Simon Wyndham

Okay then. And example of a good DV movie shot on an XL. http://www.soulsearchermovie.com I worked on this. The end result is amazing. I may be bias, but hey, you guys wanted a good example of what can be done...

The McCoy season was not the best. However it did have one of the best Dalek storylines so far.

This is the res chart from my camera BTW. http://www.simonwyndham.co.uk/reschart/tes...ndard_4_3_2.png

Out of curiosity from this site http://www.bealecorner.com/trv900/respat/ I decided to try my own resolution test using the Sony PDW510 with a Canon J17 7.7 lens. Much to my surprise, depite framing the chart right to the edge of the screen (including overscan area) the results I obtained showed far less horizontal resolution than that guys TRV900! Even when I switched to 4:3 mode whereas that TRV900 seemingly shows clear lines well over 500 lines on the chart along the vertical "I" plane (I used the letter "I" to make sure you understand which axis I am on about), mine just blurred beyond that point. Am I to seriously believe that his TRV900 produces a sharper resolution? Further to this, if he did indeed record the footage and play it back in DV25, surely it could not possibly resolve more than 500 lines across? Even further, I loaded the full size resolution chart he offers for download into Photoshop and then reduced the image to 768 pixels across. I found that even Photoshops scaling technology couldn't keep as much resolution resolving in the chart as that guys TRV900 test results!! So what exactly is going on here?

Well I've just found out the reason why. The bloody camera wasn't in 1/50 shutter! The XDCAM defaults to 1/25 shutter when progressive scan is selected. Grrrr. Silly me for not assuming that they might default it to 1/50. Oh well, at least I know how to set it up now!

I've just heard more from my colleague. He seems to be saying that the progressive mode on the 510 has more motion blur during movement than the XM1 in frame mode. Not sure how this is possible if they are both shooting at 1/50 shutter.

Hi Phil, >Assuming you had both cameras on identical shutter settings, the motion rendering >should be identical - the PDW510 will of course be sharper, have better highlight >handling, etc. Possibly your colleague would prefer the motion rendering of a shorter Yes. I have suggested to him that one of the only solutions I can think of is to change the way the performers are moving, or to have a slightly shorter shutter speed. I do like the high shutter speed look, but I just wish I could address his problems directly without him going away thinking true progressive is rubbish. He loves the frame mode on the XM1, and as you rightly say the motion characteristics should be exactly the same. >But "soft" is usually a sharpness characteristic, not a motion rendering one. Yes. I can't seem to clarify what he is saying when he uses the word 'soft'. I was shooting a f1.7 for a short DOF, so I am wondering whether he thinks that because the background might be ever so slightly out of focus even on the wide shots whether he is thinking this makes it a soft picture. Although to my eyes I can't honestly see what he is talking about. Another theory could be this. The location was heavily lit with flourescents (it's a walkway tunnel). I had set the camera up with a smoother knee roll-off. The XM1 has a naturally very contrasty image. Perhaps the smooth highlight roll-off combined with the larger contrast latitude of the 510, the reduced edge enhancement I set, and the flourescent lighting has made him think it is subjectively softer? >Most of the stuff I've seen is clearly undercranked. Most of the Hong Kong stuff they did yes. Although to my eye the US stuff such as Rush Hour hasn't. The movements are a lot 'heavier' and slower than their HK productions. If they were it must have been very subtle. There seems to be a definite change in technique too. In the Hong Kong films they would undercrank so that they could perform the movements slower (it's very easy to move too fast for the camera to see). But in the US productions I've noticed a lot more holding at the end of movements seemingly to add more definition to moves as a result of lack of undercranking. I could be wrong of course, but most of the people I know who work on martial arts film are based in HK so I don't know anyone who could tell me the practise in the US. >Yeah, well, there you go - don't listen to people! Progressive scan will make chaotioc >camerawork appear more chaotic but without necessarily making it more confusing. LOL! Yes, true. But I do wonder where the advice or 'myth' came from. >American TV and start looking like expensive American TV is a matter of not lighting it in >the way video has traditionally been lit. Abandon your carefully BBC-trained studio >camera experience here; it's a different game. Don't worry, I'm all for film style lighting. Any BBC training I have is as a freelancer so I wasn't assimilated into the studio lighting mindset ;)

I've just shot a fight sequence in a subway tunnel using progressive scan on a Sony PDW510p. Now my work colleague is complaining that frame mode on a Canon XM1 looks better and that the movement from the 510 looks 'soft' and less defined by comparison! I've tried explaining to him the characteristics of progressive and how the fighting movements need to be bigger and more defined themselves, as well as perhaps the performers slowing down their movements a touch. In Hong Kong they will often undercrank the film camera, not to speed things up so much, but so that the performers can slow down their movements to be more defined and powerful, and then played back at full speed. Problem is that with video speedups of one or two fps are not so easy to perform, and lose picture quality as they have to be done in post. Has anyone here had any experience of shooting 35mm fight sequences such as with Jackie Chan or Jet Li on their US movies? Those films don't appear to have been undercranked so I'd be interested to hearing any advice anyone could give me with regards to obtaining clear fight movements using progressive scan. I never came across this problem with the XM1 in frame mode (and to be honest I can't really see what my colleague is on about, but there must be something wrong for him to be complaining in the first place). I've already taken precautions such as not panning too fast, and being weary of heavy handheld style shots. But on the other hand I have seen films such as Spiderman 2 (the part where he is trying to avoid being fired from the pizza delivery) where there is very heavy shakey handheld camerawork. So on the one hand I am advised not to do heavy handheld work in progressive, but on the other hand I see 35mm films with quite manic camerawork (24 is another example). What am I missing here? Especially with regard to following a fighters movements during a fight what is the best advice anyone can give me when shooting in progressive mode? Also aside from the slightly artificial way the old Canon XM1 creates it's frame mode, is there really much difference between it's motion characteristic and true 25p (I'm working in PAL)? My colleague compared the true progressive mode of the 510 to the look of cheap American television when it has been converted to PAL! Help!

