Assessment of the 5D2 by the BBC (Alan Roberts)

David Heath · August 30th, 2010, 04:34 PM

My wording "somewhat soft" shouldn't be thought of the same as "soft"..........! :-) "Soft" is a relative term!

I would have expected tape Varicam to have been replaced by the 2700 on big budget stuff by now, so 960x720 by 1280x720 and DVCProHD by AVC-Intra.

Steve Phillipps · August 30th, 2010, 05:22 PM

I'm still not sure what you mean when you talk about Planet Earth and say "Varicam was seen as somewhat soft" - by who?

Steve

Pete Bauer · August 30th, 2010, 06:27 PM

David and Steve, let's let go of the "soft" comment. If the person to whom the comment was attributed wants to share his or her thoughts on the subject here, they can do so. Otherwise, it is just a hearsay comment not requiring further banter.

David Heath · August 31st, 2010, 01:56 PM

Quote:

Originally Posted by Steve Phillipps

I'm still not sure what you mean when you talk about Planet Earth and say "Varicam was seen as somewhat soft" - by who?

I was referring back to the posts in this thread between 43 and 51. Peter Burke said:

Quote:

I still maintain that the 5D2 footage is an order of magnitude better than the Planet Earth Blu-ray.

AVSFORM do not rank Planet Earth anywhere near the reference group. I often refer to the AVSFORUM PQ Teir ranknigs before I buy a Blu-Ray. One I collected a few top tier titles I quickly learnt the difference between top notch PQ, and ordinary (eg Planet Earth).

To get back on topic. I suspect the BBC would be happier if the 5D2's footage was also softened like that on Planet Earth.

The point of my post was to say that whilst Peter may be right in absolute terms about sharpness, it would be wrong to criticise the BBC. There were very valid reasons for the use of the Varicam, and they were (rightly) accepting it's 960x720 resolution. And before you say it, yes, there is a lot more to PQ than solely resolution..... :-) The 5D2 footage MAY look sharper than a Varicam - but it comes with aliasing, and apart from other considerations, I think a softer, alias free picture is preferable.

My understanding was that they wished to standardise on one camera for all the production as much as possible. In the end, they used a mixture of HDCAM and Varicam, in spite of the cost implications. I understood that was an acceptance that Varicam was most desirable in some situations, but other scenes benefitted from the higher resolution of the HDCAM recording. In other words, they recognised that the Varicam might be considered "somewhat soft" in relation to the HDCAM - that is not the same as saying "the Varicam looks soft".

Steve Phillipps · August 31st, 2010, 03:49 PM

I remember at the time the overwhelming feeling about the Varicam pictures was that without any science or number crunching they just looked superb.

A massive part of the decision would have been because of the need for slomo - it trumps just about any other consideration in wildlife filming, and I think most viewers would be unaware of just how much it is used and how vital it is. Any shot of a flying bird for instance, anything chasing anything else, but also just using say 40fps to give a more dramatic feel to walking elephants or to take a bit of the shake out of 1000mm lens shots. I think if you counted the shots in Planet Earth or Life for instance you'd find far more offspeed shots than standard sync speed ones.

I just think it's a mistake to talk about the Varicam being soft in any way, I can't remember anyone, viewer or in the industry, seeing any of the blue chip NHU stuff, shot almost exclusively on Varicam, saying "great programme, bit soft though".

Steve

Daniel Browning · August 31st, 2010, 06:42 PM

Quote:

Originally Posted by Steve Phillipps

I just think it's a mistake to talk about the Varicam being soft in any way,

I kindly disagree. When displayed at 1920x1080, the 960x720 resolution is decidedly soft in comparison with just about any 1920x1080 camera, including 1/3" one-chippers. Not only does it lack all of the fine detail that 1080p cameras can capture, but it has lower contrast at many of the overlapping levels of detail (particularly the Nyquist spatial frequency), even without sharpening. To me, that easily qualifies as "soft in any way".

