Official Sony NEX-VG10 Thread

[trez]

I've not seen anything that proves the NEX-5 has less rolling shutter or better low light performance than any of the Canon DSLRs.

Check this out:
http://www.luminous-landscape.com/reviews/cameras/sony-nex5.shtml
They say:
"My colleague Chris Sanderson and I spent a morning shooting side by side comparisons against the Canon 7D. Chris is a life-long director / cameraman / videographer, and has a professional's eye (...) One area where Sony seems to have the Canon licked was in terms of "jello" or the rolling shutter effect. Rapidly moving subjects or camera motion produced hardly any bending, while the Canon had quite a bit

I very much doubt a $2k camera is doing a full down sample of the entire sensor. For that price point, it will more than likely pixel binning, just like Canon and just like Panasonic.

I wish it did downsample, but I doubt, too. But who knows. Computational complexity of debayering process is similar, if not greater, yet all single sensor cameras do it in real-time. These chips are really fast.

The footage didn't prove anything to me. I've seen enough footage from the D90, 7D, GH1, 5D that were amazing but that they don't push all the cameras in all ways are far enough to really explain what the expect.

Not sure. I've been able to find the moire artifacts in most 7d footage on the web. I didn't spot them in VG10 footage, which doesn't prove anything, of course. We have to wait to be sure. However, I'm pretty sure they optimized the OLPF for video resolution, so we can expect much less moire. The OLPF design is not expensive - in fact you have them in every single camera on the market. It's just the matter of tuning their response properly. Why wouldn't they do it right?

I wouldn't expect manufacturers or those they work with to exploit the problems. Until this camera is in the wild, we wont know.

Yes, but it doesn't mean we can't talk, expect things and believe people who had the prototypes in their hands... Don't get me wrong - nobody said there's no (very little) moire so far. But just because it's video camera, we can expect it's optics to be tuned for video, including OLPF. This is what Panasonic will do in AF100, too - they reduce moire by OLPF (this is confirmed)

Pertaining to low pass filter, I've not seen an electronic implementation, can you explain how this is suppsed to work? An optical low pass filter, like the one used in almost all cameras, diffuses (or duplicates like a birefringent filter) the image across the pixel sites to spread out the high contrast spacial information into surrounding sites so it doesn't get lost in non-active areas.
If you tried to do this after you become electrical, the information is already gone. If you mean by pixel binning, this is a very poor way to down-convert, period.

This is (almost) true. The better definition is that the OLPF cuts the (optical) MTF as it approaches the sensor Nyquist. It's task is to filter out the spacial frequencies (details) that cannot be sampled by the sensor due to its limited spacial resolution. Those frequencies would otherwise alias. Keep in mind we (re)sample twice - first in the sensor and then when reducing the resolution to match the video raster size.
By applying OLPF which cutoff frequency has been optimized to match video raster size (no matter what real sensor resolution is), one can reduce video moire, even if in-camera downsampling is as simple as line skipping (decimation) with no additional filtering - the aliasing will simply not be introduced if there are no frequencies that can alias.
The other way of achieving similar goal is to use the digital antialias filter in-camera, prior to resampling. Of course there's still OLPF (every camera has it), but it is tuned for native sensor resolution, just like in DSLR. and not video raster size. This technique has some great advantages, but, as you know, it's complex.
In this context this two scenatios can be used to achieve similar goal - get rid of some details prior to (re)sampling to avoid aliasing:

lens->olpf (tuned for native sensor resolution)->digital lp filter->decimation
lens->olpf (tuned for video raster size)->decimation

I agree that pixel binning is very bad method of downsampling. The only reason it is used is because it's so easy. In fact, it does both digital filtering and decimation in one pass. Computationally very easy, but the filter kernel is rectangular (averaging from certain number of pixels), which is far from perfect in terms of response - poor rolloff and very significant side lobes (spectral leakage).

I would think the moire/aliasing of the sony will be more in line with the GH1. Using a stronger OLPF for the final target of 1080p will probably improve the picture from a moire/aliasing perspective, I agree,

This is exactly what I meant.

but if they use the same pixel binning, my opinion is it will look a little softer than the GH1 and 7D.

