qcms' bogus profile detection sometimes has false positives

NEW
Assigned to

Status

()

defect
10 years ago
7 years ago

People

(Reporter: jrmuizel, Assigned: jrmuizel)

Tracking

(Blocks 1 bug)

Trunk
Points:
---

Firefox Tracking Flags

(Not tracked)

Details

Attachments

(2 attachments)

Comment hidden (empty)
(Assignee)

Comment 1

10 years ago
Posted file scrs bogus profile
qcms reports this profile as bogus. The XYZ values sum to 0.950469971, 1.00003052, 1.08908081 which is not within the tolerance of our target. Is this profile actually bogus?
(Assignee)

Comment 2

10 years ago
Posted file HP bogus profile
This profile is also reported as bogus. The XYZs sum to {1.00039673, 0.999984741, 0.992767334} however the wtpt is {0.95, 1.0, 1.089}. We currently don't take the wtpt into account when checking whether a profile is bogus so perhaps the profile is not bogus?
(Assignee)

Comment 3

10 years ago
Chris can you take a look at these profiles and let me know if they should be reported as bogus or not?

Comment 4

10 years ago
The profile named "scrs bogus profile" is a bad profile:
1. The primaries add up to D65 white, which means the primaries are not properly adapted to D50. This is a requirement of the ICC spec.

2. The it has an internal ASCII name of "sRGB". Since the primaries are not at all sRGB primaries, I'd disqualify it on that basis alone.

"HP bogus profile" is also a bad profile.
1. The primaries add up to a non-D50 white. Either the primaries aren't adapted to D50 as required by the spec, or the primaries are invalid.
Blocks: 679875
Duplicate of this bug: 467404
You need to log in before you can comment on or make changes to this bug.