Closed Bug 549666 Opened 15 years ago Closed 15 years ago

[d2d] form controls are fuzzy

Categories

(Core :: Graphics, defect)

x86
Windows 7
defect
Not set
normal

Tracking

()

RESOLVED FIXED

People

(Reporter: jrmuizel, Assigned: bas.schouten)

References

Details

Attachments

(4 files, 1 obsolete file)

Only seems to happen when using D2D. Happens on d3d9 and d3d10 cards.
Attached image gdi version
Attached image d2d version
I can confirm this on a DX9 card on bugzilla comments box scroll bar and also in the options menu -> tabs, all of the check mark boxes are fuzzy.
We should always position native elements on pixel boundaries.
Assignee: nobody → bas.schouten
Status: NEW → ASSIGNED
Attachment #430192 - Flags: review?(vladimir)
As per Vlad's suggestion on IRC. We should just round the entire rectangle. We can round the mNativeRect directly, since it won't be used anymore after this.
Attachment #430192 - Attachment is obsolete: true
Attachment #430193 - Flags: review?(vladimir)
Attachment #430192 - Flags: review?(vladimir)
Pushed, changeset 204de1667911.
Status: ASSIGNED → RESOLVED
Closed: 15 years ago
Resolution: --- → FIXED
Verified fixed using Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.3a3pre) Gecko/20100304 Minefield/3.7a3pre ID:20100304213613
Status: RESOLVED → VERIFIED
Seems there's a little side-effect. Sometimes the outside pixels aren't drawn with this patch.
Status: VERIFIED → REOPENED
Resolution: FIXED → ---
Did you back this out? If not, the bug should stay FIXED.
I'm not quite sure what the problem is in your screenshot. What's mNativeRect relative to?
Nope, I didn't back it out, so the problem is the right line of the Listbox saying 'Times New Roman'.
Status: REOPENED → RESOLVED
Closed: 15 years ago15 years ago
Resolution: --- → FIXED
(In reply to comment #11) > I'm not quite sure what the problem is in your screenshot. What's mNativeRect > relative to? It's relative to the context it's drawn to I believe.
Let's file a new bug for the regression and try to get a minimal testcase.
(In reply to comment #14) > Let's file a new bug for the regression and try to get a minimal testcase. This has been done. Bug 550680 is the new bug. Still working on how to fix it.
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: