FYI - Resolution Independence

  • It appears that Res-Ind did not yet get fully turned on Leopard for
    users.  And, there's now risk that it has become completely
    unsupported.  Here's part of a bug report reply I got yesterday just
    after the GM was released:

    "After considerable deliberation, this feature has been removed from
    Mac OS X Leopard. At this time, we do not anticipate re-adding this
    feature."

    QuartzDebug still allows one to set the user-interface scaling factor
    (although not to as many settings as the version under Tiger).

    My app that was 100% res-ind savvy still is except for large cursors.
    Large cursors get completely trashed (at least on an 8-core Mac Pro
    with ATI Radeon X1900).

    I'm not angry that I spent time making my app res-ind (been doing so
    since the Tiger tech talk back in 2004).  But, I'm quite disappointed
    this feature has not made it to prime time.  I'll continue ensuring my
    app works well will other scaling factors in hopes the feature may one
    day return.  And, will still file bugs.

    For anyone else that still wants to add that ability to your app,
    write in with your problems.  I've gotten past all the common gotchas
    and can share some code techniques.

    ___________________________________________________________
    Ricky A. Sharp        mailto:<rsharp...>
    Instant Interactive(tm)  http://www.instantinteractive.com
  • On 27 Oct 2007, at 12:47, Ricky Sharp wrote:

    > I'm not angry that I spent time making my app res-ind (been doing so
    > since the Tiger tech talk back in 2004).  But, I'm quite
    > disappointed this feature has not made it to prime time.  I'll
    > continue ensuring my app works well will other scaling factors in
    > hopes the feature may one day return.  And, will still file bugs.

    I thought Apple had always made clear that resolution independence, as
    an end-user feature, would "go prime time" when high-dpi panels were
    shipping in their machines.

    I don't think it was ever the intent (at least, as far as I'm aware)
    to provide an end-user UI for it before that point, and my guess is
    that even when high-dpi panels are shipping with Apple machines,
    resolution independence will be *primarily* an automatic feature (i.e.
    you plug-in a high-dpi panel, and that panel runs with a non-1.0 scale
    factor).  It seems to me that the idea of getting us to get our apps
    to support RI *now* is that, when the high-dpi panels become readily
    available, everything will "just work" in the usual way that people
    expect with Apple kit.

    I suppose I'm surprised that you're surprised.

    (The only query I have really about RI is what happens with machines
    that have a normal panel and a high-dpi panel... presumably someone at
    Apple has thought this through, or is doing so right now.)

    Kind regards,

    Alastair.

    --
    http://alastairs-place.net
  • On Oct 27, 2007, at 7:05 AM, Alastair Houghton wrote:

    > On 27 Oct 2007, at 12:47, Ricky Sharp wrote:
    >
    >> I'm not angry that I spent time making my app res-ind (been doing
    >> so since the Tiger tech talk back in 2004).  But, I'm quite
    >> disappointed this feature has not made it to prime time.  I'll
    >> continue ensuring my app works well will other scaling factors in
    >> hopes the feature may one day return.  And, will still file bugs.
    >
    > I thought Apple had always made clear that resolution independence,
    > as an end-user feature, would "go prime time" when high-dpi panels
    > were shipping in their machines.

    Well, they already ship such panels.  I don't have the slide handy,
    but I remember looking at the entire product lineup, and I believe the
    highest dpi device is now hovering around 200 dpi.

    > I don't think it was ever the intent (at least, as far as I'm aware)
    > to provide an end-user UI for it before that point, and my guess is
    > that even when high-dpi panels are shipping with Apple machines,
    > resolution independence will be *primarily* an automatic feature
    > (i.e. you plug-in a high-dpi panel, and that panel runs with a
    > non-1.0 scale factor).  It seems to me that the idea of getting us
    > to get our apps to support RI *now* is that, when the high-dpi
    > panels become readily available, everything will "just work" in the
    > usual way that people expect with Apple kit.
    >
    > I suppose I'm surprised that you're surprised.

    I'm not surprised, just a bit disappointed.  Other than the NSCursor
    issue, framework scaling is working as advertised.

    > (The only query I have really about RI is what happens with machines
    > that have a normal panel and a high-dpi panel... presumably someone
    > at Apple has thought this through, or is doing so right now.)

    You mean dual monitors? It should "just work".  After all, if your app
    draws to points and not pixels, it is now "resolution independent".
    One square inch worth of graphics your app draws will occupy a
    physical inch on either monitor, no matter what the individual
    monitor's dpi is.

