diff --git a/filter/xbrz-config.h b/filter/xbrz-config.h
new file mode 100644
index 00000000..ae11f7a0
--- /dev/null
+++ b/filter/xbrz-config.h
@@ -0,0 +1,40 @@
+// ****************************************************************************
+// * This file is part of the HqMAME project. It is distributed under *
+// * GNU General Public License: http://www.gnu.org/licenses/gpl.html *
+// * Copyright (C) Zenju (zenju AT gmx DOT de) - All Rights Reserved *
+// * *
+// * Additionally and as a special exception, the author gives permission *
+// * to link the code of this program with the MAME library (or with modified *
+// * versions of MAME that use the same license as MAME), and distribute *
+// * linked combinations including the two. You must obey the GNU General *
+// * Public License in all respects for all of the code used other than MAME. *
+// * If you modify this file, you may extend this exception to your version *
+// * of the file, but you are not obligated to do so. If you do not wish to *
+// * do so, delete this exception statement from your version. *
+// ****************************************************************************
+
+#ifndef XBRZ_CONFIG_HEADER_284578425345
+#define XBRZ_CONFIG_HEADER_284578425345
+
+//do NOT include any headers here! used by xBRZ_dll!!!
+
+namespace xbrz
+{
+struct ScalerCfg
+{
+ ScalerCfg() :
+ luminanceWeight_(1),
+ equalColorTolerance_(30),
+ dominantDirectionThreshold(3.6),
+ steepDirectionThreshold(2.2),
+ newTestAttribute_(0) {}
+
+ double luminanceWeight_;
+ double equalColorTolerance_;
+ double dominantDirectionThreshold;
+ double steepDirectionThreshold;
+ double newTestAttribute_; //unused; test new parameters
+};
+}
+
+#endif
\ No newline at end of file
diff --git a/filter/xbrz-license.txt b/filter/xbrz-license.txt
new file mode 100644
index 00000000..94a04532
--- /dev/null
+++ b/filter/xbrz-license.txt
@@ -0,0 +1,621 @@
+ GNU GENERAL PUBLIC LICENSE
+ Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc.
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+ Preamble
+
+ The GNU General Public License is a free, copyleft license for
+software and other kinds of works.
+
+ The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works. By contrast,
+the GNU General Public License is intended to guarantee your freedom to
+share and change all versions of a program--to make sure it remains free
+software for all its users. We, the Free Software Foundation, use the
+GNU General Public License for most of our software; it applies also to
+any other work released this way by its authors. You can apply it to
+your programs, too.
+
+ When we speak of free software, we are referring to freedom, not
+price. Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+them if you wish), that you receive source code or can get it if you
+want it, that you can change the software or use pieces of it in new
+free programs, and that you know you can do these things.
+
+ To protect your rights, we need to prevent others from denying you
+these rights or asking you to surrender the rights. Therefore, you have
+certain responsibilities if you distribute copies of the software, or if
+you modify it: responsibilities to respect the freedom of others.
+
+ For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must pass on to the recipients the same
+freedoms that you received. You must make sure that they, too, receive
+or can get the source code. And you must show them these terms so they
+know their rights.
+
+ Developers that use the GNU GPL protect your rights with two steps:
+(1) assert copyright on the software, and (2) offer you this License
+giving you legal permission to copy, distribute and/or modify it.
+
+ For the developers' and authors' protection, the GPL clearly explains
+that there is no warranty for this free software. For both users' and
+authors' sake, the GPL requires that modified versions be marked as
+changed, so that their problems will not be attributed erroneously to
+authors of previous versions.
+
+ Some devices are designed to deny users access to install or run
+modified versions of the software inside them, although the manufacturer
+can do so. This is fundamentally incompatible with the aim of
+protecting users' freedom to change the software. The systematic
+pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable. Therefore, we
+have designed this version of the GPL to prohibit the practice for those
+products. If such problems arise substantially in other domains, we
+stand ready to extend this provision to those domains in future versions
+of the GPL, as needed to protect the freedom of users.
+
+ Finally, every program is threatened constantly by software patents.
+States should not allow patents to restrict development and use of
+software on general-purpose computers, but in those that do, we wish to
+avoid the special danger that patents applied to a free program could
+make it effectively proprietary. To prevent this, the GPL assures that
+patents cannot be used to render the program non-free.
+
+ The precise terms and conditions for copying, distribution and
+modification follow.
+
+ TERMS AND CONDITIONS
+
+ 0. Definitions.
+
+ "This License" refers to version 3 of the GNU General Public License.
+
+ "Copyright" also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+ "The Program" refers to any copyrightable work licensed under this
+License. Each licensee is addressed as "you". "Licensees" and
+"recipients" may be individuals or organizations.
+
+ To "modify" a work means to copy from or adapt all or part of the work
+in a fashion requiring copyright permission, other than the making of an
+exact copy. The resulting work is called a "modified version" of the
+earlier work or a work "based on" the earlier work.
+
+ A "covered work" means either the unmodified Program or a work based
+on the Program.
+
+ To "propagate" a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for
+infringement under applicable copyright law, except executing it on a
+computer or modifying a private copy. Propagation includes copying,
+distribution (with or without modification), making available to the
+public, and in some countries other activities as well.
+
+ To "convey" a work means any kind of propagation that enables other
+parties to make or receive copies. Mere interaction with a user through
+a computer network, with no transfer of a copy, is not conveying.
+
+ An interactive user interface displays "Appropriate Legal Notices"
+to the extent that it includes a convenient and prominently visible
+feature that (1) displays an appropriate copyright notice, and (2)
+tells the user that there is no warranty for the work (except to the
+extent that warranties are provided), that licensees may convey the
+work under this License, and how to view a copy of this License. If
+the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+ 1. Source Code.
+
+ The "source code" for a work means the preferred form of the work
+for making modifications to it. "Object code" means any non-source
+form of a work.
+
+ A "Standard Interface" means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of
+interfaces specified for a particular programming language, one that
+is widely used among developers working in that language.
+
+ The "System Libraries" of an executable work include anything, other
+than the work as a whole, that (a) is included in the normal form of
+packaging a Major Component, but which is not part of that Major
+Component, and (b) serves only to enable use of the work with that
+Major Component, or to implement a Standard Interface for which an
+implementation is available to the public in source code form. A
+"Major Component", in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system
+(if any) on which the executable work runs, or a compiler used to
+produce the work, or an object code interpreter used to run it.
+
+ The "Corresponding Source" for a work in object code form means all
+the source code needed to generate, install, and (for an executable
+work) run the object code and to modify the work, including scripts to
+control those activities. However, it does not include the work's
+System Libraries, or general-purpose tools or generally available free
+programs which are used unmodified in performing those activities but
+which are not part of the work. For example, Corresponding Source
+includes interface definition files associated with source files for
+the work, and the source code for shared libraries and dynamically
+linked subprograms that the work is specifically designed to require,
+such as by intimate data communication or control flow between those
+subprograms and other parts of the work.
