TV ad fails HARDING test...please help.

This is a nightmare..We have done a 20 second ad for Haispray in the UK. The ad has footage in moving boxes with moving GFX and light effects cut animated in After Effects then mastered back in FCP. It failed the Harding test which tests for flashing images and the BBC and broadcasting governing bodies in the IK are stringent on this. If we slow the animation and light effects down some could that solve this problem or could there more sinnister and less obvious problems?? How can we correct this? Thanks...

i think you need to find out exactly what bits of the ad are problematic, and how they are problematic.
very hard for us to give feedback without knowing that, let alone seeing it.


nick

We think it's the speed of the whole thing..moving boxes with lighting effects on the frames of the boxes and lighting effects moving behind titles nothing really that out of the oridinary..its fast and furious and looks great BUT...it failed a flash and pattern analysis.

Like Nick says, you need to ask them what exactly is not passing the test. The Harding test has some very specific parameters, like the number of stripes, how closely spaced they are and how much of the screen they occupy. Or it could have nothing to do with patterning and have been failed on luminance contrast. There's no point guessing what caused it to fail and fixing that. Ask them. Then fix what they want you to fix.

Well the bbc shouldn't be showing it!

You need to get a copy of the Harding report for the cx itself. This should point out your problems, but it is likely to need a rethink. I've had problems before with camera moves on people where the background is a brick wall - slowing down the move seemed to be the only solution at the time.

For flashing the general rule is keep high contrast cuts to 8 frames per shot (i.e. one high contrast cut every 8 frames). Flash photography is more of a nightmare - often you see the beeb just remove all contrast from those shots (eg celebs). Or slow them down.

Then I work on stuff that millions in the states will watch - and we can do whatever we want, there are no flashing or pattern limitations. ITV etc have to follow the Harding rules - they get fined by Ofcom if they don't.

This is interesting - being in the US, I'm unfamiliar with this, but i assume it's a test to ensure that viewers prone to seizures won't be affected by the flashing lights? Or am I stupid?

Keep up the discussion, I'm interested!

HarryD

Can you upload a demo?

Rui Barros
Editor Colorist Trainer
Lisbon, Portugal
RTP Post-Production
Apple Certified Trainer FCP 7
Apple Certified Pro FCP 7

I sent a couple of variation of the ad with all animated effects slowed down and they all failed on luminance. I am dumfounded because I put a broadcast safe filter (conservative) and lowered whites and ran through waveform videoscope. The waveform was kept below 100. I guess all I can do is take luminance levels down further and see if it passes then. I have been doing this awhile and only recently started getting flagged for broadcast safe levels. With the Harding test it is adding to my concerns..Flashing and pattern analysis.

I forgot..the after effects artist is recalibrating using AE 7's colour finesse. Rui, I could upload a clip but do not know where. Cheers Phil PS: this guy that invented the Harding test made a lot of money..it is like the nanny state that is Britain. I watch this add and think it rocks.

Quote
Harry asked:
Or am I stupid?

Sorry, Harry, can't answer that without more information.



Scott

I guess you are referring to Professor Harding.

Not sure how much cash he has made, but he certainly is the world expert on photo-sensitive epilepsy. Harding

As for the nanny state - well PSE seems to have been first discovered in Japan when a few people had seizures whilst watching a pokemon episode.

Rather more had seizures when the footage was rebroadcast as part of a news report that evening about the seizures;-)

At least that was the way I heard it. Then I read this.

'Broadcast safe' filter will only destroy your pictures to make them PAL safe and the broadcasters no longer require this. They require the individual RGB signals are within their spec. This is unrelated to the Harding test which is related to luminance changes and also patterns and motion within the frame that may cause PSE.

It is a major PITA - but it is part of the delivery requirements for UK broadcast. Charge accordingly.

My 2¢, Phil,

Personally, the QC should be done by the last person to view the spot before turning it in. If there is such a rigorous test in place, QC (Quality Control) should be just as rigorous and the spot should not have been turned in until guidlines are met.

I would love to see the ad with the offending flashes. I too never heard of this test. If it were me, I would want the specifics (numbers) on what peaks failed the test with specific timecodes & get the specific limits so as to follow them when building the spot. If I had to build projects based on such stringent specific limits of Luminance / Chrominance, I would run the test myself before turning in the tape. I'm lucky (sort of)...my studio crushes everything before the satellite uplink (Luma / Chroma) so I just build it loud & colorful so when they do crush it there's some decent color left

I can tell you that slowing the flashes down won't help...it'll just make it worse.

Hey Phil...I think this would be a fabulous topic for the FAQ. Can you post the Harding Test numbers / limitations?

Joey

When life gives you dilemmas...make dilemmanade.