    ___________________________________________________________
    Ricky A. Sharp        mailto:<rsharp...>
    Instant Interactive(tm)  http://www.instantinteractive.com
  • On Oct 27, 2007, at 7:05 AM, Alastair Houghton wrote:

    > On 27 Oct 2007, at 12:47, Ricky Sharp wrote:
    >
    >> I'm not angry that I spent time making my app res-ind (been doing
    >> so since the Tiger tech talk back in 2004).  But, I'm quite
    >> disappointed this feature has not made it to prime time.  I'll
    >> continue ensuring my app works well will other scaling factors in
    >> hopes the feature may one day return.  And, will still file bugs.
    >
    > I thought Apple had always made clear that resolution independence,
    > as an end-user feature, would "go prime time" when high-dpi panels
    > were shipping in their machines.
    >
    > I don't think it was ever the intent (at least, as far as I'm aware)
    > to provide an end-user UI for it before that point, and my guess is
    > that even when high-dpi panels are shipping with Apple machines,
    > resolution independence will be *primarily* an automatic feature
    > (i.e. you plug-in a high-dpi panel, and that panel runs with a
    > non-1.0 scale factor).  It seems to me that the idea of getting us
    > to get our apps to support RI *now* is that, when the high-dpi
    > panels become readily available, everything will "just work" in the
    > usual way that people expect with Apple kit.

    There was a 2007 WWDC session (161, Making Your Custom Controls,
    Icons, and Artwork Resolution Independent) that informed us how we
    need to make sure our apps were all resolution independent before
    Leopard shipped.
  • On Oct 27, 2007, at 7:33 AM, David Bainbridge wrote:

    > There was a 2007 WWDC session (161, Making Your Custom Controls,
    > Icons, and Artwork Resolution Independent) that informed us how we
    > need to make sure our apps were all resolution independent before
    > Leopard shipped.

    A similar session was also offered at the Leopard Tech Talks.

    ___________________________________________________________
    Ricky A. Sharp        mailto:<rsharp...>
    Instant Interactive(tm)  http://www.instantinteractive.com
  • On $BJ?@.(B 19/10/27, at 21:37, Ricky Sharp wrote:

    >
    > On Oct 27, 2007, at 7:33 AM, David Bainbridge wrote:
    >
    >> There was a 2007 WWDC session (161, Making Your Custom Controls,
    >> Icons, and Artwork Resolution Independent) that informed us how we
    >> need to make sure our apps were all resolution independent before
    >> Leopard shipped.
    >
    > A similar session was also offered at the Leopard Tech Talks.
    I was a bit dismayed at those talks, when they told us to use high-res
    bitmaps instead of vector art.... vista, as it stands, if an app uses
    the newest presentation api's can not only get resolution
    independence, but when zooming in for usability, everything is clear
    and crisp, there actually is a practical reason why thats important. A
    lot of people I know anyways, use the zoom feature, or reduce the
    resolution of their monitor and look at everything with "vaseline
    vision..."

    I guess we can only assume apple has re-prioritized things.....
    hopefully that means something "more important" has got their
    attention.....

    Andre


  • On Oct 27, 2007, at 7:59 AM, <listposter...> wrote:

    > I was a bit dismayed at those talks, when they told us to use high-
    > res bitmaps instead of vector art.

    Not sure who told you that, but that's wrong.  You can use either
    vector-based or multi-res bitmaps (basically containing at least a
    72dpi rep and 216dpi rep).

    Vector based is still preferable as it infinitely scales.  My app uses
    vector-based artwork (PDF) for everything except for the images fed to
    NSCursor in which I use multi-res TIFF.  Note that one can indeed use
    PDF artwork for NSCursor; it's just that my cursor images were 3D-
    renderings that didn't look right when vectorized.

    ___________________________________________________________
    Ricky A. Sharp        mailto:<rsharp...>
    Instant Interactive(tm)  http://www.instantinteractive.com
  • On 27 Oct 2007, at 13:33, David Bainbridge wrote:

    > There was a 2007 WWDC session (161, Making Your Custom Controls,
    > Icons, and Artwork Resolution Independent) that informed us how we
    > need to make sure our apps were all resolution independent before
    > Leopard shipped.

    AFAIK they've always talked about 2008, not specifically the release
    of Leopard, as the point when they expect this to be a mainstream
    feature.

    Kind regards,

    Alastair.

