From info-volumizer-owner@els.sgi.com Tue Mar 1 00:16:04 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j218AZn06275805; Tue, 1 Mar 2005 00:10:35 -0800 (PST) Received: from vimg3.rdg.ac.uk (vimg3.rdg.ac.uk [134.225.1.80]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j1SGlCva6290148 for ; Mon, 28 Feb 2005 08:47:17 -0800 (PST) Received: from vimp2.rdg.ac.uk ([134.225.16.91]) by vimg3.rdg.ac.uk (Exim: gateway) with esmtp id 1D5o36-0001Px-00 for info-volumizer@els.sgi.com; Mon, 28 Feb 2005 16:47:12 +0000 Received: from vimh1.rdg.ac.uk ([134.225.16.83]) by vimp2.rdg.ac.uk (Exim: virusscanner) with esmtp id 1D5o30-0004DV-00 for info-volumizer@els.sgi.com; Mon, 28 Feb 2005 16:47:06 +0000 Received: from vimb3.rdg.ac.uk ([134.225.1.43] helo=vimb3) by vimh1.rdg.ac.uk (Exim: host) with esmtp id 1D5o2z-0007Mp-00 for info-volumizer@els.sgi.com; Mon, 28 Feb 2005 16:47:05 +0000 Received: from vimb3 ([127.0.0.1] helo=rdg.ac.uk) by vimb3 (Exim: forwarder) with esmtp id 1D5o2z-0005tl-00 for info-volumizer@els.sgi.com; Mon, 28 Feb 2005 16:47:05 +0000 Content-Transfer-Encoding: 8bit User-Agent: IMHO/0.98.3 (Webmail for Roxen) X-Originating-IP: [134.225.56.20] Date: Mon, 28 Feb 2005 16:47:05 -0000 From: Content-Type: text/plain; charset=iso-8859-1 Subject: [info-volumizer] Volumizer XML To: info-volumizer@els.sgi.com MIME-Version: 1.0 Message-Id: X-Scanner: exiscan *1D5o30-0004DV-00*NRJ0bqgExwo* (The University of Reading) Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com Dear all I am using the XML file to read my volumes in Volumizer. I would like to break my volume into sections (say break the skull into a jaw and the remaining part), Can you please advise on how to do that. I am a novice user of Volumizer and I don't know how to do it within the code. I only want to do it using the XML file format if possible. Any help is appreciated. Sam From info-volumizer-owner@els.sgi.com Tue Mar 1 00:15:56 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j218AZ7J6335867; Tue, 1 Mar 2005 00:10:35 -0800 (PST) Received: from hastur.corp.sgi.com (hastur.corp.sgi.com [198.149.32.33]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j1S9W4va6281942 for ; Mon, 28 Feb 2005 01:32:11 -0800 (PST) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by hastur.corp.sgi.com (8.12.9/8.12.10/SGI_generic_relay-1.2) with ESMTP id j1S9VpVt111998768 for ; Mon, 28 Feb 2005 01:31:51 -0800 (PST) X-ASG-Debug-ID: 1109583119-7696-457-0 X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from acvstg03.intern.acv.ac.at (acvstg03.acv.ac.at [194.152.172.83]) by cuda.sgi.com (Spam Firewall) with ESMTP id CC8F9D62808F for ; Mon, 28 Feb 2005 01:32:00 -0800 (PST) MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----_=_NextPart_001_01C51D78.AA1DB2A0" X-ASG-Orig-Subj: Volumizer bizarre behavior Subject: [info-volumizer] Volumizer bizarre behavior X-MimeOLE: Produced By Microsoft Exchange V6.0.6487.1 Content-class: urn:content-classes:message Date: Mon, 28 Feb 2005 10:34:13 +0100 Message-ID: <50726031C3515F4F879593DD4E801AE622D25E@acvstg03.intern.acv.ac.at> X-MS-Has-Attach: yes X-MS-TNEF-Correlator: Thread-Topic: Volumizer bizarre behavior Thread-Index: AcUdeKoaSOnqzPunQNuI8klSk5KRpQ== From: "Oscar Serrano" To: X-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: 0.43 X-Barracuda-Spam-Status: No, SCORE=0.43 using per-user scores of TAG_LEVEL=3.5 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=1000.0 tests=HTML_MESSAGE, HTML_TAG_BALANCE_BODY X-Barracuda-Spam-Report: Code version 2.64, rules version 2.1.1381 Rule breakdown below pts rule name description ---- ---------------------- ------------------------------------------- 0.42 HTML_TAG_BALANCE_BODY BODY: HTML has unbalanced "body" tags 0.01 HTML_MESSAGE BODY: HTML included in message Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com This is a multi-part message in MIME format. ------_=_NextPart_001_01C51D78.AA1DB2A0 Content-Type: multipart/alternative; boundary="----_=_NextPart_002_01C51D78.AA1DB2A0" ------_=_NextPart_002_01C51D78.AA1DB2A0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hello, =20 We are just starting to evaluate the possibilities of volumizer but we have detected some bizarre output. We have reproduced the error in a simple program that I enclose. It just draws a shape with cube objects using the SimpleShader. The cubes are showing strange stripes that vary when moving them and we also have some strange kind of noise (I enclose a snapshot of the output). We have also experience this problem using the LUTShader and using different graphic cards. Does anyone have an idea what is wrong???=20 =20 Thank you, Oscar Serrano =20 =20 =20 ------_=_NextPart_002_01C51D78.AA1DB2A0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hello,

 

      We are just = starting to evaluate the possibilities of volumizer but we have detected some = bizarre output. We have reproduced the error in a simple program that I enclose. = It just draws a shape with cube objects using the SimpleShader. The cubes = are showing strange stripes that vary when moving them and we also have some strange kind of noise (I enclose a snapshot of the output). We have also experience this problem using the LUTShader and using different graphic = cards. Does anyone have an idea what is wrong???

 

Thank you,

        = Oscar Serrano

 

 

 

