Jump to content

Cinema5D slates the Panasonic GH5, calls V-LOG and 10bit "unusable" - They're wrong


Recommended Posts

I don't own XT2 but GH5 clearly have superior rolling shutter. Dynamic range isn't that big of a deal, and if you *need* that margin that both of them are "fighting" about you are looking at the wrong cameras anyway. Just grab the Alexa at the bottom and be done with it.
 

*edit* The artifacts could be related to the temporal noise reduction

Link to post
Share on other sites
  • Replies 188
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Commander Reid, following your dispatches all these years has been like visiting the front lines of the DSLR community where shooting truly magical cinematic images is within our grasp. You were the f

Cinema5D have put the Panasonic GH5 on test, with a few charts and have some harsh words for the new 10bit mirrorless camera. I don't agree with much of their criticism. Here's an alternativ

Some are more impartial than others, put it that way. When you have personal relations with a nice man or nice woman from a marketing company or a camera company, you're not going to risk burning

Posted Images

55 minutes ago, Ken Ross said:

I think the bottom line is that until someone has both cameras in hand, shooting the same subject material, we really don't know. We can have general impressions about a not yet released camera from some online videos, but we can't make definitive assessments on how camera A compares with camera B in some finite parameters. 

In general though, it's usually assumed a larger sensor will have a better DR, but let's wait and see. :)

Sorry to disagree with you, though your point is well taken. I do believe however, that three weeks or so from release, from viewing dozens of videos and tests, and from long experience with Panasonic cameras, as well as side by side tests I've done with the G85, in addition to comparing my own X-T2 footage with that of various reviewers and Panasonic ambassadors, not to mention the advantages of sensor size, that I can make a reasonable assessment of the image quality. Which is to say that the GH5 released at the end of this month will not differ radically from the one in reviewers' hands today. To say that I should completely disregard this experience or the numerous videos posted online by people as talented as James Miller or Mr. Neumann... well, you didn't say I should disregard anything, but actually, I don't buy the 'let's wait and see, maybe it will turn out much better than my wildest dreams' is just not my style. Illogical or not. :) And no, I'm not saying my footage is better than Mr. Neumann's, he rocks :) , I'm merely talking about image quality.

49 minutes ago, no_connection said:

I don't own XT2 but GH5 clearly have superior rolling shutter. Dynamic range isn't that big of a deal, and if you *need* that margin that both of them are "fighting" about you are looking at the wrong cameras anyway. Just grab the Alexa at the bottom and be done with it.
 

*edit* The artifacts could be related to the temporal noise reduction

Agreed, the rolling shutter rules on the GH5, and if I ever made fast camera movements, which I don't, I'd be sure to use the GH5. Disagree about DR, but in a polite and good-humored way. Just looking at the clips from my X-T2, the extra stop or so of DR is pretty darn noticeable.

@no_connection I'm not trying to put you on the spot, but what cameras do you own that you can even make such a statement? 

Link to post
Share on other sites

I don't own any cameras yet. I was super excited about the a6300 but you know how that camera went. Same for a number of other cameras. X-T20 maybe as I take still too, but RS is still unknown.

I don't disagree that DR is important, but it's so high by now that it should be good enough. And I doubt you lock the X-T2 to ISO 800 using F-log at all times. There are so many other things to consider than DR. Don't the built in profiles limit DR? so would using ISO200 right? But considering the image now looks awesome and super clean, that loss in DR flies out the window.

You may also have missed it but I made a point that DR on a chart is pretty useless and real clips is where it matters. But there is no scientific number for that.
DR tells little how highlight are handled and how colors behave close to clipping. G7 vs a6300 pool scene comes to mind. Don't remember what DR ether of those cameras have but I digress.

Link to post
Share on other sites

Here are three interesting links to augment the discussion:

This one is a pretty well done 4 way frame split between the GH5, OMD EM1 II, XT-2, A7s II.

Do us all a favor and download and view the "original" file before commenting.  I have my opinions on this but will wait for others to view before saying anything.

 

The others are two lovely short films from Martin Wellgren (Sweden, I think).  No flesh tones.  Using V-log in the second link.

LUMIX GH5 High ISO grading test [4K] - 

 

 

‘Winter’s Grip’ [GH5 10bit V-LogL footage] - 

 

Link to post
Share on other sites

Cameralabs clearly set the wrong black and white levels for the X-T2 footage in that test. Clipping the shadows. That also screws up saturation and color.
I didn't realize how fast the A7sII fell apart above ISO51200 due to noise reduction. Just look at that temporal motion NR. The dark spots just dancing around. Looks nice up to that tho.