+
+ The Corresponding Source need not include anything that users
+can regenerate automatically from other parts of the Corresponding
+Source.
+
+ The Corresponding Source for a work in source code form is that
+same work.
+
+ 2. Basic Permissions.
+
+ All rights granted under this License are granted for the term of
+copyright on the Program, and are irrevocable provided the stated
+conditions are met. This License explicitly affirms your unlimited
+permission to run the unmodified Program. The output from running a
+covered work is covered by this License only if the output, given its
+content, constitutes a covered work. This License acknowledges your
+rights of fair use or other equivalent, as provided by copyright law.
+
+ You may make, run and propagate covered works that you do not
+convey, without conditions so long as your license otherwise remains
+in force. You may convey covered works to others for the sole purpose
+of having them make modifications exclusively for you, or provide you
+with facilities for running those works, provided that you comply with
+the terms of this License in conveying all material for which you do
+not control copyright. Those thus making or running the covered works
+for you must do so exclusively on your behalf, under your direction
+and control, on terms that prohibit them from making any copies of
+your copyrighted material outside their relationship with you.
+
+ Conveying under any other circumstances is permitted solely under
+the conditions stated below. Sublicensing is not allowed; section 10
+makes it unnecessary.
+
+ 3. Protecting Users' Legal Rights From Anti-Circumvention Law.
+
+ No covered work shall be deemed part of an effective technological
+measure under any applicable law fulfilling obligations under article
+11 of the WIPO copyright treaty adopted on 20 December 1996, or
+similar laws prohibiting or restricting circumvention of such
+measures.
+
+ When you convey a covered work, you waive any legal power to forbid
+circumvention of technological measures to the extent such circumvention
+is effected by exercising rights under this License with respect to
+the covered work, and you disclaim any intention to limit operation or
+modification of the work as a means of enforcing, against the work's
+users, your or third parties' legal rights to forbid circumvention of
+technological measures.
+
+ 4. Conveying Verbatim Copies.
+
+ You may convey verbatim copies of the Program's source code as you
+receive it, in any medium, provided that you conspicuously and
+appropriately publish on each copy an appropriate copyright notice;
+keep intact all notices stating that this License and any
+non-permissive terms added in accord with section 7 apply to the code;
+keep intact all notices of the absence of any warranty; and give all
+recipients a copy of this License along with the Program.
+
+ You may charge any price or no price for each copy that you convey,
+and you may offer support or warranty protection for a fee.
+
+ 5. Conveying Modified Source Versions.
+
+ You may convey a work based on the Program, or the modifications to
+produce it from the Program, in the form of source code under the
+terms of section 4, provided that you also meet all of these conditions:
+
+ a) The work must carry prominent notices stating that you modified
+ it, and giving a relevant date.
+
+ b) The work must carry prominent notices stating that it is
+ released under this License and any conditions added under section
+ 7. This requirement modifies the requirement in section 4 to
+ "keep intact all notices".
+
+ c) You must license the entire work, as a whole, under this
+ License to anyone who comes into possession of a copy. This
+ License will therefore apply, along with any applicable section 7
+ additional terms, to the whole of the work, and all its parts,
+ regardless of how they are packaged. This License gives no
+ permission to license the work in any other way, but it does not
+ invalidate such permission if you have separately received it.
+
+ d) If the work has interactive user interfaces, each must display
+ Appropriate Legal Notices; however, if the Program has interactive
+ interfaces that do not display Appropriate Legal Notices, your
+ work need not make them do so.
+
+ A compilation of a covered work with other separate and independent
+works, which are not by their nature extensions of the covered work,
+and which are not combined with it such as to form a larger program,
+in or on a volume of a storage or distribution medium, is called an
+"aggregate" if the compilation and its resulting copyright are not
+used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit. Inclusion of a covered work
+in an aggregate does not cause this License to apply to the other
+parts of the aggregate.
+
+ 6. Conveying Non-Source Forms.
+
+ You may convey a covered work in object code form under the terms
+of sections 4 and 5, provided that you also convey the
+machine-readable Corresponding Source under the terms of this License,
+in one of these ways:
+
+ a) Convey the object code in, or embodied in, a physical product
+ (including a physical distribution medium), accompanied by the
+ Corresponding Source fixed on a durable physical medium
+ customarily used for software interchange.
+
+ b) Convey the object code in, or embodied in, a physical product
+ (including a physical distribution medium), accompanied by a
+ written offer, valid for at least three years and valid for as
+ long as you offer spare parts or customer support for that product
+ model, to give anyone who possesses the object code either (1) a
+ copy of the Corresponding Source for all the software in the
+ product that is covered by this License, on a durable physical
+ medium customarily used for software interchange, for a price no
+ more than your reasonable cost of physically performing this
+ conveying of source, or (2) access to copy the
+ Corresponding Source from a network server at no charge.
+
+ c) Convey individual copies of the object code with a copy of the
+ written offer to provide the Corresponding Source. This
+ alternative is allowed only occasionally and noncommercially, and
+ only if you received the object code with such an offer, in accord
+ with subsection 6b.
+
+ d) Convey the object code by offering access from a designated
+ place (gratis or for a charge), and offer equivalent access to the
+ Corresponding Source in the same way through the same place at no
+ further charge. You need not require recipients to copy the
+ Corresponding Source along with the object code. If the place to
+ copy the object code is a network server, the Corresponding Source
+ may be on a different server (operated by you or a third party)
+ that supports equivalent copying facilities, provided you maintain
+ clear directions next to the object code saying where to find the
+ Corresponding Source. Regardless of what server hosts the
+ Corresponding Source, you remain obligated to ensure that it is
+ available for as long as needed to satisfy these requirements.
+
+ e) Convey the object code using peer-to-peer transmission, provided
+ you inform other peers where the object code and Corresponding
+ Source of the work are being offered to the general public at no
+ charge under subsection 6d.
+
+ A separable portion of the object code, whose source code is excluded
+from the Corresponding Source as a System Library, need not be
+included in conveying the object code work.
+
+ A "User Product" is either (1) a "consumer product", which means any
+tangible personal property which is normally used for personal, family,
+or household purposes, or (2) anything designed or sold for incorporation
+into a dwelling. In determining whether a product is a consumer product,
+doubtful cases shall be resolved in favor of coverage. For a particular
+product received by a particular user, "normally used" refers to a
+typical or common use of that class of product, regardless of the status
+of the particular user or of the way in which the particular user
+actually uses, or expects or is expected to use, the product. A product
+is a consumer product regardless of whether the product has substantial
+commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+ "Installation Information" for a User Product means any methods,
+procedures, authorization keys, or other information required to install
+and execute modified versions of a covered work in that User Product from
+a modified version of its Corresponding Source. The information must
+suffice to ensure that the continued functioning of the modified object
+code is in no case prevented or interfered with solely because
+modification has been made.