------_=_NextPart_002_01C51D78.AA1DB2A0-- ------_=_NextPart_001_01C51D78.AA1DB2A0 Content-Type: application/octet-stream; name="volumizertest.cpp" Content-Transfer-Encoding: base64 Content-Description: volumizertest.cpp Content-Disposition: attachment; filename="volumizertest.cpp" Ly8gdm9sdW1pemVydGVzdC5jcHAgOgoKLy8gSW5jbHVkZSB3aW5kb3dzLmggYmVmb3JlIGluY2x1 ZGluZyBhbnl0aGluZyBlbHNlIGlmIG9uIHdpbmRvd3MuCiNpZmRlZiBfV0lORE9XUwojaW5jbHVk ZSA8d2luZG93cy5oPgojZW5kaWYKCi8vIEMgLyBDKysKI2luY2x1ZGUgPHN0ZGxpYi5oPgovLyBJ bmNsdWRlIGFwcHJvcHJpYXRlIGlvc3RyZWFtIGhlYWRlciBkZXBlbmRpbmcgdXBvbiBwbGF0Zm9y bQojaWZkZWYgX1dJTkRPV1MKI2luY2x1ZGUgPGlvc3RyZWFtPgp1c2luZyBuYW1lc3BhY2Ugc3Rk OwojZWxzZQojaW5jbHVkZSA8aW9zdHJlYW0uaD4KI2VuZGlmCi8vIE9wZW5HTAojaW5jbHVkZSA8 R0wvZ2wuaD4KI2luY2x1ZGUgPEdML2dsdXQuaD4KCi8vIFZvbHVtaXplcjIKI2luY2x1ZGUgPFZv bHVtaXplcjIvU2hhcGUuaD4KI2luY2x1ZGUgPFZvbHVtaXplcjIvQmxvY2suaD4KI2luY2x1ZGUg PFZvbHVtaXplcjIvQXBwZWFyYW5jZS5oPgojaW5jbHVkZSA8Vm9sdW1pemVyMi9QYXJhbWV0ZXJW b2x1bWVUZXh0dXJlLmg+CiNpbmNsdWRlIDxWb2x1bWl6ZXIyL1RNUmVuZGVyQWN0aW9uLmg+CiNp bmNsdWRlIDxWb2x1bWl6ZXIyL1RNU2ltcGxlU2hhZGVyLmg+CgovLyBHbG9iYWwgdmFyaWFibGVz IAp2elNoYXBlICpwU2hhcGUgPSBOVUxMOwp2elRNUmVuZGVyQWN0aW9uICpwUmVuZGVyQWN0aW9u ID0gTlVMTDsKR0xpbnQgdmlld3BvcnRbNF07CmludCBsYXN0UG9zaXRpb25bMl0gPSB7MCwgMH07 CmZsb2F0IGFuZ2xlc1syXSAgICAgPSB7MCwgMH0sIGxhc3RBbmdsZXNbMl0gPSB7MCwgMH07CgoK Ly8vLy8vLy8vLy8vLy8vLy8vLy8vLy8vIFZvbHVtaXplciAvLy8vLy8vLy8vLy8vLy8vLy8vLy8v Ly8vLy8vLy8KDQovLyBsb2FkVm9sdW1lRGF0YSAtIExvYWQgdGhlIHZvbHVtZSBkYXRhIGFuZCBp bml0aWFsaXplIHRoZSBzaGFwZSBub2RlLgp2b2lkIGxvYWRWb2x1bWVEYXRhKCkKewoJaW50IGks aixrOwoJLy9DcmVhdGUgQXJlYQoJQllURSBwYnlBcmVhWzEwMF1bMTAwXVsxMDBdOwoJY29uc3Qg aW50IHBpRGF0YURpbVszXSA9IHsxMDAsIDEwMCwgMTAwfTsJCgoJZm9yKGk9MDtpPDEwMDtpKysp CgkJZm9yKGo9MDtqPDEwMDtqKyspCgkJCWZvcihrPTA7azwxMDA7aysrKQoJCQl7CgkJCQlpZigo aT4yMCAmJiBpPDUwKSAmJiAoaj4yMCAmJiBqPDUwKSAmJiAoaz4xMCAmJiBrPDQwKSkKCQkJCQlw YnlBcmVhW2ldW2pdW2tdPTE4MDsKCQkJCWVsc2UgaWYoKGk+MjAgJiYgaTw1MCkgJiYgKGo+MjAg JiYgajw1MCkgJiYgKGs+NjAgJiYgazw5MCkpCgkJCQkJcGJ5QXJlYVtpXVtqXVtrXT0yNTU7CgkJ CQllbHNlIAoJCQkJCXBieUFyZWFbaV1bal1ba109MDsKCQkJfQoKICAgIC8vIENyZWF0ZSB0aGUg dGV4dHVyZQ0KICAgIHZ6UGFyYW1ldGVyVm9sdW1lVGV4dHVyZSAqcFZvbHVtZSA9IG5ldyB2elBh cmFtZXRlclZvbHVtZVRleHR1cmUgKA0KICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICBwaURhdGFEaW0sDQogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICh2b2lkICopIHBieUFyZWEsDQogICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgIFZaX1VOU0lHTkVEX0JZVEUsIA0KCQkJCQkJCQkJCQlWWl9MVU1JTkFO Q0UsCgkJCQkJCQkJCQkJVlpfREVGQVVMVF9JTlRFUk5BTF9GT1JNQVQpOwoKCXZ6U2hhZGVyICpw U2hhZGVyID0gbmV3IHZ6VE1TaW1wbGVTaGFkZXIoKTsKCXZ6QXBwZWFyYW5jZSAqcEFwcGVhcmFu Y2UgPSBuZXcgdnpBcHBlYXJhbmNlKHBTaGFkZXIpOwoJcEFwcGVhcmFuY2UtPnNldFBhcmFtZXRl cigidm9sdW1lIixwVm9sdW1lKTsJCglwU2hhZGVyLT51bnJlZigpOwoJcFZvbHVtZS0+dW5yZWYo KTsKCgkvLyBJbml0aWFsaXplIGdlb21ldHJ5CiAgICB2ekdlb21ldHJ5ICpwR2VvbWV0cnkgPSBu ZXcgdnpCbG9jaygpOwoKICAgIC8vIEluaXRpYWxpemUgc2hhcGUgbm9kZQogICAgcFNoYXBlID0g bmV3IHZ6U2hhcGUocEdlb21ldHJ5LCBwQXBwZWFyYW5jZSk7CgogICAgcEdlb21ldHJ5LT51bnJl ZigpOwogICAgcEFwcGVhcmFuY2UtPnVucmVmKCk7CgogICAgaWYoIXBSZW5kZXJBY3Rpb24pIHsK ICAgICAgLy8gSW5pdGlhbGl6ZSB0aGUgcmVuZGVyIGFjdGlvbgogICAgICBwUmVuZGVyQWN0aW9u ID0gbmV3IHZ6VE1SZW5kZXJBY3Rpb24oMSk7CgkgIHBSZW5kZXJBY3Rpb24tPm1hbmFnZShwU2hh cGUpOwogICAgfQp9CgovLyBEcmF3IHRoZSB2b2x1bWUgZGF0YQp2b2lkIHJlbmRlclZvbHVtZURh dGEoKQp7CiAgLy8gQmVnaW4gZHJhd2luZwogIHBSZW5kZXJBY3Rpb24tPmJlZ2luRHJhdyhWWl9S RVNUT1JFX0dMX1NUQVRFX0JJVCk7CiAgICBwUmVuZGVyQWN0aW9uLT5kcmF3KHBTaGFwZSk7IAog IHBSZW5kZXJBY3Rpb24tPmVuZERyYXcoKTsKfQoKLy8gQ2xlYW51cCB1cCB0aGUgc2hhcGUgbm9k ZSBhbmQgdGhlIHJlbmRlciBhY3Rpb24Kdm9pZCBjbGVhbnVwKCkgCnsKICAgIC8vIERlbGV0ZSB0 aGUgcmVuZGVyIGFjdGlvbiBhbmQgdW5yZWYoKSB0aGUgc2hhcGUgbm9kZQogICAgCiAgICBwUmVu ZGVyQWN0aW9uLT51bm1hbmFnZShwU2hhcGUpOwogICAgZGVsZXRlIHBSZW5kZXJBY3Rpb247CiAg ICBwU2hhcGUtPnVucmVmKCk7Cn0KCi8vLy8vLy8vLy8vLy8vLy8vLy8vLy8vIEdMVVQgY2FsbGJh Y2sgZnVuY3Rpb25zLy8vLy8vLy8vLy8vLy8vLy8vLwoKLy8gZ2x1dERpc3BsYXlGdW5jKCkgY2Fs bGJhY2sgZnVuY3Rpb24Kdm9pZCBkaXNwbGF5KCkKewoJZ2xDbGVhcihHTF9DT0xPUl9CVUZGRVJf QklUKTsKCiAgICAvLyBWaWV3cG9ydAogICAgZ2xWaWV3cG9ydCh2aWV3cG9ydFswXSwgdmlld3Bv cnRbMV0sIHZpZXdwb3J0WzJdLCB2aWV3cG9ydFszXSk7CiAgICAKICAgIC8vIFByb2plY3Rpb24g bWF0cml4CiAgICBnbE1hdHJpeE1vZGUoR0xfUFJPSkVDVElPTik7CiAgICBnbExvYWRJZGVudGl0 eSgpOwogICAgZ2xPcnRobygtMS4yLCAxLjIsIC0xLjIsIDEuMiwgLTEuMiwgMS4yKTsKICAgIAog ICAgLy8gTW9kZWx2aWV3IG1hdHJpeAogICAgZ2xNYXRyaXhNb2RlKEdMX01PREVMVklFVyk7CiAg ICBnbExvYWRJZGVudGl0eSgpOwogICAgZ2xSb3RhdGVkKCA5NiArIGFuZ2xlc1sxXSwgMSwgMCwg MCk7CiAgICBnbFJvdGF0ZWQoMTg2ICsgYW5nbGVzWzBdLCAwLCAwLCAxKTsKICAgIGdsU2NhbGVm KDEuNSwgMS41LCAxLjUpOwogICAgZ2xUcmFuc2xhdGVmKC0gMC41LCAtIDAuNSwgLSAwLjUpOwog ICAgCiAgICAvLyBFbmFibGUgYmFjay10by1mcm9udCBhbHBoYSBibGVuZGluZwogICAgZ2xFbmFi bGUoR0xfQkxFTkQpOwogICAgZ2xCbGVuZEZ1bmMoR0xfU1JDX0FMUEhBLCBHTF9PTkVfTUlOVVNf U1JDX0FMUEhBKTsJCS8vVHJhbnNwYXJlbmN5IAoKICAgIC8vIFJlbmRlciB0aGUgdm9sdW1lIGRh dGEKICAgIHJlbmRlclZvbHVtZURhdGEoKTsNCglnbHV0U3dhcEJ1ZmZlcnMoKTsKfQoKLy8gZ2x1 dEtleWJvYXJkRnVuYygpIGNhbGxiYWNrIGZ1bmN0aW9uCnZvaWQga2V5Ym9hcmQodW5zaWduZWQg Y2hhciBrZXksIGludCB4LCBpbnQgeSkKewogICAgc3dpdGNoIChrZXkpIHsKICAgIGNhc2UgMjc6 CiAgICAgICAgY2xlYW51cCgpOwoJZXhpdCgwKTsKICAgIH0KfQoKLy8gZ2x1dFJlc2hhcGVGdW5j KCkgY2FsbGJhY2sgZnVuY3Rpb24Kdm9pZCByZXNoYXBlKGludCB3aWR0aCwgaW50IGhlaWdodCkK ewogICAgLy8gVXBkYXRlIHZpZXdwb3J0CiAgICB2aWV3cG9ydFswXSA9IDA7ICAgICAgdmlld3Bv cnRbMV0gPSAwOwogICAgdmlld3BvcnRbMl0gPSB3aWR0aDsgIHZpZXdwb3J0WzNdID0gaGVpZ2h0 OwogICAgZ2x1dFBvc3RSZWRpc3BsYXkoKTsKfQoKLy8gZ2x1dE1vdXNlRnVuYygpIGNhbGxiYWNr IGZ1bmN0aW9uCnZvaWQgbW91c2UoaW50IGJ1dHRvbiwgaW50IHN0YXRlLCBpbnQgeCwgaW50IHkp CnsKICAgIGlmIChzdGF0ZSA9PSBHTFVUX0RPV04pIHsKICAgICAgICBsYXN0UG9zaXRpb25bMF0g PSB4OwogICAgICAgIGxhc3RQb3NpdGlvblsxXSA9IHk7CiAgICAgICAgbGFzdEFuZ2xlc1swXSA9 IGFuZ2xlc1swXTsKICAgICAgICBsYXN0QW5nbGVzWzFdID0gYW5nbGVzWzFdOwogICAgfQp9Cgov LyBnbHV0TW90aW9uRnVuYygpIGNhbGxiYWNrIGZ1bmN0aW9uCnZvaWQgbW90aW9uKGludCB4LCBp bnQgeSkKewogICAgYW5nbGVzWzBdID0gbGFzdEFuZ2xlc1swXSArIChsYXN0UG9zaXRpb25bMF0g LSB4KSAvIDQuMDsKICAgIGFuZ2xlc1sxXSA9IGxhc3RBbmdsZXNbMV0gKyAoeSAtIGxhc3RQb3Np dGlvblsxXSkgLyA0LjA7CiAgICBnbHV0UG9zdFJlZGlzcGxheSgpOwp9CgovLyBtYWluCmludCBt YWluKGludCBhcmdjLCBjaGFyICphcmd2W10pCnsgIAogICBnbHV0SW5pdCgmYXJnYywgYXJndik7 CiAgIGxvYWRWb2x1bWVEYXRhKCk7CiAgIAogICAvLyBJbml0aWFsaXplIHdpbmRvdwogICBnbHV0 SW5pdERpc3BsYXlNb2RlKEdMVVRfUkdCQSB8IEdMVVRfRE9VQkxFKTsKICAgZ2x1dENyZWF0ZVdp bmRvdygiU2ltcGxlIFZvbHVtZSBWaWV3ZXIiKTsKICAgCiAgIC8vIEluaXRpYWxpemUgY2FsbGJh Y2tzCiAgIGdsdXREaXNwbGF5RnVuYyhkaXNwbGF5KTsKICAgZ2x1dFJlc2hhcGVGdW5jKHJlc2hh cGUpOwogICBnbHV0S2V5Ym9hcmRGdW5jKGtleWJvYXJkKTsKICAgZ2x1dE1vdXNlRnVuYyhtb3Vz ZSk7CiAgIGdsdXRNb3Rpb25GdW5jKG1vdGlvbik7CiAgIAogICBnbHV0TWFpbkxvb3AoKTsKICAg CiAgIHJldHVybiAxOwp9Cg0K ------_=_NextPart_001_01C51D78.AA1DB2A0 Content-Type: image/jpeg; name="cubes.JPG" Content-Transfer-Encoding: base64 Content-Description: cubes.JPG Content-Disposition: attachment; filename="cubes.JPG" /9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a HBwgJC4nICIsIxwcKDcpLDAxNDQ0Hyc5PTgyPC4zNDL/2wBDAQkJCQwLDBgNDRgyIRwhMjIyMjIy MjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjL/wAARCAHhAb4DASIA AhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQA AAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3 ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWm p6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEA AwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSEx BhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElK U1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3 uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwDsvGfj OXwncabBBpkV412hIBdI9u1UJ5KnOd36Vzf/AAtjUv8AoV4f/AuL/wCN10+rIr/FTwUrqGUx3eQR kH/Rlr0P7Hbf8+0P/fArphKhCEeeF2/O3VmE1WcnySSXpf8AU8Ruvi/qFlGslz4WSNGJVSZ05I6j /V1WHxwkP/MvRf8Af9f/AI3W18e4IYtD0cxxIhN0+SqgZ+Ssw6VofiQwXnhi+sE0qG70+G509tGi We2SR0Qfv3RjI24EnJII3ZLDhuiNPDygp8mj82QpVU2myIfG6Q/8y/F/3+X/AON04fGyQ/8AMAi/ 7/L/APG6ST4d6Hc6zp9i0t+t1q93qcSTo8SxwG3kcKfKWMZBAGQCo64wOBQ07wPoV9o0GuZ1FLSX Sb29+yfaEMge3kVceZ5eNrBv7mQR1NP2WG/l/Pz8/Jj5qnc0R8aZD/zAIv8Av8v/AMbpw+M8h/5g MX/f5f8A43VnT/CdrD4X1aws7iaO31u20adPOxI8HnTlSCRtD4IJzhfT3rj/ABdovhvR/Og0rU5p b+3vZLeW2kLSfIuRv3eVGFIIIKjcORhuKcaOHk7KP5g51Er3OpHxlkP/ADAYv+/q/wDxulHxjkP/ ADAov+/q/wDxumCz0P8A4V3/AMJV9n07zP7K/sj7DsTP2rft8/d/z02/Pjbux/FjmrNl4I03S/E0 iW092brQtQ0zzJZHUpc+c65wgUGPBOeWbgY96n2eH1938X/XUOap3Ix8YZD/AMwOL/v6v/xunD4v yH/mBxf9/V/+N1JfeCdN1Txlpvnz3a/2zqGqfaNjqNvkyPt2ZU4zjnOfwrPXwZootolJ1Dzx4fTW pH89NrcjdEF8vIyM4bJxkcHHIqeG/l/MOap3Lw+Lsh/5gkX/AH8X/wCN0o+Lch/5gsX/AH8X/wCN 1V/4QbRobSfVLi9uIrBdNs7xIpJMMjTsV2tIsbZA2npHzuHTBJyNKOhaZ8QrYRSRanov2gIHu0CK UcYy4YY+QtnJAztzxnilRoO9o7eonOot2dGPizIf+YLF/wB/F/8AjdOHxXkP/MGi/wC/i/8Axup7 7w5p2nG40BGgludM0O6u5ruOCLfK7OCquGVipVcYIO7DDBHfO1Twp4Y0rVF0q41qVLyKe2STcTh0 fb5h5jCxYDbgd78DB5qVTw7+z+Y3KoupcHxVkP8AzBov++1/+N0o+KUh/wCYPF/32v8A8RRf+ANK sLPUZpL2VXsoLl5VNzESjhx9lDADI8xMnHc+nSob7wj4dsdYi0ZtVnOoC5tomQZJlEhUPx5YWMgN uHzvwMHmhU8M9l+Yc1VdSwPihIf+YRF/32v/AMRTh8TpD/zCIv8Avtf/AIiuR8QWVhp2sz2WnyXM iW7vFI1wFB3q7DjHbAXnucnA6DOWtFhaLV7fmQ61RdT0EfEyQ/8AMJi/76X/AOIpw+JUh/5hUX/f S/8AxFcAKeKf1Sj2/Fi9tPud6PiRIf8AmFRf99L/APEU4fEaQ/8AMLi/76X/AOIrgxTxS+q0e35h 7afc7sfESQ/8wyL81/8AiKUfEKQ/8wyL81/+IrhxUgo+q0e35h7afc7YeP5D/wAw2L81/wDiKcPH sh/5hsX5r/8AEVxQqQUvqtLt+Ye2n3OyHjuQ/wDMOi/8d/8AiKcPHMh/5h8X/jv/AMRXHCpBR9Vp dvzD20+5148byH/mHxf+O/8AxFOHjWQ/8uEX/jv/AMTXIrUgpfVaXb8w9tPudYPGch/5cYvyX/4m lHjGQ/8ALjF+S/8AxNcqKkFH1Wl2/MPbT7nUDxfIf+XKL8l/+Jpw8WyH/lyi/Jf/AImuYFPFH1al 2/MPbT7nSjxXIf8Alzi/75X/AOJpw8USH/l0i/75X/4mucFPFL6tS7fmHtp9zoh4mkP/AC6Rf98r /wDE04eJJD/y6xf98r/8TXPrUgo+rUu35h7ep3N0eIpD/wAu0X/fC/8AxNOHiCQ/8u0X/fC//E1h inil9WpdvzD28+5tjXpD/wAu8X/fC/8AxNOGuSH/AJYRf98L/wDE1jCnij6tS7fmHt59zXGtSH/l hF/37X/4mnDV5D/yxi/79r/hWUKkFH1el2/MXt6nc0xqsh/5ZRf9+1/wpw1OQ/8ALKL/AL9r/hWa tSCl9Xp9g9vU7mlFeTTyBEihLHJ5RB0GepFT7rr/AJ52v5xVSsP+Pn/tnJ/6AaoXfiTQrC6e1vNa 062uExvimukR1yMjIJyOCD+NctdRpysl+f8AmdNJynG7f5GtLeSQ3FvA4tRLcMyxKPKJYhSx6dAA pOTx+Yqbddf887X84qwL3/katA+t1/6TvXKaZ8PPC2uJe6lqWl+fdzaje+ZJ9olXOLmRRwrAdAO1 Y88f5fz/AMzTlff8v8j0G61FrOSCOf7KrzuUiVRGxYhSx4APYHnp+YqyTeKxVorUEcEExcV5vo1p BYR+H7O1Ty7e31LU4okyTtVZJgBk8ngd62/iDdy2c1u+nXlymuG6YafaQkst238SSR5AMe3kuSNn UEHg03HlT5fz/wAyUnzNX/L/ACOmvL17Czmu7n7JHBChd2zEcAew5P0HNcB8VbmSbSdOB2qpmYlV UKCcd8de/wCdVoruW88FeIn1G8uX1wHGoWkxKraN/CkceSBHt5Dgnf1JJ4D/AIn/APIK03/rq/8A KjRxva2v+Y9VK1y18T7vUdM1rw9qelXtnaXltHLskubiGP70canAlIDcE9M4yPauV/4WN4//AOhm 0b/wL0//AOKpf2h/9b4c/wCuc3/oMNeI1pCu4xUeVO3dClTu73Z6p4h1zxN4qggg1nW9HuI4HLxg ahZJgkY/hcVWuL3xJeeT9q8URT+TKs0Xm+IYW8uQdGXMvDDJwRzXmlFarGTSsor7ifYruz0oXGvC eGf/AISS386BpHik/t+DdG0hJcqfN4LEnJHXPNX9D1rXdASVLTWNLZWtJbSJZNagIt1kIZmixKNj ZUHI715NRQ8ZNqzS/r5h7Fd2emTSatc/avtHiCzl+17PtPma5A3nbPu78y/NjtnpUt7d61qcKw6h 4ktruJW3iO416CRQ3IzgynnBP515dRR9cn2X9fMXsY92elCG7+xfYv7Y037J5nm+R/bNvs34xu2+ ZjOOM9cVba91d4LaF/Els0VqyvboddhKxMowpUeb8pHbHSvKqKPrs+y/r5h7GPc9bttT1q0837P4 ogh86QyyeXr0K73PVjiXkn161GLnUP8AoYbL/j3+y/8AIbg/1P8Azz/1n3P9npXlFFH1yfZD9jHu z12HUNTt5kmh8S2scscIgR01yEMsQ5CAiThfbpUFyZb24e4u9Z024nfG6WXV7d2bAwMkyZPAArym ij65Psv6+YvYx7s9eN7evcT3D+IbJp7iMxTSHWoC0iEAFWPmZIwAMH0FTPqd/NYixl8S2j2gVVED a3CYwBjA2+ZjAwMfSvG6KX1yfZB7GPdnsct/dT/aPO8Q2Mn2nb5+/WoT5u37u7MnzY7Z6VYXWdRF vHbjxTbCGPbsj/tyLau0grgeZgYIGPTArxSij63Psg9jHuz2CR45pXlm1fSpJHYs7vqtuSxPJJO/ k0AWw/5imkf+DO3/APi68fop/XanZf18w9hE9jBtf+grpH/gzt//AIunhrT/AKCukf8Agzt//i68 Zoo+u1Oy/r5h7CJ7QJLT/oLaR/4M7f8A+Lp4ls/+gtpH/gzt/wD4uvFKKPrlTsv6+Yewie2iey/6 C+kf+DKD/wCLp4uLH/oL6R/4MoP/AIuvD6KPrlTsv6+Yewie5i6sP+gvpH/gyg/+Lpwu9P8A+gxp H/gyg/8Ai68KopfXKnZC9hE94F7p3/QY0j/wZQf/ABdOF9pv/QZ0j/wZQf8AxdeC0UfXKnZB7CJ7 6NQ0wf8AMZ0j/wAGMH/xdPGo6X/0GtI/8GMH/wAXXz/RR9bqdkHsIn0ENT0r/oNaR/4MYP8A4unj VNJ/6Dekf+DGH/4uvnqij63U7IPYRPogatpH/Qb0j/wYw/8AxVOGr6R/0G9I/wDBjD/8VXzrRR9b n2QewifRo1jR/wDoOaR/4MIf/iqcNa0b/oOaR/4MIf8A4qvnCil9bn2QfV4n0kNc0Uf8x3SP/BhD /wDFU8a7on/Qd0j/AMGEP/xVfNVFH1qfkH1eJ9LjX9D/AOg9pH/gwh/+Kp48QaF/0HtI/wDBhD/8 VXzLRR9an5B9XifTg8Q6D/0H9I/8D4f/AIqnjxHoH/Qf0j/wPi/+Kr5goo+tT8g+rxPqIeJPD/8A 0H9I/wDA+L/4qnDxN4e/6GDSP/A+L/4qvluil9an5B9XifU48UeHR/zMGkf+B8X/AMVTx4p8Of8A Qw6R/wCB0X/xVfKtFH1qfkH1eJ9VP4t0RIn+yeJtEinKlVkku4nC5GCcbxngnv19elczdp4fv7p7 q88S+D7m4fG+Wa0tndsDAyS+TwAPwr57orOVXnd5Jfj/AJlxp8qsmfSR1m2e9truTx3oDy23meUT 5QA3oUPSXnhs/UCs2a28NXE8k8/iHwbLNIxd5HsrVmZickkl+STXz/RU8y/lX4/5lcr7/l/kfR1j d6JaXNgW8W+Hha2TO0VvbmGBQWVgcbZMDls9KuTeJlmnkl/4T3w+m9i21RFgZPQZlzXzLRT51a1l +P8AmLld73/L/I+j77VLbUbGazufHuhNDKu1gPKBx9fNqv8AEW7tr7QdKubS4huIHlfbLDIHRscH BHB5BFfPFeuRf8kh8L/9dbn/ANGvUuWlkrFKOt2zS/aH/wBb4c/65zf+gw14jXt37Q/+t8Of9c5v /QYa8RqVsNhRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAo oooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACvX Iv8AkkPhf/rrc/8Ao168jr1yL/kkPhf/AK63P/o16QGl+0P/AK3w5/1zm/8AQYa8Rr279of/AFvh z/rnN/6DDXiNC2BhRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACi iigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKK KACvXIv+SQ+F/wDrrc/+jXryOvXIv+SQ+F/+utz/AOjXpAaX7Q/+t8Of9c5v/QYa8Rr279of/W+H P+uc3/oMNeI0LYGFFFFMAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKK KKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooo oAK9ci/5JD4X/wCutz/6NevI69ci/wCSQ+F/+utz/wCjXpAaX7Q/+t8Of9c5v/QYa8Rr279of/W+ HP8ArnN/6DDXiNC2BhRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKA CiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAK KKKACvXIv+SQ+F/+utz/AOjXryOvXIv+SQ+F/wDrrc/+jXpAaX7Q/wDrfDn/AFzm/wDQYa8Rr279 of8A1vhz/rnN/wCgw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKK ACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooA KKKKACiiigAr1yL/AJJD4X/663P/AKNevI69ci/5JD4X/wCutz/6NekBpftD/wCt8Of9c5v/AEGG vEa9u/aH/wBb4c/65zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAoooo AKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigA ooooAKKKKACiiigAr1yL/kkPhf8A663P/o168jr1yL/kkPhf/rrc/wDo16QGl+0P/rfDn/XOb/0G GvEa9u/aH/1vhz/rnN/6DDXiNC2BhRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiig AooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAC iiigAooooAKKKKACvXIv+SQ+F/8Arrc/+jXryOvXIv8AkkPhf/rrc/8Ao16QGl+0P/rfDn/XOb/0 GGvEa9u/aH/1vhz/AK5zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAoo ooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiii gAooooAKKKKACiiigAr1yL/kkPhf/rrc/wDo168jr1yL/kkPhf8A663P/o16QGl+0P8A63w5/wBc 5v8A0GGvEa9u/aH/ANb4c/65zf8AoMNeI0LYGFFFFMAooooAKKKKACiiigAooooAKUAsQAMk8ACk r0Twr4D1I2sWqS2EsgkG6L5eAPWgDkV8Nas6BhanBGeWFH/CNav/AM+jf99D/GvVzoupL1spR/wG m/2RqH/PnL/3zQB5T/wjerf8+bfmP8aQ+HNWH/Lk/wCY/wAa9YGjakRkWMx/4BTTpV+vWzmH/AaA PJz4f1Uf8uUn6Un9gar/AM+Uv6V6v/Z17/z6y/8AfNA029PS0m/74NAHlH9g6p/z4y/lTTompj/l ym/75r1g6feL1tZR/wAANN+x3Q/5d5P++TQB5R/Y2pf8+U//AHzSHR9RAybKf/vg16x9kuf+feX/ AL4NQkY4I5oA8hIKkgjBHUUlWtS/5Cd1/wBdW/nVWgAooooAKKKKACnCN2GQjEewrQ0jSJdTn4BE Kn5mx+ldzBZxW8KxRxAKowOKAPNvKk/uN+VHlv8A3G/KvTfKX/nmPypPKT/nmPyoA8z2N/dP5Um1 v7p/KvTPJj/55L/3zR5EX/PJP++aAPM9p9DRg+hr0vyIv+eSf980G2hP/LFP++aAPM8UV6X9kt/+ eEf/AHzSfZLf/nhH/wB8igDzWivSvsVsetvH/wB8iuH16JIdXmSNQq8HAoAzaKKKACiiigAooooA KKKKACiiigAooooAK9ci/wCSQ+F/+utz/wCjXryOvXIv+SQ+F/8Arrc/+jXpAaX7Q/8ArfDn/XOb /wBBhrxGvbv2h/8AW+HP+uc3/oMNeI0LYGFFFFMAooooAKKKKACiiigAooooAK+xPAV6B4Q0qJ+V +zqMfhXx3X1X4RkMfhbSz/07rQB2d/Ygr5kfKmsRlaJ63LC+DL5cnKmmX9gMb05U9DQBXsL4xsAT WjdWyXUXmx/iK51laJ607C/MbAE8UAZ88DROeKltLkxuK2bu1S5i82P8R6VgTQtE54oA6LbFfw9A JMfnWHd2Zic/LTrO7aJxzW2wjv4cjHmY/OgDmVYIrAqOh7V5Dd/8fk/++3869mvbVot3HavGLr/j 