Re: NSShadow changes on 10.5.2?

FROM : Ricky Sharp
DATE : Fri Feb 15 22:27:20 2008

On Feb 15, 2008, at 7:43 AM, Jim Correia wrote:

> On Feb 15, 2008, at 5:48 AM, Uli Kusterer wrote:
>

>> On 15.02.2008, at 11:36, João Pavão wrote:

>>> I had the impression that there was an NSShadow bug only on 10.5.0 
>>> thru 10.5.1. My shadows looked great on 10.4.x, and I had to add a 
>>> check to draw them with an extra 1pt offset on leopard to account 
>>> for that bug. This made them look exactly the same both on tiger 
>>> and on leopard.

>>
>> Well, for what it's worth, Safari (and by extension WebKit) has had 
>> this bug since 10.4.x. I haven't changed anything on my web site's 
>> text shadows in ages, and I am still running 10.4.11 on one Mac 
>> that still shows them with the bug. It still looked fine up to 
>> 10.5.1 (which I have on another Mac), but in 10.5.2 it was fixed, 
>> so all shadows designed for the old behaviour were suddenly off by 
>> one pixel down and right.
>>
>> So, I am actually able to see the matching 10.5.1 and 10.4 
>> behaviour side-by-side. Not sure about NSShadow.

>
> This is the behavior I am seeing as well. The shadows only look bad 
> on 10.5.2 (which now actually draws them at the offset you pass in.)
>
> I'd still like to derive the function
>
>     f(actual offset) = 10.5.1/10.4.x offset
>
> so that I don't have to hand tweak every shadow offset in my app so 
> that they look the same on both OS releases.



And this is why I'm really hoping they address the bug I filed (to 
revert drawing behavior to what it was, yet add an API that developers 
can choose to set if they want to in order to get new behavior). 
Nobody should have to attempt to fudge offsets based upon an OS release.

Also, for those customers with older versions of your app that choose 
to remain on those versions (e.g. maybe they don't want to go with a 
paid upgrade), they'll end up getting different output when moving to 
10.5.2 and beyond.  That's unacceptable IMO.

___________________________________________________________
Ricky A. Sharp        mailto:<email_removed>
Instant Interactive(tm)  http://www.instantinteractive.com