Quote:

Originally Posted by Steve Phillipps

I can't remember anyone, viewer or in the industry, seeing any of the blue chip NHU stuff, shot almost exclusively on Varicam, saying "great programme, bit soft though".

If that's your measure of softness, then even a 320x240 web video is not "soft". Millions of viewers sit through such tiny web videos and leave all sorts of "great programme!" comments without mentioning softness. That doesn't mean it's not soft, it just means viewers don't care (enough) to complain about it.

Content is king. If you're recording a routine science experiment and the scientist accidentally discovers cold fusion and a giant sea monster attacks during a surprise daylight meteor shower at 1000 FPS... it wont matter if it's only 960x720 (0.6 MP). Sure, it is a lot softer than the infomercial shot on 5K, but people are still going to prefer it over the infomercial. At the same time, just because it's good doesn't mean it's not soft.

Steve Phillipps · September 1st, 2010, 02:18 AM

That's not what I said Daniel. I said that I didn't remember anyone thinking it was soft when looking at the finished programmes or indeed when rushes came into the office. The only comments I've ever heard were about how lovely the images were. I'm a massive sharpness freak and wouldn't use it if it wasn't sharp.
I think there's far too much measuring, looking at numbers and side by side magnified views here. Anyone doing that really is losing track of what we're doing, making the best images for our viewers - they're the only ones that count. With this in mind I'm positive that if you set up a couple of 60" HD screens, one with a Sony EX3 and one with a Varicam and showed it to 100 people, you'd probably have a fair few of them saying they preferred the Varicam image, and I'm sure it wouldn't be a 100% vote for the EX3 and all viewers saying they didn't like the Varicam image because it just wasn't sharp and hi resolution enough.
Could be wrong though, we'll have to try it.

Steve

Ben Denham · September 2nd, 2010, 06:13 PM

"I think there's far too much measuring, looking at numbers and side by side magnified views here. Anyone doing that really is losing track of what we're doing, making the best images for our viewers - they're the only ones that count."

This also seems to be the conclusion that at least one BBC producer has come to regarding the 5DII. BBC4 has approved the use of this camera for a recent drama based on the strength of the images rather than a technical assessment of the camera. Here's the link.

HD Magazine - HD Mag - BBC Green Lights Use of VideoDSLRs

John Mercer · September 6th, 2010, 03:57 AM

You have to be careful, in my humble opinion, to quote the use of a particular pro-sumer camera for professional high end use and conclude that is the end of the story, accepted. House and many of the other productions mentioned have access to very high level equipment and post production processing for QC. Very few people using the 5D MkII today have anything like this resource.

I would put Varicam 720p footage mentioned against any of the current DSLRs for sharpness and resolution. Not only does the 5D MKII/7D footage have the aforementioned aliasing and moire issues it is, from my tests and from what I have seen, soft too (including the very nicely shot still shown earlier on in this thread). The perceptual appearance of sharpness is gained through extreme shallow depth of field, where the out of focus parts accentuate the look of the sharp bits. But from what I have seen when cut together with most pro (and much pro-sumer) HD footage it looks soft in direct comparison.

Alan Roberts is a very respected BBC boffin, whose very technical and objective analysis of cameras for professional TV use is the foundation not only for BBC recommendations but EBU ones too. He does not make the laws he only makes objective recommendations based on reliable and accepted practices.

As has been mentioned it is really immaterial (for broadcasters) the wow factor experienced when watching 5D original footage. What is critical is the way the codec and artifacts might fare through post-production and the compresion/decompression of the broadcast chain. The 5D's weak codec (MP4 with no B pictures) can fair poorly here.

There is a certain deja vu about this discussion which reminds me of '28 Days Later', where people assume the medium has arrived because the big boys are using it. I believe they are using these cameras as more of a creative tool than a de facto production standard. There is no doubt that particularly the Canon DSLRs have swept the imagination and are being used for some major production work. If the production (and crew) and story are good enough then it will obviously be shown on whatever platform. But I personally would really would like to see, with these DSLRs, less beautifully shot montages or landscapes and razor thin focus blades of grass cut to beautiful music and more real stories.

