From what I have read, GH1 has a problem with jello effect when panning fast. I tried this with my 5D2 and can't seem to see any jello effect. Plus night shot with 5D2 is really good. I wonder if the smaller 4/3 sensor will do well in low light. So far we have seen footage of GH1 night shot at low resolution. Wonder how it would look on a FullHD tv.

Actually the skew is much better in the GH-1 as compared to the D90 and 5Dll.

Edit: Check out this video. First part is handheld (shakey)...but looking at this tells me if a decent DOP got his hands on this cam like Vincent L. then there would be lots of beautiful pictures.

http://www.youtube.com/watch?v=8t7xHgOHxl8&fmt=22

You cannot mention d90 and 5dmii in the same sentence here. The d90 is much worse I believe... About the GH1 I've heard very conflicting information about this, I guess we'll just have to wait a little bit longer before we know anything at all about that one.

From what I have read, GH1 has a problem with jello effect when panning fast. I tried this with my 5D2 and can't seem to see any jello effect. Plus night shot with 5D2 is really good. I wonder if the smaller 4/3 sensor will do well in low light. So far we have seen footage of GH1 night shot at low resolution. Wonder how it would look on a FullHD tv.

In low light, I don't think there is any contest, the Canon looks to be a real benchmark here and it's never been one of the strong points four thirds. The GH1 does look like it will do pretty well in this department though, especially when compared to other camcorders.

On the Jello, it's still a case of waiting for real tests. Interestingly, I heard about a new helmet cam today, ContourHD Helmet Camera Features (http://www.vholdr.com/contourhd/features)
It's designed for fast action, but still appears to suffer from some rolling shutter problems. Cool little helmet cam though.

Adrian

I'll admit, "Writing off" the GH1 was necessary at the time for me to commit to buying the 5DMkII. It's been analyzed a lot that decision making, while as technical as we may want it to be, often becomes purely emotional.
I know how that is. Everyone does that. Like I said in another post, I've had a lot of debates with myself, some of which has involved "writing off the 5D Mk II" in order to justify giving into the temptation of committing to the GH1. By that I mean that I've stressed to myself the very disadvantages of my beloved new camera that I told myself to ignore in order to convince myself to buy it in the first place.

And the emotional response to the GH1 footage so far has been - eh, hmm, well, I don't know, well, oh wow, that looks awesome, but then a bunch of it looks pretty boring. By contrast, footage from the 5DMkII has consistently captivated and interested me from the very start.
To each his own. But again, the problem is that nobody has yet done for the GH1 what Laforet did for the 5D. There's a LOT of 5D footage that is "boring" and uninspiring to me. And I had to talk myself into accepting the very "digital" quality of its look that I hadn't seen in D90 footage.

Bottom line is think about the first time you saw night time footage shot on the 5DMkII - for me it was mindblowing. The night time footage I've seen so far from the GH1 doesn't look any more impressive than the state of the art camcorders on the market. It's "good", but I'm not seeing it look "revolutionary".
How do you know that if Laforet had used the GH1 instead of the 5D, the GH1 wouldn't look similar? Again, all of the night footage on the GH1 so far has been poorly shot. We're talking poor lighting, framing, and locations. Some guy was impressed with low-light footage of a Sony AVCHD camcorder compared to the GH1. You know the clip that I mean. He was impressed with something that I thought looked terrible. So it's all very subjective.

To that end, I like how with the 5D, I started to care less about shutter speed and ISO, and more about dialing in an aperture on my Nikon lenses, adjusting exposure with a dial, locking it and go.
Actually, to me, lack of control isn't what has bothered me most about the 5D. It's the fact that I have no idea what settings the camera is actually using in video mode. The information that it gives you when using the half-shutter-press is only an approximation for still mode, not what the camera is using for video. I recently did a camera test in extreme low light. We're talking ten tealight candles only. The camera reported a shutter speed of 1/5s, which is impossible for video. Exposure compensation wasn't working in this extremely lighting situation. I think that's when the first big crack in my love for the 5D appeared.

Well, what I do know is I can open up a .mov file off a CF card from the 5D in QuickTime Player, make a selection and save that as a new file with hardly any effort.
Why would you need to do that? Why not just drop the footage in an NLE and save a step?

Well, I was thinking also that I didn't *need* all that DOF. But I reconsidered what I was wanting to do artistically, and I came to the conclusion that I want as shallow as I can get *as an option*.
True, but also consider than in order to get a DOF closer to regular 35mm film, you have to stop-down the lens, which comes with its own pros and cons. I personally like the fact that I have to open up the aperture in order to get shallower DOF on the GH1, because more light gets collected, improving performance. Also, a wide-open lens will always have the "best" bokeh.