    --
    http://alastairs-place.net
  • On 27 Oct 2007, at 13:15, Ricky Sharp wrote:

    >> I thought Apple had always made clear that resolution independence,
    >> as an end-user feature, would "go prime time" when high-dpi panels
    >> were shipping in their machines.
    >
    > Well, they already ship such panels.  I don't have the slide handy,
    > but I remember looking at the entire product lineup, and I believe
    > the highest dpi device is now hovering around 200 dpi.

    I know it's been over 100dpi for a little while.  I haven't checked
    the specs myself though.

    > I'm not surprised, just a bit disappointed.  Other than the NSCursor
    > issue, framework scaling is working as advertised.

    I'd still expect RI to become a mainstream feature in the Leopard time-
    frame.  There have been noises for a while about high-dpi displays
    from the LCD manufacturers.  But I suppose we'll have to wait and see
    what Apple releases hardware-wise.

    >> (The only query I have really about RI is what happens with
    >> machines that have a normal panel and a high-dpi panel...
    >> presumably someone at Apple has thought this through, or is doing
    >> so right now.)
    >
    > You mean dual monitors? It should "just work".  After all, if your
    > app draws to points and not pixels, it is now "resolution
    > independent".  One square inch worth of graphics your app draws will
    > occupy a physical inch on either monitor, no matter what the
    > individual monitor's dpi is.

    Yeah, I'm not sure that's actually how things currently work in
    practice.  I think, presently, there is a lot of stuff tied to the
    idea that one window has one backing buffer with a single scale
    factor.  So if you drag a window so that it overlaps the boundary
    between the two displays, I think things will not be as one would
    really like.

    Kind regards,

    Alastair.

    --
    http://alastairs-place.net
  • On Oct 27, 2007, at 8:14 AM, Alastair Houghton wrote:

    > On 27 Oct 2007, at 13:15, Ricky Sharp wrote:
    >
    >>> You mean dual monitors? It should "just work".  After all, if your
    >>> app draws to points and not pixels, it is now "resolution
    >>> independent".  One square inch worth of graphics your app draws
    >>> will occupy a physical inch on either monitor, no matter what the
    >>> individual monitor's dpi is.
    >
    > Yeah, I'm not sure that's actually how things currently work in
    > practice.  I think, presently, there is a lot of stuff tied to the
    > idea that one window has one backing buffer with a single scale
    > factor.  So if you drag a window so that it overlaps the boundary
    > between the two displays, I think things will not be as one would
    > really like.

    You may be right, now that I think about it.  Yes, each window does
    have a single backing buffer.  I'm sure they'll work it out though.
    Probably a matter of ensuring the backing buffer contains the highest
    amount of pixels necessary for the highest dpi display.  Sort of
    reminds of the initial demos done ages ago showing a QuickTime movie
    spanning multiple monitors all at different bit depths.  I believe
    the frame buffer always contained a representation to draw on the
    highest bit-depth display and then downsampled when rendered on other
    displays.  Same should hold true for RI; downsample the portions to
    render on lower-dpi displays.

    Thankfully Cocoa (and I believe Carbon's HIView framework) uses
    framework scaling.  So there shouldn't be anything that we have to do
    in order to support this in our apps :)

    ___________________________________________________________
    Ricky A. Sharp        mailto:<rsharp...>
    Instant Interactive(tm)  http://www.instantinteractive.com
  • On 27 Oct 2007, at 14:34, Emanuele ∞ Vulcano wrote:

    > Il giorno 27/ott/07, alle ore 15:22, Ricky Sharp ha scritto:
    >
    >> I believe the frame buffer always contained a representation to
    >> draw on the highest bit-depth display and then downsampled when
    >> rendered on other displays.  Same should hold true for RI;
    >> downsample the portions to render on lower-dpi displays.
    >
    > It could also ask a view to drawRect: different portions of it with
    > different scaling coefficients, which would probably look better.
    > There are a lot of ways to work around the problem.

    Yeah, I'm not sure it's quite that simple, actually.  For instance,
    what happens if you use NSCopyBits() or similar?  And how do you deal
    with the problem of pixel-alignment rounding in such a case?  For
    instance, it might result in a horizontal line having an obvious kink
    where it goes across the displays.  There are other problems too (for
    instance, if a view stores any pixel co-ordinates, it could end up
    with some with one scale factor applied and some with a different one).