At the end of the day this is what matters creatively and it probably is important not to get too hung up on the measuring but, at the same time, it is very useful and very important for those who work directly with the major broadcasters to know the current objective technical analysis of our production gear - our jobs may depend on it!

Paul R Johnson · September 6th, 2010, 05:19 AM

When you find somebody who has been carrying out this type of review for years then there are preset standards and criteria that get implemented each test. The use of certain test charts may well reveal or not errors, but the point is that the procedures employed are standardised tests. I for one pay attention to these type of scientific tests based on evidence rather than the amazingly variable ones found all over the internet which are subjective at best, and downright biased at the worst. I've ignored ones obviously produced by the manufacturers themselves to support their list of superlatives.

As we've moved from SD to HD, we don't really have simple standards any longer. Even at the most basic level, when we had tape, in analogue - we worried about noise and the number of lines of resolution. Now we have so many variables, it's much more easy for defects - and let's be honest, this is what we're talking about - to be hidden away under a pile of exceptional specifications. I rather like a technical report - not that I understand every word, that effectively says "we got this far, and had to stop - no point going on".

Type in "professional" "HD" "camcorder" into Google and see how $199 will get you something amazing.

Words mean so little now.

What I still cannot quite get is how to interpret alternative detailing of specifications. Note how few manufacturers actually provide example files of their products so we can download a few and compare in our editors. I find it odd that almost every example of actual footage comes from Mr Bloom, bless him! Why don't the manufacturers show off their quality - not just wave figures about?

Kris Koster · September 10th, 2010, 04:47 PM

Quote:

Originally Posted by John Mercer

You have to be careful, in my humble opinion, to quote the use of a particular pro-sumer camera for professional high end use and conclude that is the end of the story, accepted. House and many of the other productions mentioned have access to very high level equipment and post production processing for QC. Very few people using the 5D MkII today have anything like this resource.

Right. And that was the point I was trying to make earlier except to add: Just because we don't have access to high level equipment and PP processing for QC pre-submission, doesn't mean the footage we put together on a 5D2 should be rejected? In the right hands, programs made with this camera could, with some of that additional high-end PP voodoo magic, be broadcast worthy?

And if this recent article (below) is to be believed, the BBC are backtracking a little in their earlier rejection of DSLR footage taking each program on a 'case by case basis'. Apparently DSLRs are in the dock right now and the beeb jury are locked in for the night! (Hopefully with a bottle of whiskey and a Canon 5DM2!!)

HD Magazine - HD Mag - BBC's Green Light for DSLRs Is OnHold

Exciting times.

John Mercer · September 14th, 2010, 06:28 AM

"Right. And that was the point I was trying to make earlier except to add: Just because we don't have access to high level equipment and PP processing for QC pre-submission, doesn't mean the footage we put together on a 5D2 should be rejected? In the right hands, programs made with this camera could, with some of that additional high-end PP voodoo magic, be broadcast worthy?"

What I meant was full pro crew and rigs with experienced focus pullers, DITs etc. Of course in the right hands, if the content warrants it, it may not be rejected. PP can only do so much with less than optimal footage.

BUT bear in mind I have so far seen the trailer for this BBC feature ("...Coronation Street") and it was awful. Aliasing and softness were dreadful. This is on an SD downconverted signal but Mad Men (shot on 35mm) shown the same way on the same channel was stellar. I'll wait to see the full show before casting final judgement though.

However it seems, at this stage to confirm the concerns about how this footage will stand up through the broadcast chain.

August 30th, 2010, 06:27 PM	#63
Pete Bauer Wrangler Join Date: May 2003 Location: Eagle River, AK Posts: 4,100	David and Steve, let's let go of the "soft" comment. If the person to whom the comment was attributed wants to share his or her thoughts on the subject here, they can do so. Otherwise, it is just a hearsay comment not requiring further banter. __________________ Pete Bauer The most beautiful thing we can experience is the mysterious. It is the source of all true art and science. Albert Einstein Trying to solve a DV mystery? You may find the answer behind the SEARCH function ... or be able to join a discussion already in progress!

