AF100 shows more aliasing and less resolution than Panasonic GH2 ?

I threw these original images into photoshop and did a huge unsharp mask and a nearest neighbor upscale so you could see what is actually going on with the detail.

af100gh2.jpg

This brings up a few questions in my mind. How is it that the GH2 seems to have almost no moire, it seems like upscaling and sharpening only brings out real detail. You can actually follow all of the lines at the right edge of the circle. In the AF100, the moire artifact pattern just gets brought out even more. It does look like there is a little aliasing in the GH2 though, but somehow it manages to have so much darn detail. Is it possible the GH2 is oversampling?

The other question I see is, is it possible the AF100 was shot interlaced? There seem to be interlacing artifacts in the AF100 chart, which are brought out even more in the modified image i attached. This could have a significant effect on the comparison, if the AF100 was shot interlaced but the GH2 was shot progressive. If it wasnt shot interlaced, then this might draw questions about what could cause those artifacts. Could be a codec problem.
 
Last edited:
nyvz said:
I threw these original images into photoshop and did a huge unsharp mask and a nearest neighbor upscale so you could see what is actually going on with the detail.

View attachment 29986

This brings up a few questions in my mind. How is it that the GH2 seems to have almost no moire, it seems like upscaling and sharpening only brings out real detail. You can actually follow all of the lines at the right edge of the circle. In the AF100, the moire artifact pattern just gets brought out even more. It does look like there is a little aliasing in the GH2 though, but somehow it manages to have so much darn detail. Is it possible the GH2 is oversampling?

The other question I see is, is it possible the AF100 was shot interlaced? There seem to be interlacing artifacts in the AF100 chart, which are brought out even more in the modified image i attached. This could have a significant effect on the comparison, if the AF100 was shot interlaced but the GH2 was shot progressive. If it wasnt shot interlaced, then this might draw questions about what could cause those artifacts. Could be a codec problem.
Click to expand...

All great points. The only way to confirm it is to retest it and upload the out of camera files. I have the GH2 if anyone wants to test it against their AF100. From the testing I have done with the GH2 and the samples I have seen from the AF100 I say there is no way the GH2 has aliasing than the AF100.
 
Faust,

Sharpening exaggerates false detail. You minimize this. But in the examples I referenced, Figure 1's false detail is essentially unnoticeable. However, turning up the sharpness, Figure 2, makes the false detail unacceptable.

The GH2 image also has aliasing that would be exaggerated by adding sharpening. The factory default settings for the AF mean nothing. As for why you like to shoot with detail on has to do with your preferences and workflow. You like sharpening, just like some people like diffusion. And if you can sharpen in camera to a look you like, then it makes no sense to wait to sharpen in post.

"Further lessening AF100 detail would equal to reach the resolution of a pixel-shifted imager, saying goodbye to FullHD and this doesn't make sense."

But you seem to believe that turning down the detail actually takes away real resolution.
 
Last edited:
nyvz said:
Could be a codec problem.
Click to expand...

I agree, I wouldn't assume that these people performed a bad/unfair test, although I second the call for a duplicate test. If anyone has a nanoflash that they could use to test the chart on the AF100 I think this would be a very helpful comparison as well. The GH2 is an amazing camera, loving everything I have shot with it so far.
 
mpgxsvcd said:
Personally I am skeptical about those results and I am a GH2 fan. I don’t believe that the GH2 has less moiré than the Af100. I sure would love to test it though.

So why don’t we test it? I have the test chart they used printed out in poster size from a higher end printer. I have the GH2 and I have finally found what I consider to be the highest resolution settings for the test chart.

If you want to meet somewhere since we are in the same place. We can test the two against each other and put an end to this discussion. We can upload the original files and let everyone else judge for themselves on their own displays.

The 14-140mm between 18 and 25mm mm is the sharpest m4/3s lens you can buy. It is even better than the 14-45mm and more consistent across the whole lens than the 7-14mm. I would love to test the 14-140mm, 20mm, and a few of the MF lenses I have on both the AF100 and the GH2. I think the results will be outstanding with both cameras.

Just let me know if you or anyone else is interested.
Click to expand...

This does not sound like a bad idea. I think I can make some time on Wednesday after work if you'd like. If you have everything set up, I can just bring my camera by, and we can use just your lenses to make sure we use the same lens on each.
 
I'm still waiting to hear Barry to chime in on this one. But ironically, I think I might pick up a GH2 as a B-Roll camera now :-)
 
Exactly.....there will be a huge write up also. He's the man. He wouldn't jump in here without having anything to back up what he's saying. So....patience is a virtue.
 
I still feel detail is VERY overrated. Otherwise the 5d would never have become the hit it has. Look at the whole package (including audio!), and then make your choice.
 
astigmatic said:
Left detail 0 Right detail -7
View attachment 29995


zeiss 35mmf/1.4 @ f2.8
Click to expand...
Same interferences among different frequencies on both circumstances.
Lowest Detail Level leads to more organic edges in the whole picture without actually solving any aliasing issue.
That's everything I was trying to point at.

Thank you very much, astigmatic.
 
Last edited:
BobbyMurcerFan said:
Well if that's the case, then it ain't detail. And I guess I kind of owe Faust an apology.
Click to expand...
Apology? OMG, there is absolutely no need for that, my friend, we're deep into an interesting discussion and the only thing I'd like to know is what makes the GH2 so better looking compared to a more sophisticated AF100 video-oriented imagery (in the German test).