One thing to remember. Content is king and the quality of the camera is secondary. If you can find an XL1 for the price range that you mention it is definitely worth considering. You may also be able to pick up a used DVX100, or the model below that the DVC80 (although it doesn't have the progressive scan option that the DVX100 has). The DVX100 would be excellent for what you are trying to achieve.

Don't forget that the Z1 'pro' version of the FX1 is swithcable between PAL and NTSC. There are quite a few people out there who have promised comparitive tests between the cameras when they have been transferred to film. Hopefully someone will release the results sometime soon.

Often I use Magic Bullet Editors for applying certain looks to video. These go down really well with the people that I work for. Although the rendering time is a right royal pain in the backside. I notice that Panasonic have a section on their broadcast site with setups for the SDX900 that not only include maximum dynamic range settings, but also some of the more popular Magic Bullet preset style looks. Does anyone have any similar settings for the MSW900 or XDCAM cameras at all, preferably PAL? I don't have access to waveform monitors (nor would I know what I was doing if I had one!) and adjusting colour via a colour matrix with all those different adjustments for different black gamma etc are rather daunting to say the least.

That's interesting. I didn't realise that the U model was different. Thanks for that clarification :)

I'm sure everybody has come across this, even if they may only have to buy one tripod plate and then keep it for a lifetime. But just why are tripod plates such as the Sony VCT-014 so damn expensive? It's a small piece of metal with a spring loaded lever in it. Now forgive me if I'm wrong, but if someone had the exact measurements I'm sure that the average school metalwork student could knock one up just as good. My question is why doesn't anyone make an alteernative? 175ukp for a small piece of metal that size is possibly the most extortionate thing I have seen yet in the pro camera circle. Does anyone have any suggestions of alternatives that fit the Sony cameras?

Yep. The NTSC XDCAM cameras have the option of a 24p board. Although they will shoot 30p without any addition. The PAL versions of course can shoot at 25p with no addition too.

The paper was about video cameras. I've tried searching and in typical fashion can't find it again! All of this was very confusing for me to begin with, because when I used to mess around with computer graphics 24-bit colour was 3x8 (RGB). I was always under the understanding that the human eye couldn't really see much more than those 256 graduations? More to the point I remember when I used to draw graphics even using what was called Tru-Color in my Atari Falcon days (16-bit colour). It used something like 165 graduations per RGB channel (scuse my poor maths and memory). Even then if I wanted to actually draw something the resolution of the screen (640x480) wouldn't allow me to use all the graduations even if I wanted to, purely because the number of pixels wouldn't allow it. To get 256 graduations of every shade in a picture would take at absolute minimum 256 pixels. That's around half the number of pixels widthways contained in SD picture data. So what actually happens to increase the quality in the real world where 10-bit is concerned on Standard Def? If the human eye can barely percieve the 256 graduations, what does 4096 gain? I'm not trying to deny that there is an advantage. Just trying to get my head around it from the point of view of the number of pixels available in the first place, and the idea of what the human eye can perceive.

Interesting. One reason I ask is that I found a technical paper elsewhere on the web that suggested that the vast majority of the extra information is used in the highlights rather than being spread evenly over the whole range. Anybody know anything about this?

That's true. How far would things need to be pushed before 8-bit really starts to fall apart by comparison? How much of the 10 bit spectrum is actually utilised by cameras that can record this amount of data?

Just thinking. Of course it is better to start with more and then downsample, but with a camera such as the DSR570 this is what it is doing anyway. My question relates to shooting footage that will be released onto DVD (PAL). DVCAM is sampled at 4:2:0, and so is a PAL DVD. So is there a large advantage shooting with DVCPRO50 or Digibeta if the final product will be released onto DVD? I've recently bought a PDW510, and it has the option of an SDI board for direct uncompressed 4:2:2 output from the camera head if I so wanted (and didn't mind having the camera attached by umbilical cord to a recording device!) But I was wondering if I was shooting something I wanted very high quality for and it was to be released on DVD would I gain anything by going to the extra trouble seeing as the colour would only have to be downsampled anyway? My finances couldn't stretch to the 530 BTW, and the lack of having to sift through tape was the reason I didn't go for the SDX.

Hi, new here! We had this discussion in the past on some of the other forums. It was eventually found out through an camera operator for the Top Gear programme that they now use Digibeta with SDI cards installed for the majority of stuff, occasionally using HD900's. Apparently they do take their time however, and take a lot of care setting up shots. Another little tidbit was that they shoot interlace and then deinterlace in post, even with the HD900's.