+
+ If you convey an object code work under this section in, or with, or
+specifically for use in, a User Product, and the conveying occurs as
+part of a transaction in which the right of possession and use of the
+User Product is transferred to the recipient in perpetuity or for a
+fixed term (regardless of how the transaction is characterized), the
+Corresponding Source conveyed under this section must be accompanied
+by the Installation Information. But this requirement does not apply
+if neither you nor any third party retains the ability to install
+modified object code on the User Product (for example, the work has
+been installed in ROM).
+
+ The requirement to provide Installation Information does not include a
+requirement to continue to provide support service, warranty, or updates
+for a work that has been modified or installed by the recipient, or for
+the User Product in which it has been modified or installed. Access to a
+network may be denied when the modification itself materially and
+adversely affects the operation of the network or violates the rules and
+protocols for communication across the network.
+
+ Corresponding Source conveyed, and Installation Information provided,
+in accord with this section must be in a format that is publicly
+documented (and with an implementation available to the public in
+source code form), and must require no special password or key for
+unpacking, reading or copying.
+
+ 7. Additional Terms.
+
+ "Additional permissions" are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions.
+Additional permissions that are applicable to the entire Program shall
+be treated as though they were included in this License, to the extent
+that they are valid under applicable law. If additional permissions
+apply only to part of the Program, that part may be used separately
+under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+ When you convey a copy of a covered work, you may at your option
+remove any additional permissions from that copy, or from any part of
+it. (Additional permissions may be written to require their own
+removal in certain cases when you modify the work.) You may place
+additional permissions on material, added by you to a covered work,
+for which you have or can give appropriate copyright permission.
+
+ Notwithstanding any other provision of this License, for material you
+add to a covered work, you may (if authorized by the copyright holders of
+that material) supplement the terms of this License with terms:
+
+ a) Disclaiming warranty or limiting liability differently from the
+ terms of sections 15 and 16 of this License; or
+
+ b) Requiring preservation of specified reasonable legal notices or
+ author attributions in that material or in the Appropriate Legal
+ Notices displayed by works containing it; or
+
+ c) Prohibiting misrepresentation of the origin of that material, or
+ requiring that modified versions of such material be marked in
+ reasonable ways as different from the original version; or
+
+ d) Limiting the use for publicity purposes of names of licensors or
+ authors of the material; or
+
+ e) Declining to grant rights under trademark law for use of some
+ trade names, trademarks, or service marks; or
+
+ f) Requiring indemnification of licensors and authors of that
+ material by anyone who conveys the material (or modified versions of
+ it) with contractual assumptions of liability to the recipient, for
+ any liability that these contractual assumptions directly impose on
+ those licensors and authors.
+
+ All other non-permissive additional terms are considered "further
+restrictions" within the meaning of section 10. If the Program as you
+received it, or any part of it, contains a notice stating that it is
+governed by this License along with a term that is a further
+restriction, you may remove that term. If a license document contains
+a further restriction but permits relicensing or conveying under this
+License, you may add to a covered work material governed by the terms
+of that license document, provided that the further restriction does
+not survive such relicensing or conveying.
+
+ If you add terms to a covered work in accord with this section, you
+must place, in the relevant source files, a statement of the
+additional terms that apply to those files, or a notice indicating
+where to find the applicable terms.
+
+ Additional terms, permissive or non-permissive, may be stated in the
+form of a separately written license, or stated as exceptions;
+the above requirements apply either way.
+
+ 8. Termination.
+
+ You may not propagate or modify a covered work except as expressly
+provided under this License. Any attempt otherwise to propagate or
+modify it is void, and will automatically terminate your rights under
+this License (including any patent licenses granted under the third
+paragraph of section 11).
+
+ However, if you cease all violation of this License, then your
+license from a particular copyright holder is reinstated (a)
+provisionally, unless and until the copyright holder explicitly and
+finally terminates your license, and (b) permanently, if the copyright
+holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+ Moreover, your license from a particular copyright holder is
+reinstated permanently if the copyright holder notifies you of the
+violation by some reasonable means, this is the first time you have
+received notice of violation of this License (for any work) from that
+copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+ Termination of your rights under this section does not terminate the
+licenses of parties who have received copies or rights from you under
+this License. If your rights have been terminated and not permanently
+reinstated, you do not qualify to receive new licenses for the same
+material under section 10.
+
+ 9. Acceptance Not Required for Having Copies.
+
+ You are not required to accept this License in order to receive or
+run a copy of the Program. Ancillary propagation of a covered work
+occurring solely as a consequence of using peer-to-peer transmission
+to receive a copy likewise does not require acceptance. However,
+nothing other than this License grants you permission to propagate or
+modify any covered work. These actions infringe copyright if you do
+not accept this License. Therefore, by modifying or propagating a
+covered work, you indicate your acceptance of this License to do so.
+
+ 10. Automatic Licensing of Downstream Recipients.
+
+ Each time you convey a covered work, the recipient automatically
+receives a license from the original licensors, to run, modify and
+propagate that work, subject to this License. You are not responsible
+for enforcing compliance by third parties with this License.
+
+ An "entity transaction" is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an
+organization, or merging organizations. If propagation of a covered
+work results from an entity transaction, each party to that
+transaction who receives a copy of the work also receives whatever
+licenses to the work the party's predecessor in interest had or could
+give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if
+the predecessor has it or can get it with reasonable efforts.
+
+ You may not impose any further restrictions on the exercise of the
+rights granted or affirmed under this License. For example, you may
+not impose a license fee, royalty, or other charge for exercise of
+rights granted under this License, and you may not initiate litigation
+(including a cross-claim or counterclaim in a lawsuit) alleging that
+any patent claim is infringed by making, using, selling, offering for
+sale, or importing the Program or any portion of it.
+
+ 11. Patents.
+
+ A "contributor" is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based. The
+work thus licensed is called the contributor's "contributor version".
+
+ A contributor's "essential patent claims" are all patent claims
+owned or controlled by the contributor, whether already acquired or
+hereafter acquired, that would be infringed by some manner, permitted
+by this License, of making, using, or selling its contributor version,
+but do not include claims that would be infringed only as a
+consequence of further modification of the contributor version. For
+purposes of this definition, "control" includes the right to grant
+patent sublicenses in a manner consistent with the requirements of
+this License.
+
+ Each contributor grants you a non-exclusive, worldwide, royalty-free
+patent license under the contributor's essential patent claims, to
+make, use, sell, offer for sale, import and otherwise run, modify and
+propagate the contents of its contributor version.
+
+ In the following three paragraphs, a "patent license" is any express
+agreement or commitment, however denominated, not to enforce a patent
+(such as an express permission to practice a patent or covenant not to
+sue for patent infringement). To "grant" such a patent license to a
+party means to make such an agreement or commitment not to enforce a
+patent against the party.