August 30th, 2010, 04:34 PM	#61
David Heath Inner Circle Join Date: Jan 2006 Posts: 2,699	My wording "somewhat soft" shouldn't be thought of the same as "soft"..........! :-) "Soft" is a relative term! I would have expected tape Varicam to have been replaced by the 2700 on big budget stuff by now, so 960x720 by 1280x720 and DVCProHD by AVC-Intra.

August 30th, 2010, 05:22 PM	#62
Steve Phillipps Inner Circle Join Date: Apr 2006 Location: Wales Posts: 2,130	I'm still not sure what you mean when you talk about Planet Earth and say "Varicam was seen as somewhat soft" - by who? Steve

August 31st, 2010, 03:49 PM	#65
Steve Phillipps Inner Circle Join Date: Apr 2006 Location: Wales Posts: 2,130	I remember at the time the overwhelming feeling about the Varicam pictures was that without any science or number crunching they just looked superb. A massive part of the decision would have been because of the need for slomo - it trumps just about any other consideration in wildlife filming, and I think most viewers would be unaware of just how much it is used and how vital it is. Any shot of a flying bird for instance, anything chasing anything else, but also just using say 40fps to give a more dramatic feel to walking elephants or to take a bit of the shake out of 1000mm lens shots. I think if you counted the shots in Planet Earth or Life for instance you'd find far more offspeed shots than standard sync speed ones. I just think it's a mistake to talk about the Varicam being soft in any way, I can't remember anyone, viewer or in the industry, seeing any of the blue chip NHU stuff, shot almost exclusively on Varicam, saying "great programme, bit soft though". Steve

September 1st, 2010, 02:18 AM	#67
Steve Phillipps Inner Circle Join Date: Apr 2006 Location: Wales Posts: 2,130	That's not what I said Daniel. I said that I didn't remember anyone thinking it was soft when looking at the finished programmes or indeed when rushes came into the office. The only comments I've ever heard were about how lovely the images were. I'm a massive sharpness freak and wouldn't use it if it wasn't sharp. I think there's far too much measuring, looking at numbers and side by side magnified views here. Anyone doing that really is losing track of what we're doing, making the best images for our viewers - they're the only ones that count. With this in mind I'm positive that if you set up a couple of 60" HD screens, one with a Sony EX3 and one with a Varicam and showed it to 100 people, you'd probably have a fair few of them saying they preferred the Varicam image, and I'm sure it wouldn't be a 100% vote for the EX3 and all viewers saying they didn't like the Varicam image because it just wasn't sharp and hi resolution enough. Could be wrong though, we'll have to try it. Steve

September 2nd, 2010, 06:13 PM	#68
Ben Denham Regular Crew Join Date: Jul 2009 Location: NSW Australia Posts: 149	"I think there's far too much measuring, looking at numbers and side by side magnified views here. Anyone doing that really is losing track of what we're doing, making the best images for our viewers - they're the only ones that count." This also seems to be the conclusion that at least one BBC producer has come to regarding the 5DII. BBC4 has approved the use of this camera for a recent drama based on the strength of the images rather than a technical assessment of the camera. Here's the link. HD Magazine - HD Mag - BBC Green Lights Use of VideoDSLRs