And at whatever focal length I want - from wide to telephoto. I know I can do that with full frame.
My problem with this argument is that there is an ambiguity about what you are using as a reference point. Are you accustomed to shooting with full-frame cameras? The whole concept of "35mm equivalent" focal lengths is using full-frame as the reference, despite relatively few people shooting full-frame in the vast majority of scenarios. The size of a 35mm film frame is roughly equivalent to APS-C. So with 35mm film movie cameras and still cameras, there has ALWAYS been a "crop factor." A 20mm lens on a film SLR is not the same as 20mm on a full-frame DSLR. So there is little substance to the phrase "whatever focal length I want," because you haven't specified what your frame of reference is.

From what I have read, GH1 has a problem with jello effect when panning fast. I tried this with my 5D2 and can't seem to see any jello effect. Plus night shot with 5D2 is really good. I wonder if the smaller 4/3 sensor will do well in low light. So far we have seen footage of GH1 night shot at low resolution. Wonder how it would look on a FullHD tv.
I can get the jello easily on my 5D. In fact, if I merely stuck a 135mm lens on it and started to pan steadily, the jello already appears. The rolling shutter on the 5D is a very good one, but not as good as high-end prosumer camcorders like the EX1. People who have extensively tested the GH1 so far have reported that the rolling shutter is EX1/RED-level. So that's amazing.

Actually, to me, lack of control isn't what has bothered me most about the 5D. It's the fact that I have no idea what settings the camera is actually using in video mode.

Yep it sucks, but on the other hand, I have an appreciation for where it steers in me in my thinking right now - choose my aperture on any Nikon lens, let the camera auto expose, adjust it brighter or darker with the wheel, lock it and record.

If I wanted more control? I may buy the GH1 when it's available, go throw together my Redrock Micro on my HVX, rent a Red One, or later this year or eventually S35 Scarlet may be in my hands on a daily basis. The 5DMkII is for a different purpose for me.


QuickTime saving - Why would you need to do that? Why not just drop the footage in an NLE and save a step?

Let's say I liked merely one part of one clip out of 10GB of footage. I can easily browse through them all in the finder in Mac OS X, then open it in QuickTime Player, then just save that part out and trash the rest.


True, but also consider than in order to get a DOF closer to regular 35mm film, you have to stop-down the lens, which comes with its own pros and cons.

Yes, and the big pro here is getting some seriously shallow dof.


My problem with this argument is that there is an ambiguity about what you are using as a reference point. Are you accustomed to shooting with full-frame cameras? The whole concept of "35mm equivalent" focal lengths is using full-frame as the reference, despite relatively few people shooting full-frame in the vast majority of scenarios. The size of a 35mm film frame is roughly equivalent to APS-C. So with 35mm film movie cameras and still cameras, there has ALWAYS been a "crop factor." A 20mm lens on a film SLR is not the same as 20mm on a full-frame DSLR. So there is little substance to the phrase "whatever focal length I want," because you haven't specified what your frame of reference is.

My frame of reference is 35mm adapter work - my own and RedRock Micro, as well as still photography (back in the day). The crop factor - yes, there's always been a crop factor, but I've gotten really used to my 20mm being wide, not normal, as it would be on the GH1. And to get that, I'd have to grab a 10mm lens, and I have not seen one single fast lens that wide (f1.8-2.8). Additionally my beloved 105mm micro nikkor is pretty useless to me on a GH1. It was already at the far end of as telephoto as I was wanting.

So, yeah, if you think differently about focal lengths and don't care about fast wides, and invest in some 4/3rds lenses, the GH1 can be pretty awesome.

Bottom line for me - I see the GH1 as a killer tool. The footage I've seen this week has been impressive and changed my mind on a few points. However, my observation on compression artifacts seems to be correct. And with that, I think I may be done with the dichotomous "this or that" part of the debate.

I may end up owning both. Or wait until my employer pays for the Scarlet S35.

-steev

Fei Meng...it absolutely looks that way. For starters check out this clip of a train. Halfway through you will see a traini zipping by very fast....take notice of the skew. It's so minimal that I'm shocked to say the least. Don't get me wrong...it's definitely present...but if that were the D90 or 5Dll...the lean would look a lot more.

YouTube - LUMIX GH1 ????? (http://www.youtube.com/watch?v=LILxnIWor-s)

And if you want to see some great shots....check out Phil Bloom's stuff on Vimeo here. Thia was shot using the slow(er) stock lens.