Link to post
Share on other sites
7 minutes ago, no_connection said:

Cameralabs clearly set the wrong black and white levels for the X-T2 footage in that test. Clipping the shadows. That also screws up saturation and color.
I didn't realize how fast the A7sII fell apart above ISO51200 due to noise reduction. Just look at that temporal motion NR. The dark spots just dancing around. Looks nice up to that tho.

I didn't download the files, but the reason the X-T2 images are so dark is because of all the manufacturers, Fuji must overstate sensitivity the most (ISO 1600 isn't ISO 1600 and so on...). There are no fewer than three different methods Japanese camera manufacturers use to calculate ISO, it's not really standardized at all. I'm guessing Gordon left the camera settings of all cameras at default, but yes, the dynamic range on the X-T2 is much better when shadows are at -2, highlights -2, but he wanted a level playing field. 

Link to post
Share on other sites

No it's definitely clipping of black level. You can see it in the saturation of the wood or any dark object being over saturated or moved colors. Try clipping black level on the other cameras and you get same effect of oversaturation of the wood.

*edit* here I clipped the other cameras to the same level. See how the wood becomes more saturated.

cameralabs_clipping.jpg

Link to post
Share on other sites

Just looked at Martin Walgren's footage of his short Film Winters grip...this footage looks as good as anything I see from top cinema  cameras...thought I'd just toss in my totally subjective little "review" of a properly filmed Gh5... my review also maybe has a place in this thread, as it seems it' easily as subjective, as the one Cinema5D did. I did not blow anything up to 400 percent, as I will not be using the camera in the way, cine5D  seems to be intent on. The images  I wanted to see were right in front of me and in normal aspect completely satisfied my needs...amazing stabilization...flawless functioning, in obviously freezing temperatures, and I hope the clip I watched was shot on Leica 12-69....just confirmed in the comment section....so my conclusion...if you want a stunning image that grades well...shoots well, under trying conditions and a system that's extremly robust, the GH5 is the right camera for you!....if you want to use a camera the way cine5D does....perhaps I'd recommend the Canon C700...it little bit more expensive!!!...but when you blow it up 400 percent, the C700 really come into it's own!!!

 

Link to post
Share on other sites
On 3/3/2017 at 10:35 PM, no_connection said:

Cameralabs clearly set the wrong black and white levels for the X-T2 footage in that test. Clipping the shadows. That also screws up saturation and color.
I didn't realize how fast the A7sII fell apart above ISO51200 due to noise reduction. Just look at that temporal motion NR. The dark spots just dancing around. Looks nice up to that tho.

I think it's incredible that a camera can record in light that is nearly invisible to the naked eye. How often do you shoot at ISO 51200 anyway? Oh, that's right, I forgot -  you don't own a camera.

10 hours ago, dbp said:

Always difficult to make heads or tails of a camera at launch time. Hell, even my own cameras that I've shot with, sometimes I remain a little unsure of how I feel about them sometimes.

 

Can you be more specific, providing details? Because after only a spending a couple weeks with a camera, I have a pretty good sense of its strengths and weaknesses.

On 3/3/2017 at 1:12 AM, no_connection said:

No it's definitely clipping of black level. You can see it in the saturation of the wood or any dark object being over saturated or moved colors. Try clipping black level on the other cameras and you get same effect of oversaturation of the wood.

*edit* here I clipped the other cameras to the same level. See how the wood becomes more saturated.

cameralabs_clipping.jpg

There's no such thing as clipping blacks. It's called crushing the blacks. And the reason the image is so dark is precisely because he's using the same settings for all cameras, and at ISO 1600, the X-T2 is probably in actuality closer to ISO 800. 

On 3/2/2017 at 4:09 AM, Philip Lipetz said:

Off topic, but Jason Lanier's work in theory of computer interactions is amazing. I assume we are talking about the same guy.

If you bothered to Google Jason Lanier, you'd see we're not talking about the same guy.

Link to post
Share on other sites
On 3/2/2017 at 10:32 PM, sanveer said:

Dynamic Range tests have become something of a joke nowadays. They remind me of the lovely shoet story titled Many Moons written by James Thurber, where every guy in the kingdom believes himself to be smarter than the other.

