HPX2700 or PMW350?

[Tom Roper]

Originally Posted by David Heath

Legacy compatibility may well have been an important reason for the choice of such as the NHU. But for anybody (such as myself) now thinking of switching from SD to HD, from tape to tapeless, etc and with few legacy issues to consider, the reasons for the NHU decision have little relevance. That's completely different to saying they were wrong.

Well said.

Switching gears, I was able to read the entire PMW350K review by Adam Wilt. I found it very postive, and very supportive of what Alister has been saying, so I think the criticism toward Alister who stated in his first post in this topic that he is a Sony fan, has nothing in it to cloud the objectivity of his own review, which is insightful.

Adam did express concern about PMW350 knee saturation, but it seems to have gone ignored that hypergammas (which don't use it) were not working on the prototype, and may not even be an issue for production cams. The kit lens also was well regarded, even though chromatic aberration correction circuit was not working on the prototype. There have been a lot of non-specific criticisms of this lens, (that you don't have to purchase) mostly owing to the fact that it's too cheap to be any good, therefore perhaps Sony could molify some people by raising the price of it to a level they feel would eliminate the cheapness concerns.

Which brings up the overall criticism of the camera package which is the price and the codec. Adam could not have known back in November when he reviewed the PMW350K that Sony would reduce the kit price by $3,100. That's a significant drop, the cam + lens now lists for $18,900. Auto focus for the first time in 2/3 inch...hmmm, that's a bargain as I see it. That just paid for the Nanoflash, so there's the 4:2:2 codec with more bit rates and options than AVC-Intra or DVCProHD, and low cost compact flash media to boot.

Lightweight, full raster, low noise, high sensitivity, low power and affordable. I think the critics are right, pass on this camera now. The animals won't like having their picture taken with the kind of detail and beauty this cam is going to make affordable.

[Steve Phillipps]

Alister "I'm sorry if I have offended or upset anyone in the course of this interesting and at times passionate debate. I do respect the views of all that have contributed, even if I don't agree with them."
Appreciate that, me too, I'm only trying to be helpful and am genuinely reporting my findings with no agenda. Also I think it might be of interest/use to some people to know what's going on in the NHU and other blue chip wildlife houses with which I have some involvement and which might not otherwise be public knowledge.
As a brief summary, I actually don't think there is any camera for moving images that I would consider a really good choice for wildlife, all of them have issues. If cash is no object I'd choose the SRW9000 at present, needing an SR recorder/player with it though as it's a tape camera, so moving the price even higher and beyond most people an common sense. At semi-sensible budgets I'd choose the Varicam, or possibly the PDW800 as when I had a PDW355 very briefly I was quite impressed with the half-vertical-rez overcranking and the 800 works in the same way but with a much better image to begin with. I'm still far from convinced about the EX3 even though I'd love to like it as it has so many big advantages for wildlife (small and light, 1 tiny 200g battery will last as long a 7kg of Varicam batteries, extra telephoto reach, cheap, good lens included, excellent wide lens option at a tiny price). Likewise the RED has a long long way to go for wildlife to for me.
Steve

[Simon Wyndham]

Steve, have you looked into the new digital Arri's that are due out this year? Around $40k, but they look really nice.

[Steve Phillipps]

Thanks Simon, which one do you mean, is it along the lines of the D21? do you have a link?
Steve

[Jeff Regan]

David,

In the U.S. the ATSC digital broadcast specs for 1080 allow for 1080/60i or 1080/60P according to a broadcast engineer friend of mine. There is no 1080/24Psf being broadcast or cablecast, according to him. Nor is there 1080/60P except for some DirecTV satellite PPV movies. Everything else is 60 fields interlace, 3:2 cadence from film source or 23.98 frame video.

I asked him if he or any engineer he knows can tell the difference between 1080/60i and 720/60P and he said, "yes, of course, the motion is much better with 720/60P". He said the resolution difference was imperceptible. He did say that when up close to a full raster display he can see diagonal lines that are smoother with 1080 vs. 720.

Jeff Regan
Shooting Star Video

[Alister Chapman]

Jeff you are only referring to US TV broadcast, what about the rest of the world? This is a global market after all. I believe that there are more 50Hz countries than 60 Hz and in these regions 25PsF is the norm for movies and common for most high end drama and documentaries. What about BluRay, web delivery (which is almost always true P) etc.