P.S. I don't think that -7 Detail Level takes AF100 resolution away but, hey, with no edge enhancement applied, it blurs!:shocked:

Shooting electronic footage without a little bit of detail sharpening or contour extraction is not a common practice, unless you're shooting for a film transfer where you don't want to see magnified artefacts...
 
Last edited:
mpgxsvcd said:
I bet Barry is busy testing this now.
Click to expand...

Would be, but I picked up a horrific cold on Thursday and have barely been up and around. Just got through shipping stuff out today and while I'd like to get to this, I've been up long enough.

I have some ideas on how to perform an ultimate test for aliasing and moire performance, presuming it works. I did try a little real-world stuff a couple of days ago, bolting the GH2 on top of the AF100 and pointing them at various brick houses with tiled roofs at varying distances, and both performed extremely well. I wasn't surprised at the AF100, as I've done that test with it before and it passed with flying colors, but I was surprised the GH2 held up so very well.

Pointed them at charts, not together but separately, and -- that told an interesting story that I don't know how to relate yet. With the 7D and GH1, the chart story translated perfectly to real-world issues, and I could make aliasing happen at will with major moire patterns. With the GH2 and AF100 it's not nearly the same.

In the meantime, here's one picture from one chart that you may find interesting, it's a chart that measures luma and chroma resolution. It's called "The Wringer" and it's designed to wring out the most from DSLRs and cameras like that... this was with both units recording to the Nano. It tells a quite different story than the german site does. I did this test a few days ago for the HDMI recording thread, but figured you guys here might find it interesting.


GH2-vs-AF100-HDMI.jpg

(and I'm not saying the German site is wrong, yet... I'm just acknowledging that there's a whole lot more going on than can be explained by a black & white chart. I can easily get aliasing on the AF100 and GH2 on a b&w chart, but cannot make it happen in real life on bricks, which is the exact opposite of my experience with the GH1 and 7D. I don't know if there's some sort of sophisticated processing going on that detects and compensates for so-called real-world images, maybe there's something in there that uses color in some sort of way... I don't know. It's odd. But on B&W I can get the AF100 to alias like shown above, yet in a color chart like The Wringer, it looks gorgeous out to 800 TVL).

Here's the deal -- I don't care if something happens or doesn't happen on a chart, I care about what happens in the real world and what happens to people's footage. Up to this moment, I've always found a direct correlation, and was always able to reproduce the nastiness that a chart shows, in real-world footage. But with the AF100, it's not that way. And I don't really know why yet. Its performance on The Wringer is absolutely admirable. Its performance on a standard b&w chart isn't that spectacular. But its performance in real-world footage is better than its chart performance. Which makes no sense.


So I've got to figure out ways to make it fail in the real world, to translate this over. With the GH1 or 7D it was easy as pie, put on a wide-angle lens and stop the iris down and point it at a brick building and -bam- ruined footage. With the AF100, you can put on the sharpest, nicest, crispest lens possible, stop it down to the deepest DOF possible, point it at a brick building with a tiled/shingled roof, and -- it looks like a really sharp, nice presentation of a brick building with a tiled/shingled roof. It doesn't mess up.

So, I don't know the answers yet. It needs some proper in-depth testing to find out what's real.

And I'll get there... I'll get to it... I just feel like trash and staying upright is not the easiest nor wisest course of action right now. :thumbsup:
 
Last edited:
Barry_Green said:
Here's the deal -- I don't care if something happens or doesn't happen on a chart, I care about what happens in the real world and what happens to people's footage. Up to this moment, I've always found a direct correlation, and was always able to reproduce the nastiness that a chart shows, in real-world footage. But with the AF100, it's not that way. And I don't really know why yet. Its performance on The Wringer is absolutely admirable. Its performance on a standard b&w chart isn't that spectacular. But its performance in real-world footage is better than its chart performance. Which makes no sense.


So I've got to figure out ways to make it fail in the real world, to translate this over. With the GH1 or 7D it was easy as pie, put on a wide-angle lens and stop the iris down and point it at a brick building and -bam- ruined footage. With the AF100, you can put on the sharpest, nicest, crispest lens possible, stop it down to the deepest DOF possible, point it at a brick building with a tiled/shingled roof, and -- it looks like a really sharp, nice presentation of a brick building with a tiled/shingled roof. It doesn't mess up.

So, I don't know the answers yet. It needs some proper in-depth testing to find out what's real.
Click to expand...
Hmmm, the Wringer seems to tell another story and that's what I'd like to believe.
Perhaps colour informations lead to a different behaviour.
GH2 seems to show a darker gamma setting with more saturated colours too, AF100 has slightly more detail (and, actually, an aliasing interference begins from the left side of the AF100 image but it's quite unnoticeable).

Time (and real world shootings) will tell...
Thank you, Barry.
 
Faust said:
...While someone here is seriously thinking about a new GH2 as MAIN camera...:happy:
Click to expand...

I think alot of weekend warriors are. :) And rightly so. Most firepower for the dollar you can get. uses every lens under the sun, Hell it will even adapt to PL easily!
 
Yep, for many people it's a money matter, undoubtedly.

If only the GH2 had a better onboard codec...
 
Last edited:
You must log in or register to reply here.
Back
Top