Return-Path: Received: from nova.lunarpages.com ([67.210.126.175] verified) by media-motion.tv (CommuniGate Pro SMTP 4.2.10) with ESMTP-TLS id 4753690 for AE-List@media-motion.tv; Thu, 21 Jun 2012 16:47:52 +0200 Received: from 70.91.57.77-busname-panjde.hfc.comcastbusiness.net ([70.91.57.77]:22313 helo=[10.1.10.112]) by nova.lunarpages.com with esmtpa (Exim 4.77) (envelope-from ) id 1Shiic-0003bT-2y for AE-List@media-motion.tv; Thu, 21 Jun 2012 07:50:46 -0700 From: James WIlson Mime-Version: 1.0 (Apple Message framework v1084) Content-Type: multipart/alternative; boundary=Apple-Mail-4--941693784 Subject: Re: [AE] Link Trapcode particle size to camera position Date: Thu, 21 Jun 2012 10:50:42 -0400 In-Reply-To: To: "After Effects Mail List" References: Message-Id: <797E7A0B-A9CE-486D-942B-FA03902A3EC6@jwmm.com> X-Mailer: Apple Mail (2.1084) X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - nova.lunarpages.com X-AntiAbuse: Original Domain - media-motion.tv X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - jwmm.com --Apple-Mail-4--941693784 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii Yes, wouldn't you want to place the spherical field at the camera's = position so that no particle will ever actually come close enough to the = lens to give that looming appearance? I haven't done this like some of = you, but that's what makes sense to me. On Jun 20, 2012, at 8:29 PM, Darby Edelen wrote: > I'm confused about the goal. >=20 > Should they ALL look like distant stars? If so, then is there a = reason you're not actually making them all distant? You could use a = spherical field for that. Although normally I'd just use Horizon to = create a distant starfield >=20 > If your stars come close to the camera they'll almost certainly look = wrong even if they're small. >=20 > -D >=20 >=20 > On Wed, Jun 20, 2012 at 5:15 PM, jarret langmeire = wrote: > Ok, just ran through those options but no cigar. The problem is that = the camera is moving through the star field resulting in those very = unrealistic BIG particle stars. With regard to the near vanish and fade, = that simply adjusts the opacity as opposed to the size. What's been = working really well for the "classic" warp fly-through is to use a = straight gradient and mapping it across X/Z on the size layer map. That = scales up the stars in the distance and whilst scaling the foreground = stars down. However, once you apply "disperse" and a fractal displace it = mixes up the direct relationship to the gradient.=20 >=20 > I was hoping that using a complex OBJ model to generate a mixed star = field would do the trick but unfortunately that option does not allow me = to map the gradient over a specific axis.=20 >=20 > Jarret >=20 --Apple-Mail-4--941693784 Content-Transfer-Encoding: 7bit Content-Type: text/html; charset=us-ascii Yes, wouldn't you want to place the spherical field at the camera's position so that no particle will ever actually come close enough to the lens to give that looming appearance? I haven't done this like some of you, but that's what makes sense to me.



On Jun 20, 2012, at 8:29 PM, Darby Edelen wrote:

I'm confused about the goal.

Should they ALL look like distant stars?  If so, then is there a reason you're not actually making them all distant?  You could use a spherical field for that. Although normally I'd just use Horizon to create a distant starfield

If your stars come close to the camera they'll almost certainly look wrong even if they're small.

-D


On Wed, Jun 20, 2012 at 5:15 PM, jarret langmeire <langmeire@gmail.com> wrote:
Ok, just ran through those options but no cigar. The problem is that the camera is moving through the star field resulting in those very unrealistic BIG particle stars. With regard to the near vanish and fade, that simply adjusts the opacity as opposed to the size. What's been working really well for the "classic" warp fly-through is to use a straight gradient and mapping it across X/Z on the size layer map. That scales up the stars in the distance and whilst scaling the foreground stars down. However, once you apply "disperse" and a fractal displace it mixes up the direct relationship to the gradient.

I was hoping that using a complex OBJ model to generate a mixed star field would do the trick but unfortunately that option does not allow me to map the gradient over a specific axis. 

Jarret


--Apple-Mail-4--941693784--