FF VG900 hands-on !

@photochris, if you want to complain about folk who make "blanket statements about brand X being garbage", it would be more convincing if you weren't in the very same thread using phrases such as "terrible" and "complete bunk" about the GH2. Just saying.
 
It (the EOSHD test) is not user error.

I just uploaded 2 new files straight from the camera:
http://www.palsomedia.com/vg900
DSC00101_boat_bridge.JPG
00031_boat_bridge.MTS
The bad aliasing is the same whether it's 24p, 60i, or 60p.
I included the full JPG just to show that it's not an issue with the lens or focus.

The other files in the same location are from my previous posts.

Thank you for your work was very helpfull...

Was a big lie from PR Sony info: VG900 have an advanced digital aliasing filter

Give it back to SoNEY :-)
 
Last edited:
@photochris, if you want to complain about folk who make "blanket statements about brand X being garbage", it would be more convincing if you weren't in the very same thread using phrases such as "terrible" and "complete bunk" about the GH2. Just saying.

In my previous post I specified the limited dynamic range, posterization in the highlights and the noisy/muddy shadow issues the GH2 has, also of note - the noise above ISO 800 or so is really bad and for me footage was unusable, 3200 on my 5d2 was far cleaner. I shoot in bright sun light a lot and the GH2 just lacks tonal gradation, for me the transitions from highlights looks so unnatural, posterization on people's faces and such. Blech. Put the sun in a shot and it looks like a video game, not what my eye sees. But it obviously works for a lot of people, so good for them.

Again, I'm not a mouthpiece for Panasonic -- I don't get pre-prod samples of cameras and I'm not trying to make money selling $30 GH2 guides, so its not in my best interest to consistently pimp the GH2 as superior while ripping Canon, Nikon and others. From his VG900 post talking about a FF sensor in a camcorder - "Frankly this attitude disgusts me and it damages the Sony brand for me."

I used a GH2 for awhile and dumped it because the 5d2 and even the Nex5n looked better to me. I don't want to overrun this thread arguing about someone's opinions posted on a personal blog, end of this discussion for me.
 
Last edited:
It (the EOSHD test) is not user error.

I just uploaded 2 new files straight from the camera:
http://www.palsomedia.com/vg900
DSC00101_boat_bridge.JPG
00031_boat_bridge.MTS
The bad aliasing is the same whether it's 24p, 60i, or 60p.
I included the full JPG just to show that it's not an issue with the lens or focus.

The other files in the same location are from my previous posts.

Aliasing seems like its always going to be an issue with line skipping cameras. The shot EOS posted wasn't even in focus.
 
It (the EOSHD test) is not user error.

I just uploaded 2 new files straight from the camera:
http://www.palsomedia.com/vg900
DSC00101_boat_bridge.JPG
00031_boat_bridge.MTS
The bad aliasing is the same whether it's 24p, 60i, or 60p.
I included the full JPG just to show that it's not an issue with the lens or focus.

The other files in the same location are from my previous posts.

Well the image certainly looks 100x better than the still from EOSHD, it's sharp. And yes, the aliasing is bad.
 
Aliasing seems like its always going to be an issue with line skipping cameras. The shot EOS posted wasn't even in focus.

Yeah it's the focus I'm wondering about. One thing I noticed when using the alpha mount on the vg900 is that auto focus is sometimes wrong. You can hold down the photo button to have it auto focus before recording, but when I tried that on a Zeiss 24mm prime lens the auto focus was sometimes wrong, and on my Sigma 50mm prime lens auto focus is almost always wrong. So I've been using manual focus now all the time as the auto focus just hasn't been reliable for me. The issue was more sinister with the Zeiss 24 lens I had tried. With my Sigma lens the auto focus is so wrong that's it's obvious, whereas with the Zeiss at a quick glance it looks like it worked ok and focused correctly, but then double checking through the viewfinder would show that it was actually slightly off. If the EOSHD user relied on auto focus then I can see why his shot came out as badly as it did. It doesn't really excuse Sony in my mind for not properly supporting auto focus on a-mount lenses mind you, but at least it would explain what we're seeing.
 
It looks like there will have to be another adapter with the SLT mirror like the LAEA2 in order for Alpha lenses to get proper AF, which is very fast and accurate with A-mount lenses. The LAEA2 will work on the VG900, but you'll be limited to crop mode. On the A99 (and other DSLR's I've owned) I've had to use the AF microadjust feature on my two lenses to get the AF to be spot on. When it showed lock, both of my lenses were slightly back focused. After a couple clicks of adjustment they're razor sharp. This is pretty common, not sure if you can tune it with the VG. Owning a m43, APS-c and FF camera, the detail a FF sensor renders seems to really magnify even the slightest focus issue. I had the same problem back in the day when I moved from a 40d to the 5d2, all my lenses initially looked soft until I corrected the AF in camera.
 
