GH2 v1.1 new firmware with 25p discussion topic

Vitaliy_Kiselev

@mpgxsvcd

I am getting better now.
Working on porting each day.
But due to the complexity of the patches and their number this is loads of work.

stoney

I hate being sick but I caught a cold. Hope you feel better. Drink Vodka :)

spacewig

Has anyone confirmed that the new & improved NR is actually worse than on FW 1.0? That's what the video in VK's post on the first page seems to show quite clearly.

Grazzi!

fatpig

@spacewig:
it was unusable before anyway. if its the same: who cares. if its worse: who cares. if its better: yay!

just shoot intra with NR -2 and you will have great results with neatvideo

ghostlayer

Here is some tests I have made using 1.1 GH2 update :

ghostlayer

@spacewig : As concern NR I find it much improved ! Footage shot à 2500 ISO seems useable (unlike before)
(NR set to 0) I think we can't judge from footage posted on Youtube, Vimeo Video quality is much better/reliable.

spacewig

@ghostlayer

Thanks for posting your test. The Video I was referring to that's on page one has a segment (between 1:23 & 1:30) specifically comparing the noise reduction at 2 different ISO settings between the stock and updated firmware. Judging from that alone I think anyone would come to the conclusion the update is for the worst (NR wise), though I am heartened to hear that some find it better as I can't see the harm in the camera having a better base NR to be further tweaked in post. I guess I will have to test moi-même.

ghostlayer

@spacewig
:) Yep judging the test on page one, the new 1.1 raw files seems the most ugly :) that's pretty amazing to me.
Anyways GH2 video mode is not raw but rather similar to jpeg so raw file only matter for photos.
We should make a complete ISO test from 160 to 3200, we may notice that the "bad ISO" have changed and the 6400 from the test may be one of them? (don't know if I'm right or it make sense)

spacewig

@ghostlayer

You're right, those are RAW photos, my mistake.

I already did an indoor test last night with SeAquake for my own use and I was planning on doing one tonight comparing the stock & upgraded firmware. I could post if anyone is interested though my knowledge of best subject for testing is limited (i just film my better half in front of the PC with dark background).

Is it OK to adjust shutter speed to compensate for brighter image, i.e. at ISO 640 start at 1/25 then gradually move up to 1/60 as the ISO increases (does the noise change at different shutter speeds for relatively static subject)?

ghostlayer

I think changing the shutter speed might be the only way to compensate the ISO rise (aperture change might be to distracting).
I even saw some tests with lens cap on (pitch black) i guess they apply a level correction above all to make it legible.

amateur

could someone make a simple video iso test but in 110% dark (with Lens cover) vimeo.com/30129774...

instructions :
Nostalgic (-2, -2, 0, -2), shutter speed at 50 and white balance at 5600K

milada

I'm uploading test...

milada

Noise test: Panasonic GH2 new firmware 1.1 HBR.
boosted mids

the original ungraded:
http://www.wupload.com/file/2617048392/Sequenza_02.mov

kanintesova

Guys, maybe start new topic on ISO/NR comparisons? Did you take into account the "ISO-bug" when doing above test @milada?

milada

no, i switched iso from 160 to 3200...

TraumManufaktur

A couple of shots from a cafe setup & interview, not graded or filtered

GH2 Firmware 1.1 HBR @3200 ISO

Smooth setting, -2,-1,0,-1 • Lumix 14mm, Canon FD 50/1.4, FD 35/2.0

Shot with a Cowboy rig

magnifico

@lucasAdamson +1
It is less crispy with 1.1 :-(
See first video with the christmas red tree. You can read letters "2011" on your left : compare. Compare it all.
I bought gh2 because image is SHARP !

We need a test here only 1080p24 1.0 and 1.1 with no hack (sorry VK your hack is improving too much quality).
Original MTS with same parameter and exposure, different darkness etc.
To check if at 24p things were changed...

May be a new thread...

Marco

Somewhere in this thread it was mentioned that the 25P HDMI out of the GH2 (FW 1.1) is useable. However, in another forum I have read that there are still the same cadence problems and that therefore the 25P HDMI out is not useable. Anybody who has done a reliable test about this?

@Vitaliy: Thanks for all your efforts of hacking the new firmware. It must be a very hard job. Did you find any changes in the new firmware (which Panasonic specifically did for you) so that it will be more difficult for you to hack it?

chef

@Gerald I agree with u...

nomad

@Marco
I tested 24p before with version 1.0 and it had a massive cadence problem with it's output in 60i.

When I tested 25p under 1.1 (obviously) there is a clean PsF coming out of the HDMI, i.e. 25 progressive frames distributed to 50i frames. There is no cadence at all, every interlaced frame clearly has the content of one single progressive frame. You just tell the NLE to use it as progressive and everything will be fine.

If you see any strangeness on a screen, check if it is really capable of 50 Hz.

danyblue

One issue of 1.0 is the 4 first frames of a clip without sound. Has this been fixed in 1.1? I can't see any info about this from Panny neither here...

Thanks

veloci

I found that there is now a problem with camera playback on HDMI for HBR recordings, videos plays smoothly, but the audio playback is intermittent, however, if I unplug the HDMI cable, video and audio plays fine on camera.

Marco

@Nomad
Thank you for your fast answer. This is the link to the forum post where flatlinepirat writes that:
"So I recorded with a BlackMagic Design Intensity, and the recorded file really showed
up as 1080 25p in Premiere Pro CS 5.5 and AE. But the usual problems were visible... dropped or black frames nearly every second, due to the crippled field cadence."

So is flatlinepirat wrong? I can't verify myself, because I don't have a HDMI recording device yet. Thanks.

nomad

Well, I have a Ninja, so I can't tell about Intensity. But if the Ninja records fine, there must be a different problem with his setup.
One important thing, though: my camera is a PAL model. If flatlinepirat has a NTSC model, he may have all kinds of cadence trouble with 25p.

ilyasolovievv

@magnifico

maybe focus was changed a bit
but it deffo looks less sharp

more tests please

Howdy, Stranger!

Categories

Tags in Topic

Top Posters