7m/3z/OgDybUf+Qndf8AXVv51Vq1qX/ITuv+urfzqrQAUUUUAFFFFAHtfwws4pvCJcxozec2SVGa 6mS0iU/6lP8AvkVy/wAK5vL8L4/6bNXdyxLIu5aAMyOG3zzBGf8AgIq6LGzmTi3iDf7oqq6FTUkM xU9aAIJtOt1P+oj/AO+aiW0tQebeP/vmtkhbhP8AaqjLEVNADorDT5Fw1rFn1xUM+kWiHi2j/KnR yFTV+ORZk2t17GgDE/s+0B5to/yqzDp2nNw1pEfwqzPAVPSq6koaAOK12KODWJ44lCxqeAK8z8Rf 8hqb6D+Vema6d2sTn1NeZeIv+Q1N+H8qAMuiiigAooooAKKKKACiiigAooooAKKKKACvXIv+SQ+F /wDrrc/+jXryOvXIv+SQ+F/+utz/AOjXpAaX7Q/+t8Of9c5v/QYa8Rr279of/W+HP+uc3/oMNeI0 LYGFFFFMAooooAKKKKACiiigAooooAK+tfDVoR4O0l8dbZD+lfJVfZfgho7rwRpMTY3C2UA/hQBB G7RPW5Y3qyJ5cnKms+9smic8VTjkaJ6ANfULDHzLyp6Gschonresr1ZY/Kk5B/Sq+oWG35lGQeho Abp9+UIBPHpVy8tEnj82Poeo9K5/5onrW0+/2kKxyD1FAGXLE0T1Zs7xonHNal7ZpNH5sfIP6Vgy xtE9AHQ3Cx3to7rjeFJI9a+frwYvZx/00b+de02940aMM/wmvFrw5vZ/+ujfzoA8l1L/AJCl1/11 b+dVatal/wAhO6/66t/OqtABRRRQAUUUUAex/DMkeGM/9NWrvbefacHpXEfDKIt4S3f9NmrreUNA GhNCHXctUHQoatW8+OD0qSeEOu5elAFaGYqetXGVZ0yPvVnMpQ1NBMVI5oAZLEVNJHIVNX3VZ0yO vpVCSMqaAL6OsybW6+tVJ4Cp6VHHIVNX0ZZ0wevY0Aeb62P+JtN9a808Rf8AIam/D+VeneIF263c L6NXmPiL/kNTfQfyoAyqKKKACiiigAooooAKKKKACiiigAooooAK9ci/5JD4X/663P8A6NevI69c i/5JD4X/AOutz/6NekBpftD/AOt8Of8AXOb/ANBhrxGvbv2h/wDW+HP+uc3/AKDDXiNC2BhRRRTA KKKKACiiigAooooAKKKKACvrHwXctF4X0vn/AJd1/lXydX1Z4XiZfCWlN626/wAqAO7jkjvodrY3 44NY97ZtE54qK1umicc1vK0d/Dg434/OgDnIpWiet2zvEnj8qXkH9Kyry0aJjxVeGVonoA09QsCh LAZHY1lAtE9dDaXSXEXlSfgfSqN/YFCSBxQBJp9/j5WOVPUVNfWSyJ5kfKmsIFonrZ0+/GNj8qeo oAxZo2j3fSvILr/j7m/3z/Ove9RslaFpY+VIJrwW74vJv98/zoA8k1H/AJCd1/11b+dVqtal/wAh S6/66t/OqtABRRRQAUUUUAe4fCl1/wCETKN085q7C4g28jpXCfDFyvhn/tq1egxSrIm1qAM/lDVy 3uMcHpTbiDaaqglTQBfngDLuXpVFlKmrlvP/AAnpSzwAjcvINAEEE5U9atuizJuXr6VmsChqxBOV PWgCKSMqaWKUqauyRrMm5evcVQkjKmgDivEDbtauG9TXmHiL/kNTfh/KvTNbP/E2m+teZ+Iv+Q1N +H8qAMuiiigAooooAKKKKACiiigAooooAKKKKACvXIv+SQ+F/wDrrc/+jXryOvXIv+SQ+F/+utz/ AOjXpAaX7Q/+t8Of9c5v/QYa8Rr279of/W+HP+uc3/oMNeI0LYGFFFFMAooooAKKKKACiiigAooo oAK+x/Cdolx4F0h0+8LVMj8K+OK+vfAl8YvCulLnj7Ov8qALM8DROeKmtLponHNbF3apcxebH+I9 KwZoWic8UAdEPLv4e3mY/OsO7tGic8UtndtE45rbIjv4cjHmY/OgDnYJmicc1v2tyl1F5Un4H0rF u7VonPFRwTtE45oAvX9gY2JA4rNVmieuitriO7h8uQjPY1m39iY2PFAE9rfA28kb8gqf5V4Vef8A H7P/ANdG/nXru5oww9jXkF3/AMfk3++386APJtS/5Cl1/wBdW/nVWrOo/wDITuv+urfzqtQAUUUU AFFFFAHsvw0Qnwrn/pq1dmjlTXL/AAsRZPB5H8XnNXVSxFTQBdilWVNrfnVe4gKmoI3KGr8ciypt agDOBKGrtvOMbW5FRTwFTVcEoaALtxACNy8iqRBQ1egnBG1ulJcQcbl6UARwTlT1qxLGsqbl/EVn EFTVq3nKmgDgNdG3WJx6GvM/EX/Iam+g/lXp/iIg65ckdN1eYeIv+Q1N9B/KgDKooooAKKKKACii igAooooAKKKKACiiigAr1yL/AJJD4X/663P/AKNevI69ci/5JD4X/wCutz/6NekBpftD/wCt8Of9 c5v/AEGGvEa9u/aH/wBb4c/65zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAr6q8KZTwrpZ/ 6d1/lXyrX154Zsj/AMIVo8gHBtUP6UAbdhflCATxVy8tEuI/Nj6dx6Vz/wA0T1q6ff7CFY5B6g0A Zk0TRP0qzZ3jROOa1L2zSaPzYuQf0rBliaJ6AOjdI7+HcMb8cj1rCurZomPFS2V40TjmtiWOO+h3 rjfjkUAc/b3DROOa34J47yHy3xu7GsK5tmic8UltcNE45oAs6hZNFv47V4ldf8fc3++f519ACaO8 s3V8bgpwfwr5o8Ua3FpU06qQ1wzttX056mgDzzUv+Qndf9dW/nVWnSSNLK0jnLMSSfem0AFFFFAB RRRQB7T8LZSnhn/ts1d+wWdMj71edfDQH/hF8/8ATVq7mCYqRzQA2WIqabHIVNaDKs6ZHWqEsZU0 AX0dZk2t1qpPAVJ4qOOQqavo6zptPWgDOVipq9BOGXa3Sq88BUnioVYqaALdxBjkdKp8oavwTBl2 t0qO4gxyOlAHn+uHOrzn3rzLxF/yGpvw/lXo/iS4is9QuZZm2qteXajd/br6S427Qx4HtQBVoooo AKKKKACiiigAooooAKKKKACiiigAr1yL/kkPhf8A663P/o168jr1yL/kkPhf/rrc/wDo16QGl+0P /rfDn/XOb/0GGvEa9u/aH/1vhz/rnN/6DDXiNC2BhRRRTAKKKKACiiigAooooAKKKKACtm38W+IL W2S3g1e6jhQYVBJwBWNRQBuf8Jj4iPXV7r/vunL418SJ93WLof8AAqwaKAOmT4g+LEGF1y7A9N1N Pj7xQ33tZuD9TXN0UAdIvj7xQvTV5/zFWYviZ4wh/wBXrlyv0NclRQB1z/E7xhIcvrMzfXFMHxI8 Vg5/tWT8hXKUUAdgnxR8XoCF1ZwDx0FcrdXU17cvcXEjSSucsx71DRQAUUUUAFFFFABRRRQBvaR4 v1XRLP7LZyIIt27DLnmtAfEfXx/y0i/74rkaKAO1j+KHiGPo8P8A3xTn+KWvv94W/wD3xXEUUAdn /wALM1vP3YP++alT4pa4nRIP++a4eigDvm+LOtuMNDbH/gNRH4o6uTzBB+VcNRQB3ifFPVkP/HvB VgfF3Vgu1rO3I/GvO6KANbXvEF1r961xOAgJyEXoKyaKKACiiigAooooAKKKKACiiigAooooAKKK KACvXIv+SQ+F/wDrrc/+jXryOvXIv+SQ+F/+utz/AOjXpAaX7Q/+t8Of9c5v/QYa8Rr279of/W+H P+uc3/oMNeI0LYGFFFFMAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKK KKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooo oAK9ci/5JD4X/wCutz/6NevI69ci/wCSQ+F/+utz/wCjXpAaX7Q/+t8Of9c5v/QYa8Rr279of/W+ HP8ArnN/6DDXiNC2BhRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKA CiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAK KKKACvXIv+SQ+F/+utz/AOjXryOvXIv+SQ+F/wDrrc/+jXpAaX7Q/wDrfDn/AFzm/wDQYa8Rr279 of8A1vhz/rnN/wCgw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKK ACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooA KKKKACiiigAr1yL/AJJD4X/663P/AKNevI69ci/5JD4X/wCutz/6NekBpftD/wCt8Of9c5v/AEGG vEa9u/aH/wBb4c/65zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAoooo AKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigA ooooAKKKKACiiigAr1yL/kkPhf8A663P/o168jr1yL/kkPhf/rrc/wDo16QGl+0P/rfDn/XOb/0G GvEa9u/aH/1vhz/rnN/6DDXiNC2BhRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiig AooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAC iiigAooooAKKKKACvXIv+SQ+F/8Arrc/+jXryOvXIv8AkkPhf/rrc/8Ao16QGl+0P/rfDn/XOb/0 GGvEa9u/aH/1vhz/AK5zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKUAk4AzQAlF KQVOCMGkoAKKXBxnHFJQAUUUUAFFFFABRSlWAyQQPpSUAFFFFABRVmOwu5YDNHbyNEoyWA4FQxQy TuEiQux7CgBlFTz2VzbS+VNC6PjO0jnFH2O5+y/avJfyN23zMcZ9M0AQUU4IzHAUk+lPjtp5UZ44 nZV6kDpQBFRSkEHBGDT1gleNpFjYov3mA4FAEdFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABR RRQAUUUUAFeuRf8AJIfC/wD11uf/AEa9eR165F/ySHwv/wBdbn/0a9IDS/aH/wBb4c/65zf+gw14 jXt37Q/+t8Of9c5v/QYa8RoWwMKKKKYBRRRQAUUUUAFFFFABRRU9rZz3km2GJ5MdQgyQKAJrDSrv Ut/2aPITqScCr48JaqRkRxn2DiuisYIdI0jz5FeAYJZXcnOfUDoTgViJcW95cPPFBOFHUBvlH0oA q/8ACL6qX2CBS3pvFS2/hvWLedZBahtp5BYYqw6G7hWK1ivWBb5i4BH4HNb00U9lpX2a3unXzACy OEOT9W5oA5m48OaxPOZGtQC3IAYcVCnhfVpCQltuI6gMDW3BpcptJAyEuQWydoH88VFpiy3aiGaN ZCh4KTbSo98UAZZ8O63FAVazdUbkg4FVW0PUUPNq9dkYwr+TGfLA53SBn3H8M1JMvkWXmTr1OA4k bj8KAOL/ALA1PaG+yvg09fDmqsMi0bH1FdXBEsrlmuFPskrDP4VDcndLssp3KAgNILk5B9OhxQBz kOi3cMjfabGV12nG0jg9qgbRtShYFrSRe4yK7u61K3WwS2RYFkK/617sLu/PrWZaIQ5lnvYHAH3X ucjHtk0AZF4mt39hBaSWKCOIkoUjUMc+pHJ6VmtompL1s5fwFdPpuryzanv3wLDE3/PXbkfXpVnV taW4LQaehjBGWZboMQc9eKAON/sjUMZ+xy/981LFot/vUyWNwU77Vro5pbuwtVknulmGBlfP+f8A I9au6BJLcJJd3Qu5LZeAouPL5/lQBzpj1u3ge3s4r1IHXDJjrWdFYapA4eK2uEYdwhrsJ7yW5vHN kb6CPOFCXpJP5YqLUdTkggWIS3UVxj70lwdx9+tAHMyRazLMZZIrpnI2lipzipxNri6aummO5Nis nmeUUON3rXQWly1pbebqOo6jKWGUDTts/HmrNjNLcTiVpXeNRykV0CfYkZNAHISm8Dny7ORCOAQh BptvcataQSQQrOiSHLAIeTXYahqfnXO1pJoCO8OM/jio7aVpS0cUt7KfVgx/WgDjobSaS4DXMM+3 PJEZNac88ltFJa2ouBaygb1MRGa7XTktow/9o/bW44MWSR+YrLbVJWkKW9xJGqk8kA5H0YcUAcxD Lbw2zhtOMsjAgZQjHvS6Da2Jvc6osqRf9cya39R1dbKJQtybi4c/eDhcfUYpttqyvErXOq+RLnGz ytxb6ELigDmdTEM+otJbWbwWwwANp/Om3UVtIUMKsnGDhTXfW9/G0bMtyJVTlsqen0Mf9azbzVvM n2acyMc9JEH8ttAHJ232SBj59tJcA+ny4pEa0E25rOUx56Zrqnur6KIE38SynpEYUH6las2s+oGJ vOubd32ksVWElPQ44/HNAHEXz20sqm0geNAOQxzzVnfpP9kun2af7dxtfd8v5YreN7qzykmWzKAZ LeXG34/KtJLqzBvKjmtPMxxJ8hUn6bePxoA5a0+zrJm6ikdPRTiop/LM7mJWWPPyhuoFeh2l5az6 UsYME+rMx+SMDB9AABRbiQmR9Vs4II0Uk4jAI/EmgDziitvW9Xgu1a0tbeIQJJuWbZh2wCMdenP8 ulYlABRRRQAUUUUAFFFFABXrkX/JIfC//XW5/wDRr15HXrkX/JIfC/8A11uf/Rr0gNL9of8A1vhz /rnN/wCgw14jXt37Q/8ArfDn/XOb/wBBhrxGhbAwooopgFFFFABRRRQAUUUUAXdP0u51Lzfs4UiJ dzFmArp7Sxj0ayimuSqODnIy4f8AADj864+CZoJA6gHByQR1rd/4SOK4hkS7s1LHGxkONuPagC7N cNqN0XmuXtbL+FZv3mPfB6VEIk2CCzmmmy2XmRdigeg6Viz6xeTs37wKjfwqoA/QVYsPENxp6MIr a2YspG50JIz364/SgDXj+yWcBTyp3WU9dncexOD+dNurhQsZaVyx+6phPyj05bH5VlRaqslt5M6O 7k53FsgnPfNN1TW7i/ZFMMEAi4XyAR/MmgDXW8nvrlLcxp5fTLxMD/Mip5okt3aKKVIrnGVPkjAH 4g1zkes3sVs0Kyn5jneclx9DUS6neoxdbqcSHq/mtn880AdHbQTWl0DLqfm7hlkLHP4A96jka4kv W33kFsozt80YJH9axtR1q61JoWkCxmJAuYy2W9zknmqbXU7rhpWPuTz+dAHTnUIxIbWOdJWP3pA5 x/wHFOeVNIn/AHNxaytIowEdyQfThh+Ncvb3MlrOk0Rwyn1Iz9cVcuNZlu7o3E9tbO5x1UgcfQ0A af2OPzftN7JbQ4YE74pHU/lmoLi8+2Egx2wt1f5fLUjPuAcnFZ82qSzjaY0SP/nmpbb+RNWLXXWt E2pp9mT/AHmD5/RgKANeyu2tnS1soUnEnzOsY38D17j6VoXrtD8os4kZ+SscbBmH51Ri8WXepmOx g0LR1d/lDJA+79XNXp400C0N1Msa3BPMW3G76e1AGZPE11GJLh7SMJ0t5JQrD/gJPNPjsEaVSlmo i2/MxXAz7dTVqw06bU5DqN5HGgI3oG9KtW9zDJeLGBK7v8sMaDiRvSgCs0MQs2iVZJMciJI/lUnv zj9KqJpzuVlxtcf33LE/hWzqEn9jeb5wUX7nmP8AuVnW2pN9neaSMs7HCP3J9BQBD9hu9Tk8h4k2 ryNw2065t57do4tkLy42MyqxIHpkkj9KL3TrhIGurnYsvZHONg+nc1j2+oTRXBt1UTMThSgC8/lQ BqRL9klfaIhIBjy9hLZ9uMU+z0+4lnL3NoSoOWPk5P49qtGL+woGvNTLG6cZVG681zM3iHU5JZWj vJokk6ojkLj0xQB1G64aRlmsUaED5VFsR+uaS51JLa1H2W2V3J2uhiYMo/z71yUer3ka7PMDpnO1 lB/+uPwq1b6rM9tKsl7JCyDfHs6u2ehP0oA0hJLMvmpAyqOWaRF4pv24zBHuraYxo3BEIIH6isdt YuJDH5qI+z1yN31watT+ITJp8lnHZxoknUmRzg+oGcfzoA2LnV2mult7VoJA69Vi2lP581Ek8Flb ubmRkmb7qvGTn6YxWRp2owWduzk3H2oMApWQgYqxd6zGl0CEivIyoJaQMGBxyAc0AXRZ/bZILqS7 YqSMq2UVR6Z7VoldJltz5urXDgNjykidkH6YNYM3im6mtxbAPFB02xSbSR9cVBHrbQRsIEkV2GN0 kpf/AAoA6K4FlDaiGzd2mccCMjP/AAIDpS2sSwIxntUBQbg4iwf1PNQaTeWN6N00lsLlhukaVCDx k4B3Y6Cs7UfER/tGT7JmW0XhFnJBx77SM0Ab66bPesl21nMEI+XbDjI/A1yWqXyyj7NC5eNHJMmT 83sB0xTLrWJbqya1MEMatIJC0ZcHvxgtjHQ9M8Cs6gAooooAKKKKACiiigAooooAK9ci/wCSQ+F/ +utz/wCjXryOvXIv+SQ+F/8Arrc/+jXpAaX7Q/8ArfDn/XOb/wBBhrxGvbv2h/8AW+HP+uc3/oMN eI0LYGFFFFMAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAfFLJDIJI3KO OhB5FOnup7pg08ryEcAsc1FRQBY+33fkiH7RJ5ajAXdwKIL+6tp4p4Z3SWE7o2B5U+1V6KALF3f3 V9dSXNzO8s0hyzsck01bqdNm2VgIzuXnoahooAt3mqXuoSF7q5klY9SxqvDNJBKskTFXU5BHamUU AXNR1W+1a48++uHnkwBuc9qp0UUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUA FFFFABXrkX/JIfC//XW5/wDRr15HXrkX/JIfC/8A11uf/Rr0gNL9of8A1vhz/rnN/wCgw14jXt37 Q/8ArfDn/XOb/wBBhrxGhbAwooopgFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUU AFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQA UUUUAFFFFABXrkX/ACSHwv8A9dbn/wBGvXkdeuRf8kh8L/8AXW5/9GvSA0v2h/8AW+HP+uc3/oMN eI17d+0P/rfDn/XOb/0GGvEaFsDCiiimAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAB RRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFF FFABRRRQAUUUUAFeuRf8kh8L/wDXW5/9GvXkdeuRf8kh8L/9dbn/ANGvSA0v2h/9b4c/65zf+gw1 4jXt37Q/+t8Of9c5v/QYa8RoWwMKKKKYBRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAF FFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUU UUAFFFFABRRRQAV65F/ySHwv/wBdbn/0a9eR165F/wAkh8L/APXW5/8ARr0gNL9of/W+HP8ArnN/ 6DDXiNfRfxZ8LS+M59KXT9U0uFrJHWUXNwVOWWMYGAemw5zivOP+FO6x/wBBzw//AOBb/wDxFCGe d0V6J/wp3WP+g54f/wDAt/8A4ij/AIU7rH/Qc8P/APgW/wD8RRcR53RXon/CndY/6Dnh/wD8C3/+ Io/4U7rH/Qc8P/8AgW//AMRRcDzuivRP+FO6x/0HPD//AIFv/wDEUf8ACndY/wCg54f/APAt/wD4 ii4HndFeif8ACndY/wCg54f/APAt/wD4ij/hTusf9Bzw/wD+Bb//ABFFwPO6K9E/4U7rH/Qc8P8A /gW//wARR/wp3WP+g54f/wDAt/8A4ii4HndFeif8Kd1j/oOeH/8AwLf/AOIo/wCFO6x/0HPD/wD4 Fv8A/EUXA87or0T/AIU7rH/Qc8P/APgW/wD8RR/wp3WP+g54f/8AAt//AIii4HndFeif8Kd1j/oO eH//AALf/wCIo/4U7rH/AEHPD/8A4Fv/APEUXA87or0T/hTusf8AQc8P/wDgW/8A8RR/wp3WP+g5 4f8A/At//iKLged0V6J/wp3WP+g54f8A/At//iKP+FO6x/0HPD//AIFv/wDEUXA87or0T/hTusf9 Bzw//wCBb/8AxFH/AAp3WP8AoOeH/wDwLf8A+IouB53RXon/AAp3WP8AoOeH/wDwLf8A+Io/4U7r H/Qc8P8A/gW//wARRcDzuivRP+FO6x/0HPD/AP4Fv/8AEUf8Kd1j/oOeH/8AwLf/AOIouB53RXon /CndY/6Dnh//AMC3/wDiKP8AhTusf9Bzw/8A+Bb/APxFFwPO6K9E/wCFO6x/0HPD/wD4Fv8A/EUf 8Kd1j/oOeH//AALf/wCIouB53RXon/CndY/6Dnh//wAC3/8AiKP+FO6x/wBBzw//AOBb/wDxFFwP O6K9E/4U7rH/AEHPD/8A4Fv/APEUf8Kd1j/oOeH/APwLf/4ii4HndFeif8Kd1j/oOeH/APwLf/4i j/hTusf9Bzw//wCBb/8AxFFwPO6K9E/4U7rH/Qc8P/8AgW//AMRR/wAKd1j/AKDnh/8A8C3/APiK Lged0V6J/wAKd1j/AKDnh/8A8C3/APiKP+FO6x/0HPD/AP4Fv/8AEUXA87or0T/hTusf9Bzw/wD+ Bb//ABFH/CndY/6Dnh//AMC3/wDiKLged0V6J/wp3WP+g54f/wDAt/8A4ij/AIU7rH/Qc8P/APgW /wD8RRcDzuivRP8AhTusf9Bzw/8A+Bb/APxFH/CndY/6Dnh//wAC3/8AiKLged0V6J/wp3WP+g54 f/8AAt//AIij/hTusf8AQc8P/wDgW/8A8RRcDzuivRP+FO6x/wBBzw//AOBb/wDxFH/CndY/6Dnh /wD8C3/+IouB53RXon/CndY/6Dnh/wD8C3/+Io/4U7rH/Qc8P/8AgW//AMRRcDzuvXIv+SQ+F/8A rrc/+jXrH/4U7rH/AEHPD/8A4Fv/APEV02raW2hfD7QdImurW4uLWWbzGtpN6fMzMOSAeh9PWgZy tFFFMQUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAF FFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQB// 2Q== ------_=_NextPart_001_01C51D78.AA1DB2A0-- From info-volumizer-owner@els.sgi.com Tue Mar 1 00:51:36 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j218n6qq6273119; Tue, 1 Mar 2005 00:49:06 -0800 (PST) Received: from hastur.corp.sgi.com (hastur.corp.sgi.com [198.149.32.33]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j218m7va6180507 for ; Tue, 1 Mar 2005 00:48:14 -0800 (PST) Received: from mtv-atc-013e--n.corp.sgi.com (mtv-atc-013e--n.corp.sgi.com [192.26.78.50]) by hastur.corp.sgi.com (8.12.9/8.12.10/SGI_generic_relay-1.2) with ESMTP id j218lfVt112408187 for ; Tue, 1 Mar 2005 00:47:41 -0800 (PST) Received: from praveenlaptop (134.14.90.76 [134.14.90.76]) by mtv-atc-013e--n.corp.sgi.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55) id FYL5PH2H; Tue, 1 Mar 2005 00:48:06 -0800 Message-ID: <056b01c51e3b$78fc1fa0$4c5a0e86@praveenlaptop> From: "Praveen Bhaniramka" To: References: <50726031C3515F4F879593DD4E801AE622D25E@acvstg03.intern.acv.ac.at> Subject: Re: [info-volumizer] Volumizer bizarre behavior Date: Tue, 1 Mar 2005 14:18:36 +0530 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_054D_01C51E69.8EC3C080" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com This is a multi-part message in MIME format. ------=_NextPart_000_054D_01C51E69.8EC3C080 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Oscar, The stripes that you see when looking at the volume data from angular = orientations are mainly due to the fact that the data is extremely high = frequency (it is a step function) and while rendering, it is = undersampled in object space with respect to the screen space = resolution. You can get rid of these artifacts by increasing the = sampling rate by using vzTMRenderAction::setSamplingRate() method. Note = that increasing the sampling rate would also increase the rendering = overhead. Volumizer uses a default sampling rate of (1, 1, 1), which = implies one slice per-voxel. Hence as the resolution of your volume data = increases with respect to your screen space resolution (which is = typically more finite), the slicing artifacts due to undersampling = reduce considerably. The programming guide discusses this issue in more = detail as well.=20 As for the noise issue, please see my posting on = http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.html. = Basically, the data pointer that you are passing to the = vzParameterVolumeTexture constructor needs to be allocated from the heap = (using malloc() or new[]). The following code is incorrect - // BYTE pbyArea[100][100][100]; const int piDataDim[3] =3D {100, 100, 100};=20 // // Create the texture vzParameterVolumeTexture *pVolume =3D new vzParameterVolumeTexture ( piDataDim, (void *) pbyArea, VZ_UNSIGNED_BYTE,=20 VZ_LUMINANCE, VZ_DEFAULT_INTERNAL_FORMAT); // Cheers, Praveen ----- Original Message -----=20 From: Oscar Serrano=20 To: info-volumizer@sgi.com=20 Sent: Monday, February 28, 2005 3:04 PM Subject: [info-volumizer] Volumizer bizarre behavior Hello, We are just starting to evaluate the possibilities of volumizer = but we have detected some bizarre output. We have reproduced the error = in a simple program that I enclose. It just draws a shape with cube = objects using the SimpleShader. The cubes are showing strange stripes = that vary when moving them and we also have some strange kind of noise = (I enclose a snapshot of the output). We have also experience this = problem using the LUTShader and using different graphic cards. Does = anyone have an idea what is wrong???=20 Thank you, Oscar Serrano ------=_NextPart_000_054D_01C51E69.8EC3C080 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi Oscar,
 
