New Firmware Out

Hey Barry Green,

Do you think that firmware could really fix the 720 alias issue? I think I remember you saying it was a hardware issue, and could not be resolved. Is this correct?

Just like, do you think if Panasonic wanted to improve the GH1 codec (not likely at this point since it's been so long), that they could do it through Firmware?
 
Barry will have an opinion I am sure, but until he does his new chart test we can't know for sure. But in my opinion it doesn't look like the firmware fixes the 720 alias issue from what I've seen so far.

As for the GH1. We've been pushing Panasonic for a firmware update to fix the poor codec since it's release in April / May. The concensous is that the GH1's hardware isn't powerful enough to fix the encoding issues as it cannot find time to encode B frames, and upping the bit rate might not be possible either with the current chipset.

It doesn't make much sense for either company to spend resources improving existing products, especially when they try to make them optimised in the first place. What we're seeing is the limits of the current hardware, and you'll most likely have to pay for a new model if you want to see any improvements. That makes more financial sense for the companies involved, as well as from an engineering point of view.

We're a bit spoiled with software updates in the IT world right now. It's optimistic to extend the same 'patching' mentality to camera hardware. This isn't the way this industry works. Manual controls on the 5D was a one off. Usually firmware updates simply fix bugs, not add new features.
 
Last edited:
gene_can_sing said:
Hey Barry Green,

Do you think that firmware could really fix the 720 alias issue? I think I remember you saying it was a hardware issue, and could not be resolved. Is this correct?
Click to expand...
We'll find out in about half an hour. But I expect no improvement.

Just like, do you think if Panasonic wanted to improve the GH1 codec (not likely at this point since it's been so long), that they could do it through Firmware?
Click to expand...
Probably not in a major way, but maybe it's possible for them to up the bandwidth from 17mbps to 24mbps... if it's a setting in the codec chip and they can change that setting, it might be possible.
 
I just updated...(I expect to be jumped on for this but..)...

Wow!!..cant talk for 720p but 1080p looks a LOT better...I have a 1080p sony bravia tv with a grill on the front that looked like some sort of dancing lightshow at a rave when I pointed the camera at it....post update it looks good (I can still get some artifacts but a great improvement)...then out to the garden to film brickwork..no crazy dancing pattern.:happy::happy::happy::happy::happy:
 
Ian-T said:
lol...I really hope it's not just some placebo effect.
Click to expand...

Exactly. Just a mental thing. That's why I'm just waiting for Barry's charts. LoL. I don't have my cameras anyway so I can't install.

>=T I don't get them back for another hour or so. GRRR~
 
I would not base my opinions on one person's test, could be flaws in their process. Barry's tests are something to consider but so are others people's findings.
 
test 109 moire.JPG

ok so 1 shot is darker..not the most scientific..but If it is placebo then I feel better with this chalk 1.09 pill.

there is still some patterning but it doesnt seem to be sustained..if u get me.
 
I really hope this test means something, but with moire just a slight difference in viewing angle, as obviously exists in these pictures, can make the difference between moire and no moire.



el presidente said:
View attachment 13365

ok so 1 shot is darker..not the most scientific..but If it is placebo then I feel better with this chalk 1.09 pill.

there is still some patterning but it doesnt seem to be sustained..if u get me.
Click to expand...
 
I think the biggest difference is visible at high iso's. I've shot quite a bit of 1280 iso footage and after the update this morning it looks noiseless and the color rendition is much better too.
 
Test has been run. Shot in "neutral" mode, with zero sharpening, using a Zeiss ZF 50mm lens at f/5.6. Tripod was locked down, absolutely nothing changed between the first shot and the second, other than the firmware. Shot in 720 mode and 1080 mode, then updated the firmware, and shot again in 720 mode and 1080 mode.

Here are pixel-for-pixel extractions of each shot:
7D-720-Trumpets-v107-v109.jpg


7D-1080-Trumpets-v107-v109.jpg



Conclusion: absolutely no differences detected. No improvement whatsoever in version 1.09 firmware, regarding aliasing or moire or pixel-binning or anything else. (nor was there claimed to be any, in the firmware release notes).
 
I stuck footage direct from camera 1080p before and after on the timeline in ae.... then scrolled through to find a similar shot...yes angle does make a difference... but with the 1.07 footage i couldnt find any "clean" shots.
with 1.09 it kind of "shimmers" rather than making a pattern.
 
Barry_Green said:
Test has been run. Shot in "neutral" mode, with zero sharpening, using a Zeiss ZF 50mm lens at f/5.6. Tripod was locked down, absolutely nothing changed between the first shot and the second, other than the firmware. Shot in 720 mode and 1080 mode, then updated the firmware, and shot again in 720 mode and 1080 mode.

Here are pixel-for-pixel extractions of each shot:
7D-720-Trumpets-v107-v109.jpg


7D-1080-Trumpets-v107-v109.jpg



Conclusion: absolutely no differences detected. No improvement whatsoever in version 1.09 firmware, regarding aliasing or moire or pixel-binning or anything else. (nor was there claimed to be any, in the firmware release notes).
Click to expand...

Thanks, Barry. Greatly appreciated.

I know Isaac said that the higher ISo's were cleaner. Any word on that? I'll try it as well to see what I get before and after firmware update. With two cameras it'll be easy to check.
 
Thanks again Barry. So enlightening.

By the way, are you going to do the rematch between the 7D and GH1 on the Chroma Du Monde and ToNi charts? Where you test the cameras side by side?
 
Hey Barry,

I'm not to familiar with the charts - What does the difference in the purple in the lower left of the picture tell you?
 
gene_can_sing said:
By the way, are you going to do the rematch between the 7D and GH1 on the Chroma Du Monde and ToNi charts? Where you test the cameras side by side?
Click to expand...
I'm gonna do that in a different way; I'm waiting for my Stouffer 14-stop chart to show up and I'll use that instead. When it arrives, yes that test will happen.
 
ScruffyLefty said:
Hey Barry,

I'm not to familiar with the charts - What does the difference in the purple in the lower left of the picture tell you?
Click to expand...
It tells you the camera is performing horribly. Because that's a black and white chart. There should be absolutely nothing but black and white. And all those wedges are made of (nearly) parallel lines. So when you see diamond patterns, and color smudges, and anything other than pure clean white lines, it means the camera has failed to actually accurately reproduce what the chart looks like.
 
You must log in or register to reply here.
Back
Top