors are doing it for me.

5 months agopermalink

@silvayn: what FF did you get?

5 months agopermalink

It's not as simple. If you are limited by required DoF (for example you need to have three people in focus) so you are at f/2.8 on m43, you would need ~f/4 on aps-c and f/5.6 on FF. In that case you would benefit from potentially better DR on larger sensor unless you also have to raise ISO because of aperture change. In that case they would be mostly equal.
If you have more than enough DoF as it is - larger sensor will be better. Also, it's much easier to make a f/5.6 lens than f/2.8 one plus there are no f/1.4 zooms for m43. But that's why I said majority and hardly.

5 months agopermalink

"If you cannot get compelling results with one format, just switching to another will rarely help."

I see. That is why all sports photographers shoot with m43.

5 months agopermalink

@CaPi I have been using APS-C for years and I am considerung to "upgrade" to FF. Following your reasoning.. wont that give me more possibilities?

Well for example if all you do is shoot cats you'll get a bigger file of those cats vs getting features that allow you to expend photography like say a Tilt/FAS screen, smaller weight and foot print which fits into smaller lighter bag and say less weight and less accessories so you spend more time with the camera

But if all you want is bigger files go FF nuits...

5 months agopermalink

Man, wish the ignore list worked in comments section too... Well, have to in full manual.

5 months agopermalink

Revenant

"I see. That is why all sports photographers shoot with m43."

Actually, the IQ of m43 is perfectly fine for most sports photography and photojournalism, where critical IQ is rarely the primary concern. It's mostly about getting the shot, and delivering it to your employer before anyone else does.

So, if there was a m43 camera with the same features, performance and build quality as a D5 or a 1DX II, if the right lenses were available, and if the manufacturers offered the same kind of support as Canon and Nikon do, then I think many sports photographers would be happy to shoot with one.

5 months agopermalink

Checat

You really didn’t see a difference between N7 and A6300? Personally, I see a huge difference in all camera aspects (even between 6000 and 6300 was a big step). It makes me wonder if you really used cameras you mention ;).

5 months agopermalink

https://www.dxomark.com/cameras/launched-between-2002-and-2017/launch_price-from-0-to-45200-usd/sensor_format-sensor_micro43-sensor_apsc#hideAdvancedOptions=false&viewMode=list&yDataType=rankDxo

Checat, I was talking about RAW image quality. JPEGs improved considerably but if you expose similarly, RAW qualities are very similar. Sensor development from NEX7 to A6300 is not a leap forward, but is natural development in technology.
It has been different in m43: From EM5 II to EM1 II, I see real development (sometimes up to 1.5 stops), and 16-20MP is not a bad increase.
I think Sony is putting too much effort to speed the readout and developing FF sensors. 2018 should be the year that we see some real improvement in APSC sensors.
Or have we come to the limit for APSC?

5 months agopermalink

Dante Birchen

The best m43 in this list is right next to the Sony A580 which has the same sensor as the Nikon D7000 and Pentax K-5 (both score higher). It's just physics. Any improvements in m43 can also be applied in larger formats. There is no way to catch up.

5 months agopermalink

Wow, NX500 still top APS-C sensor? (and highest resolution to boot) Shame they pulled out...
Regarding progress - there was visible progress in every line of Samsung NX, from slow NX100 with 14MP to NX210 to NX300 to fast NX1/NX500 with 28MP, PDAF, 4k, etc. I imagine you could trace the advances for other manufacturers too (Fuji also went from slow lower res and useless video to faster higher res very decent video, for example).
Regarding breakthroughs - we are all waiting for usable global shutter (or fast enough electronic readout) so we can say goodbye to curtains, slow flash sync, rolling shutter in video (and stills), etc.

5 months agopermalink

Dante, my point is not "APSC cannot catch up", my point is "m43 has improved but APSC stayed nearly the same". Of course APSC will improve further (I guess).
By the way I am not talking based on charts and number, I'm talking from my real experience and real pictures that I have actually taken.

5 months agopermalink

Checat

ozturert- There is a huge difference in RAW quality (who cares about JPG?). Subjective- a lot better ISO, sharper image, better dynamic range, better color, e-shutter, to be honest, the only thing that staid the same is resolution… objective- for a lack of better option, just check DXO measurements, for example, dynamic range chart gives most useful information- on ISO 3200 you have 2evf gain, what do you expect in this time to improve? (m4/3 has similar gain/time)

So what I want to say in general is… Pretty nice theory (apsc stagnancy/m4/3 gain), but there are no facts behind it.

5 months agopermalink

"you'll get a bigger file"

because as we all know, a larger sensor makes a bigger file...

5 months agopermalink

Thanks for your feedback @all!

5 months agopermalink

SylvainP

@capi a K1

5 months agopermalink

SylvainP

@dante are you using the overall score to compare ? this include resolution in the comparison and some weighting on the numbers to be able to rank but instead compare the 3 main rating like the dr, noise etc and the em1 mkii is close to the recent aps-c while offering some advantages like better ibis, lighter shutter, smaller lenses etc. It may not be for everyone but the difference in real life is not what lots of people think it is.

A friend of mine shoot with a 7d mkii and he was very surprised how much more dr and lower noise at higher iso I got shooting side by side em1.2. My previous panasonic G85 was also better then the 7d mkii. When it's time to work on highlight and shadow the 7d mkii is not even close.

When I got the fuji xt-2 I had hope it would be a big difference between it and my G85 and ended up returning the fuji. Not that the fuji was bad but I wasn't getting much more while losing on other aspects.

5 months ago*permalink

@Revenant: Are you trying to say that you cannot get compelling results with m43 and sports photography for whatever reason? Because this is what I implied as well.

5 months agopermalink

SylvainP, I used 7D Mark II as well. At ISO6400 and above 7D II was better than EM1 II but below EM1 II was better. AF-C on 7D II was slightly better (especially when there is low light) but I don't take birds or sports much in dusk or evening. At good light, EM1 II's AF is really very good. I sold 7D II and kept EM1 II.
Now I'm testing 6D II. EM1 II's AF-C with 12-100mm f4 Pro is looks slightly better than 6D II and 24-70mm f2.8 L II, on the other hand Canon combination can focus in near darkness (AF-S).

5 months agopermalink

Dante Birchen

Close but no sigar. Als long as there is a cheap ass D3400 that almost tops the rankings and easily beats any m43 chock-full of noise reduction measures it shows physics cannot be beat. You can throw a lot of money at it, but you can also get a larger sensor for better IQ. I prefer the latter because it is more effective. When you are on a budget Nikon, Fujifilm and Pentax have some great offerings. For a little more you can now get a Full frame Sony A7.

5 months agopermalink

D3400 has a good sensor, but horrible body with miserable AF. You can't get it all, unfortunately.

5 months agopermalink

Dante Birchen

We weren't talking about that. You can get horrible bodies and lame AF in mirrorless m43 and aps-c DSLR/mirrorless alike.

5 months ago*permalink

OK OK I give up :)

5 months agopermalink

:) I own a D3400 - do you have a recommendation for travel with better AF and maybe a bit more iso capability?

5 months agopermalink

D500? D750?

5 months agopermalink