Julian

Don't know why it's too bad.. I love the sensor.

 

My last GH4 test video (now I have to wait for the camera that I ordered... :)).

This time I changed the White Output Levels to 235

 

I used this card during my tests:

 

 

Didn't have any problem with 4K. Not sure if 1080p 200mbps works well for long recording times, haven't tested it.

I tested a number of cards (Sony 95mb/s read 30 mb/s write), all 'old' but fast. Pretty much everything worked fine for 4K recording.

Geez, I'm confused again... Looking at some of my clips on Vimeo and in Media Player Classic (and Premiere CC) I see differences both in the black and white levels. Now you mentioned it, I remember those old topics. I'm with Matt on not fully understanding the ins and outs of it...

 

But apparently the black levels are also a thing with Vimeo:

 

 

There's plenty of freaking native files out there.

I wonder if this is true for GX7 as well, which is supposedly using the same sensor.

 

This has been debated before, but I think it is not the same sensor.

GX7 = Panasonic made.

 

I have strong reasons to believe the GH4 is Sony-based, like the GH3. Panasonic Europe told me it is based on the GH3, and NOT the sensor from the Panasonic PDF. Panasonic Japan told Andrew something else though.

 

But...

I tried opening the raw files from the GH4, there is no raw support in Adobe Camera Raw. What I usually do, and what usually works, is change the EXIF and make ACR think it is an older camera. This worked for example with the D4S (change it to D4), and all the Canon camera's (1200D, change to 650D or whatever).

 

So, I tried changing the camera type to GX7 in the exif, it wouldn't open the raw. When I changed the model type to GH3, it would let me open the GH4 raw. I think this pretty much proves the GH4 sensor is of the same origin as the GH3. Which is supposedly Sony.

 

Anyway, I think camera's with dont have luma settings record with the industry standard 16-235 settings. But it'd be interesting if someone could try.

I think the latter, because you will be able to record more highlight detail. You just have to make sure you actually use it, in post. This does make me wonder if using 0-255 makes sense to get more shadow detail.

This might be obvious to some of you. It wasn't to me. By default the GH4 records in 16-255. So that's what I shot my video's in. Turns out, you have to do some stuff to recover the 'super whites', if you don't do it you end up with crushed highlights.

 

This guy took some original footage from Vimeo (yes, I'm flattered by the compliments I got on my cat ;)) and made a tutorial about it.

 

Awesome stuff, must see! The bad news, I have to re-edit all my videos and upload them again ;)

 

 

Also check out his website: http://www.florisliesker.nl/

This guy makes great stuff.

Indecisive if you can have flaresex? no way! :P

Just saying, not saying it because I think you are calling me a fanboy :P

 

I mean, I also 'need' (read want) dynamic range, low light performance, 1x crop... I just can't see a solution that is double the price (at least!) as direct competition. Yes, the A7S + external recorder might still be relatively cheap, but I'm not willing to spend that much money on it for what I do with it. If the A7S would have internal 4K and would cost lets say €2000, then I'd have to think about my next purchase. Now I don't and in a way, I like that simplicity ;)

It's still bewildering that 1/4 actually = 1080p!

These 4K files are not an easy task to edit in realtime.

You don't need a mac pro either. Like i said, no problems on my desktop that was slightly high end 3 years ago.

@Quirky: it's my personal opinion. For the GH4 I don't need an external recorder. I can record 4K internally. I don't need anything extra, €1499,95 gets me everything needed for 4K shooting. I can always buy a recorder later if they are widely available and cheaper for even better quality.

 

Even if the A7S body gets the same price - or a bit lower - than the GH4, the base cost for 4K shooting is a lot higher. Of course it's the closest competitor right now, but I don't have to doubt for a second. Even if they'd be both available now. Not because I'm a fanboy. I just want the best for the least ammount of money.

If you can adapt a lens to Nikon mount, you can use it with the right adapter on the Speed Booster (with Nikon mount). Just think of it as giving your camera a new mount.

 

Of course the lens must have a image circle that is big enough. But I don't think there is any Nikon lens that has a image circle that's too small for the BMPCC + Speed Booster. Even with the BMPCC speed booster you still have a ~aps-c-crop.

I think it's already shipping in the US. Some people already got it.

Do you still have the GH3? Would like to see a comparison of both regarding motion cadence. It's something I haven't been able to put my finger on, so curious if the difference would be obvious to me.

 

Congrats on receiving the camera! I'm sure you'll enjoy it.