I was blown away by how bad the vg-900 looked against the rest in the eoshd test. I wonder if the aps-c crop mode was set to 'on'? I thought the eoshd test was a good one. i find Andrew's articles enjoyable to read and honest - it's obvious because he is not afraid to tread on toes and lower his chances of receiving test cameras and freebies. I get the feeling people who find his blog threatening are the ones who wish they had the get up and go and drive to make the effort to undertake in depth tests such as these. No hopers who get jealous and defensive of people who are unafraid to voice their opinions and break a few eggs while making an omelette. shame the results were as they turned out. I thought I wanted a vg-900 before that.
 
Hey guys I have a semi-noob question regarding manual focus and prime lenses on the vg-900. As far as technique goes, since on this camera the digital zoom on prime lenses just does a sensor crop and doesn't adjust anything on the lens itself, does that mean that zooming in with digital zoom on a prime lens, focusing, then zooming back out is an acceptable practice to get more accurate focus? I understand that this method doesn't work with typical optical zoom on lenses that aren't parfocal, but in this case since it's a prime lens and the zoom is purely a digital crop then it would work, correct? I ask because while I have a 2x magnifier on the external monitor that I use, I mostly use this camera in sunlight so I find I'm using the viewfinder 99% of the time to get my focus as usually the external display is just too washed out.
 
I tested the 2x digital zoom using my VG900 with my Pentax PF80-ED spotting scope and PF-CA35 camera adapter that has about a 1000mm FL and it helps to see the focus a bit better.

8170719353_5e982e392b_c.jpg
 
I tested the 2x digital zoom using my VG900 with my Pentax PF80-ED spotting scope and PF-CA35 camera adapter that has about a 1000mm FL and it helps to see the focus a bit better.

What creative style settings did you use to get the picture looking flatter ? :)
 
Aliasing seems like its always going to be an issue with line skipping cameras. The shot EOS posted wasn't even in focus.
The 5D3 and GH2 do not suffer from aliasing as bad as the VG900.

The EOSHD VG900 image is not out of focus.
I can confirm from my testing that it is soft and has problems resolving details.

I uploaded test files from the VG900 and GH2 to compare the resolution and aliasing.
I also uploaded a VG900 still image to show that the lens is in focus and that it can resolve tiny details.
The GH2 sharpness was set to -2 (lowest).
The VG900 does not have any sharpness settings.

http://www.palsomedia.com/vg900
test1_gh2.MTS
test1_vg900.MTS
test1_vg900_JPG_Full.JPG

Edit:
I also uploaded 5D3 vs VG900 test using the same lens (Nikon-mount Zeiss ZF 100 F2):
5d3_vg900.m4
 
Last edited:
The 5D3 and GH2 do not suffer from aliasing as bad as the VG900.

I can confirm from my testing that it is soft and has problems resolving details.

I've done more testing, some wide shots with my A99 are doing the same thing, but some are just stunning. Not sure what's going on. Seems like there are some compression issues. Sometimes. It doesnt always happen. Odd. The stills from the A99 have so much depth, they're just beautiful images. Sometimes the video really blows me away, sometimes it's just meh. I'm going to rent an external recorder later in December to test and see if its just AVCHD or something else.

This video shows great detail in the wide shots, sometimes I get this level of detail and sometimes it's not so great. Everything I've shot has been with the 24-70z and 70-400g so it's not the lens as these guys are using all Sony lenses. http://vimeo.com/53924162
 
I've done more testing, some wide shots with my A99 are doing the same thing, but some are just stunning. Not sure what's going on. Seems like there are some compression issues. Sometimes. It doesnt always happen. Odd. The stills from the A99 have so much depth, they're just beautiful images. Sometimes the video really blows me away, sometimes it's just meh. I'm going to rent an external recorder later in December to test and see if its just AVCHD or something else.

Is it possible the f stop and lens length are what makes the difference? I've only been using my vg-900 with a 50mm full frame lens, and all my footage so far has been shot at f 1.4 to 2.0. I bought the camera specifically for very shallow depth of field portrait type work hence why that's all I've been filming with it. Maybe that's why I'm not bothered with some of the issues others see, because maybe they are filming very wide with deep depth of field and that's what brings out the limits of this camera? I do notice that when people point out bad footage it seems to be either wide or telephoto with deep dof, both of which are scenarios I don't really film on this camera, at least not right now anyways.
 
I tested the 2x digital zoom using my VG900 with my Pentax PF80-ED spotting scope and PF-CA35 camera adapter that has about a 1000mm FL and it helps to see the focus a bit better.

View attachment 62946

did this lens cover the FF or was it used in crop mode?

it would be nice to see a few more tests with the same FF lens, but tested both in FF mode, and in crop mode
 
I have only used my VG900 in FF mode and the only APS-C lens I tested was my 10-24mm Tamron and it showed vignetting from 10-13mm and was mostly gone at 14mm or more. I also tested it with my smaller Orion EON 72 (430mm F6 ED ) scope.
8170750340_b8e9073c6e_c.jpg
 
which lens mount adapters do you have joe? i live over in queens and i have plenty of canon and nikon lenses to try out. if you make it up north at all we can bang out a bunch of tests.
 
Back
Top