Producers making programmes for Discovery are some of my biggest clients. I deliver 1080P material to them most months. It may be broadcast within a 60i stream, but it's still progressive.

Anecdotal references from people that claim they cannot see the difference with "average" screen sizes at "average" distances are just individual opinions. The irrefutable fact remains that 1080i or 1080P has significantly more horizontal resolution than 720P. 1080P also has higher significantly higher vertical resolution. These are facts. It is also a fact that the temporal motion in 1080i60 and 720p30 is the same, the refresh rate is after all the same, the problem is that progressive displays often have a hard time dealing with interlace.

David's and my comments on the resolution of the human eye and theoretical viewing distances and screen sizes are also based on well proven, sound optical science, not opinions or hearsay.

Even if some people struggle to see a difference, for whatever reason, there are plenty of others that can, perhaps they choose to sit closer than the "average" viewing distance, perhaps they have a larger than "average" screen, perhaps they choose to watch on a computer screen just a few inches away.

So who should we cater for? Should we just make programmes that Mr Average viewing an "average" screen from the "average" viewing distance thinks is OK or should we strive to produce the best that we can? Should we not bother to produce programmes for those that can see the difference or should we just let them suffer ;-). Should we be producing material that is acceptable to some broadcasters in some region, or all broadcasters?

Even if this is just marketing hype (which I don't believe it is), should we not be producing 1920x1080 programmes to make the end customers feel good about their 1920x1080 equipment investment. After all these people are not just the customers of the TV manufacturers, but they are our customers too, at the end of the day these are people that help keep us employed.

If we took the view of simply producing programmes for Mr Average and ignored everyone else we would probably still have 405 line or NTSC TV. I'm quite sure that 10 years ago Mr Average thought that SD TV was just fine, I know my wife did. I also remember many, many people claiming that there was little difference between SD and HD in the early days. Thankfully though there were enough people that wanted something better. These were not Mr Average but people striving to improve the quality of their viewing experience. It's only by catering for the "few" and pushing technology forwards that things get improved. If we all took the view that "average" is as good as we need we would get no-where. We would all be driving "average" cars. Lets face it if you drive normally an "average" car well get you to your destination just as quickly as a luxury car, yet strangely many people like to buy luxury cars.

1080P is a logical forwards progression from 720P. Even if the difference is small, even if Mr Average can't see a difference at "average" viewing distances there are many good reasons for shooting at the highest resolution you can. Even if your end product will be 720P by shooting at 1080P you can crop and zoom in post with no discernible quality loss. You could use this to remove an errant boom mic or simply to compensate for a lens that's not quite long enough. It's excellent for green screen where any small keying issues disappear when squashed down to 720P. You can also composite a green screen scene at 1080 and then do a pan-and-scan within the scene while still retaining 720P resolution.

I'm afraid I have yet to hear a truly valid reason in resolution terms at least, for shooting 720P over 1080P. In an ideal world we would have 1080P60 or P50. Were not there yet, but I'm sure that will come. Buy a 720P camera and that's it. Buy a 1080P camera and in most cases you can choose to shoot 1080P or 720P.

One thing I don't think many people realise is that an EX recording overcrank at 720P60 records at 70Mb/s. Yes it's 4:2:0 but in progressive the difference between 4:2:0 and 4:2:2 is tiny (yes I would prefer 4:2:2). Also consider this: I think we all agree that an EX at 1920 x 1080 35Mb/s produces a good image. What about an EX at 1280 x 720 35Mb/s? Same bit rate but now only recording less than half as much data. My opinion is that at 720P the EX pictures are about as good as it gets for a 720P camera.

[Simon Wyndham]

Thanks Simon, which one do you mean, is it along the lines of the D21? do you have a link?
Steve

Everything is about to change... | ARRI Digital

I think that these may be the dark horse of this year that might take a few by surprise.

What about an EX at 1280 x 720 35Mb/s? Same bit rate but now only recording less than half as much data. My opinion is that at 720P the EX pictures are about as good as it gets for a 720P camera.

I've wondered about this a few times, about whether the compressor just records at a lower bitrate rather than keeping the same as it would for 1080. After all a lot of the time the EX records well under the 35Mb/s specified top rate.

But I agree, we should always strive to produce the best possible.

