GH4 Panasonic GH5 coming in 2017

brudney said:
I'd expect a Panasonic LUMIX Luminary to know more or less how to use the damn camera... :P
Click to expand...

It's pretty obvious to me - having grown up in Atlanta and coming of age long before it became an international city - that "Hillbillyinagriptruck" "Floop" is just playing his character to the hilt in his revealing "GH5 vs GH4 low light 800 12800 High ISO 8bit 10bit halfass comparison Rig Job". That's the role-playing which allowed some wiley, brilliant southern senators to simply sandbag their opponents. About Panasonic's granting him "Luminary" status and trusting him with a GH5, ipsa re loqitur, which I'm pretty sure Floop could translate for you!

Sure, the folksy, self-deprecating vibe it's off-puttin' to some, but the piece is well-done IMO, with final conclusions awaiting, as he says, the production firmware. I imaging that he's accustomed to working with much more pricey sytems by his comments about DR.

So where did you lose the Southern/Appalachian accent, Floop?
 
Last edited:
I wonder why current users are not testing 18 Mpixel video. It is the best GH5 video mode. It has more than double the image data compared to 4k modes.
 
Peter Berry said:
It's pretty obvious to me - having grown up in Atlanta and coming of age long before it became an international city - that "Hillbillyinagriptruck" "Floop" is just playing his character to the hilt in his revealing "GH5 vs GH4 low light 800 12800 High ISO 8bit 10bit halfass comparison Rig Job". That's the role-playing which allowed some wiley, brilliant southern senators to simply sandbag their opponents. About Panasonic's granting him "Luminary" status and trusting him with a GH5, ipsa re loqitur, which I'm pretty sure Floop could translate for you!

Sure, the folksy, self-deprecating vibe it's off-puttin' to some, but the piece is well-done IMO, with final conclusions awaiting, as he says, the production firmware. I imaging that he's accustomed to working with much more pricey sytems by his comments about DR.

So where did you lose the Southern/Appalachian accent, Floop?
Click to expand...
I think he's referring to Kai's video...I know I was.
 
I pre-ordered GH5 on announcement day, and I know that with the GH4 you had to purchase Vlog via snail mail. Is this still the case, or is it just an instant delivery serial code now?

I ask because I am wondering, if I want vlog on launch, do I need to pre-order vlog now as well via snail mail or can I just do it when the camera arrives?
 
philiplipetz said:
V Log is flawed according to them
Click to expand...

Don't need a "Lab test" to see that. Just look at any video on YouTube (even) and you see how the image essentially gets worse when people move from 8 bit to 10 bit.

And this is exactly the type of thing that keeps me away from this whole segment of cameras. I'm not sure this is related to pre production. This might quite possibly never get fixed.

I don't know... it feels like there are several camera brands out there who "mean well", but they just don't seem 'grown up' enough to get there. Or at least not critical enough about how they approach it all.

I hope I'm wrong and that the GH5 can deliver on all it's specs.
 
Last edited:
Personally my use of the GH5 is for 60P UHD, never mentioned in the review maybe because the only competing cameras are several times the cost making the GH5 a clear winner. Unless I get a Shogun Inferno I will record internally limited to 8bit at UHD 60P. I am sure there will be improvements in performance as times goes on this year. A7 is not on my list as it cannot do 60P UHD. If the only competing cameras are the FS7 and Canon costing many times as much I think the GH5 is a winner until Panasonic upgrades the DVX200!!! Look forward to getting mine hopefully the end of March.
 
stoneinapond said:
Jeezus wept, it's a pre-production version. They even keep saying it. Constantly. Yet continue testing against finished cameras. Lost some respect there....
Click to expand...

My optimistic side agrees and wants this to be true. My realist side is cautiously skeptical, as I've been around enough "betas" a month or so out from release to know how indicative a pre-production version of anything can be of the final. It just depends. If this is just a few tweaks, will definitely be fixed. If it's a fundamental issue in the encoder or sensor, this could be incredibely representative of final. There's no way to tell, and one month out from full launch is not very far. It seems something this central to the camera would be 90-95% worked out at this point in the process...

ade4all said:
yes the 10 bit should be greatly improved when it gets the 400 Mbit firmware update
Click to expand...