Cinema5D rated the A7S at 12 stops of dynamic range, so rating the GH5 at 10-11 is actually pretty good. They probably remove 30% of the manufacturer's stated figures for the RED cameras. RED says 18 stops on the RED Weapon and Cinema5D will conveniently say12 stops. Or less. 

I feel, that they have a curious bias against Panasonic Cameras. I personally hate anything Panasonic right now,  since my GX85 fiasco refuses to end. But, for the price, there isn't a better camera manufacturer for Indie Filmmakers out there. Panasonic gets way too many things right. Having 1 stop less of dynamic range doesn't even matter. Cinema5D praised the Sony RX100 to no end. And yet there was no mention of testing of great cameras like the G7, GX85, G85 etc. They definitely have a bias against Panasonic.

And unlike me, they don't really gave against justifiable reason. 

They performed a test, they shared the results. No reason to give any justification. And the fact that they have not reviewed the GX85 or G85 is not evidence that Cinema 5D has a bias against anybody, from an epistimological perspective. :)

Edit: I should add that I have no bias whatsoever against Sanveer, he's one cool guy. But shooting with the G85 and the X-T2, whatever the difference is in dynamic range between those two cameras, one or two stops or whatever, it's immediately apparent when I compare files from shots taken on the very same day, the same time and under the same lighting conditions, and it's pretty dramatic - I don't need to enlarge the images 400% to see that the X-T2 kills the G85 in dynamic range.

Link to post
Share on other sites
On 3/2/2017 at 5:38 AM, cantsin said:

No. You probably mean Jaron Lanier. (It always seemed to me that Jason was cashing in on this easy-to-be-made confusion.)

No need to cash in on any confusion - his work speaks for itself.

 

Link to post
Share on other sites
34 minutes ago, jonpais said:

There's no such thing as clipping blacks. It's called crushing the blacks. And the reason the image is so dark is precisely because he's using the same settings for all cameras, and at ISO 1600, the X-T2 is probably in actuality closer to ISO 800. 

Crushing would be smashing them together. Clipping just cut them off, just like you can do with highlights. But terminology aside.
You are wrong. The reason it's darker is because they are cutting information away at the bottom. What should be an elevated shade of gray is now black. As you can see in the screenshot they are all the same brightness when this is done to the rest of the cameras. Except the sony that has a flatter profile. it has nothing to do with sensitivity or ISO. Cutting away at the bottom will darken the entire image and screw up colors.

Link to post
Share on other sites
Just now, no_connection said:

Crushing would be smashing them together. Clipping just cut them off, just like you can do with highlights. But terminology aside.
You are wrong. The reason it's darker is because they are cutting information away at the bottom. What should be an elevated shade of gray is now black. As you can see in the screenshot they are all the same brightness when this is done to the rest of the cameras. Except the sony that has a flatter profile. it has nothing to do with sensitivity or ISO. Cutting away at the bottom will darken the entire image and screw up colors.

My god, you're right! I have to stop smoking opiates. All the cameras tested have identical values at ISO 1600. 

Link to post
Share on other sites
1 hour ago, Philip Lipetz said:

Jonpais when the moderator starts to post sacarcasric comments about several forum members it sets the wrong tone. Are you sure this  is the example you want to make? I really hope that this just reflects a bad day. And I hope that you soon feel better 

 

 

If someone is going to claim that a stop or two of dynamic range is unimportant; if another accuses a website of bias without adequate proof; and another, without any justification whatsoever, attacks a working professional whose body of work would be the envy of many, you bet I'm going to respond.

Link to post
Share on other sites

I did not say that dynamic range was unimportant or that a stop or two does not matter. Buy if it's everything then the GH4 (before v-log) could not make anything useful (as it looses 2 stops not having v-log(taken from headline)) and it's a miracle that anything was ever made with lower dynamic range. My opinion is that it's not a be all end all parameter and I'm sticking too it, and I respect that you want and need all the dynamic range you can get. But why defend that stop or two when the Alexa just walks all over the rest of them combined?


When I see them messing up black level for footage I called it out, what more proof do you want than waveform? You can easily spot it in the original footage. And if they didn't mess up and the camera was indeed crushing the blacks, then that is even worse as it makes the camera unusable, and the fought over DR is just a bogus number. So telling me I'm wrong kind of dooms the camera for any use.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...