+
+ If you convey a covered work, knowingly relying on a patent license,
+and the Corresponding Source of the work is not available for anyone
+to copy, free of charge and under the terms of this License, through a
+publicly available network server or other readily accessible means,
+then you must either (1) cause the Corresponding Source to be so
+available, or (2) arrange to deprive yourself of the benefit of the
+patent license for this particular work, or (3) arrange, in a manner
+consistent with the requirements of this License, to extend the patent
+license to downstream recipients. "Knowingly relying" means you have
+actual knowledge that, but for the patent license, your conveying the
+covered work in a country, or your recipient's use of the covered work
+in a country, would infringe one or more identifiable patents in that
+country that you have reason to believe are valid.
+
+ If, pursuant to or in connection with a single transaction or
+arrangement, you convey, or propagate by procuring conveyance of, a
+covered work, and grant a patent license to some of the parties
+receiving the covered work authorizing them to use, propagate, modify
+or convey a specific copy of the covered work, then the patent license
+you grant is automatically extended to all recipients of the covered
+work and works based on it.
+
+ A patent license is "discriminatory" if it does not include within
+the scope of its coverage, prohibits the exercise of, or is
+conditioned on the non-exercise of one or more of the rights that are
+specifically granted under this License. You may not convey a covered
+work if you are a party to an arrangement with a third party that is
+in the business of distributing software, under which you make payment
+to the third party based on the extent of your activity of conveying
+the work, and under which the third party grants, to any of the
+parties who would receive the covered work from you, a discriminatory
+patent license (a) in connection with copies of the covered work
+conveyed by you (or copies made from those copies), or (b) primarily
+for and in connection with specific products or compilations that
+contain the covered work, unless you entered into that arrangement,
+or that patent license was granted, prior to 28 March 2007.
+
+ Nothing in this License shall be construed as excluding or limiting
+any implied license or other defenses to infringement that may
+otherwise be available to you under applicable patent law.
+
+ 12. No Surrender of Others' Freedom.
+
+ If conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot convey a
+covered work so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you may
+not convey it at all. For example, if you agree to terms that obligate you
+to collect a royalty for further conveying from those to whom you convey
+the Program, the only way you could satisfy both those terms and this
+License would be to refrain entirely from conveying the Program.
+
+ 13. Use with the GNU Affero General Public License.
+
+ Notwithstanding any other provision of this License, you have
+permission to link or combine any covered work with a work licensed
+under version 3 of the GNU Affero General Public License into a single
+combined work, and to convey the resulting work. The terms of this
+License will continue to apply to the part which is the covered work,
+but the special requirements of the GNU Affero General Public License,
+section 13, concerning interaction through a network will apply to the
+combination as such.
+
+ 14. Revised Versions of this License.
+
+ The Free Software Foundation may publish revised and/or new versions of
+the GNU General Public License from time to time. Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+ Each version is given a distinguishing version number. If the
+Program specifies that a certain numbered version of the GNU General
+Public License "or any later version" applies to it, you have the
+option of following the terms and conditions either of that numbered
+version or of any later version published by the Free Software
+Foundation. If the Program does not specify a version number of the
+GNU General Public License, you may choose any version ever published
+by the Free Software Foundation.
+
+ If the Program specifies that a proxy can decide which future
+versions of the GNU General Public License can be used, that proxy's
+public statement of acceptance of a version permanently authorizes you
+to choose that version for the Program.
+
+ Later license versions may give you additional or different
+permissions. However, no additional obligations are imposed on any
+author or copyright holder as a result of your choosing to follow a
+later version.
+
+ 15. Disclaimer of Warranty.
+
+ THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
+APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
+HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
+OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
+THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
+IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
+ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+ 16. Limitation of Liability.
+
+ IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
+THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
+GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
+USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
+DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
+PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
+EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGES.
+
+ 17. Interpretation of Sections 15 and 16.
+
+ If the disclaimer of warranty and limitation of liability provided
+above cannot be given local legal effect according to their terms,
+reviewing courts shall apply local law that most closely approximates
+an absolute waiver of all civil liability in connection with the
+Program, unless a warranty or assumption of liability accompanies a
+copy of the Program in return for a fee.
+
+ END OF TERMS AND CONDITIONS
diff --git a/filter/xbrz.cpp b/filter/xbrz.cpp
new file mode 100644
index 00000000..5fbf0701
--- /dev/null
+++ b/filter/xbrz.cpp
@@ -0,0 +1,1184 @@
+// ****************************************************************************
+// * This file is part of the HqMAME project. It is distributed under *
+// * GNU General Public License: http://www.gnu.org/licenses/gpl.html *
+// * Copyright (C) Zenju (zenju AT gmx DOT de) - All Rights Reserved *
+// * *
+// * Additionally and as a special exception, the author gives permission *
+// * to link the code of this program with the MAME library (or with modified *
+// * versions of MAME that use the same license as MAME), and distribute *
+// * linked combinations including the two. You must obey the GNU General *
+// * Public License in all respects for all of the code used other than MAME. *
+// * If you modify this file, you may extend this exception to your version *
+// * of the file, but you are not obligated to do so. If you do not wish to *
+// * do so, delete this exception statement from your version. *
+// ****************************************************************************
+
+#include "xbrz.h"
+#include
+#include
+
+namespace
+{
+template inline
+unsigned char getByte(uint32_t val) { return static_cast((val >> (8 * N)) & 0xff); }