    It's actually quite a hard problem to solve in a good way.  Ricky's
    solution is a possibility, though I think we all know that downscaling
    automatically wouldn't produce the best results; looking at the
    NSWindow API, I believe Apple has already used a similar solution for
    the case where a window overlaps screens that have different display
    depths.

    Kind regards,

    Alastair.

    --
    http://alastairs-place.net
  • On Oct 27, 2007, at 8:07 AM, Alastair Houghton wrote:

    > On 27 Oct 2007, at 13:33, David Bainbridge wrote:
    >
    >> There was a 2007 WWDC session (161, Making Your Custom Controls,
    >> Icons, and Artwork Resolution Independent) that informed us how we
    >> need to make sure our apps were all resolution independent before
    >> Leopard shipped.
    >
    > AFAIK they've always talked about 2008, not specifically the release
    > of Leopard, as the point when they expect this to be a mainstream
    > feature.

    See http://developer.apple.com/leopard/overview/ and search for
    Resolution.
  • On 27 Oct 2007, at 16:01, David Bainbridge wrote:

    > On Oct 27, 2007, at 8:07 AM, Alastair Houghton wrote:
    >
    >> On 27 Oct 2007, at 13:33, David Bainbridge wrote:
    >>
    >>> There was a 2007 WWDC session (161, Making Your Custom Controls,
    >>> Icons, and Artwork Resolution Independent) that informed us how we
    >>> need to make sure our apps were all resolution independent before
    >>> Leopard shipped.
    >>
    >> AFAIK they've always talked about 2008, not specifically the
    >> release of Leopard, as the point when they expect this to be a
    >> mainstream feature.
    >
    > See http://developer.apple.com/leopard/overview/ and search for
    > Resolution.

    Sure, it's listed as "Leopard Technology".  So what?

    Remember, the version that's been released is just the first version
    of Leopard.  There's nothing stopping Apple from providing an end-user
    resolution independence UI in one of the inevitable updates.

    Anyway, this is getting off topic now.

    Kind regards,

    Alastair.

    --
    http://alastairs-place.net
  • on 10/27/07 6:14 AM, Alastair Houghton at <alastair...>
    wrote:

    > On 27 Oct 2007, at 13:15, Ricky Sharp wrote:
    >
    >>> I thought Apple had always made clear that resolution independence,
    >>> as an end-user feature, would "go prime time" when high-dpi panels
    >>> were shipping in their machines.
    >>
    >> Well, they already ship such panels.  I don't have the slide handy,
    >> but I remember looking at the entire product lineup, and I believe
    >> the highest dpi device is now hovering around 200 dpi.
    >
    > I know it's been over 100dpi for a little while.  I haven't checked
    > the specs myself though.

    Don't forget that the iPhone runs at 163 dpi and therefore Apple already has
    a shipping computer with a native high resolution display. So while
    resolution independence might not be all that useful for shipping *Macs*
    right now.... :-)

    Jeffrey
  • At least in my quick tests here, some apps work perfectly with RI on
    Leopard GM (Calculator seemed to work, your app, dock), some had some
    minor issues that shouldn't take much to fix (iCal, iChat, AB). My
    thinking is that they just ran out of time to fully implement RI in
    the shipping apps. I know Apple has said 2008, but its interesting on
    what Apple said about "after much deliberation the feature was
    removed". Is that just their terse way of saying its dead or to say it
    was moved to 10.6, or to say that since its not ready in Leopard yet
    (but maybe a future 10.5.x or LCD Monitor update).

    > Message: 14
    > Date: Sat, 27 Oct 2007 08:05:39 -0500
    > From: Ricky Sharp <rsharp...>
    > Subject: Re: FYI - Resolution Independence
    > To: <listposter...>
    > Cc: Cocoa Dev <cocoa-dev...>
    > Message-ID: <A46AA1B3-9D97-4F80-AC6E-42F97C5C1DE0...>
    > Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
    >
    >
    > On Oct 27, 2007, at 7:59 AM, <listposter...> wrote:
    >
    >> I was a bit dismayed at those talks, when they told us to use high-
    >> res bitmaps instead of vector art.
    >
    > Not sure who told you that, but that's wrong.  You can use either
    > vector-based or multi-res bitmaps (basically containing at least a
    > 72dpi rep and 216dpi rep).
    >
    > Vector based is still preferable as it infinitely scales.  My app uses
    > vector-based artwork (PDF) for everything except for the images fed to
    > NSCursor in which I use multi-res TIFF.  Note that one can indeed use
    > PDF artwork for NSCursor; it's just that my cursor images were 3D-
    > renderings that didn't look right when vectorized.
    >
    > ___________________________________________________________
    > Ricky A. Sharp        mailto:<rsharp...>
    > Instant Interactive(tm)  http://www.instantinteractive.com

