Response to Cinema5D GH5 Criticism
Photo Moment - March 02, 2017
A rather unpleasant article was released about the GH5 by Cinema5D yesterday, and we're gonna talk about that. Plus, I'll clarify a mistake from yesterday's electronic vs mechanical shutter discussion.
PRODUCTS MENTIONED IN TODAY'S PHOTO MOMENT (MANY ARE AFFILIATE LINKS):
Panasonic LUMIX GH5: [B&H - https://bhpho.to/2kQ9LT3 ] [Amazon - http://amzn.to/2lE4X1U ] [Adorama - http://jal.bz/2lNYpQv ]
VIDEOS MENTIONED IN TODAY'S PHOTO MOMENT:
GH5 Bit Depth Difference Electronic vs Mechanical Shutter?: https://youtu.be/IfH07v5Tpwc
OTHER LINKS MENTIONED IN TODAY'S PHOTO MOMENT:
Cinema5D GH5 Article: https://www.cinema5d.com/panasonic-gh5-lab-review-10-bit-flaws-vlog-free…
EOSHD response to Cinema5D: http://www.eoshd.com/2017/03/cinema5d-slate-panasonic-gh5-call-v-log-10b…
Check out my unbiased demo of the GH5! AF used for whole video... https://www.youtube.com/watch?v=HUm49kFxmMw
Hi Joseph, what's your opinion about image quality in 180fps? According to cinema 5D that was crap too and looks like 720p rather than 1080p. Their article seems to be offline now btw.
+Marco Valk Their article was pretty flawed it seems. Judge for yourself and look at my sample footage https://youtu.be/ylURMDgOliY
Hi PhotoJoseph, any follow up on the Cinema5D and Panasonic?
+SwitchBack Events I haven't seen it if there was. Anyone else?
Simply love your videos :) Regards from Denmark :) Waiting for my GH5 to come and love the videos to be prepared for it
This is off topic, but I don't remember seeing anyone evaluate the in-camera audio for the GH-5. Same as GH-4?
+Todd Hanson Sean had this question in part 2 of the interview. Yes, essentially the same.
...it's a bit beyond my scope. OK got it.
Cinema5D has never provided anything of value. I'm surprised anyone pays attention to Cinema5D's self-serving garbage.
I don't want a cake that isn't ready yet. So I will wait for the GH5. Even though I was really bummed yesterday when the camera store where I ordered the GH5 shipped the Leica DG Vario-Elmarit 12-60mm "GH5" kit lens but not the camera. ;-) (because it hasn't been released in the Netherlands yet)
Yes, it is a little bit torture. But on the other hand I see it as a positive sign that the new Lumix G lenses have started arriving from the factory. Which probably means the production of the GH5 body isn't far behind.
+Floris You'll have the lens but no camera?! Torture!!
Which site is showing ship date of 4/4/17? I ordered mine from Best Buy and like B&H, shows will ship By: 3/30? Will some places get the GH5 sooner then others?
I think the GH5 is being released everywhere in the world at the same time. But what I gather from my supplier (one of the biggest in the Netherlands) Panasonic hasn't given stores an exact date yet. So it's just an estimate from Best Buy and B&H just like the 20/3/2017 that is being listed on Amazon.co.uk Although some stores will get the camera sooner than others. As there seems to be already a waiting list.
Thanks for the video. I don't expect the GH5 to grade like pro res. I shoot mainly weddings and corporate gigs and the GH4 I have right now works great for my applications. Can't wait for the GH5.
You are such a panasonic fanboy it's sad. GH5 is not the second coming of christ. It's nothing but an overpriced MFT camera that does video somewhat well but still falls far behind with stills.
+masmole1 Alright alright, I accept. And the C5D tests were flawed. Badly. Unfortunately as another commenter has pointed out quite astutely, misinformed tests like that are dangerous because they are a respected name and reports like that have a habit of not going away. Should I stand by and say nothing? Their headline is practically click bait, and that shit works.