Honest question: will it be? I was under the impression the 400mbit was there to allow for Intraframe codec instead of IPB, with the extra bitrate going entirely towards encoding each frame. I'm not a pro here but my understanding is that this would solely benefit motion, and wouldn't seem to have any impact on the color science or other problems exhibited here?
 
filmguy123 said:
Barry, what do you say?
Click to expand...

No way to know unless I know what all the variables are. I do think it's silly to compare a $2,000 pre-production camera against a $5,000 DSLR, a $10,000 S35 cine camera, and a $28,000 cinema camera, and then be surprised when the camera that costs 1/3, 1/5, or 1/14 as much, doesn't do as well. I mean, come on.

But there are too many things we don't know. Why were the tones flattened? And why do they immediately blame V-Log? Frankly that's the kind of thing I could easily see happening from noise reduction, or from the recording format's compression engine. What format was it recorded in, at what bitrate?

People keep clamoring for 10-bit 4:2:2, but if you don't greatly increase the data rate you're recording at, you're just going to make the image look worse. 100 megabits seems to be about the bare minimum for UHD 8-bit 4:2:0 at 24p. You probably need a good 200 megabits to handle UHD 10-bit 4:2:2.

What would it have looked like to an external recorder? Maybe it'd be brilliant to an external, but the internal codec was smushing it? Who knows? No way to know.

The responsible thing to do is to immediately forward the files to Panasonic, instead of publishing it. I believe they said they did forward them, so we'll see what comes of it.

That said, like I said before, we don't know what we don't know. It's possible the guys who conducted the test did absolutely everything proper and professionally, and that there really is a problem. Or perhaps they didn't know enough to consider other elements and eliminate those variables (like turning NR down or off, or recording externally, etc). Or maybe they were recording 4K/60p, in which case it's always 4:2:0 8-bit, and they didn't know that?

As far as I'm concerned, it's a data point, but not the whole conclusion. I'd wait to see what Adam Wilt says about it when he gets an opportunity to put it through its paces, since I've known Adam for years and trust that he knows exactly what to do and will know how to properly interpret what he's seeing.
 
Barry_Green said:
But there are too many things we don't know. Why were the tones flattened? And why do they immediately blame V-Log? Frankly that's the kind of thing I could easily see happening from noise reduction, or from the recording format's compression engine. What format was it recorded in, at what bitrate?
Click to expand...

I think in the Cinema5D article they said their GH5 tests were recorded in 10-bit 4:2:2 at 150 Mbps (H.264), although I'm not sure they mentioned the frame rate.

In any case the cameras they compared it against are both capable of recording in much higher bitrates internally. The FS7 is capable of recording at up to 300 Mbps (H.264 at 30fps) or 600 Mbps (H.264 at 60fps), while the C700 is capable of recording in ProRes, which clocks in at 880 Mbps for 2160p30 in ProRes 422 HQ.

From these limited tests, it was very difficult to determine how much of what they saw was due to inherent limitations of the camera, and how much was due to choice of recording mode, codec, camera settings, and other test parameters. I probably wouldn't draw too many firm conclusions based on this set of tests alone.
 
They definitely should have compared the internal to the external recording, thus eliminating the codec as the variable.

To make a wild guess, they probably didn't expose properly, then pushed the grade too far and thusly generated the perceived problems.
 
davedv said:
From these limited tests, it was very difficult to determine how much of what they saw was due to inherent limitations of the camera, and how much was due to choice of recording mode, codec, camera settings, and other test parameters.
Click to expand...
Cinema5D's main complaint was the color averaged macroblocks. That's a function of the h264 codec. You hit the nail on the head with the data rates: a 150Mbps signal is gonna be way more compressed than an 880Mbps signal! Great sleuthing, guys!
 
Cinema 5d was critical because he compared it to the c700 and fs7 - aside from dynamic range, gh5 was better then a7s ii. I was VERY surprised that he said to get the a7sii instead of gh5.
 
You must log in or register to reply here.
Back
Top