    > Date: Sat, 27 Oct 2007 14:07:44 +0100
    > From: Alastair Houghton <alastair...>
    > Subject: Re: FYI - Resolution Independence
    > To: David Bainbridge <dbainbridge...>
    > Cc: Cocoa Development <cocoa-dev...>
    > Message-ID: <B9896844-8527-4956-B758-F57A4D839202...>
    > Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
    >
    > On 27 Oct 2007, at 13:33, David Bainbridge wrote:
    >
    >> There was a 2007 WWDC session (161, Making Your Custom Controls,
    >> Icons, and Artwork Resolution Independent) that informed us how we
    >> need to make sure our apps were all resolution independent before
    >> Leopard shipped.
    >
    > AFAIK they've always talked about 2008, not specifically the release
    > of Leopard, as the point when they expect this to be a mainstream
    > feature.
    >
    > Kind regards,
    >
    > Alastair.
    >
    > --
    > http://alastairs-place.net
    Alex Kac - President and Founder
    Web Information Solutions, Inc. - Microsoft Certified Partner

    "It is useless for sheep to pass a resolution in favor of
    vegetarianism while wolves remain of a different opinion."
    -- William Randolph Inge
  • On Oct 27, 2007, at 10:21 AM, Alex Kac <alex...> wrote:
    > I know Apple has said 2008, but its interesting on
    > what Apple said about "after much deliberation the feature was
    > removed".

    I got a similar response regarding a mistake in a comment in the
    TextEdit sample code - "after much deliberation", they removed the
    comment. So I don't read too much into these form-letter bug responses.

    --
    David Catmull
    <uncommon...>
    http://www.uncommonplace.com/
  • On $BJ?@.(B 19/10/27, at 22:05, Ricky Sharp wrote:

    >
    > On Oct 27, 2007, at 7:59 AM, <listposter...> wrote:
    >
    >> I was a bit dismayed at those talks, when they told us to use high-
    >> res bitmaps instead of vector art.
    >
    > Not sure who told you that, but that's wrong.  You can use either
    > vector-based or multi-res bitmaps (basically containing at least a
    > 72dpi rep and 216dpi rep).
    >
    > Vector based is still preferable as it infinitely scales.  My app
    > uses vector-based artwork (PDF) for everything except for the images
    > fed to NSCursor in which I use multi-res TIFF.  Note that one can
    > indeed use PDF artwork for NSCursor; it's just that my cursor images
    > were 3D-renderings that didn't look right when vectorized.
    I was told that for icons and toolbar items etc that I should use
    multi-resolution bitmaps, and for simple things like arrows (abstract
    icons) that I should use vector art.
    With vector icons, there are certain things such as line width etc
    that can have different meanings when scaled up and down (e.g do I
    want the border of this icon to always be 1pt in width even when
    scaled) not to mention anti aliasing..... when some icons are scaled
    down, the anti-aliasing is sometimes downright atrocious... (I have
    the same opinion about vista's anti-aliasing btw).
    Apple's vector anti-aliasing is not satisfactory IMO (I do like their
    font ant-aliasing though because it preserves the typeface - sorry Mr
    Splosky)

    In any case, the whole backing buffer thing is a hard problem to
    crack......

    >
    > ___________________________________________________________
    > Ricky A. Sharp        mailto:<rsharp...>
    > Instant Interactive(tm)  http://www.instantinteractive.com
    >
  • On Oct 27, 2007, at 5:15 AM, Ricky Sharp wrote:

    > Well, they already ship such panels.  I don't have the slide handy,
    > but I remember looking at the entire product lineup, and I believe
    > the highest dpi device is now hovering around 200 dpi.

    No, not even close. The highest pixel density display that Apple's
    shipping is on the iPhone, and that's 160 DPI.

    -jcr
  • >> Well, they already ship such panels.  I don't have the slide
    >> handy, but I remember looking at the entire product lineup, and I
    >> believe the highest dpi device is now hovering around 200 dpi.
    >
    > No, not even close. The highest pixel density display that Apple's
    > shipping is on the iPhone, and that's 160 DPI.

    iPod nano: 204 dpi
    But - I know - this has nothing to do with Resolution Independency of
    Leopard discussed here. ;-)

    Mani
    --
    http://www.mani.de
    iVolume - Loudness adjustment for iTunes.
    LittleSecrets - The encrypted notepad.
  • As another data point, I filed a bug report awhile back
    suggesting/complaining that the Quartz Debug tool didn't allow scaling
    below 1.0, pointing out that scaling below 1.0 would be something
    potentially useful (and something I would like) for small screens like
    on the 12" Powerbook.