September 6th, 2010, 03:57 AM	#69
John Mercer Regular Crew Join Date: Jul 2004 Location: Spain Posts: 73	You have to be careful, in my humble opinion, to quote the use of a particular pro-sumer camera for professional high end use and conclude that is the end of the story, accepted. House and many of the other productions mentioned have access to very high level equipment and post production processing for QC. Very few people using the 5D MkII today have anything like this resource. I would put Varicam 720p footage mentioned against any of the current DSLRs for sharpness and resolution. Not only does the 5D MKII/7D footage have the aforementioned aliasing and moire issues it is, from my tests and from what I have seen, soft too (including the very nicely shot still shown earlier on in this thread). The perceptual appearance of sharpness is gained through extreme shallow depth of field, where the out of focus parts accentuate the look of the sharp bits. But from what I have seen when cut together with most pro (and much pro-sumer) HD footage it looks soft in direct comparison. Alan Roberts is a very respected BBC boffin, whose very technical and objective analysis of cameras for professional TV use is the foundation not only for BBC recommendations but EBU ones too. He does not make the laws he only makes objective recommendations based on reliable and accepted practices. As has been mentioned it is really immaterial (for broadcasters) the wow factor experienced when watching 5D original footage. What is critical is the way the codec and artifacts might fare through post-production and the compresion/decompression of the broadcast chain. The 5D's weak codec (MP4 with no B pictures) can fair poorly here. There is a certain deja vu about this discussion which reminds me of '28 Days Later', where people assume the medium has arrived because the big boys are using it. I believe they are using these cameras as more of a creative tool than a de facto production standard. There is no doubt that particularly the Canon DSLRs have swept the imagination and are being used for some major production work. If the production (and crew) and story are good enough then it will obviously be shown on whatever platform. But I personally would really would like to see, with these DSLRs, less beautifully shot montages or landscapes and razor thin focus blades of grass cut to beautiful music and more real stories. At the end of the day this is what matters creatively and it probably is important not to get too hung up on the measuring but, at the same time, it is very useful and very important for those who work directly with the major broadcasters to know the current objective technical analysis of our production gear - our jobs may depend on it!

September 6th, 2010, 05:19 AM	#70
Paul R Johnson Inner Circle Join Date: Nov 2005 Location: Lowestoft - UK Posts: 4,044	When you find somebody who has been carrying out this type of review for years then there are preset standards and criteria that get implemented each test. The use of certain test charts may well reveal or not errors, but the point is that the procedures employed are standardised tests. I for one pay attention to these type of scientific tests based on evidence rather than the amazingly variable ones found all over the internet which are subjective at best, and downright biased at the worst. I've ignored ones obviously produced by the manufacturers themselves to support their list of superlatives. As we've moved from SD to HD, we don't really have simple standards any longer. Even at the most basic level, when we had tape, in analogue - we worried about noise and the number of lines of resolution. Now we have so many variables, it's much more easy for defects - and let's be honest, this is what we're talking about - to be hidden away under a pile of exceptional specifications. I rather like a technical report - not that I understand every word, that effectively says "we got this far, and had to stop - no point going on". Type in "professional" "HD" "camcorder" into Google and see how $199 will get you something amazing. Words mean so little now. What I still cannot quite get is how to interpret alternative detailing of specifications. Note how few manufacturers actually provide example files of their products so we can download a few and compare in our editors. I find it odd that almost every example of actual footage comes from Mr Bloom, bless him! Why don't the manufacturers show off their quality - not just wave figures about?

September 14th, 2010, 06:28 AM	#72
John Mercer Regular Crew Join Date: Jul 2004 Location: Spain Posts: 73	"Right. And that was the point I was trying to make earlier except to add: Just because we don't have access to high level equipment and PP processing for QC pre-submission, doesn't mean the footage we put together on a 5D2 should be rejected? In the right hands, programs made with this camera could, with some of that additional high-end PP voodoo magic, be broadcast worthy?" What I meant was full pro crew and rigs with experienced focus pullers, DITs etc. Of course in the right hands, if the content warrants it, it may not be rejected. PP can only do so much with less than optimal footage. BUT bear in mind I have so far seen the trailer for this BBC feature ("...Coronation Street") and it was awful. Aliasing and softness were dreadful. This is on an SD downconverted signal but Mad Men (shot on 35mm) shown the same way on the same channel was stellar. I'll wait to see the full show before casting final judgement though. However it seems, at this stage to confirm the concerns about how this footage will stand up through the broadcast chain.