And to be clear… I am genuinely that excited about the camera. Not because I'm a Luminary but because it's awesome. I am not going to discuss any shortcomings on a camera that isn't finished yet. I think my "cake in the oven" analogy is fair. Criticism can come once a finished product is in people's hands. Until then… reserve negative judgement on things that may change.
And to be clear… I am genuinely that excited about the camera. Not because I'm a Luminary but because it's awesome. I am not going to discuss any shortcomings on a camera that isn't finished yet. I think my "cake in the oven" analogy is fair. Criticism can come once a finished product is in people's hands. Until then… reserve negative judgement on things that may change.
Joseph Linaschke Who says I'm so against the GH5 and mft in general? I own a GH4 and actually have a GH5 preordered. But I also shoot Sony, Red, and even Nikon for some studio work. I'm just saying you're a fanboy that's all. Yea yea I know you're a "lumix luminary" so I guess an official fanboy by default ready to come to the GH5's rescue when someone like Cinema5D happens to say some not so good things about it. Then giggling like a school girl when EosHD rebuts Cinema5D lol. You gush over this camera a bit too much that perhaps some blinders are on in the form of a strong bias to forgiving much of the GH5's potential but yet-to-be revealed shortcomings. Yes I watch your videos, because they happen to come up when I scour the web for more info on this GH5 I have preordered. And just because I'm calling you out as a Lumix fanboy, doesn't mean I don't like you or your fanboy videos. I enjoy them quite a bit. Just own it man :)
+masmole1 And you know this how… you've tested a production model extensively? Oh wait… no, you haven't. No one has. Because they don't exist yet. And before you excite yourself with your extreme cleverness of calling me a fan boy, read the description in every single one of my videos. Oh and if you're so against the GH5 or MFT in general, why are you watching my videos?
Hi Joseph, I've really been enjoying your videos - so much information, and at rate far higher than you can read online, and more current, too.
Since you are dealing with misinformation:-)
There seems to be some confusion around DFD. It has been suggested by users of one brand of m4/3 cameras, that DFD works by utilising the poor axial CA inherent in DFD capable lenses (and possibly other defects) which is why DFD does not work with other brand lenses, because those other brand lenses are optically, so superior and lack the defects that make DFD feasible.
Can Panasonic (ie, Sean Robinson, perhaps) address this suggestion, but also go into some detail as to how DFD does work; or point to any articles that don't just reproduce marketing materials with pseudo explanations? Would be appreciated, thanks.
Since you are dealing with misinformation:-)
There seems to be some confusion around DFD. It has been suggested by users of one brand of m4/3 cameras, that DFD works by utilising the poor axial CA inherent in DFD capable lenses (and possibly other defects) which is why DFD does not work with other brand lenses, because those other brand lenses are optically, so superior and lack the defects that make DFD feasible.
Can Panasonic (ie, Sean Robinson, perhaps) address this suggestion, but also go into some detail as to how DFD does work; or point to any articles that don't just reproduce marketing materials with pseudo explanations? Would be appreciated, thanks.
Thanks guys, much appreciated - keep up the good work.
+smrphotographyllc Thanks mate :-)
+jr hinde. So here is as straight forward explanation as I can give. DFD does not rely on "poor" optical characteristics. In fact, LUMIX & LUMIX Leica lenses are designed to some of the highest modern standards for optic design.
DFD uses all optical information available from our lenses. Simply put, we use things like OOF blur at any given focus point, optical sharpness, etc to determine where the lens is at any given focus point. That information combined with the new vector tracking in the GH5 produce a focusing & tracking system that rival most PDAF systems (In my opinion after shooting with the camera for a few months now). Cameras prior to the GH5 don't have the vector tracking system, so tracking will not be as good as it is on the GH5.