[Gary Nattrass]

For info if any people are not familiar with what cameras are currently approved for HD at the BBC there is a list on-line: BBC - Commissioning - Producing High-Definition TV

Of course there are exceptions and I know of one definite programme that was all shot on an EX-1 Liz Smiths cruise prog was done this way by an independent and I spoke to the director and got the in-story of the shoot more reading here:http://forums.dvdoctor.net/showthrea...ight=liz+smith

As previously mentioned I am using a 2700 as main camera for broadcast and have now been told that it is OK to use the 301's as b-cameras.

[Cris Daniels]

Well I am not a broadcast engineer and I can very clearly see the difference between 720p and 1080p footage on a full resolution display WHEN the footage was shot with a full raster 1080 camera.

I shoot with a guy that runs a Panasonic at 1080p/23.976 and his footage is clearly not up to the 1080p resolution of the Sony, it looks more like 720p in terms of detail. Thats because his camera is using pixel shifting to achieve 1080 resolution. And it doesn't substitute for a true 1080 full raster images. So yes, his 1080p really isn't any better than his 720p. But he can only speak for himself, even on a lowly Sony EX1, the difference between 720p and 1080p is striking even on a smallish 37" LCD TV.

So I don't doubt some people are seeing little difference between 720p and 1080p resolutions.

The difference in quality is so big that I almost cringe kicking down to 720p to overcrank, I already know that in post I will see the resolution drop when I intercut that footage into my native 1080p files.

[Alister Chapman]

Originally Posted by Cris Daniels

The difference in quality is so big that I almost cringe kicking down to 720p to overcrank, I already know that in post I will see the resolution drop when I intercut that footage into my native 1080p files.

That's my experience too. I'd almost rather produce the whole show at 720P than intercut 1080P and 720P, the difference is not insignificant.

[Gary Nattrass]

I agree I either do 720p or 1080p and never mix and match, my first question to a director is always do you need slo mo? if so then I shoot at 720p.

Of course with the usual chaos of production bods who struggle to name the frame rate required it is never a perfect world and looking at some of the mix of definitions on certain programmes it will take time for things to settle down.

I heard a story lately that progressive shooting was not to be done at all on a certain BBC house buying type programme and it puzzled me.

[Ron Evans]

(I heard a story lately that progressive shooting was not to be done at all on a certain BBC house buying type programme and it puzzled me.)

Well if the viewers are like me, panning around a house at 30p is very disturbing because of the judder. Since there isn't 1080P60 yet there is no options at 1080 other than interlace to get smooth motion. In this case 720P60 would work!!!!
I have recognized that a lot of programs are now 30P and I find this disturbing to the point of not watching. For steady camera movement its close to acceptable but unfortunately most shoots are not that steady.
With most all of the selling being directed to 1080p sets the sooner we get 1080P60 the better.
I may be very sensitive having shot film and video since the early 60's all as an amateur. But I can clearly see differences in resolution and frame rate and I am sure so can many others. Quality differences are more evident on my Panasonic Plasma as I think the 240HZ Sony masks a lot of issues. Its sad that the display has to make up for the source!!!!

Ron Evans

[Sander Vreuls]

Actually, the current HD camera's the company I work for use, can shoot 1080p50/60. The camera is the Grass Valley LDK8000 Elite. Although that is the top section that Grass Valley produce, it does show Grass Valley have gotten atleast some demands for it and are putting it in their current camera's allready.

However there is not much demand yet for those formats in the day to day use around here.

[Jeff Regan]

Originally Posted by Cris Daniels

I shoot with a guy that runs a Panasonic at 1080p/23.976 and his footage is clearly not up to the 1080p resolution of the Sony, it looks more like 720p in terms of detail. Thats because his camera is using pixel shifting to achieve 1080 resolution. And it doesn't substitute for a true 1080 full raster images. So yes, his 1080p really isn't any better than his 720p.

Cris,

You don't mention which Panasonic camera model, but no native 720P Panasonic camera, Varicam 27H, HDX900, HPX2000, HPX2700, does pixel shifting. My guess is that you are talking about an HPX500 or HVX200/A, HVX200, HPX170. Those cameras use 540x960 SD CCD's and do a pixel offset to derive 1080P. Please don't compare that approach to a native 720P camera such as a Varicam, apples to oranges. The P2 720P cameras scale to 1080 in a 32 bit processing environment and are recorded onto a full sample AVC-Intra codec, not DVCPRO HD that sub-samples 720P to 960 horizontal rectangular pixels.

