W3C home > Mailing lists > Public > public-comments-wcag20@w3.org > December 2013

Flash threshold definition appears wrong

From: <noreply@w3.org>
Date: Mon, 23 Dec 2013 14:39:45 +0000
Message-Id: <E1Vv6fZ-00037Z-Hl@stuart.w3.org>
To: public-comments-wcag20@w3.org
Name: Mark Rogers
Email: mark.rogers@powermapper.com
Affiliation: Powermapper Software Limited
Document: W2
Item Number: Success Criterion 2.3.1
Part of Item: 
Comment Type: technical
Summary of Issue: Flash threshold definition appears wrong
Comment (Including rationale for any proposed change):
The source for the original flash thresholds seems to be UK technical guidance for TV here:
http://stakeholders.ofcom.org.uk/binaries/broadcast/guidance/gn_flash.pdf

This specifies luminosity differences in candelas, an absolute physical unit:

"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"

That's pretty easy to understand - there's a problem if flashes get brighter and darker by an absolute amount.

In WCAG 2 the definition has been translated from absolute units to units relative to the maximum relative luminance:
http://www.w3.org/TR/WCAG20/#general-thresholddef

"A general flash is defined as a pair of opposing changes in relative luminance of 10% or more of the maximum relative luminance where the relative luminance of the darker image is below 0.80;"

It's not entirely clear from the definition what the "maximum relative luminance" is the luminance of, but the only thing that seems to make sense is that it's brightest part of the flash.

There seems to be a problem here because "maximum relative luminance" is not an absolute, so it's possible to have flashing images with a low maximum relative luminance which:

- fail the WCAG definition; but
- don't fail the 20 candelas limit

For example, flashing from luminosity=0.0010 and luminosity=0.0012 (i.e. both colors in the flash are nearly black). That seems to fail the definition:

- change is luminosity = 0.0002 which is more than 10% of maximum relative luminance (10% * 0.0012 = 0.00012)
- luminance of darker image is below 0.80

Although it fails, the the flash is very hard to see because it's so dim. It appears the test gets more sensitive as the image pairs in the flash get darker.

In summary:

- the definition of "maximum relative luminance" is hard to understand; and
- there seems to be a technical error in the "general flash" definition 

Best Regards
Mark

Mark Rogers - mark.rogers@powermapper.com
PowerMapper Software Ltd - www.powermapper.com 
Registered in Scotland No 362274 Quartermile 2 Edinburgh EH3 9GL 












Proposed Change:
Think this would need to some debate and research to come up with an alternative.
Received on Monday, 23 December 2013 14:39:47 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:11:17 UTC