Not necessarily. The highest part of the MTF does not affect the sharpness at all. On Panavision site there's nice presentation on it. It's the middle of the MTF that determines the perceived sharpness. The highs only affect the texture perception, and, in my opinion, moire destroys textures much more than some gentle smoothing.

The only reason I think people are complaining about 24p is because it's one of the only things we don't have to wait and see that they didn't implement.

Don't forget they're preparing 'pro' inexpensive s35 camera. I fully understand people complaining about lack of 24p, but I understand Sony, too. As long as there's no real competition, they want to maximize their income...

 

[Rick Burnett]

Check this out:
http://www.luminous-landscape.com/reviews/cameras/sony-nex5.shtml
They say:
"My colleague Chris Sanderson and I spent a morning shooting side by side comparisons against the Canon 7D. Chris is a life-long director / cameraman / videographer, and has a professional's eye (...) One area where Sony seems to have the Canon licked was in terms of "jello" or the rolling shutter effect. Rapidly moving subjects or camera motion produced hardly any bending, while the Canon had quite a bit

If that is the case, people will be happy, but they also went on to talk about other problems with the video mode. Most of the complaints compared to the 7D are not things that would get better with an OLPF, like they mentioned it was softer. Further, I don't see how that proves ANYTHING about it being good in low light. In fact, looking at their sample images at different ISOs, I couldn't believe how much noise was present at ISO800. I can't speak for the GH1, but with the 7D I can use all the way up to ISO1600 easily, and ISO2000 when I am careful. The 5D and 1D are even better than the 7D. That said, I'd really have to test the camera to compare. It's also worth pointing out, most of their tests were with picture mode, not video mode. I've seen plenty of tests with the 7D where the dynamic range in video is not as high as in picture mode. I think a lot of it was contributed to the compression, so I can't say what we'd expect on the Sony.

I wish it did downsample, but I doubt, too. But who knows. Computational complexity of debayering process is similar, if not greater, yet all single sensor cameras do it in real-time. These chips are really fast.

That is true, I don't think the problem has been the computational aspect of it though, it's the actual process of getting the data off the sensor fast enough. It has two problems, amount of data and heat generated. Let's say you read every 3 lines of a sensor (after pixel binning has occured), you'll increase the skew horizontally 3x alone if you read them all. Imagine the 7D with 3x the skew as now!!!

Not sure. I've been able to find the moire artifacts in most 7d footage on the web. I didn't spot them in VG10 footage, which doesn't prove anything, of course. We have to wait to be sure. However, I'm pretty sure they optimized the OLPF for video resolution, so we can expect much less moire. The OLPF design is not expensive - in fact you have them in every single camera on the market. It's just the matter of tuning their response properly. Why wouldn't they do it right?

It's not a matter of 'doing it right' it's a matter of trade-offs. With a bayer pattern, the density of red/blue is half of green. If you pick a OLPF that is for the lowest density, red/blue, you soften your image more than you might want. If you go look at the resolution charts of the Alexa, you will see that even it has moire as well. Sure, they too could tune out the moire completely, but then the picture would be extra soft.

I am sure Sony has to make that same decision. It's why the red cameras shoot higher than 1080p so when you downsample from the full image, you can get rid of all the moire and aliasing. I've yet to see a native resolution bayer pattern sensor not produce moire or not be crazy soft.

Better than the 7D? Without question. The 7D is the worst (as all the Canons) because of the rainbow moire. I think it's a problem with their pixel binning and NOT their OLPF. Whatever they are doing, they need to rethink because it looks horrible. The moire in the GH1 is much much more pleasant. But still, I don't like it at all.

Yes, but it doesn't mean we can't talk, expect things and believe people who had the prototypes in their hands... Don't get me wrong - nobody said there's no (very little) moire so far. But just because it's video camera, we can expect it's optics to be tuned for video, including OLPF. This is what Panasonic will do in AF100, too - they reduce moire by OLPF (this is confirmed)

As I said, if you wanted to, you could definitely remove ALL moire in ANY video camera by a strong enough OLPF solution, but it comes at a price. How soft an image you want. I understand this, but given they will pixel bin, it think it will be on par with the GH1 if they want the same sharpness.

