seku

Walking through the debayering and log curve in Excel is genius!

I need to give it a try as well, thanks for the great video!

Guys ... please don't confuse those talks about +2stops or +3.5 stops on the GH5s with dynamic range. Those reviewers just say that the GH5s is that much cleaner than the GH5 with ISO pushed. not DR.

As stated multiple times :

Vlog can store 14+ stops. EVA1 stiff.

Vlog-l can store 12 stops. end of it. for GH5 and GH5s.

But there is a difference of usable stops in Vlog-l and stored stops. A few months back i found Wolfcrow's IRE tests on the GH5 quite illuminating. I recommend watching the whole video before debating Vlog-l. And i can fully see the GH5s go deeper, pushing a lot more signal/noise out of the low IRE. (for the GH5, his tests show no image information below 15-20 IRE, valuable detail above 30 IRE). If the GH5s can push deeper, you will get your enhanced DR, but still in the 12 stop envelope.

I am eagerly awaiting his tests.

18 minutes ago, sfc said:

Better than EVA-1 in Vlog? 14 stops of DR? No IBIS because it jitters on tripods even when it is off? 

Really does sound like a no excuses camera. Nothing at all like that outdated GH5 camera.

I wonder, why does that 14stops number get thrown around?

Last time i checked, Vlog-L maxes out at 12 stops of DR. That's why i thought it was such a bummer that they didn't include real Vlog. 

I will be waiting for some reputable DR tests ... i won't get up my hopes yet.

mh.... i would love to compare a gh5s clip with the improved highlight rolloff to @Sage's ghalex LUT. also a shadow region comparison would be nice

very clean ... but using a Metabones XL does feel a bit like cheating :D

yeah... no IBIS and worst of all, no full V-log...

I just read the techreport.com webcached article , and i see no mention of IBIS or lack of it.

43rumors mention it in their latest post, but that image looks weird to me... nothing like Panasonic would show (zoom in on the GH5s in that picture, it got Dummy written all over it).

Is there some information I'm missing?

56 minutes ago, Cliff Totten said:

Anybody that thinks that the GH5's MFT sensor does not compete with Sony's APS-C or full frame models....is lying to themselves.

the physical point is ... that theoretically they cant

But as you said, i really wonder what this Sony&Panasonc cooperation will bring to bear. (and we will keep hoping for that crazy dual exposure chip)

This is getting quite out of hand ... and i fail to see why.

Some people need AF, some don't. Please don't crap on a camera which is awesome just because it doesn't have the one feature you want. Buy a different camera.

Calling ML hacked crap is callous at best. They did a lot for low-budget shooters, including opening up 3k 21:9 RAW on the 5dmk3. I sold mine a few weeks ago, not because of autofocus, but because of lacking exposure tools. I'm grateful to ML, and will continue with them on the truly open Axiom camera.

This is also a hobby to me, i'm a one-man band. Right now i am looking at the GH5 or GH5s which seems like an awesome compromise of picture quality (higher res/non-RAW) and a bag of tools i sorely miss (waveform for correct expo on skintones, highlights, etc). Also vectorscope.

Of course i'm lusting after a C200. It got the RAW i got so used to. Also the exposure tools i want. But this is a pro-cinema cam, which boasts your coveted AF, pro-XLR sound, internal RAW, wonderful color science, high DR, and is good in low light. And i think the price tag is right. Getting such a package under 10 grand at 1.5kg is unheard of. That's the weight of the 1DC!

So the C200 is on my personal shortlist. So is the GH5 and GH5s, and  i can barely wait to hear more on january 8th

also remember that 1080p RAW is less than 1080p video in resolution... debayering eats resolution.

But ML RAW's forté lies elsewhere, as we know

as they say, the proof is in the pudding? let's just wait and see how this will play out

and invest in great lenses meanwhile. great glass never gets old

True, but Luke is no silly boy. If he posted pictures with a different resolution, we would be talking about a new sensor immediately. In his case, i would resize :D

In 3 weeks we will know more. I will buy the camera anyway. Be it a better DR/highlight treatment with the existing sensor, or  better sensor.

4 minutes ago, Jonesy Jones said:

This is not going to happen. 

Same sensor as current GH5.

Why not? Does it have to be the same sensor?

Rumor upgraded to FT5 now : https://www.43rumors.com/ft5-panasonic-gh5s-shoots-240-fps-slow-motion-movies/

heh. they don't loose much time, do they.

32 minutes ago, DBounce said:

I read that story. If true it would seem to cast some doubts on the speculation about a Sony STARVIS sensor being used.

It does seem close.

The rumor you refer to quotes : 

Quote

Effective pixels: 10,280,000 pixels

Which is really close to the 10289712 pixels quoted in the starvis documentation (all pixels scan, 4:3 ratio 3704x2778)

32 minutes ago, DBounce said:

Total number of pixels: 11,930,000

 

In the spec sheet too, it only mentions : 

17:9 : 4168 (H) × 2176 (V) : 8.93M pixels

4:3  : 3792 (H) × 2824 (V) approx. 10.71M pixels

Just for kicks, if i do 4168 x 2824, i get : 11.77Mpixels ... which is still not 11.93 ... but close 

@Don Kotlos : i didn't take into account the lens factor.

