- From: Rick Byers <rbyers@google.com>
- Date: Wed, 22 Jan 2014 20:22:01 -0500
- To: Jacob Rossi <Jacob.Rossi@microsoft.com>
- Cc: "Asir Vedamuthu Selvasingh (MS OPEN TECH)" <asirveda@microsoft.com>, "public-pointer-events@w3.org" <public-pointer-events@w3.org>
- Message-ID: <CAFUtAY9UjWz3VvOmxh6yeurt4pM=xu0YEg9D4fvjoFt8EJEboQ@mail.gmail.com>
Yep, I'm very familiar with all those places geometry can get confused - we fixed multiple issues in Chrome on Android and ChromeOS :-). Thanks for looking into it! Rick On Wed, Jan 22, 2014 at 4:00 PM, Jacob Rossi <Jacob.Rossi@microsoft.com>wrote: > There is some variance from device to device as digitizer manufacturers > haven’t standardized on the sensitivity level for this metric. Touch > geometry is also not currently a part of the Windows Hardware Certification > Kit [1]. But I wouldn’t expect values that small. > > > > I’m looking into whether we do have a scaling problem in IE. Digitizer > resolutions don’t necessarily match the screen resolutions. So it’s > possible there’s an issue there. There’s also DPI scaling to account for as > well. Also keep in mind that there is no such thing as a physical unit in > CSS and browsers optimize scaling for consistent apparent visual size, not > consistent physical size. I was under the impression we validated all this > for width/height when we updated IE11, but I’ll double check. > > > > I wouldn’t be too concerned about sites compensating for IE. Geometry > hasn’t been used much in the wild yet and there are Windows devices that do > report appropriately sized widths and heights. I don’t believe there’s a > consistent offset or anything that a site might compensate for. > > > > I’ll report back with what I find. > > > > -Jacob > > > > [1] http://msdn.microsoft.com/en-us/library/windows/hardware/jj124227.aspx > > > > *From:* Rick Byers [mailto:rbyers@google.com] > *Sent:* Friday, January 17, 2014 1:11 PM > *To:* Jacob Rossi; Asir Vedamuthu Selvasingh (MS OPEN TECH) > *Cc:* public-pointer-events@w3.org > *Subject:* IE11 reporting incorrect scale for touch width/height > > > > Hi, > > On IE11 on my SurfaceRT, the reported width and height values seem way too > small to be correct in terms of CSS pixels (as specified by the standard > and MSDN document). Eg. with http://www.rbyers.net/eventTest.html I > typically see values of width/height=1, but with a really fat finger I can > get values up to 5. > > > > Is this just a bug in IE or the Surface? Are any other devices known to > properly support touch dimensions with pointer events? Any known web sites > that use this information? > > > > I ask because I'm a little nervous to have Polymer+Chrome accurately > report width/height values if there are sites already compensating for the > scale in IE not matching the documentation. > > > > Thanks, > > Rick >
Received on Thursday, 23 January 2014 01:22:48 UTC