This is (almost) true. The better definition is that the OLPF cuts the (optical) MTF as it approaches the sensor Nyquist. It's task is to filter out the spacial frequencies (details) that cannot be sampled by the sensor due to its limited spacial resolution. Those frequencies would otherwise alias. Keep in mind we (re)sample twice - first in the sensor and then when reducing the resolution to match the video raster size.
By applying OLPF which cutoff frequency has been optimized to match video raster size (no matter what real sensor resolution is), one can reduce video moire, even if in-camera downsampling is as simple as line skipping (decimation) with no additional filtering - the aliasing will simply not be introduced if there are no frequencies that can alias.
The other way of achieving similar goal is to use the digital antialias filter in-camera, prior to resampling. Of course there's still OLPF (every camera has it), but it is tuned for native sensor resolution, just like in DSLR. and not video raster size. This technique has some great advantages, but, as you know, it's complex.
In this context this two scenatios can be used to achieve similar goal - get rid of some details prior to (re)sampling to avoid aliasing:

lens->olpf (tuned for native sensor resolution)->digital lp filter->decimation
lens->olpf (tuned for video raster size)->decimation

I agree that pixel binning is very bad method of downsampling. The only reason it is used is because it's so easy. In fact, it does both digital filtering and decimation in one pass. Computationally very easy, but the filter kernel is rectangular (averaging from certain number of pixels), which is far from perfect in terms of response - poor rolloff and very significant side lobes (spectral leakage).

But like I mentioned, with a Bayer pattern, you have two different densities of B/R and G sites. This means one has to choose where you want that filter to fall. Likewise, the binning method also affects the OLPF because the ultimate usage of the sites will determine how far you need to spread that spacial material hitting the sensor, which of course will introduce softening. Moire is not going away until the resolution of the bayer pattern is higher than the target resolution. I forget the conversion but I think it was 70 or 80% resolving power of a bayer pattern.

Maybe I am over sensitive to moire! But for me, I can't wait to be rid of it, it's so distracting.

Not necessarily. The highest part of the MTF does not affect the sharpness at all. On Panavision site there's nice presentation on it. It's the middle of the MTF that determines the perceived sharpness. The highs only affect the texture perception, and, in my opinion, moire destroys textures much more than some gentle smoothing.

It really depends on how far you need to push the original image. Too strong an OLPF is going to really soften the image because if you have to push every bit of spacial information 2 or 3 lines in every horizontal and vertical direction and you are at resolution already, it is going to soften edges. If they are using the same sensor, and given the bayer patterns affect on pixel binning, If they put too much it will soften.

Zoom in 1:1 on a picture off of a red one. it's not necessarily that sharp. It has aliasing (especially on chrome reflections).

I'm just not holding out a bunch of hope in the resolution area on any of the pixel binned sensors. Even if they have less moire and aliasing (as they will surely be less), I think the images will resolve to the same amount (given the same sensors) as all the OLPF will introduce is more spreading of the image and less false detail.

Like I mentioned, most sensors use a birefringent OLPF filter (multiple layers of it). All this does is duplicate the image in one direction at an offset. Multiple layers of the material create multiple duplicates, which is intended to make sure that the image at the high contrast portions is spread out to the surrounding sites and not lost in an area not sampled. This spreads EVERYTHING out.

I think if people are happy with the GH1 moire (which I would be compared to my 7D) it will be on par with the Sony.

You might not know it, but Panasonic announced that the AF100 will not have the same sensor as the GH1. No other details have been announced so I cannot say what is different, but clearly they must have felt there was a reason not to use the one in the GH1. (this was just recently as originally they said it did).

Don't forget they're preparing 'pro' inexpensive s35 camera. I fully understand people complaining about lack of 24p, but I understand Sony, too. As long as there's no real competition, they want to maximize their income...