+
+inline unsigned char getRed (uint32_t val) { return getByte<2>(val); }
+inline unsigned char getGreen(uint32_t val) { return getByte<1>(val); }
+inline unsigned char getBlue (uint32_t val) { return getByte<0>(val); }
+
+template inline
+T abs(T value)
+{
+ //static_assert(std::is_signed::value, "");
+ return value < 0 ? -value : value;
+}
+
+const uint32_t redMask = 0xff0000;
+const uint32_t greenMask = 0x00ff00;
+const uint32_t blueMask = 0x0000ff;
+
+template inline
+void alphaBlend(uint32_t& dst, uint32_t col) //blend color over destination with opacity N / M
+{
+ //static_assert(N < 256, "possible overflow of (col & redMask) * N");
+ //static_assert(M < 256, "possible overflow of (col & redMask ) * N + (dst & redMask ) * (M - N)");
+ //static_assert(0 < N && N < M, "");
+ dst = (redMask & ((col & redMask ) * N + (dst & redMask ) * (M - N)) / M) | //this works because 8 upper bits are free
+ (greenMask & ((col & greenMask) * N + (dst & greenMask) * (M - N)) / M) |
+ (blueMask & ((col & blueMask ) * N + (dst & blueMask ) * (M - N)) / M);
+}
+
+
+//inline
+//double fastSqrt(double n)
+//{
+// __asm //speeds up xBRZ by about 9% compared to std::sqrt
+// {
+// fld n
+// fsqrt
+// }
+//}
+//
+
+
+inline
+uint32_t alphaBlend2(uint32_t pix1, uint32_t pix2, double alpha)
+{
+ return (redMask & static_cast((pix1 & redMask ) * alpha + (pix2 & redMask ) * (1 - alpha))) |
+ (greenMask & static_cast((pix1 & greenMask) * alpha + (pix2 & greenMask) * (1 - alpha))) |
+ (blueMask & static_cast((pix1 & blueMask ) * alpha + (pix2 & blueMask ) * (1 - alpha)));
+}
+
+
+uint32_t* byteAdvance( uint32_t* ptr, int bytes) { return reinterpret_cast< uint32_t*>(reinterpret_cast< char*>(ptr) + bytes); }
+const uint32_t* byteAdvance(const uint32_t* ptr, int bytes) { return reinterpret_cast(reinterpret_cast(ptr) + bytes); }
+
+
+//fill block with the given color
+inline
+void fillBlock(uint32_t* trg, int pitch, uint32_t col, int blockWidth, int blockHeight)
+{
+ //for (int y = 0; y < blockHeight; ++y, trg = byteAdvance(trg, pitch))
+ // std::fill(trg, trg + blockWidth, col);
+
+ for (int y = 0; y < blockHeight; ++y, trg = byteAdvance(trg, pitch))
+ for (int x = 0; x < blockWidth; ++x)
+ trg[x] = col;
+}
+
+inline
+void fillBlock(uint32_t* trg, int pitch, uint32_t col, int n) { fillBlock(trg, pitch, col, n, n); }
+
+
+#ifdef _MSC_VER
+#define FORCE_INLINE __forceinline
+#elif defined __GNUC__
+#define FORCE_INLINE __attribute__((always_inline)) inline
+#else
+#define FORCE_INLINE inline
+#endif
+
+
+enum RotationDegree //clock-wise
+{
+ ROT_0,
+ ROT_90,
+ ROT_180,
+ ROT_270
+};
+
+//calculate input matrix coordinates after rotation at compile time
+template
+struct MatrixRotation;
+
+template
+struct MatrixRotation
+{
+ static const size_t I_old = I;
+ static const size_t J_old = J;
+};
+
+template //(i, j) = (row, col) indices, N = size of (square) matrix
+struct MatrixRotation
+{
+ static const size_t I_old = N - 1 - MatrixRotation(rotDeg - 1), I, J, N>::J_old; //old coordinates before rotation!
+ static const size_t J_old = MatrixRotation(rotDeg - 1), I, J, N>::I_old; //
+};
+
+
+template
+class OutputMatrix
+{
+public:
+ OutputMatrix(uint32_t* out, int outWidth) : //access matrix area, top-left at position "out" for image with given width
+ out_(out),
+ outWidth_(outWidth) {}
+
+ template
+ uint32_t& ref() const
+ {
+ static const size_t I_old = MatrixRotation::I_old;
+ static const size_t J_old = MatrixRotation::J_old;
+ return *(out_ + J_old + I_old * outWidth_);
+ }
+
+private:
+ uint32_t* out_;
+ const int outWidth_;
+};
+
+
+template inline
+T square(T value) { return value * value; }
+
+
+/*
+inline
+void rgbtoLuv(uint32_t c, double& L, double& u, double& v)
+{
+ //http://www.easyrgb.com/index.php?X=MATH&H=02#text2
+ double r = getRed (c) / 255.0;
+ double g = getGreen(c) / 255.0;
+ double b = getBlue (c) / 255.0;
+
+ if ( r > 0.04045 )
+ r = std::pow(( ( r + 0.055 ) / 1.055 ) , 2.4);
+ else
+ r /= 12.92;
+ if ( g > 0.04045 )
+ g = std::pow(( ( g + 0.055 ) / 1.055 ) , 2.4);
+ else
+ g /= 12.92;
+ if ( b > 0.04045 )
+ b = std::pow(( ( b + 0.055 ) / 1.055 ) , 2.4);
+ else
+ b /= 12.92;
+
+ r *= 100;
+ g *= 100;
+ b *= 100;
+
+ double x = 0.4124564 * r + 0.3575761 * g + 0.1804375 * b;
+ double y = 0.2126729 * r + 0.7151522 * g + 0.0721750 * b;
+ double z = 0.0193339 * r + 0.1191920 * g + 0.9503041 * b;
+ //---------------------
+ double var_U = 4 * x / ( x + 15 * y + 3 * z );
+ double var_V = 9 * y / ( x + 15 * y + 3 * z );
+ double var_Y = y / 100;
+
+ if ( var_Y > 0.008856 ) var_Y = std::pow(var_Y , 1.0/3 );
+ else var_Y = 7.787 * var_Y + 16.0 / 116;
+
+ const double ref_X = 95.047; //Observer= 2°, Illuminant= D65
+ const double ref_Y = 100.000;
+ const double ref_Z = 108.883;
+
+ const double ref_U = ( 4 * ref_X ) / ( ref_X + ( 15 * ref_Y ) + ( 3 * ref_Z ) );
+ const double ref_V = ( 9 * ref_Y ) / ( ref_X + ( 15 * ref_Y ) + ( 3 * ref_Z ) );
+
+ L = ( 116 * var_Y ) - 16;
+ u = 13 * L * ( var_U - ref_U );
+ v = 13 * L * ( var_V - ref_V );
+}
+*/
+
+inline
+void rgbtoLab(uint32_t c, unsigned char& L, signed char& A, signed char& B)
+{
+ //code: http://www.easyrgb.com/index.php?X=MATH
+ //test: http://www.workwithcolor.com/color-converter-01.htm
+ //------RGB to XYZ------
+ double r = getRed (c) / 255.0;
+ double g = getGreen(c) / 255.0;
+ double b = getBlue (c) / 255.0;
+
+ r = r > 0.04045 ? std::pow(( r + 0.055 ) / 1.055, 2.4) : r / 12.92;
+ r = g > 0.04045 ? std::pow(( g + 0.055 ) / 1.055, 2.4) : g / 12.92;
+ r = b > 0.04045 ? std::pow(( b + 0.055 ) / 1.055, 2.4) : b / 12.92;
+
+ r *= 100;
+ g *= 100;
+ b *= 100;
+
+ double x = 0.4124564 * r + 0.3575761 * g + 0.1804375 * b;
+ double y = 0.2126729 * r + 0.7151522 * g + 0.0721750 * b;
+ double z = 0.0193339 * r + 0.1191920 * g + 0.9503041 * b;
+ //------XYZ to Lab------
+ const double refX = 95.047; //
+ const double refY = 100.000; //Observer= 2°, Illuminant= D65
+ const double refZ = 108.883; //
+ double var_X = x / refX;
+ double var_Y = y / refY;
+ double var_Z = z / refZ;
+
+ var_X = var_X > 0.008856 ? std::pow(var_X, 1.0 / 3) : 7.787 * var_X + 4.0 / 29;
+ var_Y = var_Y > 0.008856 ? std::pow(var_Y, 1.0 / 3) : 7.787 * var_Y + 4.0 / 29;
+ var_Z = var_Z > 0.008856 ? std::pow(var_Z, 1.0 / 3) : 7.787 * var_Z + 4.0 / 29;
+
+ L = static_cast(116 * var_Y - 16);
+ A = static_cast< signed char>(500 * (var_X - var_Y));
+ B = static_cast< signed char>(200 * (var_Y - var_Z));
+};
+
+
+inline
+double distLAB(uint32_t pix1, uint32_t pix2)
+{
+ unsigned char L1 = 0; //[0, 100]
+ signed char a1 = 0; //[-128, 127]
+ signed char b1 = 0; //[-128, 127]
+ rgbtoLab(pix1, L1, a1, b1);
+
+ unsigned char L2 = 0;
+ signed char a2 = 0;
+ signed char b2 = 0;
+ rgbtoLab(pix2, L2, a2, b2);
+
+ //-----------------------------
+ //http://www.easyrgb.com/index.php?X=DELT
+
+ //Delta E/CIE76
+ return std::sqrt(square(1.0 * L1 - L2) +
+ square(1.0 * a1 - a2) +
+ square(1.0 * b1 - b2));
+}
+
+
+/*
+inline
+void rgbtoHsl(uint32_t c, double& h, double& s, double& l)
+{
+ //http://www.easyrgb.com/index.php?X=MATH&H=18#text18
+ const int r = getRed (c);
+ const int g = getGreen(c);
+ const int b = getBlue (c);
+
+ const int varMin = numeric::min(r, g, b);
+ const int varMax = numeric::max(r, g, b);
+ const int delMax = varMax - varMin;
+
+ l = (varMax + varMin) / 2.0 / 255.0;
+
+ if (delMax == 0) //gray, no chroma...