The reason we state that non Panasonic or Panasonic Leica lenses may not perform as well as our lenses is because we do not profile the other optics with the DFD system. Its as simple as that. Not because of optical faults, not because x-brand makes better lenses or not. Those notions couldn't be further from the truth.
I hope this helps clear up some of the misinformation being spread by some people.
DFD uses all optical information available from our lenses. Simply put, we use things like OOF blur at any given focus point, optical sharpness, etc to determine where the lens is at any given focus point. That information combined with the new vector tracking in the GH5 produce a focusing & tracking system that rival most PDAF systems (In my opinion after shooting with the camera for a few months now). Cameras prior to the GH5 don't have the vector tracking system, so tracking will not be as good as it is on the GH5.
The reason we state that non Panasonic or Panasonic Leica lenses may not perform as well as our lenses is because we do not profile the other optics with the DFD system. Its as simple as that. Not because of optical faults, not because x-brand makes better lenses or not. Those notions couldn't be further from the truth.
I hope this helps clear up some of the misinformation being spread by some people.
+rj hinde I'll see what I can come up with.
Scan line removal is Syncro Scan
Compression artifacts does exist in log gamma curve, the problem is the misinformation that C5D is spreading in a matter that not many understand, which they aggravate even further since many regard them as a source of information.
I have followed C5D for a while and every now and then they commit slips and mistakes like this, even more issues that require more technical knowledge, which is the case - there are often comments pointing out their mistakes. In my mind this is quite terrible, I'm not condemning mistakes but if you don't know something or you are not knowledgeable enough, just say so or ask for help, don't keep going and creating more misinformation.
Saying this is a color depth issue before even doing an external recording test to rule that out is just irresponsible. More importantly, this has all the signs of compression macro blocking, how could they not know it? This really put in check their technical knowledge, but again, the worst part is that there are already few places talking more in-depth about matters like color depth, chroma subsampling, codecs, type of compressions, etc. and then to spread misinformation, this will only makes matter more complicated.
Those that don't know much will take their word for it and keep spreading more misinformation. There are already plenty of bad tests of 10-bit 422, we don't need more articles perpetuating these mistakes. A guy by the name of James Miller that have been doing GH5 tests lately was quite more careful and did a lot of internal and external tests and he was able to eliminate these same compression issues and and even in 8-bit, proving it's not a color depth issue.
C5D should really do a proper test and correct themselves. But this also shows that people that need this level os result and pixel peeping will really take advantage of 400Mbps codec that should be based on the H.264 All-I Varicam's division AVC Ultra codec, a much more robust codec and then, as long as Lumix engineers can match it, there shouldn't be compression artifacts like the one found in the IPB 150Mbps codec.
I have followed C5D for a while and every now and then they commit slips and mistakes like this, even more issues that require more technical knowledge, which is the case - there are often comments pointing out their mistakes. In my mind this is quite terrible, I'm not condemning mistakes but if you don't know something or you are not knowledgeable enough, just say so or ask for help, don't keep going and creating more misinformation.
Saying this is a color depth issue before even doing an external recording test to rule that out is just irresponsible. More importantly, this has all the signs of compression macro blocking, how could they not know it? This really put in check their technical knowledge, but again, the worst part is that there are already few places talking more in-depth about matters like color depth, chroma subsampling, codecs, type of compressions, etc. and then to spread misinformation, this will only makes matter more complicated.
Those that don't know much will take their word for it and keep spreading more misinformation. There are already plenty of bad tests of 10-bit 422, we don't need more articles perpetuating these mistakes. A guy by the name of James Miller that have been doing GH5 tests lately was quite more careful and did a lot of internal and external tests and he was able to eliminate these same compression issues and and even in 8-bit, proving it's not a color depth issue.
C5D should really do a proper test and correct themselves. But this also shows that people that need this level os result and pixel peeping will really take advantage of 400Mbps codec that should be based on the H.264 All-I Varicam's division AVC Ultra codec, a much more robust codec and then, as long as Lumix engineers can match it, there shouldn't be compression artifacts like the one found in the IPB 150Mbps codec.
