User Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Firefox/52.0 Build ID: 20161011030212 Steps to reproduce: 1. Start Nightly 2. Go to any sites (e.g. about:home) 3. Enable RDM (Ctrl+Shift+M) 4. Select device "Laptop (1920 x 1080)" 5. Conform viewport-dimension Actual results: Laptop (1920 x 1080) is different from viewport-dimension (1280 x 720) Expected results: Same

Thanks for reporting this! I don't think it should block enabling RDM, but it does point out a confusing scenario. The data[1] we have for the device "Laptop (1920 x 1080)" is: { "name": "Laptop (1920 x 1080)", "width": 1280, "height": 720, "pixelRatio": 1.5, ... }, This means we set the CSS visual viewport to 1280 x 720 (innerWidth/Height will give these values) and override the devicePixelRatio to 1.5. This means it is approximating a device with the physical size 1920 x 1080 as stated in the name (CSS visual width * dPR = physical device width). Will this be confusing to follow in the UI? Right now, I can see how it would be pretty confusing because nothing states the dPR on screen at all, but bug 1276971 will soon resolve that part so the dPR value is at least displayed. Even with the dPR on screen, do we think users will be confused that the "supposed" size suggested in the device name (really the physical size) doesn't match the viewport's dimensions on screen? If we do think it's confusing, we could rename the device in some way, such as: * State the CSS size instead * Use a more abstract word for the size * Add the dPR value into the name somehow (maybe 1280x720@1.5dppx or something...?) [1]: https://github.com/mozilla/simulated-devices/blob/a505b6a73f1c07d6b9ab27c78b893ea829929bc0/devices.json#L523-L533

I think the disconnect is hard to follow. You're right about this being a good bug! I have a question—what is CSS size? I think the biggest disconnect comes from selecting laptop (which has a defined pixel value), and then not seeing that pixel value reflected along the bottom. Whatever the solution is, I think it makes sense to have those two values match exactly, with the dPR value along the top in its section, which might mean changing the laptop label. Welcome to other interpretations, though—this is challenging, like most RDM bugs.

(In reply to Helen V. Holmes (:helenvholmes) (:✨) from comment #2) > I have a question—what is CSS size? If you're asking about the option I purposed of "State the CSS size instead", I was thinking the device label could change to "Laptop (1280 x 720)" since that's what innerWidth/Height will say and also the dimension inputs around the viewport. This problem doesn't come up for most devices because: * most are just names like "Apple iPhone 5" * the viewport just takes on the appropriate size, whatever it may be * there's no size data in the device's name For the "Use a more abstract word for the size" option, that could be something like "Laptop (Medium)".

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #3) > This problem doesn't come up for most devices because: > > * most are just names like "Apple iPhone 5" > * the viewport just takes on the appropriate size, whatever it may be > * there's no size data in the device's name Ahh, I understand now. I think this suggestion is a good one, then: > For the "Use a more abstract word for the size" option, that could be > something like "Laptop (Medium)". Medium laptops forever!

Luckily I was so goddamn detailed in my research that I can lookup the exact reasoning for such a choice! bug 1247302 comment 11 Those are the Surface Pro specs. I agree that we don't want a disconnect between the values, it obviously causes confusion. I'm not sure what the best solution would be. My first suggestion was to make sure we always displayed the effective size and if that was created with a dpr that we make that info available as well. For mobile devices this could still be confusing if someone knows what the resolution is supposed to read yet we are showing something else due to the dpr. Here's an example of what I was thinking: ________/\__________ | 1920 x 1080 | | 1280x720@1.5dppx | -------------------- #asciiartftw Changing the title feels like it will be confusing still because the page should behave like a different resolution than is being displayed.

(In reply to Bryan Clark (DevTools PM) [@clarkbw] from comment #5) > My first suggestion was to make sure we always displayed the effective size > and if that was created with a dpr that we make that info available as well. > For mobile devices this could still be confusing if someone knows what the > resolution is supposed to read yet we are showing something else due to the > dpr. > > Here's an example of what I was thinking: > > ________/\__________ > | 1920 x 1080 | > | 1280x720@1.5dppx | > -------------------- Is that a drawing of what would appear under the viewport (where the width x height editable field appear today), or would it go elsewhere? > Changing the title feels like it will be confusing still because the page > should behave like a different resolution than is being displayed. When you say "title" here you're referring to the device name that appears in the selector?

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #6) > Is that a drawing of what would appear under the viewport (where the width x > height editable field appear today), or would it go elsewhere? Yeah, drawing of what would appear under the viewport. > When you say "title" here you're referring to the device name that appears > in the selector? Right, going to medium feels like curing the symptom, not the underlying cause.

People continue to be confused by this. I went ahead and took the cheap approach of correcting the names: https://github.com/mozilla/simulated-devices/commit/23027a11df3a3993f8859f7b1d81778a16c2b7b8 Now they'll state both the size and dppx explicitly for these cases. (This data is loaded from a CDN, so deploying an update "fixes" all active versions of Firefox.)