+ {
+ h = 0;
+ s = 0;
+ }
+ else
+ {
+ s = l < 0.5 ?
+ delMax / (1.0 * varMax + varMin) :
+ delMax / (2.0 * 255 - varMax - varMin);
+
+ double delR = ((varMax - r) / 6.0 + delMax / 2.0) / delMax;
+ double delG = ((varMax - g) / 6.0 + delMax / 2.0) / delMax;
+ double delB = ((varMax - b) / 6.0 + delMax / 2.0) / delMax;
+
+ if (r == varMax)
+ h = delB - delG;
+ else if (g == varMax)
+ h = 1 / 3.0 + delR - delB;
+ else if (b == varMax)
+ h = 2 / 3.0 + delG - delR;
+
+ if (h < 0)
+ h += 1;
+ if (h > 1)
+ h -= 1;
+ }
+}
+
+inline
+double distHSL(uint32_t pix1, uint32_t pix2, double lightningWeight)
+{
+ double h1 = 0;
+ double s1 = 0;
+ double l1 = 0;
+ rgbtoHsl(pix1, h1, s1, l1);
+ double h2 = 0;
+ double s2 = 0;
+ double l2 = 0;
+ rgbtoHsl(pix2, h2, s2, l2);
+
+ //HSL is in cylindric coordinatates where L represents height, S radius, H angle,
+ //however we interpret the cylinder as a bi-conic solid with top/bottom radius 0, middle radius 1
+ assert(0 <= h1 && h1 <= 1);
+ assert(0 <= h2 && h2 <= 1);
+
+ double r1 = l1 < 0.5 ?
+ l1 * 2 :
+ 2 - l1 * 2;
+
+ double x1 = r1 * s1 * std::cos(h1 * 2 * numeric::pi);
+ double y1 = r1 * s1 * std::sin(h1 * 2 * numeric::pi);
+ double z1 = l1;
+
+ double r2 = l2 < 0.5 ?
+ l2 * 2 :
+ 2 - l2 * 2;
+
+ double x2 = r2 * s2 * std::cos(h2 * 2 * numeric::pi);
+ double y2 = r2 * s2 * std::sin(h2 * 2 * numeric::pi);
+ double z2 = l2;
+
+ return 255 * std::sqrt(square(x1 - x2) + square(y1 - y2) + square(lightningWeight * (z1 - z2)));
+}
+*/
+
+
+inline
+double distRGB(uint32_t pix1, uint32_t pix2)
+{
+ const double r_diff = static_cast(getRed (pix1)) - getRed (pix2);
+ const double g_diff = static_cast(getGreen(pix1)) - getGreen(pix2);
+ const double b_diff = static_cast(getBlue (pix1)) - getBlue (pix2);
+
+ //euklidean RGB distance
+ return std::sqrt(square(r_diff) + square(g_diff) + square(b_diff));
+}
+
+
+inline
+double distNonLinearRGB(uint32_t pix1, uint32_t pix2)
+{
+ //non-linear rgb: http://www.compuphase.com/cmetric.htm
+ const double r_diff = static_cast(getRed (pix1)) - getRed (pix2);
+ const double g_diff = static_cast(getGreen(pix1)) - getGreen(pix2);
+ const double b_diff = static_cast(getBlue (pix1)) - getBlue (pix2);
+
+ const double r_avg = (static_cast(getRed(pix1)) + getRed(pix2)) / 2;
+ return std::sqrt((2 + r_avg / 255) * square(r_diff) + 4 * square(g_diff) + (2 + (255 - r_avg) / 255) * square(b_diff));
+}
+
+
+inline
+double distYCbCr(uint32_t pix1, uint32_t pix2, double lumaWeight)
+{
+ //http://en.wikipedia.org/wiki/YCbCr#ITU-R_BT.601_conversion
+ //YCbCr conversion is a matrix multiplication => take advantage of linearity by subtracting first!
+ const int r_diff = static_cast(getRed (pix1)) - getRed (pix2); //we may delay division by 255 to after matrix multiplication
+ const int g_diff = static_cast(getGreen(pix1)) - getGreen(pix2); //
+ const int b_diff = static_cast(getBlue (pix1)) - getBlue (pix2); //substraction for int is noticeable faster than for double!
+
+ const double k_b = 0.0722; //ITU-R BT.709 conversion
+ const double k_r = 0.2126; //
+ const double k_g = 1 - k_b - k_r;
+
+ const double scale_b = 0.5 / (1 - k_b);
+ const double scale_r = 0.5 / (1 - k_r);
+
+ const double y = k_r * r_diff + k_g * g_diff + k_b * b_diff; //[!], analog YCbCr!
+ const double c_b = scale_b * (b_diff - y);
+ const double c_r = scale_r * (r_diff - y);
+
+ //we skip division by 255 to have similar range like other distance functions
+ return std::sqrt(square(lumaWeight * y) + square(c_b) + square(c_r));
+}
+
+
+inline
+double distYUV(uint32_t pix1, uint32_t pix2, double luminanceWeight)
+{
+ //perf: it's not worthwhile to buffer the YUV-conversion, the direct code is faster by ~ 6%
+ //since RGB -> YUV conversion is essentially a matrix multiplication, we can calculate the RGB diff before the conversion (distributive property)
+ const double r_diff = static_cast(getRed (pix1)) - getRed (pix2);
+ const double g_diff = static_cast(getGreen(pix1)) - getGreen(pix2);
+ const double b_diff = static_cast(getBlue (pix1)) - getBlue (pix2);
+
+ //http://en.wikipedia.org/wiki/YUV#Conversion_to.2Ffrom_RGB
+ const double w_b = 0.114;
+ const double w_r = 0.299;
+ const double w_g = 1 - w_r - w_b;
+
+ const double u_max = 0.436;
+ const double v_max = 0.615;
+
+ const double scale_u = u_max / (1 - w_b);
+ const double scale_v = v_max / (1 - w_r);
+
+ double y = w_r * r_diff + w_g * g_diff + w_b * b_diff;//value range: 255 * [-1, 1]
+ double u = scale_u * (b_diff - y); //value range: 255 * 2 * u_max * [-1, 1]
+ double v = scale_v * (r_diff - y); //value range: 255 * 2 * v_max * [-1, 1]
+
+#ifndef NDEBUG
+ const double eps = 0.5;
+#endif
+ assert(std::abs(y) <= 255 + eps);
+ assert(std::abs(u) <= 255 * 2 * u_max + eps);
+ assert(std::abs(v) <= 255 * 2 * v_max + eps);
+
+ return std::sqrt(square(luminanceWeight * y) + square(u) + square(v));
+}
+
+
+inline
+double colorDist(uint32_t pix1, uint32_t pix2, double luminanceWeight)
+{
+ if (pix1 == pix2) //about 8% perf boost
+ return 0;
+
+ //return distHSL(pix1, pix2, luminanceWeight);
+ //return distRGB(pix1, pix2);
+ //return distLAB(pix1, pix2);
+ //return distNonLinearRGB(pix1, pix2);
+ //return distYUV(pix1, pix2, luminanceWeight);
+
+ return distYCbCr(pix1, pix2, luminanceWeight);
+}
+
+
+enum BlendType
+{
+ BLEND_NONE = 0,
+ BLEND_NORMAL, //a normal indication to blend
+ BLEND_DOMINANT, //a strong indication to blend
+ //attention: BlendType must fit into the value range of 2 bit!!!
+};
+
+struct BlendResult
+{
+ BlendType
+ /**/blend_f, blend_g,
+ /**/blend_j, blend_k;
+};
+
+
+struct Kernel_4x4 //kernel for preprocessing step
+{
+ uint32_t
+ /**/a, b, c, d,
+ /**/e, f, g, h,
+ /**/i, j, k, l,
+ /**/m, n, o, p;
+};
+
+#define dist(col1, col2) colorDist(col1, col2, cfg.luminanceWeight_)
+/*
+input kernel area naming convention:
+-----------------
+| A | B | C | D |
+----|---|---|---|
+| E | F | G | H | //evalute the four corners between F, G, J, K
+----|---|---|---| //input pixel is at position F
+| I | J | K | L |
+----|---|---|---|
+| M | N | O | P |
+-----------------
+*/
+FORCE_INLINE //detect blend direction
+BlendResult preProcessCorners(const Kernel_4x4& ker, const xbrz::ScalerCfg& cfg) //result: F, G, J, K corners of "GradientType"
+{
+ BlendResult result = {};
+
+ if ((ker.f == ker.g &&
+ ker.j == ker.k) ||
+ (ker.f == ker.j &&
+ ker.g == ker.k))
+ return result;
+
+ //auto dist = [&](uint32_t col1, uint32_t col2) { return colorDist(col1, col2, cfg.luminanceWeight_); };
+
+ const int weight = 4;
+ double jg = dist(ker.i, ker.f) + dist(ker.f, ker.c) + dist(ker.n, ker.k) + dist(ker.k, ker.h) + weight * dist(ker.j, ker.g);
+ double fk = dist(ker.e, ker.j) + dist(ker.j, ker.o) + dist(ker.b, ker.g) + dist(ker.g, ker.l) + weight * dist(ker.f, ker.k);
+
+ if (jg < fk) //test sample: 70% of values max(jg, fk) / min(jg, fk) are between 1.1 and 3.7 with median being 1.8
+ {
+ const bool dominantGradient = cfg.dominantDirectionThreshold * jg < fk;
+ if (ker.f != ker.g && ker.f != ker.j)
+ result.blend_f = dominantGradient ? BLEND_DOMINANT : BLEND_NORMAL;
+
+ if (ker.k != ker.j && ker.k != ker.g)
+ result.blend_k = dominantGradient ? BLEND_DOMINANT : BLEND_NORMAL;
+ }
+ else if (fk < jg)
+ {
+ const bool dominantGradient = cfg.dominantDirectionThreshold * fk < jg;
+ if (ker.j != ker.f && ker.j != ker.k)
+ result.blend_j = dominantGradient ? BLEND_DOMINANT : BLEND_NORMAL;
+
+ if (ker.g != ker.f && ker.g != ker.k)
+ result.blend_g = dominantGradient ? BLEND_DOMINANT : BLEND_NORMAL;
+ }
+ return result;
+}
+
+struct Kernel_3x3
+{
+ uint32_t
+ /**/a, b, c,
+ /**/d, e, f,
+ /**/g, h, i;
+};
+
+#define DEF_GETTER(x) template uint32_t inline get_##x(const Kernel_3x3& ker) { return ker.x; }
+//we cannot and NEED NOT write "ker.##x" since ## concatenates preprocessor tokens but "." is not a token
+DEF_GETTER(a) DEF_GETTER(b) DEF_GETTER(c)
+DEF_GETTER(d) DEF_GETTER(e) DEF_GETTER(f)
+DEF_GETTER(g) DEF_GETTER(h) DEF_GETTER(i)
+#undef DEF_GETTER
+
+#define DEF_GETTER(x, y) template <> inline uint32_t get_##x(const Kernel_3x3& ker) { return ker.y; }
+DEF_GETTER(a, g) DEF_GETTER(b, d) DEF_GETTER(c, a)
+DEF_GETTER(d, h) DEF_GETTER(e, e) DEF_GETTER(f, b)
+DEF_GETTER(g, i) DEF_GETTER(h, f) DEF_GETTER(i, c)
+#undef DEF_GETTER
+
+#define DEF_GETTER(x, y) template <> inline uint32_t get_##x(const Kernel_3x3& ker) { return ker.y; }
+DEF_GETTER(a, i) DEF_GETTER(b, h) DEF_GETTER(c, g)
+DEF_GETTER(d, f) DEF_GETTER(e, e) DEF_GETTER(f, d)
+DEF_GETTER(g, c) DEF_GETTER(h, b) DEF_GETTER(i, a)
+#undef DEF_GETTER
+
+#define DEF_GETTER(x, y) template <> inline uint32_t get_##x(const Kernel_3x3& ker) { return ker.y; }
+DEF_GETTER(a, c) DEF_GETTER(b, f) DEF_GETTER(c, i)
+DEF_GETTER(d, b) DEF_GETTER(e, e) DEF_GETTER(f, h)
+DEF_GETTER(g, a) DEF_GETTER(h, d) DEF_GETTER(i, g)
+#undef DEF_GETTER
+
+
+//compress four blend types into a single byte
+inline BlendType getTopL (unsigned char b) { return static_cast(0x3 & b); }
+inline BlendType getTopR (unsigned char b) { return static_cast(0x3 & (b >> 2)); }
+inline BlendType getBottomR(unsigned char b) { return static_cast(0x3 & (b >> 4)); }
+inline BlendType getBottomL(unsigned char b) { return static_cast(0x3 & (b >> 6)); }
+
+inline void setTopL (unsigned char& b, BlendType bt) { b |= bt; } //buffer is assumed to be initialized before preprocessing!