Quote
Hey Phil...I think this would be a fabulous topic for the FAQ. Can you post the Harding Test numbers / limitations?

Not quite sure it fits in the FAQ. First time in my life I've heard of it and I'd suspect after this thread is done we wont hear of it for another 5 years. Interesting topic I got to admit.

Michael Horton
-------------------

PSE = Photosensitive Epilepsy, caused by flashes of red/blue and luminance, and certain bar patterns. Lots of interesting stuff about it - It may be that TV viewing accounts for the onset of most of it, and that up to 75% of individuals remain photosensitive for life. Suspected to affect 95000 people in the EU.

The following article summarizes the the technical requirements in the UK for PSE:

[www.broadcastpapers.com]


Graham Harding is a honcho in the world of Neurophysiology, and is the researcher behind defining the causes of it in TV viewing. His work has been nominated for major technical awards in the field of PSE and television.

A Cambridge University undertaking is the Harding Flash and Pattern Analyser (FPA). They offer a software program called Harding FPA that checks a video program on a frame by frame basis looking for luminance, flashes and patterns that Harding has identified with PSE.

[www.hardingfpa.com]


As for the BBC and PSE: check this out -

[www.ofcom.org.uk]

scroll down to page 102 - Draft Rule 2.10

This policy is thought to reduce the incidence from 1 in 4000 to 1 in 3 million.

A serious matter, no doubt.

IMO, as a practical matter this is no different than the things we in the US are required to do for broadcast. If anyone has produced for PBS, then you've seen their spec manual. And they're rigid about it, too.

I conjecture that this is different because of the PAL system - my read thus far is that the 50Hz system may be worse in inducing PSE, which may explain why we don't have this requirement in the US.

BTW - my wife recently left work in the Dept of Neurology at Johns Hopkins. Her work is in MS, not PSE, but from her contacts she's put me onto some articles; if I find anything of interest I'll post it.

Fascinating - still looking into it.

HarryD

Quote
Horton:
I'd suspect after this thread is done we wont hear of it for another 5 years

I dunno. I'll put a dollar on a prediction that we'll see more, not less, about this in the coming months.

Scott

Yes...facinating indeed.

Being the Father of a Special Needs Child for 9 years, I thought I was fairly savvy in the fields of neuro & chemical imbalance disorders & causes. I have never heard of this Harding test. I would still like someone from the UK to please kindly post the "Rules" for this test (luminance / chrominance levels, specs (numbers), etc). I am interested now to see if I am producing work that also fails this test.

Why hasn't this been implemented in the US?

Anybody??

My opinion: Because the FDA (Food & Drug Administration) hasn't figured out how to make $$$ on it yet.

Joey

When life gives you dilemmas...make dilemmanade.

Its the speed of the flashes, the percentage of screen covered by a disturbing pattern and the contrast level between shots that are the main concerns. I know that in those shots where a celeb is being photographed and there are a lot of flashes going off that slowing it down can help pass.

The luminance one is not really about how bright your picture is, so broadcast safing won't fix it. Its about the difference between the brightness of one shot and the brightness of the second shot, and how quickly the shots are switched. So if you have a black frame and a white frame changing every frame, you're sure to get slapped for it. If it's a light grey frame and a dark grey frame, it's less likely to be a problem.

I'm not an expert on this. I still say get a proper report that tells you exactly what you need to change to pass.

Still a perplexing situation..for GrafixJoe..if you google the harding test there is a demo of the actual program (PC based) in action. The BBC rules and regulations have some comments on it also. I guess I can tone it all down but I'm concerned over what Jude said regarding the speed of clips changing. It is late in the day, we have never had this problem before and we cannot miss the TV spot otherwise we are done for. As for my nanny state comments they were purely reactionary. Thanks everyone till next time

Here's the specs from the OFCOM guide, avaiable as a pdf here : [www.bbc.co.uk]



Ofcom Guidance Note on Flashing
Images and Regular Patterns in Television
Re-issued as Ofcom Notes (25 July 2005)
Revised and re-issued by the ITC in July 2001
Editorial amendment June 2002

1. Flickering or intermittent images and certain types of regular pattern can cause
problems for some viewers who have photosensitive epilepsy. These guidelines have
been drawn up following consultation with leading medical opinion in this area with
the aim of reducing the risk of exposure to potentially harmful stimuli.

2. Television is by nature a flickering medium. In Europe each transmitted picture is
refreshed 50 times each second and interlaced scanning generates flicker 25 times
each second. It is therefore impossible to eliminate the risk of television causing
convulsions in viewers with photosensitive epilepsy. To reduce risk the following
guidelines on visual content should be applied when flashing or regular patterns are
clearly discernible in normal domestic viewing conditions. It should be noted that the
level of any cumulative risk arising from successive sequences of ?potentially harmful?
flashes over a prolonged period is unknown. If, as medical opinion suggests, the risk
of seizures increases with the duration of flashing, broadcasters should note that it is
possible that a sequence of flashing images lasting more than 5 seconds might
constitute a risk even when it complies with the guidelines below.

