Michael Ma

That rolling shutter at the 55-57 seconds in...Doesn't look good.

Canon is being Canon. No DPAF in 4K and ridiculous crop. I wonder how terrible the rolling shutter is. I'm surprised how much hype is around this camera that has useless 4K.

Everyone here I imagine would know the difference. But it sounds like you don't know the quirks of the GH5? All UHS-I cards labeled 95MB/sec, even the best ones, won't sustain at 400mbits/sec (50MB/sec) on the GH5 because Panasonic has made it that way. Panasonic also has set an arbitrary restriction that it won't do dual (redundant) 400mbits/sec recording on UHS-II cards unless it identifies itself as a v90 card. So the Sandisk 300MB/sec UHS-II card will do 400mbps on a single card, but if you have 2 of these cards on the GH5, it won't do dual (redundant) recording because the SanDisk 300MB/sec not seeing it as a v90 card.

Jon, by any chance, do you own 2 Panasonic 128GB v90 cards? The rumor is Pansonic v90 cards are the only cards that are actually recognized as v90 cards on the GH5 and thus the only cards you can do backup recording at 400mbps. Some SD card firmware programmer chimed in once at one of the GH5 facebook groups I visit. He said it needs a special hard baked flag on the card which many v90 card manufacturers haven't adopted. And since then, I've only talked to one person who have bought 2 Panasonic v90 cards to confirm this. I've heard many people say that 2 Angelbird v90 or 2 ADATA v90 didn't allow backup recording at 400mbps. My second curiosity is that, if Panasonic is willing to do that, are my 300MB/sec cards not being utilized by the GH5 in other ways? If you have some time to spare, can you test if raw photos are written faster on a Panasonic v90 card versus another non-Panasonic 300MB/sec card? The GH5 has the buffer for about 61-62 raw photos which will be written to my 2000x Lexar at about 100MB/sec. Give or take 5MB/sec. Would a Panasonic v90 card be able to clear the buffer faster?

Hopefully this video will show dissuade you from buying a 5D4 if you need 4k. From my personal tests, the rolling shutter is so bad you can even see the lag when doing a closeup on someone's face even when you're not panning. They compare the rolling shutter to the 1DX2 which is far better in 4K when it comes to rolling shutter.

You missed the part people have been using 8bit C-Log for years with very little issues with banding. Are you saying GH5 has better LOG because it is 10bits?

I think if you want video there are better options from Panasonic, Canon, and Sony. All these cameras could use canon lenses. I don't think D850 solves any particular problem in the video market. Maybe if they had DPAF and a fully articulating swivel screen, and clean 4K like the GH5, then it would offer a solid argument for people to switch their entire set of lenses with the blind faith that Nikon will keep up to date with video tech in the future. It's a risky leap of faith. I think it's a great camera for Nikon shooters and people who that aren't heavily invested in one system

If you remember the original launch of the 5D Mark II, it didn't include 1080p 24p. Only after a public outcry, Canon caved to the public demand and released a firmware update that included 24p. https://prolost.com/blog/2010/3/1/canon-adds-24p-to-the-5d-mark-ii-and-i-blame-you.html So looking back, Canon has been trying to limit video capabilities to be on their photo cameras, even on the launch of the 5D Mark II. But they cannot stop the momentum of the market demand that was forced on them, which they had a big part in creating.

From my observation, Panasonic doesn't have much to gain by putting arbitrary limitations on SD cards, and since it seems like they are on the side of the customer, I'm hopeful these will work fine when the 400mb/s update hits. I'm not saying people all should run out and buy the Lexar 2000x cards. But I do firmly believe Lexar, a subsidiary of Micron, is a better brand than a brand like Delkin, who seems to be the only manufacturer selling v90 cards at the moment, who is not even a part of the SD Association. I'm posting information for people who already have the 2000x cards who've ordered a GH5, pre-buying other slower cards just because someone has said they don't work. I'm just saying not only do they work, but they work well, so they can save their money buying slower cards for the GH5.

Thanks for the hack! Works great on my GX85. Just for kicks, I tried it on my GF7, and it had the very opposite effect. The contrast felt like it was turned to 11, and all shadows were nearly black while half the sky was clipping to cyan.

I'm using Lexar 2000x cards right now on the GH5 and they work great. I have a few of them including the 128GB and the 32GB ones. They all work great on the GH5. My tests say that when they are freshly formatted, the GH5 writes 100Mbytes/sec (about 800Mbits/sec). And a fragmented nearly full card will still not dip below 70MBytes/sec (560Mbits/sec). They do read/write much faster on a computer, but those are the speeds on my GH5. I read the page where a guy tested a plethora of SD cards including the Lexar 2000x cards and reported that he got 35MB/sec. Definitely no problems like that on my side. As a matter of fact, my 2000x Lexar's seems to outperform every card he tested according to his results.

Where the director wants you to focus on when looking at the image rather than your eyes hunting for meaning, subject, bright spots of the image. When the subject or narrative is not clear, the eyes will wander looking at extraneous detail such as sharpness, noise, etc.

I really like this guy's videos. If somebody is taking a magnifying glass to the dark shadow areas the image, the real failure is the composition of the shot. Noise only becomes a problem when it steals away focus from the message. ISO 6400 is clean enough where it should never happen. At 12800, I think it the composition needs to be strong for things like b-roll.

It sounds great. Same crop with the GH5 with metabones XL. But if you don't already own the metabones, that puts the GH5 + metabones at nearly the same price range as the 5D4 if you go gray market. And I'm going to guess 8-bit C-LOG 4:2:2 is more or less on par, and maybe easier to grade than 10-bit V-Log 4:2:2. GH5 Advanages (even after 5D4 update) IBIS Fully articulating swivel screen Video tools such as focus peaking, waveform, vectorscope Legendary Panasonic Battery life (Canon's video recording battery consumption rate is quite high). 4K 60p (vs 3K 60p on the 5D4) No 30-minute recording limit Less rolling shutter 5D4 Advantages: DPAF Canon skin tones Full frame 1080p (unchanged) Full frame pictures Conclusion: GH5 is still a great purchase. I wouldn't cancel your preorder.

I'm going to roughly guess about 7 meters. I've attached the raw file if you're interested.P1410745.RW2 As a side note, I exported to jpg with Adobe Camera RAW, which adds 25 sharpening by default. But the blurry image was processed the same way.

Ah yes, 42.5mm! The repair guy did the same type of subjective test initially. If it wasn't for your assurance that it wasn't normal, I might still have the blurry center. Thanks!

Just an update. I finally got my Pan Leica 85mm 1.2 Nocticron back from Panasonic. It was a long process (took approximately 10 weeks). Initially, they did tests and thought nothing was wrong with the lens and on my second visit, I had to show them the difference with a floor model and how all the edges were excellent except the very center which looked fuzzy at 1.2 although it cleared up very quickly when stopped down to 1.8. Before. 100% crop at the center at f/1.2: Now, 100% crop at the center at f/1.2 Definitely very happy with the replacement lens. Maybe even slightly sharper than their floor model.

It's like you invite someone over to a prescreening of a movie that hasn't been audio mastered, and instead of telling you that there were issues with audio and hearing what you have to say about it, they publish an article to the world saying the movie is unwatchable. Then they make assumptions and state that the audio issues probably can't be fixed. Should anyone ever let this guy prescreen movies ever?

I would like to know how they graded the 10-bit image, and what software and version they used. I sure hope they didn't use a LUT for their "rough rec.709 grade". Almost sounds like it. If that's the case, it's most likely a major source of the problem. It does bring up the point if IPB could be inadequate for 10-bit 4K LOG. I would like to see further on this. I hope Panasonic sends Andrew a GH5 soon!

I would love raw. And it would make me rethink my GH5 order because raw makes 10-bit 4:2:2 look very compromised. But watching the Panasonic interviews, a reason why they don't implement things is that some features that people want causes overheating issues. Not sure if this would, but let's say Sony does allow raw recording. Looking at their track record, I would absolutely make sure RAW video doesn't cause some crazy impracticality issues, like overheating.

Ah, I missed that! I hope the firmware doesn't have adverse effects. Love it so far.

I'm surprised no one is talking about it because the update is a huge deal for the owners of the GX85/G85. And I should clarify, you can still hear it if you put your ear to the camera, but it is a huge improvement. I had to double check to see if the IBIS was still working. I've also been wondering how many people were going to be disappointed once they realized the GH5's stabilization makes as much noise as it does (assuming it would be as loud as the G85). I guess it won't be an issue on the GH5 which is fantastic news. http://av.jpn.support.panasonic.com/support/global/cs/dsc/download/fts/index2.html#dsc_dl

Thanks jon. I definitely would love this lens...actually, I don't think a day hasn't gone by where I haven't thought about getting one but there have been some patent filings for a Leica 25mm 1.2....so I'm going to hold out for a year or two. I know it really makes no sense image quality wise but I gotta build some anticipation in my life somehow.

I had my suspicions that a lot of the motivation to have a particular device rated, and not rate competitions on dxomark had to do with sponsorships.

This is always the problem with A and B comparisons. It is better not to look at them, unless you own the better camera.