Had to return my test sample today... :(

Already missing it. Going to get one ASAP :)

 

I'm curious how long it will take before there is a real competitor (same price range, internal 4K recording, hybrid stills/video). I can't even see the A7S as a direct competitor because it will be at least twice the price with the required external recorder, that only will ship by the end of the year.

 

Looks like Panasonic has a great year ahead.

Nope... you just have to frame 4K and do the framing/panning afterwards. Maybe you can figure out something with the zoomed view for manual focusing to compose your HD frame within the shot. You'd have to calculate which zoom level gives the right crop.

 

But it should be pretty easy to guess where your full HD frame is, as it is fits exactly 4x inside of the image.

 

/Edit: just noticed, there is a custom guideline where you can place two lines wherever you want. If you put them exactly in the middle have your 4K frame devided in 4x 1080p.

WTF, UK only, why?

It's a test sample from Panasonic Nederland. Testing it for www.digifotopro.nl. Early may it'll hit the shelves.

What are the full specs of your laptop?

I'm just saying that I don't notice a big difference with 1080p or 4K at reasonable bitrates. So maybe with the right software your laptop can handle it.

For instance if I had a p6 to nikon mount adapter, could I then use them on the nikon speedbooster?  

 

Yep, you can!

I'm confused. Is the Iscorama that small or is the Trump that huge? :)

In Premiere CC, I dont have any difficulty editing 4K 100Mbps files from the GH4. I hardly notice a difference with 1080p. Haven't tried grading in resolve yet, probably this will be a bit more intensive, will try soon.

 

I'm running a 3 year old desktop: Core i5 2400, 16GB ram, Intel 180GB SSD (programs), 4TB internal HDD (video files), Nvidia GTX660.

Great job with the tests, Julian.  That rolling shutter skew seems very minimal given the speed of travel, was it shot in 4K or FHD?

 

Cinema 4K

 

Hah. I immediately recognized the song.

 

I also found that song on soundcloud a year or so ago, favourited it and thought "This would be a lovely song for a video" :)

 

Hehe. I get everything from the Vimeo Music Store (Creative Commons, Attribution only). Always end up with the same artists...

Always a pita to pick some music. I know half of the tracks there by heart by now :unsure:

You are likely right about the technical aspect, but  seeing the pictures comparing the amount of light was very similar on the 50m f1.4 vs the 50mm medium format lens at f/4 to my eye. So unless the images are fake, there is something to his statement...

 

Saw the images. The Nikon is definitely brighter. Make crop of the middle and you'll see a big difference. It doesn't look like 3 stops brighter (it should be: f/4 > 2.8 > 2 > 1.4). It's hard to say from one test image, without the necessary parameters. Was the shutter speed / iso exactly the same? was the light from outside constant?

 

I agree with Rich, a MF lens will transmit more light onto the sensor, but that only changes the corner brightness. Anyway, would love to be proven wrong with a proper test :)

 

Btw, MF lenses don't come with fast apertures. So yes, if you get the 50mm f/4 Flektogon and use it wide open on a fullframe dslr, it won't vignette. But a 'normal' 50mm f/1.4 or f/1.8 won't vignette either at f/4. Same goes for 80mm's... A 85mm 1.4/1.8 doesn't vignette much when stopped down to 2.8.

 

Anyway, not saying you should not go for these lenses, but you shouldn't be doing it for the brightness...

Re the amount of light, i'm sure you are right that the fstop doesn't magically change, but here is a great write up on theses lenses by a dp, he shows a side by side comparison of a 50mm nikon f41.4 wide open and a Jenna f4 wide open and the amount of light is similar. So there must be something to his statement "When you look at the numbers, they don´t seem to be real fast, but don´t be fooled. Because of the huge diameter of those beasts they are almost twice as fast as the numbers suggest.

 

Sadly the comparison images seem to be gone. Anyway, it is simply impossible. That DP can't beat physics either. Yes, they gather more light, but they distribute it to a bigger image circle because they have to cover medium format film. So you will loose that light. If there would be some kind of speed booster for medium format, you could bundle that light onto a full frame sensor and actually win an F-stop.

 

On the other hand, a 50mm 1.4 on a full frame camera has a lot of light fall of in the covers (vignetting). A medium format lens on a fullframe sensor wont have much light fall off wide open because you are only using the middle of the image circle (which is the brightest). So maybe that's the difference he's seeing. Claiming f/4 is like f/2, that is a two stop difference. A f/1.4 lens might have close to 2 stops light fall of in the corners, so there you'll probably see a difference, but that's gonna be it.