I like to see that my message on Twitter has paid off in video form, hehehe, well. In cinema5D I have already left the link to a video where it is shown how the camera records externally. https://vimeo.com/205130607
I like how 2 websites are talking about the GH5 and both came from "Canon" sites. EOS and 5D in name :D They didnt choose their name future proof
Thanks for the story of the Nikon, mistakes we learn not to do again. When my tripod fell with my GH4 on it, it was a quite windy day and I had the 12-35mm lens on with a variable ND and UV/IR cut filter on the lens and it fell onto concrete ground, straight down on the front of the lens. Luckily the lens was 100% fine with no damage because of the filters. The filters on the lens only had damage on the edge as I had a lens cap on at the time, but the step up ring I used was pushed in and was hard to get off the lens,thankfully I got it off the lens and didn't damage the lenses filter thread. At least I still have some time on my insurance time left on it before it expires so can get it fixed easily.
About V-Log L and Cinema5D, I think EOSHD are right. The problems with V-log L could be partially down to the bit rate and h.264 codec which can have macroblocking which can be fixed with noise reduction like neat video in editing or if you render to a codec with has a deblocking option. Also knowing how to work with its limitations can really help as pushing it hard can cause issues if the bit rate isn't high enough or the best codec is used. If Cinema5D had used an external recorder in 10-bit to test the V-Log L footage, then that could have tested V-Log L properly in a robust codec like DNxHR or ProRes at higher bitrates
I know many had issues with V-Log L on the GH4, but with my Atomos Ninja 2 and GH4, I found aside from the noise which made it hard to use in areas with dark shadows and some banding in skies (not always), it was quite a bit usable if exposed to +2 stops and from seeing V-Log L footage on the GH5 from Neumannfilms in an older pre production camera, it looked quite good to my eye and better than other profiles in both color and dynamic range.
About V-Log L and Cinema5D, I think EOSHD are right. The problems with V-log L could be partially down to the bit rate and h.264 codec which can have macroblocking which can be fixed with noise reduction like neat video in editing or if you render to a codec with has a deblocking option. Also knowing how to work with its limitations can really help as pushing it hard can cause issues if the bit rate isn't high enough or the best codec is used. If Cinema5D had used an external recorder in 10-bit to test the V-Log L footage, then that could have tested V-Log L properly in a robust codec like DNxHR or ProRes at higher bitrates
I know many had issues with V-Log L on the GH4, but with my Atomos Ninja 2 and GH4, I found aside from the noise which made it hard to use in areas with dark shadows and some banding in skies (not always), it was quite a bit usable if exposed to +2 stops and from seeing V-Log L footage on the GH5 from Neumannfilms in an older pre production camera, it looked quite good to my eye and better than other profiles in both color and dynamic range.
+Dave's Nature Productions I look forward to getting a GH5 again and shooting some more and eventually playing with V-Log too, which I've still never done!
I liked your N2020 story. I also had one in the eighties. Before I knew not to put anything fragile in my checked luggage, I put my N2020 in my checked luggage on our return flight to Seattle. When I arrived home the lens was snapped in two. Lesson learned. However when my GH3 fell off the tripod last fall, camera and lens were both fine- a real tribute to Panasonic workmanship!
+John Packard ouch! And yeah the Panasonic GH cameras are tanks, no doubt.
Do you know if we can expect some update for our beloved GX8 ? Some new feature, like olympus did with the em1. Will love auto iso in M mode and that fast fn button setting you talk about some video ago. And yeah, cinema 5d jump the gun pretty quick on this one...
+Private Blop No idea what is or isn't planned for the GX8, sorry. The long-press on the Fn button though isn't new; I just didn't know about it! Try it on your GX8. It should work!
Sorry, I mean auto iso with exposure compensation in M mode
Comments from YouTube