The stripes that you see when looking at the volume data from = angular=20 orientations are mainly due to the fact that the data is extremely high=20 frequency (it is a step function) and while rendering, it=20 is undersampled in object space with respect to the screen space=20 resolution. You can get rid of these artifacts by increasing the = sampling=20 rate by using vzTMRenderAction::setSamplingRate() method. Note that = increasing the sampling rate would also increase the rendering overhead. = Volumizer uses a default sampling rate of (1, 1, 1), which implies one = slice=20 per-voxel. Hence as the resolution of your volume data increases with = respect to=20 your screen space resolution (which is typically more finite), the = slicing=20 artifacts due to undersampling reduce considerably. The programming = guide=20 discusses this issue in more detail as well.
 
As for the noise issue, please see my posting on http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.= html.=20 Basically, the data pointer that you are passing to the=20 vzParameterVolumeTexture constructor needs to be allocated from the = heap=20 (using malloc() or new[]). The following code is incorrect -
 
//<snip>
 BYTE pbyArea[100][100][100];
 const int piDataDim[3] = =3D {100,=20 100, 100}; 
//<snip>
    // Create the texture
   =20 vzParameterVolumeTexture *pVolume =3D new vzParameterVolumeTexture=20 (
           &n= bsp;           &nb= sp;           &nbs= p;       =20 piDataDim,
          = ;            =             &= nbsp;        =20 (void *)=20 pbyArea,
          &= nbsp;           &n= bsp;           &nb= sp;        =20 VZ_UNSIGNED_BYTE,=20
           VZ_= LUMINANCE,
          = ; VZ_DEFAULT_INTERNAL_FORMAT);
//<snip>
 
Cheers,
Praveen
 
 
----- Original Message -----
From:=20 Oscar=20 Serrano
Sent: Monday, February 28, 2005 = 3:04=20 PM
Subject: [info-volumizer] = Volumizer=20 bizarre behavior

Hello,

 

      We are just = starting to=20 evaluate the possibilities of volumizer but we have detected some = bizarre=20 output. We have reproduced the error in a simple program that I = enclose. It=20 just draws a shape with cube objects using the SimpleShader. The cubes = are=20 showing strange stripes that vary when moving them and we also have = some=20 strange kind of noise (I enclose a snapshot of the output). We have = also=20 experience this problem using the LUTShader and using different = graphic cards.=20 Does anyone have an idea what is wrong???

 

Thank you,

        = Oscar=20 Serrano

 

 

 