Panasonic Lumix GH1. First footage on Vimeo (http://vimeo.com/4405518?pg=embed&sec=)

Imagine Panasonic decided to put this in their camcorders.

Now I'm a bit confuse with DOF in FF and Cropped Sensor:

I did two test with a 5D MKII

ImageShack - Image Hosting :: 5dmkii200mmf28.jpg (http://img24.imageshack.us/my.php?image=5dmkii200mmf28.jpg)

and with a Cropped Sensor 400D

ImageShack - Image Hosting :: 400d200mmf28.jpg (http://img134.imageshack.us/my.php?image=400d200mmf28.jpg)

Maybe I'm wrong but I'm not getting the same DOF in booths Test. The Focus is in 20mm

Thanks

I think you posted this in the wrong section of the forum.

But to answer your question, I don't think DOF has anything to do with what size sensor you use. As long as your lens aperture is set the same, the same DOF effect will happen no matter what size sensor it is. Just that the field of view of a larger sensor is more than that of a smaller sensor. Imagine the 400D image is cropped from part of the 5DmkII image. That's what it is essentially.

Just my thoughts.

I think you posted this in the wrong section of the forum.

I agree. This discussion is about 5DMkII vs GH1.

But to answer your question, I don't think DOF has anything to do with what size sensor you use.

This is false. Sensor size has a strong relation to depth of field. Why do you think a 1/3" sensor and a FF sensor vary in depth of field so much?

To bring this back on topic, I had started earlier on in this thread about comparing the dof on GH1's smaller sensor size vs the 5DMkII's larger sensor size beginning here:

http://www.dvinfo.net/conf/1035594-post5.html

It's been hashed out already. All things being equal, the depth of field will be shallower on the 5DMkII with the larger sensor. And that can be good or bad depending on your goal.

p.s.

Now I'm a bit confuse with DOF in FF and Cropped Sensor. Maybe I'm wrong but I'm not getting the same DOF in booths Test.

You're right that the DOF is not the same. However, you have an error in your test. You'd want to compare equivalent focal length lenses compensated for the crop factor. The field of view of a 200mm lens on the 5DMkII would be around 122mm on the 400D, since the 400D's sensor is .6111 the size of the 5DMkII sensor size.

My math:
22mm sensor width on 400D divided by 36mm 5DMkII sensor size width = .6111
200mm lens x.6111 = 122mm

This is false. Sensor size has a strong relation to depth of field. Why do you think a 1/3" sensor and a FF sensor vary in depth of field so much?


No, indeed Alex is right (and I never thought of that before Alex mentionned it), you will always get the same DOF as long as your absolute aperture is the same. Picture the physic behind that, shallow DOF emerge from light coming from different direction. Only aperture has an influence on that, not sensor size.
In the industry when you have bigger sensor, you tend to have bigger aperture (easier to build it that way beause of optic manufacturing limitations), but in fact only aperture matters for DOF control.

On a technical level, looks like I'm wrong.

Depth of Field and the Small-Sensor Digital Cameras - photo.net (http://photo.net/learn/optics/dofdigital/)

Digital Camera Sensor Sizes: How it Influences Your Photography (http://www.cambridgeincolour.com/tutorials/digital-camera-sensor-size.htm)

Meanwhile, it's been far easier to think in terms of sensor sizes relating to dof. My 1/3" and 1/2" sensor size cameras have a deeper dof than my 35mm sensor size cameras.

Have fun figuring it all out!

The way I see it, you can use a normal 35mm lens to create an image on the sensor, but it would mean the focal length of the lens in relation to the sensor would be say for a 20mm FL will be probably 4 times that (just a random figure plugged out of the air) for the sensor. Just like a 20mm FL lenses will be approximately 32mm on a cropped 1.6x sensor.

That 80mm FL on the 1/3" sensor might not be practical for normal shooting.

For instance on my 350D and using a 50mm lens, we are talking about 80mm equivalent. I had to move back more than 2m just to get the subject in frame. Imagine how far you will have to move away from the subject in order to frame subject on a 1/3" sensor.

I hazard to say the design of the lens on a 1/3" sensor would be different as compared to a DSLR to take into account of the smaller sensor.

Anyway I could be way wrong but it make sense to me.

This is false. Sensor size has a strong relation to depth of field. Why do you think a 1/3" sensor and a FF sensor vary in depth of field so much?
Yeah, that's why if you crop an FF image you'll get deeper focus... Oh, wait a minute?!

It's the focal length and aparture. Nothing else. But a 50mm focal length is extreme tele on smaller sensors. (cropped full frame and zoomed in)