Alister,

I was not referencing an average viewer, but a seasoned broadcast television engineer who works in network affiliates, sat trucks, microwave trucks. I trust his eyes more than any average viewer.

I do agree with your points about the flexibility of 1080, but, again your experience and mine don't match. Live action, same scene at 720 and 1080, not visible to me in normal conditions, not visible to TV display reveiwers, not visible to experienced broadcast engineer, not visible to DLP projection engineer, never mind small screens, I'm talking large screens.

Given two CCD's of the same size, one with full raster 1080, the other with full raster 720, the latter will have better light sensitivity or less noise for the same amount of sensitivity.

XDCAM EX in 1080/60i is very poor, so that limits any camera using that codec. EX1 and EX3 don't use a low pass filter in front of the sensors and are known to produce lots of IR contamination. Only circular polarizer filters should be used, not linear. I hope the 350 has a low pass filter in it, every other 2/3" camera I know of does.

As far as an EX1 being as good as 720P cameras get, not bloody likely.

Jeff Regan
Shooting Star Video

[David Heath]

Originally Posted by Jeff Regan

In the U.S. the ATSC digital broadcast specs for 1080 allow for 1080/60i or 1080/60P according to a broadcast engineer friend of mine.

Well, yes, but from a transmission point of view there is no difference between 1080i/30 and 1080psf/30 apart maybe from setting a data flag to tell the receiver. That's the whole point of psf! It allows progressive carriage over an interlace transmission medium. (Obviously limited to the system framerate- 30fps in the US, 25fps in Europe.)

There is no 1080/24Psf being broadcast or cablecast, according to him..........Everything else is 60 fields interlace, 3:2 cadence from film source or 23.98 frame video.

Errr, but then what is psf if not "60 fields interlace, 3:2 cadence from film source or 23.98 frame video"!?! If I was asked for a definition of 23.98psf, then I'd be hard pushed to come up with a better one than that!

I'll admit my experience is based on 50Hz standards and equipment, but I'd expect the principles to be the same - just without the 3:2 pulldown. Of course, it's conceivable that as the UK didn't broadcast HD until well after the US, the delay may have meant that more subleties were incorporated into European systems than in the US, and psf recognition at reception could be one. That doesn't change the point of all this - that if properly sourced, transmitted and displayed, the resolution difference between 720p and 1080p should be easily visible to most people.

I asked him if he or any engineer he knows can tell the difference between 1080/60i and 720/60P and he said, "yes, of course, the motion is much better with 720/60P".

I've never previously heard of anyone able to tell the difference in motion portrayal between a 60p and an 60i system. I can only suspect he must have thought you were still referring to 720p/60 and 1080psf/24.

But as Alister has said, the resolution issue all depends on source and display. A colleague of mine said very similar to Jeffs engineer friend a while back, and on closer questioning it turned out he was referring to an HVX200 or 171, and displayed on a screen with less than 1920x1080 resolution. He also said the resolution difference was slight.

We repeated the exercise with a full HD screen, and 720 and 1080 modes on a borrowed EX1. He now no longer questions the difference in principle between 720 and 1080. (To the extent that he now uses an EX1 himself.)

Those cameras use 540x960 SD CCD's and do a pixel offset to derive 1080P. Please don't compare that approach to a native 720P camera such as a Varicam, apples to oranges.

That's completely missing the point of what Cris was saying. Yes, a 1280x720 chipped camera such as the 2700 will be better than a 960x540 one with pixel shifting, but the resolution difference between it in 1080 mode and 720 mode will be slight if anything - it's the chips that are the limiting factor, not the recording mode.

Repeat the exercise with a 1920x1080 chipped camera and it is a completely different story. The 1080 mode will be far sharper than the 720 mode. (Which I think is what Cris was implying?)

As far as the 960x540 goes, then pixel shift predicts a resolution boost of about 1.5x for luminance only, compared to non use. In the Panasonic case, using it for both H&V means about sq rt 1.5x on each axis, or about 1.2x. That means you can expect some detail up to about 1150x650, so respectable for a camera using a 720p recording mode, but it should give an idea why true 1280x720 chips are better. (And that's before we even think about mtf's.......)