73 lines
2.9 KiB
Text
73 lines
2.9 KiB
Text
|
From sbaker@link.com Tue May 26 07:55:03 1998
|
||
|
|
X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil]
|
||
|
|
["1540" "Tue" "26" "May" "1998" "07:54:04" "-0500" "Steve Baker" "sbaker@link.com" nil "40" "Re: Question" "^From:" nil nil "5" nil nil nil nil nil]
|
||
|
|
nil)
|
||
|
|
X-VM-Last-Modified: (13793 54640 649954)
|
||
|
|
X-VM-IMAP-Retrieved: nil
|
||
|
|
X-VM-POP-Retrieved: nil
|
||
|
|
X-VM-Message-Order:
|
||
|
|
(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)
|
||
|
|
X-VM-Summary-Format: "%n %*%a %-17.17F %-3.3m %2d %4l/%-5c %I\"%s\"\n"
|
||
|
|
X-VM-Labels: ("r")
|
||
|
|
X-VM-VHeader: ("Resent-" "From:" "Sender:" "To:" "Apparently-To:" "Cc:" "Subject:" "Date:") nil
|
||
|
|
X-VM-Bookmark: 20
|
||
|
|
Received: from lfkw10.bgm.link.com (bgm.link.com [130.210.2.10])
|
||
|
|
by meserv.me.umn.edu (8.8.8/8.8.8) with ESMTP id HAA28563
|
||
|
|
for <curt@me.umn.edu>; Tue, 26 May 1998 07:55:02 -0500 (CDT)
|
||
|
|
Received: from borgus.bgm.link.com (borgus.bgm.link.com [130.210.236.13])
|
||
|
|
by lfkw10.bgm.link.com (8.8.6/HTI-Hack-8.8.4) with SMTP
|
||
|
|
id HAA00068 for <curt@me.umn.edu>; Tue, 26 May 1998 07:54:55 -0500 (CDT)
|
||
|
|
X-Sender: steve@borgus.bgm.link.com
|
||
|
|
Reply-To: Steve Baker <sbaker@link.com>
|
||
|
|
In-Reply-To: <199805230429.XAA10965@kenai.me.umn.edu>
|
||
|
|
Message-ID: <Pine.SGI.3.96.980526074643.2282C-100000@borgus.bgm.link.com>
|
||
|
|
MIME-Version: 1.0
|
||
|
|
Content-Type: TEXT/PLAIN; charset=US-ASCII
|
||
|
|
From: Steve Baker <sbaker@link.com>
|
||
|
|
To: "Curtis L. Olson" <curt@me.umn.edu>
|
||
|
|
Subject: Re: Question
|
||
|
|
Date: Tue, 26 May 1998 07:54:04 -0500 (CDT)
|
||
|
|
|
||
|
|
On Fri, 22 May 1998, Curtis L. Olson wrote:
|
||
|
|
|
||
|
|
> I'm running into floating point problems again where my scenery tiles
|
||
|
|
> sometimes have a pixel gap between them. I was wondering if you had
|
||
|
|
> any ideas ...
|
||
|
|
|
||
|
|
It's a problem that plagues us too.
|
||
|
|
|
||
|
|
I can't say that I have a magic solution.
|
||
|
|
|
||
|
|
One thing I have *considered* is to do everything in integers (for
|
||
|
|
the terrain skin at least).
|
||
|
|
|
||
|
|
This sounds counter-intuitive - but the source data for your
|
||
|
|
terrain is certainly not accurate to better than a meter - and
|
||
|
|
integers have the huge advantage that they don't suffer from
|
||
|
|
roundoff error providing all you do is add and subtract them.
|
||
|
|
|
||
|
|
However, I havn't had the nerve to try this yet - right now, we
|
||
|
|
kludge the edges of each terrain tile to make it a teeny-tiny bit
|
||
|
|
too big. The resulting overlap is very small - but enough to get
|
||
|
|
rid of the cracks.
|
||
|
|
|
||
|
|
> Anyways, what this forces me to do is for each tile I must
|
||
|
|
> glTranslate() by center_of_tile_to_be_drawn - center_of_tile_I'm_in.
|
||
|
|
> But, up until this point everything was done in double's so I was
|
||
|
|
> hoping the final translate about which is on the order of 20,000 -
|
||
|
|
> 30,000 meters would be ok.
|
||
|
|
|
||
|
|
But that translate could be done in integer meters.
|
||
|
|
|
||
|
|
> Anyways, I just thought I'd see if you had any brilliant thoughts on
|
||
|
|
> the subject.
|
||
|
|
|
||
|
|
Nope - it's a continual problem for me too.
|
||
|
|
|
||
|
|
Steve Baker (817)619-8776 (Vox/Vox-Mail)
|
||
|
|
Raytheon Systems Inc. (817)619-4028 (Fax)
|
||
|
|
Work: SBaker@link.com http://www.hti.com
|
||
|
|
Home: SJBaker1@airmail.net http://web2.airmail.net/sjbaker1
|
||
|
|
|