But just for fun, a short recap of distance vs head size (seems i am really bored tonight) :

if nothing else, this should remind us that open gate 4/3 anamorphic gets us damn close to 16:9 S35 in vertical field of view...

I had some calculation fun, with @Don Kotlos's excellent male head height figures...

Taking into account @Neumann Films picture of the crazy color graded guy from chin to upper forehead ... while knowing that it was filmed 30" away

This is what some easy calculations came up with (don't worry bout feet or meter, i just used inches in the calculation): 

Link to the Calculator

Which ... if our guy has an average head ... then this was filmed on a Micro 4/3 sensor, cropped to 16:9

Even if our guy has a BIG BIIIIG head (99th percentile), and even slightly cropped, he barely makes it into open gate 4/3 territory :

So i would really rule out any sensor larger than that.... S35 seems completely improbable.

Just in case, the calculation for S35 in 16:9 cropped to 4:3(best case scenario) with largest head :

Even with Mr. Bighead, we are getting further away from the 30" @Neumann Films mentioned.

I'll stick with the 4/3 rumor

I was trying to have some fun with the filename : 5a2d83138fdae_CrazyGrade.jpg.583bf8d5fe25b8a0f3936335a1b5e286.jpg

That clearly looks like hexadecimal to me ... i tried converting the hexadecimal parts to ASCII... just gibberish. Then i thought they could be hexadecimal color codes, but  the length doesn't match up (group of 6)

Damn, what a tease 

thanks for the comparison between daylight and tungsten. very educative

24 minutes ago, maxotics said:

Companies are NOT ethical, they're goal is to succeed in their business.  Net neutrality doesn't make companies ethical or non ethical.  Again, a faulty argument to me.  IF people really want net neutrality than the government should set up enough network so that every citizen is guaranteed a certain amount of bandwidth and any content creator can use that same bandwidth to provide content.  The current system sort of works, but you point out that much of the population doesn't have much choice in bandwidth.  My guess is they have what the cable companies and telecoms have already wired decades ago.  

So you are arguing that ISPs should be nationalised? Not a bad idea that.

Just to put things into perspective : 10 years ago, my country was rather slow internet wise. Then the government passed a law that forced the ISPs to upgrade their network, so that the country would be very competitive. i have 200/100mbit for around 50 euro now. unlimited (they don't care about anything below 3 terabyte/month... bandwidth is cheap).

Quote

I have a hotspot which I use for business.  It's part of my cell plan and I pay $20 for the device.  It gives me unlimited data, but 15gig fast, then slows down to 600K?  Why, because most who use it will end up watching Netflix

How is that an argument? You consumed 15gigs, and then they throttle you. They do that with everyone who consumes that much, not because others watch Netflix. Blame your hegemony of providers and their hold on the market. Re-establish competition, because right now it is a cartell. Or nationalise the stuff.

As you said, Net neutrality doesn't make companies ethical or non ethical. I concur. But net neutrality is/was a safety lock that prevented unethical businesses to do even shadier stuff. Which wasn't legal before. Now it is.

cantsin nails it with his post.

I cannot fathom how anyone on this site... content creators... can agree with their biggest showcases (youtube, vimeo, ...) being possible paywalled. Or they'll add a monthly limit. I also cannot see how deregulating an oligarchy of ISPs is going to do any good for customers.

Maybe i have that view because i work in IT. For myself i don't care, as i know my ways around ISP induced limitations (look up tunneling through https header packaged traffic to your own VPS) ... but frankly, you (the US) will loose so much. Not only in media creation, but also for new startups, who will find they'll eventually have to pay the big providers to even get speed or connection. That's where you are headed.

Hi Sage,

just a silly comprehension question ... why are there 2 LUTs? let's say i whitebalanced my shots with graycards, why does it still matter if i use the tungsten or skylight LUT?

i feel like i don't understand something quite important here.

I was wondering about that as well... I guess there are multiple solutions :

1. Two subpixel just expose for a longer time than the other, up to the max shutter speed defined, whereas the 2 other exposes way shorter. In that case, where the longer exposure blows out, you use the 2 short exposure subpixels to "recover" detail in the highlights. If this is the case, i kind of worry about motion blur artifacts, as we would be running different exposure times. As the sensor site talk about "Integration time"... i think this is what we will be getting.

2. I would kill for this : something similar to Magic Lantern  Dual ISO : Each pair of subpixels can have its own ISO set. No motion blur artifacts. and you "keep" the good signal (low iso for highlights, higher ISO for shadow detail). This only works if the sensor does not scale linearly with ISO amplification, like Canons. I do not know about how the Sony sensors behave.

3. Similarly, maybe there's an ND over a pair of subpixels? (ok, that is far-fetched)

Yep, it does. Thanks a lot for that link, will help me a lot!

don't take it too seriously, this is a fun speculation thread, with some good information interspersed

If there is no big announcement... i'll buy the GH5, or if i drink too much during festivities, i'll order the EVA1. And if you have a partner that would buy a cam for you... damn am i jelly.