[/QUOTE]

I understand wanting to maximize your income, but this method of doing it will make me not shop with a company. Shooting 24p is easier than 30p. It takes less bandwidth. What I understand is that they are alienating a WHOLE bunch of people that aren't even looking at something in the pro range (whatever that is now). And if they can't prove a pro camera has enough warrant of price OTHER than 24p, then that is just a bad business decision. Why not lead the pack here? With a LESSER feature than 30p that people want. These people, like me, who don't want their pro camera, will just shop elsewhere. It would be different if we were asking for a feature that would cost a lot of money to implement. Clearly it can do 25p. What is this 'pro' inexpensive camera? The costs I've seen tossed around for the one they showed at NAB I think it was were anything but inexpensive.

 

[Gweilo66]

Assuming Sony will do the usual expanded audio options and features of this for the pro side? Maybe the 24p as well. Looks like this one at least has a stero mini mic in.
SPECS

I wish this had a servo zoom. if you want to do any more home, educational or industrial bits, sure nice to have. Hard to have a shot list that includes a slow establishing zoom and have to say..uh..can't do it. (Yes...we will only be making indy classics with these

Anyone know if the new Panny will have a servo on the kit lens?

(ah..doesn't come as a kit)

 

[Gearhead]

Hi!

There is already a test of the VG10 from last week: http://www.slashcam.de/artikel/Test/Sony-NEX-VG10.html

For those who don't speak german a small summary:
- it's technically more or less a NEX-5 in a new housing
- no audio level display or manual audio gain
- no zebra
- no internal ND-filters, no XLR for audio
- no focus transitions, no focus assist, no magnification while focussing (!)
- menu as on the nex-3/5
- crops image from 2:3-sensor to 16:9 video left and right, so the 27mm gets 32,5mm (FF equiv.)
- similar rolling shutter and moire as NEX-5
- positive: manual video control (with dedicated buttons for WB, shutter, aperture, gain, focus mode)
- a very good onboard microphone
- a good viewfinder
- optional histogram on screen

I hope I didn't made any mistakes describing the features... but I think the next tests will follow soon. They claim, that this was a production model.

Oliver

 

[trez]

Further, I don't see how that proves ANYTHING about it being good in low light. In fact, looking at their sample images at different ISOs, I couldn't believe how much noise was present at ISO800.

I guess we saw the same pictures (it was some foliage) - I was surprised by the amount of noise at ISO800, too. On the other hand I watched one test footage shot at night with only available light at it looked great (the shooter confirmed it did very good low-light job). Don't know what to think about it...

I can't speak for the GH1, but with the 7D I can use all the way up to ISO1600 easily, and ISO2000 when I am careful. The 5D and 1D are even better than the 7D. That said, I'd really have to test the camera to compare.

That's right. We have to wait. By the way - ISO800 is not bad at all. I can live with it, provided the noise level is low. I got used to shooting with DOF adapter - the actual ISO around 100 or even less... PITA, had to invest in superfast glass and 1.2 HMI pars to avoid the genny

It's also worth pointing out, most of their tests were with picture mode, not video mode. I've seen plenty of tests with the 7D where the dynamic range in video is not as high as in picture mode. I think a lot of it was contributed to the compression, so I can't say what we'd expect on the Sony.

I agree. My experience with 7d (and I own one) is that that the dynamic range is significantly lower in video mode. I have to control the lighting ratios and adjust the exposure very carefully (not easy thing to do if you don't have waveform monitor) to get good footage. I didn't measure the dynamic range myself, but have the impression that my highlights get clipped too often (or the shadows are flooded by pure black). EX1 with DOF adapter is way more forgiving (to some extent this might be due to the highlights getting diffused by the gg, which lowers the contrast)

That is true, I don't think the problem has been the computational aspect of it though, it's the actual process of getting the data off the sensor fast enough. It has two problems, amount of data and heat generated.

True. It's not the downsampling process itself, but rather sensor readout issues.

Let's say you read every 3 lines of a sensor (after pixel binning has occured), you'll increase the skew horizontally 3x alone if you read them all. Imagine the 7D with 3x the skew as now!!!

Very good point. Haven't thought about it before.