3. A potentially harmful flash occurs when there is a pair of opposing changes in
luminance (i.e., an increase in luminance followed by a decrease, or a decrease
followed by an increase) of 20 candelas per square metre (cd.m-2) or more (see
notes 1 and 2). This applies only when the screen luminance of the darker image is
below 160 cd.m-2. Irrespective of luminance, a transition to or from a saturated red is
also potentially harmful.

3.1.1. Isolated single, double, or triple flashes are acceptable, but a sequence of
flashes is not permitted when both the following occur:

i. the combined area of flashes occurring concurrently occupies more than
one quarter of the displayed (see note 3) screen area; and

ii. there are more than three flashes within any one-second period. For
clarification, successive flashes for which the leading edges are separated by
9 frames or more are acceptable, irrespective of their brightness or screen
area.

4. Rapidly changing image sequences (e.g. fast cuts) are provocative if they result
in areas of the screen that flash, in which case the same constraints apply as for
flashes.

5. A potentially harmful regular pattern contains clearly discernible stripes when
there are more than five light-dark pairs of stripes in any orientation. The stripes may
be parallel or radial, curved or straight, and may be formed by rows of repetitive
elements such as polka dots. If the stripes change direction, oscillate, flash or
reverse in contrast they are more likely to be harmful than if they are stationary. If the
patterns obviously flow smoothly across, into, or out of the screen in one direction
they are exempt from restriction.

5.1. Potentially harmful patterns are not permitted when either of the
following two conditions apply:

i. the stripes are stationary and the pattern occupies more than 40% of the
displayed screen area; or

ii. the stripes change direction, oscillate, flash, or reverse in contrast and the
pattern occupies more than twenty five per cent of screen area; and in
addition to either of the above two conditions applying, when

iii. the screen luminance of the darker bars in the pattern is below 160 cd.m-2
and differs from the lighter bars by 20 cd.m-2 or more (see notes 1 and 2).

Footnotes:
1. Video waveform luminance is not a direct measure of display screen brightness.
Not all domestic display devices have the same gamma characteristic, but a display
with a gamma of 2.2 may be assumed for the purpose of determining electrical
measurements made to check compliance with these guidelines (see appendix I).

2. For the purpose of measurements made to check compliance with these
guidelines, pictures are assumed to be displayed in accordance with the ?home
viewing environment? described in Recommendation ITU-R BT.500 in which peak
white corresponds to a screen illumination of 200 cd.m-2.

3. It may be assumed that overscan on modern domestic television receiver displays
will normally be in the range 3.5% ± 1% of the overall picture width or height (as
indicated in EBU Technical recommendation R95-2000).

First issued as an ITC Guidance Note November 1994, first revised September 1999,
revised and re-issued July 2001. Re-issued by Ofcom (date tbc)

Further References

Relevant Codes

The ITC Programme Code, Autumn 1998, Section 7.3 revised September
1999,Section1.12(iii) revised Spring 2001

Ofcom Broadcasting Code, Harm and Offence, Section 2.10 (date tbc)

The ITC Advertising Standards Code, September 2002

External Publications

Harding, Graham F.A., & Jeavons Peter M. Photosensitive Epilepsy (1994) ISBN:
0898683 02 6

Hey Jude, thankyou for this...P

And this is the box we have in the edit suite to alert us to images that are likely to fail PSE guidlines, but everything goes through Harding test before air too. Station managers have some discretion to allow material that fails Harding to still air.

The idea that 60Hz is better than 50Hz for PSE is an interesting one - but I would have thought it just meant that edits had to be every 10 frames rather than every 8 frames.

Why it isn't implemented in the US is probably down to the networks figuring out how much it would cost them to abide by the regulations versus how much they are likely to have to pay out in lawsuits.

Did you ever post a .mov of the offending spot? Probably wouldn't be too tricky to spot the offending images. (i.e. flashes taking up more than 50% of the screen occuring more than 3 times a second OR rotatating patterns).

T

Hi guys,
We have just launched a website where you can test your video for broadcast compliance (OfCom, ITU recommendations).

www.HardingTestOnline.com

Just upload your files amd wait for the results - 100% online (24h/7days ).

Enjoy!

Michael Horton Wrote:
-------------------------------------------------------
> >
> I'd suspect after
> this thread is done we wont hear of it for another
> 5 years.

Pretty spot on!