------=_NextPart_000_054D_01C51E69.8EC3C080-- From info-volumizer-owner@els.sgi.com Tue Mar 1 01:12:15 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j219AjQC6210214; Tue, 1 Mar 2005 01:10:45 -0800 (PST) Received: from hastur.corp.sgi.com (hastur.corp.sgi.com [198.149.32.33]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j2191vva6241264 for ; Tue, 1 Mar 2005 01:02:03 -0800 (PST) Received: from mtv-atc-013e--n.corp.sgi.com (mtv-atc-013e--n.corp.sgi.com [192.26.78.50]) by hastur.corp.sgi.com (8.12.9/8.12.10/SGI_generic_relay-1.2) with ESMTP id j2191ZVt112228404 for ; Tue, 1 Mar 2005 01:01:35 -0800 (PST) Received: from praveenlaptop (134.14.90.76 [134.14.90.76]) by mtv-atc-013e--n.corp.sgi.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55) id FYL5PHJZ; Tue, 1 Mar 2005 01:01:56 -0800 Message-ID: <05a501c51e3d$67e582e0$4c5a0e86@praveenlaptop> From: "Praveen Bhaniramka" To: References: Subject: Re: [info-volumizer] Volumizer XML Date: Tue, 1 Mar 2005 14:32:26 +0530 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com Hi Sam, I believe that we had discussed this issue before - http://els.sgi.com/Archives_Public/info-volumizer/200405/msg00009.html. Can you please be specific as to what you are looking for? >From what I can tell, you would need to do the following - 1. Load the Volume data. 2. Process the volume data so that you isolate the jaw from the rest of the skull. (I assume you know how to do this since Volumizer would not help you accomplish this kind of data segmentation). 3. Create two separate shapes, one for each segmented object, and then load and render them using Volumizer. Steps 1 and 2 can be done as a preprocess. You can use this technique to compute different segments in the volume data and store the information about each of these 'objects' on the disk. Then you can create an XML file where you describe each of these different segments as separate shapes and load them into an application. hth, Praveen ----- Original Message ----- From: To: Sent: Monday, February 28, 2005 10:17 PM Subject: [info-volumizer] Volumizer XML > Dear all > I am using the XML file to read my volumes in Volumizer. I would like to > break my volume into sections (say break the skull into a jaw and the > remaining part), Can you please advise on how to do that. I am a novice > user > of Volumizer and I don't know how to do it within the code. I only want to > do it using the XML file format if possible. Any help is appreciated. > Sam From info-volumizer-owner@els.sgi.com Wed Mar 2 01:10:49 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j2295nok5987278; Wed, 2 Mar 2005 01:05:49 -0800 (PST) Received: from acvstg03.intern.acv.ac.at (acvstg03.acv.ac.at [194.152.172.83]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j228xiva6371265 for ; Wed, 2 Mar 2005 00:59:49 -0800 (PST) Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----_=_NextPart_001_01C51F06.7BD981FD" Subject: AW: [info-volumizer] Volumizer bizarre behavior X-MimeOLE: Produced By Microsoft Exchange V6.0.6487.1 Date: Wed, 2 Mar 2005 10:01:55 +0100 Message-ID: <50726031C3515F4F879593DD4E801AE622D262@acvstg03.intern.acv.ac.at> X-MS-Has-Attach: yes X-MS-TNEF-Correlator: Thread-Topic: [info-volumizer] Volumizer bizarre behavior Thread-Index: AcUePtXnW3ZTIYsSSP2i0ZzAvojzbQAu8v/A From: "Oscar Serrano" To: Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com This is a multi-part message in MIME format. ------_=_NextPart_001_01C51F06.7BD981FD Content-Type: multipart/alternative; boundary="----_=_NextPart_002_01C51F06.7BD981FD" ------_=_NextPart_002_01C51F06.7BD981FD Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Thank you very much for your answer Praveen, you were totally right and = now the rendering using the simpleShader looks much better. With the = LUTShader we are still having some problems but we think that are caused = because of the interpolation and I don't know whether there is a = solution for them without changing the input data.=20 =20 In our output when using the two cubes from the mail before using a LUT = with just 3 RGBA mappings (255,0,0,255 for the luminance associated with = cube1; 0,255,0,255 for the luminance associated with cube 2 and 0,0,0,0 = for background) shows that the cube with lower luminance is drawn = correctly but in the other you get a mix of colors when looking at the = volume data from angular orientations (I enclose a snapshot from one = output and the new loadVolumeData function). Changing the color table or = the luminance of the input cubes changes the output but is not improving = much. Is this problem related with the extremely high frequency data = that we are using??. Can the interpolation be avoided somehow (assuming = that this is this is what is causing our problem)?? . And the most = important question... There will be any way of showing the two cubes = with different colors without that mixing??. =20 Thanks and regards Cheers, Oscar =20 P.D. By the way... I was changing the pbyArea in the code from the mail = before. But also the piDataDim just in the case it could cause problems = but I think that what you mean is that I should change just the pbyArea. = I don=B4t think that allocating the dimension is really needed right?? =20 =20 -----Urspr=FCngliche Nachricht----- Von: owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im Auftrag von Praveen = Bhaniramka Gesendet: Dienstag, 1. M=E4rz 2005 09:49 An: info-volumizer@els.sgi.com Betreff: Re: [info-volumizer] Volumizer bizarre behavior =20 Hi Oscar, =20 The stripes that you see when looking at the volume data from angular = orientations are mainly due to the fact that the data is extremely high = frequency (it is a step function) and while rendering, it is = undersampled in object space with respect to the screen space = resolution. You can get rid of these artifacts by increasing the = sampling rate by using vzTMRenderAction::setSamplingRate() method. Note = that increasing the sampling rate would also increase the rendering = overhead. Volumizer uses a default sampling rate of (1, 1, 1), which = implies one slice per-voxel. Hence as the resolution of your volume data = increases with respect to your screen space resolution (which is = typically more finite), the slicing artifacts due to undersampling = reduce considerably. The programming guide discusses this issue in more = detail as well.=20 =20 As for the noise issue, please see my posting on = http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.html. = Basically, the data pointer that you are passing to the = vzParameterVolumeTexture constructor needs to be allocated from the heap = (using malloc() or new[]). The following code is incorrect - =20 // BYTE pbyArea[100][100][100]; const int piDataDim[3] =3D {100, 100, 100};=20 // // Create the texture vzParameterVolumeTexture *pVolume =3D new vzParameterVolumeTexture ( piDataDim, (void *) pbyArea, VZ_UNSIGNED_BYTE,=20 VZ_LUMINANCE, VZ_DEFAULT_INTERNAL_FORMAT); // =20 Cheers, Praveen =20 =20 ----- Original Message -----=20 From: Oscar Serrano =20 To: info-volumizer@sgi.com=20 Sent: Monday, February 28, 2005 3:04 PM Subject: [info-volumizer] Volumizer bizarre behavior =20 Hello, =20 We are just starting to evaluate the possibilities of volumizer = but we have detected some bizarre output. We have reproduced the error = in a simple program that I enclose. It just draws a shape with cube = objects using the SimpleShader. The cubes are showing strange stripes = that vary when moving them and we also have some strange kind of noise = (I enclose a snapshot of the output). We have also experience this = problem using the LUTShader and using different graphic cards. Does = anyone have an idea what is wrong???=20 =20 Thank you, Oscar Serrano =20 =20 =20 ------_=_NextPart_002_01C51F06.7BD981FD Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Thank you very much for your answer Praveen, you were totally right and now the = rendering using the simpleShader looks much better. With the LUTShader we are = still having some problems but we think that are caused because of the = interpolation and I don’t know whether there is a solution for them without = changing the input data.

 

In our output when using the two cubes from the mail before using a LUT with just 3 = RGBA mappings (255,0,0,255 for the luminance associated with cube1; 0,255,0,255 for = the luminance associated with cube 2 and 0,0,0,0 for background) shows that = the cube with lower luminance is drawn correctly but in the other you get a = mix of colors when looking at the volume data from angular orientations (I enclose a = snapshot from one output and the new loadVolumeData function). Changing the color = table or the luminance of the input cubes changes the output but is not improving = much. Is this problem related with the extremely high frequency data that we = are using??. Can the interpolation be avoided somehow (assuming that this is this is = what is causing our problem)?? . And the most important question… There = will be any way of showing the two cubes with different colors without that = mixing??.

 

Thanks and regards Cheers,

=A0=A0=A0=A0=A0=A0=A0=A0 = Oscar

 

P.D. =A0By the way… I was changing the pbyArea in the code from the mail before. = But also the piDataDim just in the case it could cause problems but I think that = what you mean is that I should change just the pbyArea. I don=B4t think that = allocating the dimension is really needed right??

 

=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 =

-----Urspr=FCngliche Nachricht-----
Von: owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im Auftrag von Praveen Bhaniramka
Gesendet: Dienstag, 1. M=E4rz 2005 09:49
An: info-volumizer@els.sgi.com
Betreff: Re: [info-volumizer] Volumizer bizarre = behavior

 

Hi = Oscar,

 

The stripes = that you see when looking at the volume data from angular orientations are mainly due to = the fact that the data is extremely high frequency (it is a step function) and = while rendering, it is undersampled in object space with respect to the screen space = resolution. You can get rid of these artifacts by increasing the sampling = rate by using vzTMRenderAction::setSamplingRate() method. Note that increasing the = sampling rate would also increase the rendering overhead. Volumizer uses a = default sampling rate of (1, 1, 1), which implies one slice per-voxel. Hence as the = resolution of your volume data increases with respect to your screen space = resolution (which is typically more finite), the slicing artifacts due to undersampling = reduce considerably. The programming guide discusses this issue in more detail as well. =

 

As for the = noise issue, please see my posting on http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.= html. Basically, the data pointer that you are passing to the = vzParameterVolumeTexture constructor needs to be allocated from the heap (using malloc() or = new[]). The following code is incorrect -

 

//<snip>

 BYTE = pbyArea[100][100][100];
 const int piDataDim[3] =3D {100, 100, = 100}; 

//<snip>

    // Create the texture
    vzParameterVolumeTexture *pVolume =3D new = vzParameterVolumeTexture (
            &= nbsp;           &n= bsp;           &nb= sp;       piDataDim,
            &= nbsp;           &n= bsp;           &nb= sp;       (void *) pbyArea,
            &= nbsp;           &n= bsp;           &nb= sp;       VZ_UNSIGNED_BYTE,
           VZ_LUMI= NANCE,
           VZ_DEFA= ULT_INTERNAL_FORMAT);

//<snip>

 

Cheers,

Praveen

 

 

----- Original = Message -----

Sent: Monday, February 28, 2005 = 3:04 PM

Subject: [info-volumizer] Volumizer = bizarre behavior

 

Hello,

 

      We are just starting to evaluate the possibilities of volumizer but we = have detected some bizarre output. We have reproduced the error in a simple = program that I enclose. It just draws a shape with cube objects using the = SimpleShader. The cubes are showing strange stripes that vary when moving them and we = also have some strange kind of noise (I enclose a snapshot of the output). We = have also experience this problem using the LUTShader and using different = graphic cards. Does anyone have an idea what is wrong???

 

Thank you,

        Oscar Serrano

 

 

 

------_=_NextPart_002_01C51F06.7BD981FD-- ------_=_NextPart_001_01C51F06.7BD981FD Content-Type: image/jpeg; name="cubes2.JPG" Content-Transfer-Encoding: base64 Content-Description: cubes2.JPG Content-Disposition: attachment; filename="cubes2.JPG" /9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a HBwgJC4nICIsIxwcKDcpLDAxNDQ0Hyc5PTgyPC4zNDL/2wBDAQkJCQwLDBgNDRgyIRwhMjIyMjIy MjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjL/wAARCAFHATQDASIA AhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQA AAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3 ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWm p6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEA AwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSEx BhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElK U1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3 uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwDsvGfj OXwncabBBpkV412hIBdI9u1UJ5KnOd36Vzf/AAtjUv8AoV4f/AuL/wCN10+rIr/FTwUrqGUx3eQR kH/Rlr0P7Hbf8+0P/fArphKhCEeeF2/O3VmE1WcnySSXpf8AU8Ruvi/qFlGslz4WSNGJVSZ05I6j /V1XHxvkP/Mvxf8Af9f/AI3Wz8e4IYtD0cxxIhN0+SqgZ+Ss06XofiQw3nhi+sE0qK70+G509tGi We2SR0Qfv3RjI24EnJII3ZLDhuiNPDygp8mj82QpVU2myMfGuQ/8wCL/AL/L/wDG6cPjRIf+YDF/ 3+X/AON0knw80S51nT7Fpb9brV7vU4knR4ljgNvI4U+UsYyCAMgFR1xjoKWneCdCvtHg1vOopaS6 Ve3v2X7QhkD28irjzPLxtYN/cyCOpp+zw38v5/5+THzVO5oj4yyH/mBRf9/V/wDjdPHxikP/ADA4 v+/q/wDxurGn+FLaHwzqthZ3EyW+tW+jTr52HaDzpypBI2h8EE5wvp71yXivRvDmkebBpepSyX9v eSW8ttIWf5Bkb93lIFIIIKjcORhuKI0cPJ2UfzBzqJXudQPi9If+YJF/39X/AON04fFuQ/8AMFi/ 7+L/APG6atpov/Cvx4o+z6f5n9l/2V9i2Jn7Tv2+fu/56bfnxjOO+Oas2XgzTtM8SSJbzXRutFv9 N8ySRlKXHnOucIFBTBP95umPel7PD6+7+LDmqdyMfFeQ/wDMGi/7+L/8bp4+Kkh/5g8X/fa//G6k vPB2n6l4u0/zprpf7Xv9S8/Yyjb5Uj7duV4zjnOfwqkvhHRxbxqTfeeNDXWHfzk2tyN0YXZkZGcN k4yODjkVPD9vzDmqdy4PijIf+YRF/wB9r/8AEU8fE2Q/8wmL/vtf/iKrDwZpMVrNqU95PHZDT7W7 WN3wVMzFcNIsbZA2nonO4dMEnM0w6Np/jm3Ebx6hpHnBA90m1SjjGWDD+EtnJAztzx2ao0He0dvU TqVFuzfHxKkP/MKi/wC+l/8AiKcPiPIf+YXF/wB9L/8AEVYvNBsLAz6IhhluNO0e5upbpIY90jsw KqwYMVIXGDndhuCO9LUPDXh7TdRGmzatIl1HNbpJuJw6tt3nlAseA24fM3A5qVToP7P5jc6i6lgf EOQ/8wyL81/+Ip4+IEh/5hsX5r/8RS3fgrTrK1vpXu5A1pDcPIpuIyUbcPs4YAZG9efc+nSmXfhn Q7PVI9KOozG9E9vGVGT5gfbv42AJgNkfM3oeaSp4d7L8wcqq6ko8eSH/AJh0X5r/APEU4eOZD/zD 4v8Ax3/4iud1m1s7HVJrSyed1hd43aYAHcHYcY7Yxz3OTgdKqLWiw1Fq9vzIdaoup2A8bSH/AJcI v/Hf/iaePGUh/wCXGL8l/wDia5JalWj6rS7fmL20+51Y8XyH/lyi/Jf/AImnDxZIf+XOL8l/+Jrm FqVaPq1Lt+Ye2n3OlHiiQ/8ALpF/3yv/AMTTx4lkP/LrF/3yv/xNc6tTLS+rUu35h7efc3x4hkP/ AC7Rf98L/wDE08a9If8Al3i/74X/AOJrCWplpfVqXb8w9vPubI1uQ/8ALCL/AL4X/wCJp41eQ/8A LGL/AL9r/hWQtTLR9XpdvzF7ep3NQapIf+WUX/ftf8KsQ3FxOrMkUGFwCWWNev1HsayVrRtf+PKb /ron8mrKrRpwg5JfmaU6k5SSbLO66/552v5xVDHeSS3c1sgtWmgVGkA8rCh923npk7W468e4rJtP EmhX90lrZ61p1zcPnZFDdI7tgZOADk8An8Kwtd/5Bfj/AP7Bkf8A6JnrjUl/L+f+Z08r7/l/kdzu uv8Anna/nFVdNQeTUJLFBbNcxosjoojIVTuxk4wPunjP8xXm/iT4Y+D7Dwtq95a6R5dxb2U0sT/a ZjtZUJBwXweR3rrtH/5HjWv+vaL/ANFy1UHGT279/wDMUlJLf8v8jot11/zztfziqtdai1nJBHP9 lV53KRKojYsQpY8AHsDz0/MV594r1LV7PxDcJ4VuLy7uzbA6narH9ojtE/gljVnULMRnEQyHAyRx k34JLGUeEn02/mv7RjMY7qeZpZJf3TZLM3O7OcjjB4wMYBBxlJJx/P8AzCSai3f8jkPiVK8vi+Qu c4hQAAYAGOw7UUz4jf8AI3S/9ck/lRULYo6v4n3eo6ZrXh7U9KvbO0vLaOXZJc3EMf3o41OBKQG4 J6Zxke1cr/wsbx//ANDNo3/gXp//AMVS/tD/AOt8Of8AXOb/ANBhrxGt4V3GKjyp27oiVO7vdnqn iHXPE3iqCCDWdb0e4jgcvGBqFkmCRj+FxVee+8SXnk/a/FEU/kyLNF5viGFtjjoy5l4YZOCOa8zo rVYyaVlFfcT7Fd2emLc66J4Z/wDhJLbzoGkeKT+3oN0bSHLlT5vBYk5I655rR0PW9a0FJUtdY0tl a0ltYlk1mAi3WQhmaLEo2NlQcjvXkVFDxk2rNL+vmHsV3Z6jLNqlx9p+0eILOX7Xs+0b9cgbztv3 d+Zfmx2z0qa8vtW1KFYdQ8S2t3Erbwk+uwyKGwRnBlPPJ/OvKKKPrk+y/r5i9jHuz1NROLP7H/bO m/ZfM83yP7Yt9m/GN23zMZxxmrh1HUnht4X8S2rRWzK0CHXISsRXhSo8z5SO2OleQUUfXJ9l/XzD 2Me7PZbbV9UtfM8jxRbw+bIZZPL12Fd7nqxxJyT60i390OviGx/1H2b/AJDMP+p/55/6z7n+z0rx uil9bn2Qexj3Z7VDqt5BKksXiWzjkSIQI6a1CCsY5CAiThfbpTJ7lbudp7rW9MnmfG6SXVoGY4GB kl89K8Yoo+uT7IPYR7s9wOqSNPNO/iKwM06eXLIdYhLSLgDax8zJGAOD6VK2tSy2Ys5fE1k9qFCi FtZhKYHQbfMxgYH5V4VRR9bn2Qewj3Z7u+sed5/neI7B/tG3zt+rwnzNv3d37znHbPSrC+IZBAkP /CU2Yij27E/tmLau3BXA8zjGBj0xXgFFL61Psg9jHuz3ltQspZGkl1vSnkclmZtTgJYnqSd9OW+0 0f8AMZ0j/wAGUH/xdeCUU/rdTsg9hE9/XUdMH/MZ0j/wYwf/ABdSLqelD/mNaR/4MYP/AIuvnyij 63U7IXsIn0QNW0kf8xvSP/BjD/8AFU9dY0cf8xzSP/BhD/8AFV86UUfW59kHsIn0gutaMP8AmOaR /wCDCH/4qpF13RB/zHdI/wDBhD/8VXzXRS+tT8g+rxPphdf0Mf8AMe0j/wAGEP8A8VUi+IdBH/Mf 0j/wPh/+Kr5joo+tT8g+rxPqBfEmgD/mP6R/4Hxf/FVKvibw8P8AmYNI/wDA+L/4qvlqil9an5B9 XifVK+KfDo/5mHSP/A6L/wCKqO88VaY9uIrDxdodrlgzs88UpOAcAfvAB1Pr26d/lqilKvKSs0hx oxi7pn0JaJ4fsLpLqz8S+D7a4TOyWG0tkdcjBwQ+RwSPxq3/AGnYn7dv8beHZPtyolwJBEwdUDgL tMuMEOc8c4Hvn5worLmX8q/H/M05X3/L/I99+weFv+g54K/8ALX/AOLrbsdb06zub+8PjPQpL26j SMSmSIKgAYZ2+ZycMe46DrXzPRTU0tkvx/zE4t9fy/yPpf8A4SFP+h/0H8of/jlNTV9Nl1Kzu7/x todwtqzMqLJFGSWUr18w+vp2r5qooUkndJfj/mDi2rN/ke2fEb/kbpf+uSfyoo+I3/I3S/8AXJP5 UVmtix37Q/8ArfDn/XOb/wBBhrxGvbv2h/8AW+HP+uc3/oMNeI0LYGFFFFMAooooAKKKKACiiigA ooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAPb PiN/yN0v/XJP5UUfEb/kbpf+uSfyopLYB37Q/wDrfDn/AFzm/wDQYa8Rr279of8A1vhz/rnN/wCg w14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKK ACiiigAooooAKKKKACiiigAooooA9s+I3/I3S/8AXJP5UUfEb/kbpf8Arkn8qKS2Ad+0P/rfDn/X Ob/0GGvEa9u/aH/1vhz/AK5zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiii gAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooA9s+I3/I3S/wDXJP5UUfEb /kbpf+uSfyopLYB37Q/+t8Of9c5v/QYa8Rr279of/W+HP+uc3/oMNeI0LYGFFFFMAooooAKKKKAC iiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKK KKAPbPiN/wAjdL/1yT+VFHxG/wCRul/65J/KiktgHftD/wCt8Of9c5v/AEGGvEa9u/aH/wBb4c/6 5zf+gw14jQtgYUUUUwCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAoooo AKKKKACiiigAooooAKKKKACiiigAooooA9s+I3/I3S/9ck/lRR8Rv+Rul/65J/KiktgHftD/AOt8 Of8AXOb/ANBhrxGvbv2h/wDW+HP+uc3/AKDDXiNC2BhRRRTAKKKKACiiigAooqxYkC9iyqNk4w67 gcjHTB59ODzik3ZXLpw55qLdrsr0V6xB8MdEfMcl5fedGFEgSWMjJGeBtyB16+lTt8LdEdixutQy Tk4eMD8gleS87widrv7jw6ufYKlJwk3f06PVP5q1jyCivYj8M9H8oxi71FVbG7bIgDAYxkBMHGP1 NW7X4eeHLaOMG1lmkQ58553DE5yPukDj2HaolnuFSvq/kZVOJMBG3K5P5f5v+vxPEqVVZ2CqpZmO AAMkmvcW8A+GXYs2mlmY5JNxKST/AN9U3/hX/hf/AKBn/kxL/wDFVP8Ab+G/ll9y/wAzL/WjBX+G X3L/ADPFJLaeEkSwyIQu4hlIwM4z9M8VFXsuqfD/AEc6bKdMsvLu0/eRjzC3mEA4Q7yQAc/TOCcg YPkuoWf2K5aMEkA4OQQVbupyByM8/wAh0Hbg8wpYq/Juv68z2sDjsPj6UquHuuV2adrq+z80/wA9 +hUooorvOkKkiglnYrDE8jAZIRSTj8Kjrc8K/wDIUl/64n/0Jazqz5IOS6HZl+GWKxMKEnZSdjPj 0rUJZAi2cwJ/vIVH5nip/wDhH9U/59f/ACIv+NdzRXmPMKnRI+4hwfhEvfnJ+ll+jOG/4R/VP+fX /wAiL/jR/wAI/qn/AD6/+RF/xruaKX9oVey/r5l/6oYH+ef3r/5E45fC1+ygl4FJGSpY5HtwKX/h Fb7/AJ62/wD303+FdhRU/XqxsuFcu7P7zgdQ0m603YZgrI3R0JIz6fWqNejXVtHeWslvKDscYODg j0Nef3VtJZ3UlvKBvQ4ODkH0Nd+ExPtVaW6PlM/yVZfNTpa05fg+36r59iGiiius+dCiiigAoooo AKKKKAPbPiN/yN0v/XJP5UUfEb/kbpf+uSfyopLYB37Q/wDrfDn/AFzm/wDQYa8Rr279of8A1vhz /rnN/wCgw14jQtgYUUUUwCiiigAooooAKnshm8iHy8tj5l3A+2MNz+BqCp7IZvIh8vLY+ZdwPtjD c/gamXws1ofxY+q/M99G9SWhkjjeNdrF0/dqoIypxjleqk4yD7k0xrrUVdEaLDPnb+5XnHb/AFvX /A+lLvVGC+YEaMHHy5WIDjpjtuAz8uUbOeM1ozKzxlQqMCCCjjhhg8e3bsfpX585cu6Ph6+K+rSj 7SlGSlezlHa3Z2Sav3b7XirMq5v8J+8i3MM7fs546Zyd+OM+vY4zQP7QzGN8XzDJPkfc+v7z+Wau 1FGFjYxqjIP4f7uAAOOwHtx0PHes+fy/A8iGYSlF2hC+v2I/5Lbfrtrpcgb7crovnQndn5hbsQD7 /P8A5/KkkGpAYiktmbGRuiZV7dTuJ/TtVqYkRlhII9oJLMPl6Hr7d+o6U+jne+n3ErMJxUZ8kH3X JHX191d1sygv9pvGjrNZMGIOQjEbSOSDnn+vqK4j4haNHFbNqdxKrSTsIyFJT51VipAJI6AjGR1P U4r0NQVkYeWAGOdynrwBz7/nwPwrjvid/wAixEP3f/HyOH7/ACP0PY9/fGOc4Pdl1SSxUEtLntZN mtWGYU404wSno7Rjqmtm1r91n5dDx2iiivuT7IK3PCv/ACFJf+uJ/wDQlrDrc8K/8hSX/rif/Qlr DFfwZHq5H/yMaPqdhRRRXgH64FFFFABRRRQAVzHi1VDWjbRuIcE45IGMfzP5109cz4u/5c/+B/8A stdOD/jx+f5HhcSr/hMq/wDbv/pSOZooor3T8rCiiigAooooAKKKKAPbPiN/yN0v/XJP5UUfEb/k bpf+uSfyopLYB37Q/wDrfDn/AFzm/wDQYa8Rr279of8A1vhz/rnN/wCgw14jQtgYUUUUwCiiigAo oooAKltxunVcbs5G3GS3HQcHk9Bx1xUVS26lp1VU3k5G3jJGOgz39PfGOaUtmaUvjj6o9/cvGnAj jKLlN54XHHAxwoJ2k4HykHdWiyq6lWUMrDBBGQRWXIuIn+RVypObjknHGSOeBnax7qQd1aUJBjCh XXaACH5I4Hfv9cmvzuotEz87zWFqcKkN03qvl1u9n56N7R2CMj5l2uCCT83PUnv/AE7DHSiQldre YEGQMMODkj9ew+vehiVkU+YAGONrDrwTx7/nwPxp9Z9bnjyly1FU3vr39d1/n6sKZGCu5fLCDJOV PByT+vc/XvRCCIwpjEe0ABVPy9B09u3QdKZIyxyb2Z0UDls/KQA3X0x1zx257UW6FRptuVNa9uu3 o7betvLViycOp/eDtuXkdRwR7+uOBnkVyPxK/feFoTH5cga4DDPOR5bnIPqAM/h3zg9kwJGAxXkH IrifiQWk8KQk7JAblSDHzvHlscjg47nr0B55rsy7/eYep6mR2li6LfSX4O/l+F+/uvW/kFFFFfen 6EFbnhX/AJCkv/XE/wDoS1h1ueFf+QpL/wBcT/6EtYYr+DI9XI/+RjR9TsKKKK8A/XAooooAKKKK ACuZ8Xf8uf8AwP8A9lrpq5nxd/y5/wDA/wD2WunB/wAePz/I8PiT/kV1f+3f/SkczRRRXun5UFFF FABRRRQAUUUUAe2fEb/kbpf+uSfyoo+I3/I3S/8AXJP5UUlsA79of/W+HP8ArnN/6DDXiNe3ftD/ AOt8Of8AXOb/ANBhrxGhbAwooopgFFFFABRRRQAVLbrvnVfL8zORs3YJ47e/p744PSoqmtYmmuFR YTMSD8gOCQASce/p/I9KUtmaUVepFLuv62f5P0PfGjZlKrGAzDgT/MWP3eRz0ztJxypB3VeiypRS 5GVyI3OWGMDrnnHfryetUI2imiR408yOZQyFiH3joCevGDtLANlWU7jWkYwI1QKHCleHOehHOTnJ HX61+d1OzPz3M6mkIS6t9vTdt7fK2qXJqhZF3ROuxXypG1uh9j7UqkkZKleSMGhSWUEqVJGdp6j8 qiRlWQqWdSx4Vz1OWPB75wTjPAA4FZ9DxVBypyj/AC69/Xrb52+eyZuVJiWZ0BP8R+Vj8oHPb0A4 yc8HrT5iRGWEgj2gksw+Xoevt36jpRISu1vMCDIGGHByR+vYfXvTl3FRuADY5AOQD9aPMty92FTt o/6a7ettNtEC7iPmAByehzx2riPiYn/FPx/uutyDlT9792eTx14zn+6pGRnFdkgEchBjK7jxtOVP LH8D3Jx1IGTXIfE5A3hqE4JIuR8o7/I59D0xnPHAIyM125dpi4ep6mS/us0ppbP9V5PXXTd+ab0P HqKKK+8P0EK3PCv/ACFJf+uJ/wDQlrDrY8NXEcGrAOceahRT2zkH+mKwxKbpSsenkslHMKTk+qO0 ooorwD9eCiiigAooooAK5nxd/wAuf/A//Za6auZ8Xf8ALn/wP/2WunB/x4/P8jw+JP8AkV1f+3f/ AEpHM0UUV7p+VBRRRQAUUUUAFFFFAHtnxG/5G6X/AK5J/Kij4jf8jdL/ANck/lRSWwDv2h/9b4c/ 65zf+gw14jXt37Q/+t8Of9c5v/QYa8RoWwMKKKKYBRRRQAUUUUAFOjkkhkDxOyOOjKcEU2igabTu j0/wb4hW6b7BJHPsmYrbjbmMELnyjjPG3PJB46BQgr0evnC2uDAWH8L4yecoQeGXBHzDt9T612Mf xS1sZD29gRyc+U5PsPvjjt/jXzeYZRUqVOeieDxDlNTHzhWw9ru91tZ6bdLfkesQso+Tc4bHCSHk AAd+/UZOTyetLIWVgVdctwqPwGOCeD1z+fA6V5I/xQ1xoZUENkjv911jb93xjgEnJzg85/HtUm8e a2YWVdWllLDG17OFVx3zwc/T3/PjWR4lvWy/r0PKhw3i6k3UbilbW7/LS/6/3tbntdQwgJ+78sxn HQHKjAA49B2xx0JxXisfj/xKmQdTYjk/6mPOT9VPGe34cUsfxA8SKTu1EsGYMT5UecdMD5cDP0/r mv7BxK6x+9/5Ew4ZxSUo88bPzl02vpb8/lue1SFY2EjOyD+L+7gAnnsB78dBz2rk/iUu7w1ENiv/ AKRnaeDxG5yD2IxntnBHfFeczeLdVuLyWabUdR2uoAWO6Me0jHICgL2P8Pf15rOn1rVLuIxXOo3V xGeqTTM6/XBJGfeurC5LVpVY1JSWh6OByCeFxFKvOopcr1tf9UvxvpuuhRooor6Q+jCiiigDt9D1 X+0bcpKy/aY/vAfxD+9j/P61q15vFPLAxaGV42IwSjEHH4USzyzsGmleRgMAuxJx+NedPAKU24uy Ps8LxdKlh4wqw5prre1/wep6RRXmdFT/AGd/e/D/AIJr/rn/ANOP/Jv/ALU9KkkjhjLyuqIOrMcA VX/tKx/5/bf/AL+r/jXntFNZcusjOfGVRv3KSXq7/oj0L+0rH/n9t/8Av6v+Nc94ouYLj7J5M0cu 3fnYwbH3fSueorWlgo05qaexwZhxLVxuGlh5QSUra3fRp/oFFFFdp80FFFFABRRRQAUUUUAe2fEb /kbpf+uSfyoo+I3/ACN0v/XJP5UUlsA79of/AFvhz/rnN/6DDXiNe3ftD/63w5/1zm/9BhrxGhbA wooopgFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAB RRRQAUUUUAFFFFABRRRQB7Z8Rv8Akbpf+uSfyoo+I3/I3S/9ck/lRSWwDv2h/wDW+HP+uc3/AKDD XiNe3ftD/wCt8Of9c5v/AEGGvEaFsDCiiimAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFF FABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAHtnxG/5G6X/rkn8qKPiN/yN0v/ AFyT+VFJbAO/aH/1vhz/AK5zf+gw14jXt37Q/wDrfDn/AFzm/wDQYa8RoWwMKKKKYBRRRQAUUUUA FFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAU UUUAe2fEb/kbpf8Arkn8qKPiN/yN0v8A1yT+VFJbAO/aH/1vhz/rnN/6DDXiNe3ftD/63w5/1zm/ 9BhrxGhbAwooopgFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUU UUAFFFFABRRRQAUUUUAFFFFABRRRQB7Z8Rv+Rul/65J/Kij4jf8AI3S/9ck/lRSWwDv2h/8AW+HP +uc3/oMNeI17d+0P/rfDn/XOb/0GGvEaFsDCiiimAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUU UAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAHtnxG/wCRul/65J/Kij4j f8jdL/1yT+VFJbAO/aH/ANb4c/65zf8AoMNeI17d+0P/AK3w5/1zm/8AQYa8RoWwMKKKKYBRRRQA UUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABR RRQAUUUUAe2fEb/kbpf+uSfyoo+I3/I3S/8AXJP5UUlsA79of/W+HP8ArnN/6DDXiNe3ftD/AOt8 Of8AXOb/ANBhrxGhbAwooopgFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFF ABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQB7Z8Rv+Rul/wCuSfyoo+I3/I3S/wDXJP5UUlsA 79of/W+HP+uc3/oMNeI17d+0P/rfDn/XOb/0GGvEaFsDCiiimAUUUUAFFFFABRRRQAUUUUAFFFFA BRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAHtnxG/5G6X/r kn8qKPiN/wAjdL/1yT+VFJbAanjfUPBXjeaz+33eqILJWWNrZFUNkKCTuU/3BjpXJ/8ACJ/Dj/oI +IP/ACH/APEUUUWAP+ET+HH/AEEfEH/kP/4ij/hE/hx/0EfEH/kP/wCIooosAf8ACJ/Dj/oI+IP/ ACH/APEUf8In8OP+gj4g/wDIf/xFFFFgD/hE/hx/0EfEH/kP/wCIo/4RP4cf9BHxB/5D/wDiKKKL AH/CJ/Dj/oI+IP8AyH/8RR/wifw4/wCgj4g/8h//ABFFFFgD/hE/hx/0EfEH/kP/AOIo/wCET+HH /QR8Qf8AkP8A+IooosAf8In8OP8AoI+IP/If/wARR/wifw4/6CPiD/yH/wDEUUUWAP8AhE/hx/0E fEH/AJD/APiKP+ET+HH/AEEfEH/kP/4iiiiwB/wifw4/6CPiD/yH/wDEUf8ACJ/Dj/oI+IP/ACH/ APEUUUWAP+ET+HH/AEEfEH/kP/4ij/hE/hx/0EfEH/kP/wCIooosAf8ACJ/Dj/oI+IP/ACH/APEU f8In8OP+gj4g/wDIf/xFFFFgD/hE/hx/0EfEH/kP/wCIo/4RP4cf9BHxB/5D/wDiKKKLAH/CJ/Dj /oI+IP8AyH/8RR/wifw4/wCgj4g/8h//ABFFFFgD/hE/hx/0EfEH/kP/AOIo/wCET+HH/QR8Qf8A kP8A+IooosAf8In8OP8AoI+IP/If/wARR/wifw4/6CPiD/yH/wDEUUUWAP8AhE/hx/0EfEH/AJD/ APiKP+ET+HH/AEEfEH/kP/4iiiiwB/wifw4/6CPiD/yH/wDEUf8ACJ/Dj/oI+IP/ACH/APEUUUWA P+ET+HH/AEEfEH/kP/4ij/hE/hx/0EfEH/kP/wCIooosAf8ACJ/Dj/oI+IP/ACH/APEUf8In8OP+ gj4g/wDIf/xFFFFgLXjDVbXWtfe9s2YxNGq/MuCCKKKKYH//2Q== ------_=_NextPart_001_01C51F06.7BD981FD Content-Type: application/rtf; name="loadVolumeData.rtf" Content-Transfer-Encoding: base64 Content-Description: loadVolumeData.rtf Content-Disposition: attachment; filename="loadVolumeData.rtf" e1xydGYxXGFuc2lcYW5zaWNwZzEyNTJcZGVmZjBcZGVmbGFuZzEwMzF7XGZvbnR0Ymx7XGYwXGZu aWxcZnBycTFcZmNoYXJzZXQwIENvdXJpZXIgTmV3O317XGYxXGZzd2lzc1xmY2hhcnNldDAgQXJp YWw7fX0NCntcY29sb3J0YmwgO1xyZWQwXGdyZWVuMTI4XGJsdWUwO1xyZWQwXGdyZWVuMFxibHVl MjU1O30NCntcKlxnZW5lcmF0b3IgTXNmdGVkaXQgNS40MS4xNS4xNTA3O31cdmlld2tpbmQ0XHVj MVxwYXJkXGNmMVxmMFxmczIwIC8vIEdsb2JhbCB2YXJpYWJsZXMgXHBhcg0KXGNmMCB2elNoYXBl ICpwU2hhcGUgPSBOVUxMO1xwYXINCnZ6VE1SZW5kZXJBY3Rpb24gKnBSZW5kZXJBY3Rpb24gPSBO VUxMO1xwYXINCkdMaW50IHZpZXdwb3J0WzRdO1xwYXINClxjZjIgaW50XGNmMCAgbGFzdFBvc2l0 aW9uWzJdID0gXHswLCAwXH07XHBhcg0KXGNmMiBmbG9hdFxjZjAgIGFuZ2xlc1syXSAgICAgPSBc ezAsIDBcfSwgbGFzdEFuZ2xlc1syXSA9IFx7MCwgMFx9O1xwYXINClxwYXINClxwYXINClxjZjEg Ly8vLy8vLy8vLy8vLy8vLy8vLy8vLy8vIFZvbHVtaXplciAvLy8vLy8vLy8vLy8vLy8vLy8vLy8v Ly8vLy8vLy9ccGFyDQpccGFyDQovLyBsb2FkVm9sdW1lRGF0YSAtIExvYWQgdGhlIHZvbHVtZSBk YXRhIGFuZCBpbml0aWFsaXplIHRoZSBzaGFwZSBub2RlLlxwYXINClxjZjIgdm9pZFxjZjAgIGxv YWRWb2x1bWVEYXRhKClccGFyDQpce1xwYXINClx0YWJcY2YyIGludFxjZjAgIGksaixrO1xwYXIN Clx0YWJcY2YxIC8vQ3JlYXRlIEFyZWFccGFyDQpcY2YwXHRhYiBCWVRFICpwYnlBcmVhID0gXGNm MiBuZXdcY2YwICBCWVRFWzEwMCoxMDAqMTAwXTtccGFyDQpcdGFiIEJZVEUgKnBieUxvb2t1cFRh YmxlQ29sb3IgPSBcY2YyIG5ld1xjZjAgIEJZVEVbMjU2KjRdO1xwYXINClx0YWJcY2YyIGludFxj ZjAgICpwaURhdGFEaW0gPSBcY2YyIG5ld1xjZjAgIFxjZjIgaW50XGNmMCBbM107XHBhcg0KXHRh YlxjZjIgY29uc3RcY2YwICBcY2YyIGZsb2F0XGNmMCAgc2FtcGxpbmdSYXRlWzNdID0gXHs1LCA1 LCA1XH07XHBhcg0KXHRhYlxjZjIgaW50XGNmMCAgKnBpUHRyPXBpRGF0YURpbTtccGFyDQpccGFy DQpcdGFiICpwaVB0cj0xMDA7XHBhcg0KXHRhYiAqcGlQdHIrKztccGFyDQpcdGFiICpwaVB0cj0x MDA7XHBhcg0KXHRhYiAqcGlQdHIrKztccGFyDQpcdGFiICpwaVB0cj0xMDA7XHRhYlxwYXINClxw YXINClx0YWIgQllURSogcGJ5UnVubmluZ1B0ciA9IHBieUFyZWE7XHBhcg0KXHRhYlxjZjIgZm9y XGNmMCAoaT0wO2k8MTAwO2krKylccGFyDQpcdGFiXHRhYlxjZjIgZm9yXGNmMCAoaj0wO2o8MTAw O2orKylccGFyDQpcdGFiXHRhYlx0YWJcY2YyIGZvclxjZjAgKGs9MDtrPDEwMDtrKyspXHBhcg0K XHRhYlx0YWJcdGFiXHtccGFyDQpcdGFiXHRhYlx0YWJcdGFiXGNmMiBpZlxjZjAgKChpPjIwICYm IGk8NTApICYmIChqPjIwICYmIGo8NTApICYmIChrPjEwICYmIGs8NDApKVxwYXINClx0YWJcdGFi XHRhYlx0YWJcdGFiICpwYnlSdW5uaW5nUHRyPTE7XHBhcg0KXHRhYlx0YWJcdGFiXHRhYlxjZjIg ZWxzZVxjZjAgIFxjZjIgaWZcY2YwICgoaT4yMCAmJiBpPDUwKSAmJiAoaj4yMCAmJiBqPDUwKSAm JiAoaz42MCAmJiBrPDkwKSlccGFyDQpcdGFiXHRhYlx0YWJcdGFiXHRhYiAqcGJ5UnVubmluZ1B0 cj0yNTU7XHBhcg0KXHRhYlx0YWJcdGFiXHRhYlxjZjIgZWxzZVxjZjAgIFxwYXINClx0YWJcdGFi XHRhYlx0YWJcdGFiICpwYnlSdW5uaW5nUHRyPTA7XHBhcg0KXHRhYlx0YWJcdGFiXHRhYiBwYnlS dW5uaW5nUHRyKys7XHBhcg0KXHRhYlx0YWJcdGFiXH1ccGFyDQpccGFyDQpcdGFiXGNmMiBmb3Jc Y2YwICAoXGNmMiBpbnRcY2YwICBpID0gMDsgaTwyNTY7IGkrKykgXHtccGFyDQogICAgICAgIFxj ZjIgaWZcY2YwICAoaSA9PSAxKVx7XHBhcg0KICAgICAgICAgICAgcGJ5TG9va3VwVGFibGVDb2xv cltpICogNF0gPSAgICAgMjU1O1x0YWJcdGFiXHRhYlxjZjEgLy9SXHBhcg0KXGNmMCAgICAgICAg ICAgICBwYnlMb29rdXBUYWJsZUNvbG9yW2kgKiA0ICsgMV0gPSAwO1x0YWJcdGFiXHRhYlx0YWJc Y2YxIC8vR1xwYXINClxjZjAgICAgICAgICAgICAgcGJ5TG9va3VwVGFibGVDb2xvcltpICogNCAr IDJdID0gMDtcdGFiXHRhYlx0YWJcdGFiXGNmMSAvL0JccGFyDQpcY2YwICAgICAgICAgICAgIHBi eUxvb2t1cFRhYmxlQ29sb3JbaSAqIDQgKyAzXSA9IDI1NTtcdGFiXHRhYlx0YWJcY2YxIC8vT3Bh Y2l0eVxwYXINClxjZjAgICAgICAgICBcfVxwYXINClx0YWJcdGFiXGNmMiBlbHNlXGNmMCAgXGNm MiBpZlxjZjAgIChpID09IDI1NSlce1xwYXINCiAgICAgICAgICAgIHBieUxvb2t1cFRhYmxlQ29s b3JbaSAqIDRdID1cdGFiICAwO1xwYXINCiAgICAgICAgICAgIHBieUxvb2t1cFRhYmxlQ29sb3Jb aSAqIDQgKyAxXSA9IDI1NTtccGFyDQogICAgICAgICAgICBwYnlMb29rdXBUYWJsZUNvbG9yW2kg KiA0ICsgMl0gPSAwO1xwYXINCiAgICAgICAgICAgIHBieUxvb2t1cFRhYmxlQ29sb3JbaSAqIDQg KyAzXSA9IDI1NTtccGFyDQogICAgICAgIFx9XHBhcg0KICAgICAgICBcY2YyIGVsc2VcY2YwXHtc cGFyDQogICAgICAgICAgICBwYnlMb29rdXBUYWJsZUNvbG9yW2kgKiA0XSA9XHRhYiAgMDtccGFy DQogICAgICAgICAgICBwYnlMb29rdXBUYWJsZUNvbG9yW2kgKiA0ICsgMV0gPSAwO1xwYXINCiAg ICAgICAgICAgIHBieUxvb2t1cFRhYmxlQ29sb3JbaSAqIDQgKyAyXSA9IDA7XHBhcg0KICAgICAg ICAgICAgcGJ5TG9va3VwVGFibGVDb2xvcltpICogNCArIDNdID0gMDtccGFyDQogICAgICAgIFx9 XHBhcg0KICAgIFx9XHBhcg0KXHBhcg0KXHBhcg0KICAgIFxjZjEgLy8gQ3JlYXRlIHRoZSB0ZXh0 dXJlXHBhcg0KXGNmMCAgICAgdnpQYXJhbWV0ZXJWb2x1bWVUZXh0dXJlICpwVm9sdW1lID0gXGNm MiBuZXdcY2YwICB2elBhcmFtZXRlclZvbHVtZVRleHR1cmUgKFxwYXINCiAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgcGlEYXRhRGltLFxwYXINCiAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgKFxjZjIgdm9pZFxjZjAgICopIHBieUFy ZWEsXHBhcg0KICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICBWWl9V TlNJR05FRF9CWVRFLCBccGFyDQpcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJc dGFiXHRhYiBWWl9MVU1JTkFOQ0UsXHBhcg0KXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFiXHRhYlx0 YWJcdGFiXHRhYlx0YWIgVlpfREVGQVVMVF9JTlRFUk5BTF9GT1JNQVQpO1xwYXINClxwYXINClx0 YWIgdnpQYXJhbWV0ZXJMb29rdXBUYWJsZSAqcExVVCA9IFxjZjIgbmV3XGNmMCAgdnpQYXJhbWV0 ZXJMb29rdXBUYWJsZShccGFyDQpcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJc dGFiIDI1NixccGFyDQpcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFiIHBi eUxvb2t1cFRhYmxlQ29sb3IsIFxwYXINClx0YWJcdGFiXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFi XHRhYlx0YWIgVlpfVU5TSUdORURfQllURSwgXHBhcg0KXHRhYlx0YWJcdGFiXHRhYlx0YWJcdGFi XHRhYlx0YWJcdGFiXHRhYiBWWl9SR0JBKTtccGFyDQpccGFyDQpcdGFiIHZ6VE1MVVRTaGFkZXIg KnBTaGFkZXIgPSBcY2YyIG5ld1xjZjAgIHZ6VE1MVVRTaGFkZXIoKTtccGFyDQogICAgdnpBcHBl YXJhbmNlICpwQXBwZWFyYW5jZT0gXGNmMiBuZXdcY2YwICB2ekFwcGVhcmFuY2UocFNoYWRlcik7 XHBhcg0KICAgIHBTaGFkZXItPnVucmVmKCk7XHBhcg0KICAgIHBBcHBlYXJhbmNlLT5zZXRQYXJh bWV0ZXIoInZvbHVtZSIsIHBWb2x1bWUpO1xwYXINCiAgICBwVm9sdW1lLT51bnJlZigpO1xwYXIN CiAgICBwQXBwZWFyYW5jZS0+c2V0UGFyYW1ldGVyKCJsb29rdXBfdGFibGUiLCBwTFVUKTtccGFy DQogICAgcExVVC0+dW5yZWYoKTtccGFyDQpccGFyDQpcdGFiXGNmMSAvLyBJbml0aWFsaXplIGdl b21ldHJ5XHBhcg0KXGNmMCAgICAgdnpHZW9tZXRyeSAqcEdlb21ldHJ5ID0gXGNmMiBuZXdcY2Yw ICB2ekJsb2NrKCk7XHBhcg0KXHBhcg0KICAgIFxjZjEgLy8gSW5pdGlhbGl6ZSBzaGFwZSBub2Rl XHBhcg0KXGNmMCAgICAgcFNoYXBlID0gXGNmMiBuZXdcY2YwICB2elNoYXBlKHBHZW9tZXRyeSwg cEFwcGVhcmFuY2UpO1xwYXINClxwYXINCiAgICBwR2VvbWV0cnktPnVucmVmKCk7XHBhcg0KICAg IHBBcHBlYXJhbmNlLT51bnJlZigpO1xwYXINClxwYXINCiAgICBcY2YyIGlmXGNmMCAoIXBSZW5k ZXJBY3Rpb24pIFx7XHBhcg0KICAgICAgXGNmMSAvLyBJbml0aWFsaXplIHRoZSByZW5kZXIgYWN0 aW9uXHBhcg0KXGNmMCAgICAgICBwUmVuZGVyQWN0aW9uID0gXGNmMiBuZXdcY2YwICB2elRNUmVu ZGVyQWN0aW9uKDEpO1xwYXINClx0YWIgICBwUmVuZGVyQWN0aW9uLT5zZXRTYW1wbGluZ1JhdGUo c2FtcGxpbmdSYXRlKTtccGFyDQpcdGFiICAgcFJlbmRlckFjdGlvbi0+bWFuYWdlKHBTaGFwZSk7 XHBhcg0KICAgIFx9XHBhcg0KXH1cZjFccGFyDQp9DQoA ------_=_NextPart_001_01C51F06.7BD981FD-- From info-volumizer-owner@els.sgi.com Wed Mar 2 02:29:55 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j22AQpAF6397925; Wed, 2 Mar 2005 02:26:51 -0800 (PST) Received: from hastur.corp.sgi.com (hastur.corp.sgi.com [198.149.32.33]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j22AF2va6235762 for ; Wed, 2 Mar 2005 02:15:03 -0800 (PST) Received: from mtv-atc-013e--n.corp.sgi.com (mtv-atc-013e--n.corp.sgi.com [192.26.78.50]) by hastur.corp.sgi.com (8.12.9/8.12.10/SGI_generic_relay-1.2) with ESMTP id j22AEwVt112806977 for ; Wed, 2 Mar 2005 02:14:58 -0800 (PST) Received: from praveenlaptop (134.14.90.76 [134.14.90.76]) by mtv-atc-013e--n.corp.sgi.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2655.55) id FYL5PJ2L; Wed, 2 Mar 2005 02:15:00 -0800 Message-ID: <018901c51f10$c72c1560$4c5a0e86@praveenlaptop> From: "Praveen Bhaniramka" To: References: <50726031C3515F4F879593DD4E801AE622D262@acvstg03.intern.acv.ac.at> Subject: Re: [info-volumizer] Volumizer bizarre behavior Date: Wed, 2 Mar 2005 15:45:30 +0530 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0184_01C51F3E.DCFB5760" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com This is a multi-part message in MIME format. ------=_NextPart_000_0184_01C51F3E.DCFB5760 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Oscar, Yes, your hypothesiss is correct. This error is mainly due to trilinear = interpolation in the OpenGL pipeline. It only shows up with the higher = value since there exists a visible range of voxels between voxels with = this value and voxels with a value 0. e.g. Your data looks like the following (looking at just a row of voxels) - 0 0 0 0 0 255 255 255 0 0 0 0 1 1 1 0 0 0 0 During rasterization, if any fragment's texture coordinate maps to a = location between the voxels with 0 and 255 above, you will get an = interpolated value between these two values. This interpolated value CAN = be 1 depending on the distance of the sample point from the two adjacent = voxels. This interpolated value then gets looked up into the LUT, which = assigns it a color value of Red. Note that this is extremely dependent = on the view and hence you see this only from certain viewing angles and = as you rotate, the colors seem to change. With the smaller value also = this interpolation does occur, however any interpolated value between 0 = and 1 gets mapped to either Red or is completely transparent. Hence this = artifact does not show up with the other cube. Hope this makes sense.=20 Disabling linear interpolation - The vzTMRenderAction currently assumes = linear interpolation for all the built-in shaders inside Volumizer. = However, just like all other things in Volumizer, you can override this = default setting by writing a custom shader of your own. (see vzTMShader = class for details - it is discussed further in Chapter 5 of the pguide). = After invoking the vzTMShader::bindVolumeTextureCB method to bind the = 'volume' parameter, simply make the appropriate GL calls to set the = interpolation mode to GL_LINEAR. What graphics hardware are you running = this on? If the graphics hardware supports ARB_fragment_program, you can = re-use some of the fragment programs that we ship under the src/shaders = directory and use the vzTMFragmentProgram class interface as well. With = future Volumizer releases, we do plan on adding different interpolation = modes as an optional hint that you can pass to various shaders.=20 Dimensions - Yes, you do not need to allocate memory for these entities = since the values are copied and stored internally. This is usually = evident from the declaration for the various methods. For larger data = like volume data and lookup tables, performing a deep copy operation is = wasteful and unnecessary. Hence, Volumizer uses shallow pointer copies = consistenly thru the API.=20 - Praveen ----- Original Message -----=20 From: Oscar Serrano=20 To: info-volumizer@els.sgi.com=20 Sent: Wednesday, March 02, 2005 2:31 PM Subject: AW: [info-volumizer] Volumizer bizarre behavior Thank you very much for your answer Praveen, you were totally right = and now the rendering using the simpleShader looks much better. With the = LUTShader we are still having some problems but we think that are caused = because of the interpolation and I don't know whether there is a = solution for them without changing the input data.=20 In our output when using the two cubes from the mail before using a = LUT with just 3 RGBA mappings (255,0,0,255 for the luminance associated = with cube1; 0,255,0,255 for the luminance associated with cube 2 and = 0,0,0,0 for background) shows that the cube with lower luminance is = drawn correctly but in the other you get a mix of colors when looking at = the volume data from angular orientations (I enclose a snapshot from one = output and the new loadVolumeData function). Changing the color table or = the luminance of the input cubes changes the output but is not improving = much. Is this problem related with the extremely high frequency data = that we are using??. Can the interpolation be avoided somehow (assuming = that this is this is what is causing our problem)?? . And the most = important question. There will be any way of showing the two cubes with = different colors without that mixing??. Thanks and regards Cheers, Oscar P.D. By the way. I was changing the pbyArea in the code from the mail = before. But also the piDataDim just in the case it could cause problems = but I think that what you mean is that I should change just the pbyArea. = I don=B4t think that allocating the dimension is really needed right?? =20 -----Urspr=FCngliche Nachricht----- Von: owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im Auftrag von Praveen = Bhaniramka Gesendet: Dienstag, 1. M=E4rz 2005 09:49 An: info-volumizer@els.sgi.com Betreff: Re: [info-volumizer] Volumizer bizarre behavior Hi Oscar, The stripes that you see when looking at the volume data from angular = orientations are mainly due to the fact that the data is extremely high = frequency (it is a step function) and while rendering, it is = undersampled in object space with respect to the screen space = resolution. You can get rid of these artifacts by increasing the = sampling rate by using vzTMRenderAction::setSamplingRate() method. Note = that increasing the sampling rate would also increase the rendering = overhead. Volumizer uses a default sampling rate of (1, 1, 1), which = implies one slice per-voxel. Hence as the resolution of your volume data = increases with respect to your screen space resolution (which is = typically more finite), the slicing artifacts due to undersampling = reduce considerably. The programming guide discusses this issue in more = detail as well.=20 As for the noise issue, please see my posting on = http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.html. = Basically, the data pointer that you are passing to the = vzParameterVolumeTexture constructor needs to be allocated from the heap = (using malloc() or new[]). The following code is incorrect - // BYTE pbyArea[100][100][100]; const int piDataDim[3] =3D {100, 100, 100};=20 // // Create the texture vzParameterVolumeTexture *pVolume =3D new vzParameterVolumeTexture = ( piDataDim, (void *) pbyArea, VZ_UNSIGNED_BYTE,=20 VZ_LUMINANCE, VZ_DEFAULT_INTERNAL_FORMAT); // Cheers, Praveen ----- Original Message -----=20 From: Oscar Serrano=20 To: info-volumizer@sgi.com=20 Sent: Monday, February 28, 2005 3:04 PM Subject: [info-volumizer] Volumizer bizarre behavior Hello, We are just starting to evaluate the possibilities of = volumizer but we have detected some bizarre output. We have reproduced = the error in a simple program that I enclose. It just draws a shape with = cube objects using the SimpleShader. The cubes are showing strange = stripes that vary when moving them and we also have some strange kind of = noise (I enclose a snapshot of the output). We have also experience this = problem using the LUTShader and using different graphic cards. Does = anyone have an idea what is wrong???=20 Thank you, Oscar Serrano ------=_NextPart_000_0184_01C51F3E.DCFB5760 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi Oscar,
 
Yes, your hypothesiss is correct. This error is mainly due to = trilinear=20 interpolation in the OpenGL pipeline. It only shows up with the higher = value=20 since there exists a visible range of voxels between voxels with this = value and=20 voxels with a value 0. e.g.
 
Your data looks like the following (looking at just a row of = voxels)=20 -
0 0 0 0 0 255 255 255 0 0 0 0 1 1 1 0 0 0 0
 
During rasterization, if any fragment's texture coordinate maps to = a=20 location between the voxels with 0 and 255 above, you will get an = interpolated=20 value between these two values. This interpolated = value CAN be 1=20 depending on the distance of the sample point from the two adjacent = voxels. This=20 interpolated value then gets looked up into the LUT, which assigns it a = color=20 value of Red. Note that this is extremely dependent on the view and = hence you=20 see this only from certain viewing angles and as you rotate, the colors = seem to=20 change. With the smaller value also this interpolation does occur, = however any=20 interpolated value between 0 and 1 gets mapped to either Red or is = completely=20 transparent. Hence this artifact does not show up with the other cube. = Hope this=20 makes sense.
 
Disabling linear interpolation - The vzTMRenderAction currently = assumes=20 linear interpolation for all the built-in shaders inside Volumizer. = However,=20 just like all other things in Volumizer, you can override this default = setting=20 by writing a custom shader of your own. (see vzTMShader class for = details - it=20 is discussed further in Chapter 5 of the pguide). = After invoking the=20 vzTMShader::bindVolumeTextureCB method to bind the 'volume' = parameter,=20 simply make the appropriate GL calls to set the interpolation mode to = GL_LINEAR.=20 What graphics hardware are you running this on? If the graphics hardware = supports ARB_fragment_program, you can re-use some of the fragment = programs that=20 we ship under the src/shaders directory and use the vzTMFragmentProgram = class=20 interface as well. With future Volumizer releases, we do plan on adding=20 different interpolation modes as an optional hint that you can pass to = various=20 shaders.
 
Dimensions - Yes, you do not need to allocate memory for these = entities=20 since the values are copied and stored internally. This is usually = evident=20 from the declaration for the  various methods. For larger data like = volume=20 data and lookup tables, performing a deep copy operation is wasteful and = unnecessary. Hence, Volumizer uses shallow pointer copies consistenly = thru the=20 API.
 
- Praveen
 
----- Original Message -----
From:=20 Oscar=20 Serrano
Sent: Wednesday, March 02, 2005 = 2:31=20 PM
Subject: AW: [info-volumizer] = Volumizer=20 bizarre behavior

Thank you very = much for your=20 answer Praveen, you were totally right and now the rendering using the = simpleShader looks much better. With the LUTShader we are still having = some=20 problems but we think that are caused because of the interpolation and = I don=92t=20 know whether there is a solution for them without changing the input = data.=20

 

In our output = when using the=20 two cubes from the mail before using a LUT with just 3 RGBA mappings=20 (255,0,0,255 for the luminance associated with cube1; 0,255,0,255 for = the=20 luminance associated with cube 2 and 0,0,0,0 for background) shows = that the=20 cube with lower luminance is drawn correctly but in the other you get = a mix of=20 colors when looking at the volume data from angular orientations (I = enclose a=20 snapshot from one output and the new loadVolumeData function). = Changing the=20 color table or the luminance of the input cubes changes the output but = is not=20 improving much. Is this problem related with the extremely high = frequency data=20 that we are using??. Can the interpolation be avoided somehow = (assuming that=20 this is this is what is causing our problem)?? . And the most = important=20 question=85 There will be any way of showing the two cubes with = different colors=20 without that mixing??.

 

Thanks and = regards=20 Cheers,

        =20 Oscar

 

P.D.  By = the way=85 I was=20 changing the pbyArea in the code from the mail before. But also the = piDataDim=20 just in the case it could cause problems but I think that what you = mean is=20 that I should change just the pbyArea. I don=B4t think that allocating = the=20 dimension is really needed right??

 

            =

-----Urspr=FCngliche = Nachricht-----
Von:=20 owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im=20 Auftrag von Praveen Bhaniramka
Gesendet: Dienstag, 1. M=E4rz 2005=20 09:49
An: info-volumizer@els.sgi.com
Betreff: Re: = [info-volumizer]=20 Volumizer bizarre behavior

 

Hi = Oscar,

 

The stripes that you see when = looking at=20 the volume data from angular orientations are mainly due to the fact = that the=20 data is extremely high frequency (it is a step function) and while=20 rendering, it is undersampled in object space with respect = to the=20 screen space resolution. You can get rid of these artifacts by = increasing the=20 sampling rate by using vzTMRenderAction::setSamplingRate() = method. Note=20 that increasing the sampling rate would also increase the rendering = overhead.=20 Volumizer uses a default sampling rate of (1, 1, 1), which implies one = slice=20 per-voxel. Hence as the resolution of your volume data increases with = respect=20 to your screen space resolution (which is typically more finite), the = slicing=20 artifacts due to undersampling reduce considerably. The programming = guide=20 discusses this issue in more detail as well.

 

As for the noise issue, = please see my=20 posting on http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.= html.=20 Basically, the data pointer that you are passing to the=20 vzParameterVolumeTexture constructor needs to be allocated from = the heap=20 (using malloc() or new[]). The following code is incorrect=20 -

 

//<snip>

 BYTE=20 pbyArea[100][100][100];
 const int piDataDim[3] =3D {100, 100, = 100}; 

//<snip>

    // Create = the=20 texture
    vzParameterVolumeTexture *pVolume =3D = new=20 vzParameterVolumeTexture=20 = (
           &n= bsp;           &nb= sp;           &nbs= p;       =20 = piDataDim,
          = ;            =             &= nbsp;        =20 (void *)=20 = pbyArea,
          &= nbsp;           &n= bsp;           &nb= sp;        =20 VZ_UNSIGNED_BYTE,=20 =
           VZ_= LUMINANCE,
          = ; VZ_DEFAULT_INTERNAL_FORMAT);

//<snip>

 

Cheers,

Praveen

 

 

----- Original Message = -----=20

From: Oscar Serrano=20

Sent: Monday, February 28, 2005 3:04=20 PM

Subject: [info-volumizer] Volumizer = bizarre=20 behavior

 

Hello,

 

      We are just = starting=20 to evaluate the possibilities of volumizer but we have detected some = bizarre=20 output. We have reproduced the error in a simple program that I = enclose. It=20 just draws a shape with cube objects using the SimpleShader. The = cubes are=20 showing strange stripes that vary when moving them and we also have = some=20 strange kind of noise (I enclose a snapshot of the output). We have = also=20 experience this problem using the LUTShader and using different = graphic=20 cards. Does anyone have an idea what is wrong???

 

Thank = you,

        = Oscar=20 Serrano

 

 

 

------=_NextPart_000_0184_01C51F3E.DCFB5760-- From info-volumizer-owner@els.sgi.com Wed Mar 2 20:00:25 2005 Received: (from majordom@localhost) by els.sgi.com (8.12.10/8.12.2/els-2.2) id j233umGN6343365; Wed, 2 Mar 2005 19:56:48 -0800 (PST) Received: from acvstg03.intern.acv.ac.at (acvstg03.acv.ac.at [194.152.172.83]) by els.sgi.com (8.12.10/8.12.2/els-2.2) with ESMTP id j22BD3va6358206 for ; Wed, 2 Mar 2005 03:13:07 -0800 (PST) Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C51F19.1CDE2803" Subject: AW: [info-volumizer] Volumizer bizarre behavior X-MimeOLE: Produced By Microsoft Exchange V6.0.6487.1 Date: Wed, 2 Mar 2005 12:15:16 +0100 Message-ID: <50726031C3515F4F879593DD4E801AE622F9A8@acvstg03.intern.acv.ac.at> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [info-volumizer] Volumizer bizarre behavior Thread-Index: AcUfFms4Fgv6oMyqTcOzgt1NA8Y72QAAjqqg From: "Oscar Serrano" To: Sender: owner-info-volumizer@els.sgi.com Precedence: bulk Reply-To: info-volumizer@els.sgi.com This is a multi-part message in MIME format. ------_=_NextPart_001_01C51F19.1CDE2803 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Again thank you very much... You have explained it really good and I = think we will try to write the shader reusing the fragment programs as = you say. =20 Regards, Oscar =20 -----Urspr=FCngliche Nachricht----- Von: owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im Auftrag von Praveen = Bhaniramka Gesendet: Mittwoch, 2. M=E4rz 2005 11:16 An: info-volumizer@els.sgi.com Betreff: Re: [info-volumizer] Volumizer bizarre behavior =20 Hi Oscar, =20 Yes, your hypothesiss is correct. This error is mainly due to trilinear = interpolation in the OpenGL pipeline. It only shows up with the higher = value since there exists a visible range of voxels between voxels with = this value and voxels with a value 0. e.g. =20 Your data looks like the following (looking at just a row of voxels) - 0 0 0 0 0 255 255 255 0 0 0 0 1 1 1 0 0 0 0 =20 During rasterization, if any fragment's texture coordinate maps to a = location between the voxels with 0 and 255 above, you will get an = interpolated value between these two values. This interpolated value CAN = be 1 depending on the distance of the sample point from the two adjacent = voxels. This interpolated value then gets looked up into the LUT, which = assigns it a color value of Red. Note that this is extremely dependent = on the view and hence you see this only from certain viewing angles and = as you rotate, the colors seem to change. With the smaller value also = this interpolation does occur, however any interpolated value between 0 = and 1 gets mapped to either Red or is completely transparent. Hence this = artifact does not show up with the other cube. Hope this makes sense.=20 =20 Disabling linear interpolation - The vzTMRenderAction currently assumes = linear interpolation for all the built-in shaders inside Volumizer. = However, just like all other things in Volumizer, you can override this = default setting by writing a custom shader of your own. (see vzTMShader = class for details - it is discussed further in Chapter 5 of the pguide). = After invoking the vzTMShader::bindVolumeTextureCB method to bind the = 'volume' parameter, simply make the appropriate GL calls to set the = interpolation mode to GL_LINEAR. What graphics hardware are you running = this on? If the graphics hardware supports ARB_fragment_program, you can = re-use some of the fragment programs that we ship under the src/shaders = directory and use the vzTMFragmentProgram class interface as well. With = future Volumizer releases, we do plan on adding different interpolation = modes as an optional hint that you can pass to various shaders.=20 =20 Dimensions - Yes, you do not need to allocate memory for these entities = since the values are copied and stored internally. This is usually = evident from the declaration for the various methods. For larger data = like volume data and lookup tables, performing a deep copy operation is = wasteful and unnecessary. Hence, Volumizer uses shallow pointer copies = consistenly thru the API.=20 =20 - Praveen =20 ----- Original Message -----=20 From: Oscar Serrano =20 To: info-volumizer@els.sgi.com=20 Sent: Wednesday, March 02, 2005 2:31 PM Subject: AW: [info-volumizer] Volumizer bizarre behavior =20 Thank you very much for your answer Praveen, you were totally right and = now the rendering using the simpleShader looks much better. With the = LUTShader we are still having some problems but we think that are caused = because of the interpolation and I don't know whether there is a = solution for them without changing the input data.=20 =20 In our output when using the two cubes from the mail before using a LUT = with just 3 RGBA mappings (255,0,0,255 for the luminance associated with = cube1; 0,255,0,255 for the luminance associated with cube 2 and 0,0,0,0 = for background) shows that the cube with lower luminance is drawn = correctly but in the other you get a mix of colors when looking at the = volume data from angular orientations (I enclose a snapshot from one = output and the new loadVolumeData function). Changing the color table or = the luminance of the input cubes changes the output but is not improving = much. Is this problem related with the extremely high frequency data = that we are using??. Can the interpolation be avoided somehow (assuming = that this is this is what is causing our problem)?? . And the most = important question... There will be any way of showing the two cubes = with different colors without that mixing??. =20 Thanks and regards Cheers, Oscar =20 P.D. By the way... I was changing the pbyArea in the code from the = mail before. But also the piDataDim just in the case it could cause = problems but I think that what you mean is that I should change just the = pbyArea. I don=B4t think that allocating the dimension is really needed = right?? =20 =20 -----Urspr=FCngliche Nachricht----- Von: owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im Auftrag von Praveen = Bhaniramka Gesendet: Dienstag, 1. M=E4rz 2005 09:49 An: info-volumizer@els.sgi.com Betreff: Re: [info-volumizer] Volumizer bizarre behavior =20 Hi Oscar, =20 The stripes that you see when looking at the volume data from angular = orientations are mainly due to the fact that the data is extremely high = frequency (it is a step function) and while rendering, it is = undersampled in object space with respect to the screen space = resolution. You can get rid of these artifacts by increasing the = sampling rate by using vzTMRenderAction::setSamplingRate() method. Note = that increasing the sampling rate would also increase the rendering = overhead. Volumizer uses a default sampling rate of (1, 1, 1), which = implies one slice per-voxel. Hence as the resolution of your volume data = increases with respect to your screen space resolution (which is = typically more finite), the slicing artifacts due to undersampling = reduce considerably. The programming guide discusses this issue in more = detail as well.=20 =20 As for the noise issue, please see my posting on = http://els.sgi.com/Archives_Public/info-volumizer/200411/msg00013.html. = Basically, the data pointer that you are passing to the = vzParameterVolumeTexture constructor needs to be allocated from the heap = (using malloc() or new[]). The following code is incorrect - =20 // BYTE pbyArea[100][100][100]; const int piDataDim[3] =3D {100, 100, 100};=20 // // Create the texture vzParameterVolumeTexture *pVolume =3D new vzParameterVolumeTexture = ( piDataDim, (void *) pbyArea, VZ_UNSIGNED_BYTE,=20 VZ_LUMINANCE, VZ_DEFAULT_INTERNAL_FORMAT); // =20 Cheers, Praveen =20 =20 ----- Original Message -----=20 From: Oscar Serrano =20 To: info-volumizer@sgi.com=20 Sent: Monday, February 28, 2005 3:04 PM Subject: [info-volumizer] Volumizer bizarre behavior =20 Hello, =20 We are just starting to evaluate the possibilities of volumizer = but we have detected some bizarre output. We have reproduced the error = in a simple program that I enclose. It just draws a shape with cube = objects using the SimpleShader. The cubes are showing strange stripes = that vary when moving them and we also have some strange kind of noise = (I enclose a snapshot of the output). We have also experience this = problem using the LUTShader and using different graphic cards. Does = anyone have an idea what is wrong???=20 =20 Thank you, Oscar Serrano =20 =20 =20 ------_=_NextPart_001_01C51F19.1CDE2803 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Again thank you = very much… You have explained it really good and I think we will try to write the = shader reusing the fragment programs as you say.

 

Regards,

=A0=A0=A0=A0=A0=A0= =A0 Oscar

 

-----Urspr=FCngliche = Nachricht-----
Von: owner-info-volumizer@els.sgi.com = [mailto:owner-info-volumizer@els.sgi.com] Im Auftrag von Praveen = Bhaniramka
Gesendet: Mittwoch, 2. = M=E4rz 2005 11:16
An: = info-volumizer@els.sgi.com
Betreff: Re: = [info-volumizer] Volumizer bizarre behavior

 

Hi = Oscar,

 

Yes, your = hypothesiss is correct. This error is mainly due to trilinear interpolation in the = OpenGL pipeline. It only shows up with the higher value since there exists a = visible range of voxels between voxels with this value and voxels with a value = 0. e.g.

 

Your data = looks like the following (looking at just a row of voxels) -

0 0 0 0 0 255 = 255 255 0 0 0 0 1 1 1 0 0 0 0

 

During = rasterization, if any fragment's texture coordinate maps to a location between the voxels = with 0 and 255 above, you will get an interpolated value between these two values. This interpolated value CAN be 1 depending on the distance of the sample point from the two adjacent voxels. This = interpolated value then gets looked up into the LUT, which assigns it a color value = of Red. Note that this is extremely dependent on the view and hence you see this = only from certain viewing angles and as you rotate, the colors seem to = change. With the smaller value also this interpolation does occur, however any = interpolated value between 0 and 1 gets mapped to either Red or is completely = transparent. Hence this artifact does not show up with the other cube. Hope this = makes sense.