It's not a matter of 'doing it right' it's a matter of trade-offs. With a bayer pattern, the density of red/blue is half of green. If you pick a OLPF that is for the lowest density, red/blue, you soften your image more than you might want. If you go look at the resolution charts of the Alexa, you will see that even it has moire as well. Sure, they too could tune out the moire completely, but then the picture would be extra soft.

I fully agree that it's the tradeoff. I simply think that those OLPF could be tweaked to cut lower, to filter more moire. I would trade some sharpness for less moire - I hate it. I see it everywhere, every tree branch, high contrast edge or fine texture suffers from it - maybe I'm oversensitive. Panasonic attempted to do it by applying more aggresive optical filter in AF100. You are right they finally decided to use different sensor, but they tried...

I am sure Sony has to make that same decision. It's why the red cameras shoot higher than 1080p so when you downsample from the full image, you can get rid of all the moire and aliasing. I've yet to see a native resolution bayer pattern sensor not produce moire or not be crazy soft.

Of course oversampling is much superior - you can apply the low pass digital filter prior to resampling and get rid of the moire without affecting the ultimate (video raster) MTF at all. And must admit your remarks about bayer sensor and OLPF make sense, too. In case of three-chip cameras, there's less of a tradeoff, indeed.

Better than the 7D? Without question. The 7D is the worst (as all the Canons) because of the rainbow moire. I think it's a problem with their pixel binning and NOT their OLPF.

Definitely. Their OLPF is tuned for native sensor resolution, so it's not capable of reducing much of video moire

Whatever they are doing, they need to rethink because it looks horrible. The moire in the GH1 is much much more pleasant. But still, I don't like it at all.

Agree.

As I said, if you wanted to, you could definitely remove ALL moire in ANY video camera by a strong enough OLPF solution, but it comes at a price. How soft an image you want. I understand this, but given they will pixel bin, it think it will be on par with the GH1 if they want the same sharpness.

You might be right. I'm not sure, though. Panny tried to reduce moire by softening the image in AF100. They change the sensor - true; maybe the reason was the softening was too pronounced, probably not only. People control their moire by using those Caprock filters - you can find some resolution charts on this forum showing that the loss of resolving power is marginal, while moire is greatly reduced. I truly believe that by properly designed filter the moire can be significantly improved (reduced) without affecting sharpness too much.

But like I mentioned, with a Bayer pattern, you have two different densities of B/R and G sites. This means one has to choose where you want that filter to fall. Likewise, the binning method also affects the OLPF because the ultimate usage of the sites will determine how far you need to spread that spacial material hitting the sensor, which of course will introduce softening. Moire is not going away until the resolution of the bayer pattern is higher than the target resolution. I forget the conversion but I think it was 70 or 80% resolving power of a bayer pattern.

Yes, 70-80% is probably correct value. The numbers are the estimates, or averages (calculated for average scene, whatever it is).

Maybe I am over sensitive to moire! But for me, I can't wait to be rid of it, it's so distracting.

So am I! I hate moire, too. That's why I'm ready to trade some sharpness for it.

It really depends on how far you need to push the original image. Too strong an OLPF is going to really soften the image because if you have to push every bit of spacial information 2 or 3 lines in every horizontal and vertical direction and you are at resolution already, it is going to soften edges. If they are using the same sensor, and given the bayer patterns affect on pixel binning, If they put too much it will soften.
Zoom in 1:1 on a picture off of a red one. it's not necessarily that sharp. It has aliasing (especially on chrome reflections).

Again - agree. I only think that some softening the image is beneficial. It controls moire. Sure it adversely affects the top of the MTF. But I can live with that. I don't watch my footage 1:1. If poor MTF at the top of the range affects the sharpness it probably means I'm pixel-peeping. I move my seat back or try to find a way to get 4K system. In proper screening conditions it's the middle of the MTF that should determine perceived sharpness - you need some margin. Applying more aggressive optical filter will mostly affect fine textures, which are destroyed by moire anyway.

I'm just not holding out a bunch of hope in the resolution area on any of the pixel binned sensors. Even if they have less moire and aliasing (as they will surely be less), I think the images will resolve to the same amount (given the same sensors) as all the OLPF will introduce is more spreading of the image and less false detail.

