Nikon D4 Passes but the D800 Fails the EBU/BBC Broadcast Quality Test

[Sareesh Sudhakaran]

The EBU/BBC Test reports on the Nikon D4 and D800 are out, and the results are very surprising, at least to me.

The D4 barely makes the grade, while the D800 fails completely. The D800E wasn't tested. Major reason: Aliasing.

Read my notes on the report here: Nikon D4 Passes but the D800 Fails the EBU/BBC Broadcast Quality Test | wolfcrow

I have also included links to the actual reports in PDF for download.

Thoughts, anyone?

[Adrian Frearson]

Thanks for sharing. The D4s dynamic range at 13 stops, seems impressive.

[Glen Vandermolen]

What's surprising to me is that any DSLR passed the EBU/BBC quality test.

[Mark Watson]

Higher resolution has to come with some challenges, maybe the aliasing is somehow related to the high pixel count. I bet it shoots amazing stills. About using a recorder, I think the damage is done at the filter/sensor stage, so I doubt an external recorder is going to help here. The $3,000 price of the camera didn't seem bad, but then I checked the price of the best Nikon wide angle zoom, 14-24mm, I think, and it was going for $2,000. Guess I'll be waiting on Canon to counter-strike...

[David Heath]

Originally Posted by Sareesh Sudhakaran

The EBU/BBC Test reports on the Nikon D4 and D800 are out, and the results are very surprising, at least to me.

The D4 barely makes the grade, while the D800 fails completely. .

Thoughts, anyone?

Firstly, thank you for the links!

Most important to note though is that the EBU (or BBC) doesn't simply "pass" or "fail" cameras - they rate them on a tiering system. With reference to your blog, you say "The EBU – TECH 3335 describes measurement procedures for assessing the quality of video cameras used in television production. Their recommendations are usually the law in these parts,......"

That's true - but it's only half the story. 3335 describes the measuring process - but it's the sister document R118 that describes how to interpret those results - and which defines the law! (See EBU TECHNICAL - News - Are your cameras tiered enough for HD? )

As far as the D4 goes, then I think it's misleading to say that the D4 "passes the EBU quality test", without qualification, and if you look at R118 you can see why. The very first section ( 1.1.1 !!) deals with resolution in very direct terms:

1.1.1 Resolution
The camera should achieve the full resolution of the recording/production system. This requires sensors in 3-sensor cameras to have pixel-dimensions which are not less than those of the transmission system, or for a single-sensor camera to have significantly more pixels in order to do a satisfactory decode of the colour-patch pattern.

If the 3335 tests only give a result for resolution of "....the resolution does not substantially exceed 1,355×762…(at 1080p)" there is no way it can be seen to "pass" the EBU requirements. Or at least, no way will it pass on a high tier. Also worth noting that without an external recorder, the codec alone would disqualify it from approval on a higher level.

From the relevant section (3.1.4) I'd say the D4 should be considered in the "Tier 3" level, and I suspect that is what is meant by the test comment of "could be acceptable". It's acceptable for Tier 3, whilst (on account of the aliasing) the D800 isn't even acceptable for that.

[Tony Davies-Patrick]

It would be far easier for the BBC to actually accept videos on the quality of the final product that is sent to them, not on judgement of the tools used to capture the footage.

Another factor is that all the HD material, whether the original footage is full 1080P or not, is actually only broadcast at 1080i anyway (generally upscaled 720P footage), and a huge amount of that material actually stems from BBC cameras with either old film stock or digital video camcorders capable of only maximum 720P resolution.

[Sareesh Sudhakaran]

Originally Posted by David Heath

As far as the D4 goes, then I think it's misleading to say that the D4 "passes the EBU quality test", without qualification, and if you look at R118 you can see why. The very first section ( 1.1.1 !!) deals with resolution in very direct terms:
...
From the relevant section (3.1.4) I'd say the D4 should be considered in the "Tier 3" level, and I suspect that is what is meant by the test comment of "could be acceptable". It's acceptable for Tier 3, whilst (on account of the aliasing) the D800 isn't even acceptable for that.

Excellent information, David! You are absolutely right. Thanks for sharing.

[Joachim Hoge]

Originally Posted by Tony Davies-Patrick

It would be far easier for the BBC to actually accept videos on the quality of the final product that is sent to them, not on judgement of the tools used to capture the footage.

Another factor is that all the HD material, whether the original footage is full 1080P or not, is actually only broadcast at 1080i anyway (generally upscaled 720P footage), and a huge amount of that material actually stems from BBC cameras with either old film stock or digital video camcorders capable of only maximum 720P resolution.

They need to safeguard against the "abuse" the footag/codec will go through after they receive it. Ie generation loss, the actual broadcast etc. This is one of the main reasons they set these standards

[Brian Drysdale]

Originally Posted by Tony Davies-Patrick

It would be far easier for the BBC to actually accept videos on the quality of the final product that is sent to them, not on judgement of the tools used to capture the footage.

To a certain extent the BBC has always selected the equipment that can be used to shoot their programmes. In the past, if you were freelance crew shooting film for them you couldn't use a CP16 R (a 16mm news camera) to shoot programmes, you had to use an Arriflex, Aaton or Eclair. I'm sure it did occasionally happen, because the CP16 can be used for double system sound, but it wasn't on their list of requirements for joining their freelance crew lists.

Knowing some producers, it's probably a lot easier to police if they're informed that the cameras have to met a certain spec. BBC usually commissions programmes, so it's their call as to the technical specification and if one has unusual requirements this can be negotiated with them (eg the "Our War" series) . Although, they may require a certain post production workflow as part of it.

News is rather different to programmes, so what is technically accepted there may not pass as the main production camera on programmes. In the past BBC news used Betacam, while the programmes used Betacam SP and later Digital Betacam.

[David Heath]

Originally Posted by David Heath

From the relevant section (3.1.4) I'd say the D4 should be considered in the "Tier 3" level, and I suspect that is what is meant by the test comment of "could be acceptable". It's acceptable for Tier 3, whilst (on account of the aliasing) the D800 isn't even acceptable for that.

Having looked again, the criteria would seem to place a basic D4 in Tier 4 - not tier 3 - on account of the codec. To get Tier 3 status it needs to be 24 Mbs AVC-HD or better. What the tests seem to be implying is that *with an external recorder* the basic D4 is good enough to be counted as tier 3, whereas the D800 is not.

[David Heath]

Originally Posted by Tony Davies-Patrick

It would be far easier for the BBC to actually accept videos on the quality of the final product that is sent to them, not on judgement of the tools used to capture the footage.

I disagree. The point of such as these documents is to provide clarity in advance of any filming or production taking place. You should see them as a way of knowing that if you use camera A exactly whether you will automatically pass the quality threshold (at least in technical terms! :-) ) or whether your material is likely to have restrictions on it's use.

It's also important to see them as "recommendations" to a broadcaster - not "laws", and this is clearly stated at the front of the document. A bit like looking both ways before crossing a road - you may not get prosecuted for not doing so...... but it's a bit silly not to. This is important as it means non compliant footage MAY be used, but you can't guarantee it will automatically be acceptable. Hence such as archive or unique footage is likely to be OK, but something shot specifically on a sub-standard camera won't be.

The other aspect is that "quality of the final product" can be difficult to assess, and that's especially true regarding aliasing and compression artifacts. They may seem OK - but have bad effects in the transmission chain. The analogy I've used before about aliases is that it's a bit like food poisoning - the food may taste great and nothing seem wrong..... but just wait a few hours!! To quote from R118 again:

1.1.4 Spatial Aliasing
Ideally, no spatial-frequency content in the scene should cause aliasing in the camera. Aliasing causes motion-dependent video compression to fail in the extreme, since the aliased frequencies cancel motion detection...............

[Sareesh Sudhakaran]

I know many people who shoot for members of the EBU (don't ask me whom) with cameras you wouldn't believe - including the HV30, the 7D, etc.

There are also cases of them being 'convinced' to choose DSLR footage for major shows, but who knows what happens behind closed doors? When they contact you for work, they usually aren't that stringent.

However, if you are an independent producer pitching for a show, choosing a 'rejected' camera is the worst thing you can do.

In any case, the BMCC and its clout in the industry might change matters for good.

[Rikki Bruce]

Interesting.

The latest edition of Nikon Pro mag here in the UK stated that the D800 passed this test!

[Sareesh Sudhakaran]

Funny, the test report is available for download for all to see. Maybe the new D600 will "pass".

[Sabyasachi Patra]

Thanks David. Some very nice info and analysis.

I am wondering whether the 4k of the Canon 1D C will pass. Since it is 4K with the new Ulta HD tag, I am wondering whether it will pass BBC or not.

Will the BMC pass?
Sabyasachi

Originally Posted by David Heath

I disagree. The point of such as these documents is to provide clarity in advance of any filming or production taking place. You should see them as a way of knowing that if you use camera A exactly whether you will automatically pass the quality threshold (at least in technical terms! :-) ) or whether your material is likely to have restrictions on it's use.

It's also important to see them as "recommendations" to a broadcaster - not "laws", and this is clearly stated at the front of the document. A bit like looking both ways before crossing a road - you may not get prosecuted for not doing so...... but it's a bit silly not to. This is important as it means non compliant footage MAY be used, but you can't guarantee it will automatically be acceptable. Hence such as archive or unique footage is likely to be OK, but something shot specifically on a sub-standard camera won't be.

The other aspect is that "quality of the final product" can be difficult to assess, and that's especially true regarding aliasing and compression artifacts. They may seem OK - but have bad effects in the transmission chain. The analogy I've used before about aliases is that it's a bit like food poisoning - the food may taste great and nothing seem wrong..... but just wait a few hours!! To quote from R118 again: