FROM : Ricky Sharp
DATE : Sat Nov 24 14:35:31 2007
On Nov 24, 2007, at 12:30 AM, Chris Hanson wrote:
> On Nov 23, 2007, at 1:41 PM, Ricky Sharp wrote:
>
>> As John pointing out, the control could have been already modified
>> to support Hi-DPI (aka resolution independence). My guess is that
>> Apple is using bitmapped images and are just providing multi-res
>> flavors (72 dpi and 288 dpi).
>>
>> But, IMO, that's bad! If I were implementing the control, I would
>> allow any size and would either vector-based images (e.g. PDF) or
>> drawing primitives for the content.
>
> Really, whether a control supports high DPI and whether it is
> implemented using vectors/primitives or using bitmaps is largely
> irrelevant. A bitmap sized for 72 DPI can serve just fine for
> higher resolutions, as if the *interface* is running at a higher
> resolution it will still be appropriately sized.
My point here about being bad was that due to the bitmapped graphics,
the control can only be sized to a certain maximum. To keep the math
simple, let's say this maximum is 1/2 square inch. On a 72 dpi
display, that's 36 x 36 pixels. On a 288 dpi display, that's 144 x
144 pixels. Thus, a multi-res TIFF with two images (36 x 36 and 144 x
144) will indeed work, but it is still designed to only look right up
to a maximum of 1/2 square inch. In fact, this is what res-ind does
for us; we design stuff based upon points and not pixels.
If using vector-based artwork, the control has the advantage to
infinitely scale its content. In the specific case of the progress
indicator, the content artwork need not be bitmapped based. Thus,
progress indicators of 1 sq inch, 2 sq inches, etc. will all have
properly rendered content.
___________________________________________________________
Ricky A. Sharp mailto:<email_removed>
Instant Interactive(tm) http://www.instantinteractive.com
DATE : Sat Nov 24 14:35:31 2007
On Nov 24, 2007, at 12:30 AM, Chris Hanson wrote:
> On Nov 23, 2007, at 1:41 PM, Ricky Sharp wrote:
>
>> As John pointing out, the control could have been already modified
>> to support Hi-DPI (aka resolution independence). My guess is that
>> Apple is using bitmapped images and are just providing multi-res
>> flavors (72 dpi and 288 dpi).
>>
>> But, IMO, that's bad! If I were implementing the control, I would
>> allow any size and would either vector-based images (e.g. PDF) or
>> drawing primitives for the content.
>
> Really, whether a control supports high DPI and whether it is
> implemented using vectors/primitives or using bitmaps is largely
> irrelevant. A bitmap sized for 72 DPI can serve just fine for
> higher resolutions, as if the *interface* is running at a higher
> resolution it will still be appropriately sized.
My point here about being bad was that due to the bitmapped graphics,
the control can only be sized to a certain maximum. To keep the math
simple, let's say this maximum is 1/2 square inch. On a 72 dpi
display, that's 36 x 36 pixels. On a 288 dpi display, that's 144 x
144 pixels. Thus, a multi-res TIFF with two images (36 x 36 and 144 x
144) will indeed work, but it is still designed to only look right up
to a maximum of 1/2 square inch. In fact, this is what res-ind does
for us; we design stuff based upon points and not pixels.
If using vector-based artwork, the control has the advantage to
infinitely scale its content. In the specific case of the progress
indicator, the content artwork need not be bitmapped based. Thus,
progress indicators of 1 sq inch, 2 sq inches, etc. will all have
properly rendered content.
___________________________________________________________
Ricky A. Sharp mailto:<email_removed>
Instant Interactive(tm) http://www.instantinteractive.com
| Related mails | Author | Date |
|---|---|---|
| slasktrattenator | Nov 23, 17:10 | |
| Andreas Mayer | Nov 23, 17:18 | |
| John Stiles | Nov 23, 19:53 | |
| slasktrattenator | Nov 23, 20:14 | |
| slasktrattenator | Nov 23, 22:11 | |
| Ricky Sharp | Nov 23, 22:41 | |
| slasktrattenator | Nov 23, 23:00 | |
| slasktrattenator | Nov 24, 00:40 | |
| Chris Hanson | Nov 24, 07:30 | |
| Ricky Sharp | Nov 24, 14:35 |






Cocoa mail archive