Resolve - yes. But resolving power is not sharpness, given proper screening conditions.

I think if people are happy with the GH1 moire (which I would be compared to my 7D) it will be on par with the Sony.

It seems that moire is very bad on Sony, unfortunately (the latest german test say so). It looks like the camera is a crap. It doesn't improve on anything compared to DSLR The only real benefit, besides form factor, is full raster HDMI output while recording. Well, 'only' is not the right word. Half year ago people were doing crazy things to bypass the 7d codec by capturing its (already downscaled) HDMI output, removing pulldown and erasing the red 'record' dot by replacing part of the frame with compressed footage. Now there's a camera that outputs uncompressed and everybody says it's a crap (including me)

 

[Rick Burnett]

I used to us an HV30 with a DIY 35mm I built. I am NOT willing to give up any of the larger range of use I get with modern cameras. The next camera I get will have to be good at ISO1600 to at least the same level as the 7D.

And yeah, I saw when people were recording out of the HDMI port! That was just crazy, as it was yet again, ANOTHER piece of gear to drag around with you!

My monitors in my studio are 2560x1600 I believe (Apple Cinema 30") so I do see everything at full 1:1 resolution. I can say with certainty that the 7D at 1080p just isn't that great. That's why in all the projects *I* am in control of the final output, I release in 720p. For a downconverted image, it's great!

 

[Rick Burnett]

Oh, and I forgot to add, I actually do have the Caprock 2.0 filter. As Barry can attest, it's actually a pain in the neck to use because it is SO dependent on where you are focused, what focal length you are using. Sometimes it works, most times it does not. And knowing if it worked or not is almost impossible unless you have an external monitor to look at. I stopped using it and just started avoiding the moire

 

[trez]

Oh, and I forgot to add, I actually do have the Caprock 2.0 filter. As Barry can attest, it's actually a pain in the neck to use because it is SO dependent on where you are focused, what focal length you are using. Sometimes it works, most times it does not. And knowing if it worked or not is almost impossible unless you have an external monitor to look at. I stopped using it and just started avoiding the moire

Absolutely. The main reason is because it's located before taking lens. Just like those grad nds... PITA sometimes. But you say it works in certain configurations. The possibility to put it after rear lens element would help a lot, I guess.

 

[Otis Grapsas]

At some point, cameras with rolling shutters should spec the amount of rolling shutter present in the images. How much time from top to bottom in 24p? That would be a very useful spec.

 

[Crosby!]

Review on Luminous Landscape and a Vimeo VG10 group.

 

[bbinnard]

I thought the Luminous Landscape review was excellent. I especially liked the explanation of 24P/30P/60i because, frankly, I have never quite understood all the clamor about 24P. Probably that is because amateurs like me show their videos on a TV set, and we have no desire or inclination to get into any comparison with real moviemakers.

What will be interesting to see is to what degree Sony updates the VG10 firmware to accommodate some of the shortcomings pointed out in the review. My sense is they will never incorporate RAW mode for stills - people who want that will want to use DSLRs - but the lack of visible focus confirmation and a live histogram is something that should be fixed.

What I'm really looking forward to is the GH2 announcement. That one might end up being the VG10's strongest competitor.

 

[Camera Expert]

The GH1 especially with the hack is already a strong competitor. From what I hear, the manual features of the GH1 is much better. I still wouldn't mind owning a VG10. I just wish Sony was more generous when it comes to putting features on their consumer camcorders.

 

[Satsuei]

Just got a hands on with the NEX-VG10 at the Sony Showroom here in Tokyo. Unfortunately, they wouldn't let me put my own memory card into the camera and do some test shooting.

But overall here are some impressions:
- Well built. Pretty solid.
- The lens feels good in the hand. Nice quality -- not super cheap feeling like some low end SLR lenses tend to be.
- Good close focus ability.
- The camera is pretty front heavy because of the lens. This gets noticable after holding it for a while.
- Pretty menu, but somewhat weird since you are using arrow keys and control dial at a 90-degree angle to the LCD screen.
- Switching the control dial between shutter speed and iris in M mode is fiddly and may not be good for on-the-fly shooting adjustments. May be better if using manual aperture lenses via an adapter so you have physically separate aperture and shutter speed controls?
- Weird position for the 1/8" mic input -- on the side of the top handle right in the middle. Gets in the way of actually using the top handle to hold the camera.

Very tempting, but still wondering if I should pre-order one of these. Or maybe hold out a bit more for the GH2 announcement.

 

[CPM Camera Rigs]

Just noticed this review;

http://www.videomaker.com/article/15150/

Just don't understand why they left out the XLR. A1U was an after thought but atleast it has them.

 

[FuGZ]

Hi guys,

i lurk alot here. But i have a lil somethin to share on the Sony NEX-VG10. A lil hands on of my own.

http://www.point8cam.com/blog/?p=1115

It's a consumer camcorder afterall. However, it's difficult to define what a consumer film maker is.

 

[Otis Grapsas]

Hi guys,

It's a consumer camcorder afterall. However, it's difficult to define what a consumer film maker is.

Someone who shoots 100 tests and a few home videos, all in 24P and shown with judder on 60p TVs and 60hz computer monitors.

 

[Evro]

Consumer or not, you can't deny that this camera will be awesome for people who shoot weddings for a living and want rich bokeh but are aren't willing to push a square peg into a round hole via the fiddly DSLR form-factor (many a colleague have run into overheating problems, intermittent shut-downs & loss of coverage due to the limitations of shooting live events on HDSLR.)

For my Aussie (PAL 50i) mates out there, Ryda's On-line shop in Sydney are selling them at $2,888AUD, well below the $3,499AUD list price. And for those of you that have already invested thousands of dollars in professional Nikon & Canon lenses, ebay has a heap of sellers offering Sony E mount adaptors & high capacity FV100 batteries

I paid for mine in full, now the long 2-week wait till the camera lands in Australia.

 

[Chris Watson]

Maybe this has been brought up already but does anyone know what the flange to focal plane distance is on the Sony E-Mount vs. the Micro 4/3 mount? I ask because if the two are the same, that opens up mounting c-mount TV lenses to the camera which is very exciting. Those zoom lenses can open up to as much as 1.8. Very intriguing to say the least!

 

[Satsuei]

The Sony NEX-VG10 went on sale in Japan today.

I played with one in the store for quite a while and got some footage on my own memory card to examine.

Good news is that interlacing artifacts are not to be found. I can take the VG10 footage and drop it onto a 1080p30 timeline with other progressive footage and there is no problem editing in Premiere CS4.

I also tried shooting some seriously underexposed footage -- around minus 3-stops. And the image held up very well. Color was a bit muddied, but there was still a good amount of detail.

Also tried the 16mm F2.8 lens and the camera becomes VERY lightweight with this combination. It's too bad the top handle is not removable for guerilla shooting... but then I guess you'd be better off just using some other compact handycam for that.

Overall, a very nice camera. If I were rich I'd buy one right now. But since I can only pick up one camera for personal use I'll probably wait to see what Panasonic announces at Photokina on Sept. 21. It's going to be a long 1-1/2 weeks!

 

[Paul V Doherty]

Consumer or not, you can't deny that this camera will be awesome for people who shoot weddings for a living and want rich bokeh but are aren't willing to push a square peg into a round hole via the fiddly DSLR form-factor (many a colleague have run into overheating problems, intermittent shut-downs & loss of coverage due to the limitations of shooting live events on HDSLR.)

For my Aussie (PAL 50i) mates out there, Ryda's On-line shop in Sydney are selling them at $2,888AUD, well below the $3,499AUD list price. And for those of you that have already invested thousands of dollars in professional Nikon & Canon lenses, ebay has a heap of sellers offering Sony E mount adaptors & high capacity FV100 batteries

I paid for mine in full, now the long 2-week wait till the camera lands in Australia.

I would be very appreciative of your assessment of the 25p performance and general PAL goodness produced by this camera, especially any comments or empirical observations on latitude and the quality of burned out highlights