    My bug report was closed a little over a month ago with the comment:

    "After careful consideration, there are currently no plans for this to
    be implemented, but we will continue to reference this issue in the
    future."
  • On Oct 29, 2007, at 6:12 AM, E. Wing wrote:

    > As another data point, I filed a bug report awhile back
    > suggesting/complaining that the Quartz Debug tool didn't allow scaling
    > below 1.0, pointing out that scaling below 1.0 would be something
    > potentially useful (and something I would like) for small screens like
    > on the 12" Powerbook.
    >
    > My bug report was closed a little over a month ago with the comment:
    >
    > "After careful consideration, there are currently no plans for this to
    > be implemented, but we will continue to reference this issue in the
    > future."

    FWIW, I have been able to run Xcode 2.x's QuartzDebug under Leopard;
    gives you a larger selection of scalings to choose from.

    ___________________________________________________________
    Ricky A. Sharp        mailto:<rsharp...>
    Instant Interactive(tm)  http://www.instantinteractive.com
  • > I thought Apple had always made clear that resolution independence, as
    > an end-user feature, would "go prime time" when high-dpi panels were
    > shipping in their machines.

    I would consider jamming 1920x1200 into a laptop to be high-dpi. I played
    with resolution independence a while back with an early seed, and wow, it
    was just gorgeous (the bits that worked, that is).

    --
    Scott Ribe
    <scott_ribe...>
    http://www.killerbytes.com/
    (303) 722-0567 voice
  • > FWIW, I have been able to run Xcode 2.x's QuartzDebug under Leopard;
    > gives you a larger selection of scalings to choose from.

    Ref my earlier comment about how good things looked--I also learned
    something else important: DO NOT BOOT BACK INTO TIGER WITH THAT SCALING
    PREFERENCE ACTIVE! This only applies on a system with 10.4 & 10.5 volumes
    and a shared home directory. But Tiger picked up the preference, tried to be
    resolution independent, and much hilarity ensued.

    --
    Scott Ribe
    <scott_ribe...>
    http://www.killerbytes.com/
    (303) 722-0567 voice
  • On 30 Oct 2007, at 02:03, Scott Ribe wrote:

    >> I thought Apple had always made clear that resolution independence,
    >> as
    >> an end-user feature, would "go prime time" when high-dpi panels were
    >> shipping in their machines.
    >
    > I would consider jamming 1920x1200 into a laptop to be high-dpi. I
    > played
    > with resolution independence a while back with an early seed, and
    > wow, it
    > was just gorgeous (the bits that worked, that is).

    The high-res 17" screens are about 133 dpi, which is getting close to
    twice the notional 72 dpi "point-to-pixel equivalence" that is stated
    in Apple's "Coordinate Systems and Transforms" documentation.  Having
    just bought a new MacBook last week I am enjoying the higher density
    screen but I have more than once felt that resolution independence is
    becoming a "need" rather than a "want" with this type of display.

    Nicko
  • On Nov 3, 2007, at 7:10 AM, Nicko van Someren wrote:

    > On 30 Oct 2007, at 02:03, Scott Ribe wrote:
    >
    >>> I thought Apple had always made clear that resolution
    >>> independence, as
    >>> an end-user feature, would "go prime time" when high-dpi panels were
    >>> shipping in their machines.
    >>
    >> I would consider jamming 1920x1200 into a laptop to be high-dpi. I
    >> played
    >> with resolution independence a while back with an early seed, and
    >> wow, it
    >> was just gorgeous (the bits that worked, that is).
    >
    > The high-res 17" screens are about 133 dpi, which is getting close
    > to twice the notional 72 dpi "point-to-pixel equivalence" that is
    > stated in Apple's "Coordinate Systems and Transforms"
    > documentation.  Having just bought a new MacBook last week I am
    > enjoying the higher density screen but I have more than once felt
    > that resolution independence is becoming a "need" rather than a
    > "want" with this type of display.

    Which aligns with Apple guidance on this topic at last WWDC... aka
    getting close being a need.

    -Shawn
previous month october 2007 next month
MTWTFSS
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31        
Go to today