+inline void setTopR (unsigned char& b, BlendType bt) { b |= (bt << 2); }
+inline void setBottomR(unsigned char& b, BlendType bt) { b |= (bt << 4); }
+inline void setBottomL(unsigned char& b, BlendType bt) { b |= (bt << 6); }
+
+inline bool blendingNeeded(unsigned char b) { return b != 0; }
+
+template inline
+unsigned char rotateBlendInfo(unsigned char b) { return b; }
+template <> inline unsigned char rotateBlendInfo(unsigned char b) { return ((b << 2) | (b >> 6)) & 0xff; }
+template <> inline unsigned char rotateBlendInfo(unsigned char b) { return ((b << 4) | (b >> 4)) & 0xff; }
+template <> inline unsigned char rotateBlendInfo(unsigned char b) { return ((b << 6) | (b >> 2)) & 0xff; }
+
+
+#ifndef NDEBUG
+int debugPixelX = -1;
+int debugPixelY = 84;
+bool breakIntoDebugger = false;
+#endif
+
+#define eq(col1, col2) (colorDist(col1, col2, cfg.luminanceWeight_) < cfg.equalColorTolerance_)
+
+/*
+input kernel area naming convention:
+-------------
+| A | B | C |
+----|---|---|
+| D | E | F | //input pixel is at position E
+----|---|---|
+| G | H | I |
+-------------
+*/
+template
+FORCE_INLINE //perf: quite worth it!
+void scalePixel(const Kernel_3x3& ker,
+ uint32_t* target, int trgWidth,
+ unsigned char blendInfo, //result of preprocessing all four corners of pixel "e"
+ const xbrz::ScalerCfg& cfg)
+{
+#define a get_a(ker)
+#define b get_b(ker)
+#define c get_c(ker)
+#define d get_d(ker)
+#define e get_e(ker)
+#define f get_f(ker)
+#define g get_g(ker)
+#define h get_h(ker)
+#define i get_i(ker)
+
+#ifndef NDEBUG
+ if (breakIntoDebugger)
+ __debugbreak(); //__asm int 3;
+#endif
+
+ const unsigned char blend = rotateBlendInfo(blendInfo);
+
+ if (getBottomR(blend) >= BLEND_NORMAL)
+ {
+ //auto eq = [&](uint32_t col1, uint32_t col2) { return colorDist(col1, col2, cfg.luminanceWeight_) < cfg.equalColorTolerance_; };
+ //auto dist = [&](uint32_t col1, uint32_t col2) { return colorDist(col1, col2, cfg.luminanceWeight_); };
+
+ bool doLineBlend = true;
+ if (getBottomR(blend) >= BLEND_DOMINANT)
+ doLineBlend = true;
+ //make sure there is no second blending in an adjacent rotation for this pixel: handles insular pixels, mario eyes
+ else if (getTopR(blend) != BLEND_NONE && !eq(e, g)) //but support double-blending for 90° corners
+ doLineBlend = false;
+ else if(getBottomL(blend) != BLEND_NONE && !eq(e, c))
+ doLineBlend = false;
+ //no full blending for L-shapes; blend corner only (handles "mario mushroom eyes")
+ else if (eq(g, h) && eq(h , i) && eq(i, f) && eq(f, c) && !eq(e, i))
+ doLineBlend = false;
+ else
+ doLineBlend = true;
+
+ const uint32_t px = dist(e, f) <= dist(e, h) ? f : h; //choose most similar color
+
+ OutputMatrix out(target, trgWidth);
+
+ if (doLineBlend)
+ {
+ const double fg = dist(f, g); //test sample: 70% of values max(fg, hc) / min(fg, hc) are between 1.1 and 3.7 with median being 1.9
+ const double hc = dist(h, c); //
+
+ const bool haveShallowLine = cfg.steepDirectionThreshold * fg <= hc && e != g && d != g;
+ const bool haveSteepLine = cfg.steepDirectionThreshold * hc <= fg && e != c && b != c;
+
+ if (haveShallowLine)
+ {
+ if (haveSteepLine)
+ Scaler::blendLineSteepAndShallow(px, out);
+ else
+ Scaler::blendLineShallow(px, out);
+ }
+ else
+ {
+ if (haveSteepLine)
+ Scaler::blendLineSteep(px, out);
+ else
+ Scaler::blendLineDiagonal(px,out);
+ }
+ }
+ else
+ Scaler::blendCorner(px, out);
+ }
+
+#undef a
+#undef b
+#undef c
+#undef d
+#undef e
+#undef f
+#undef g
+#undef h
+#undef i
+}
+
+
+template //scaler policy: see "Scaler2x" reference implementation
+void scaleImage(const uint32_t* src, uint32_t* trg, int srcWidth, int srcHeight, const xbrz::ScalerCfg& cfg, int yFirst, int yLast)
+{
+ yFirst = std::max(yFirst, 0);
+ yLast = std::min(yLast, srcHeight);
+ if (yFirst >= yLast || srcWidth <= 0)
+ return;
+
+ const int trgWidth = srcWidth * Scaler::scale;
+
+ //"use" space at the end of the image as temporary buffer for "on the fly preprocessing": we even could use larger area of
+ //"sizeof(uint32_t) * srcWidth * (yLast - yFirst)" bytes without risk of accidental overwriting before accessing
+ const int bufferSize = srcWidth;
+ unsigned char* preProcBuffer = reinterpret_cast(trg + yLast * Scaler::scale * trgWidth) - bufferSize;
+ std::fill(preProcBuffer, preProcBuffer + bufferSize, 0);
+ //static_assert(BLEND_NONE == 0, "");
+
+ //initialize preprocessing buffer for first row: detect upper left and right corner blending
+ //this cannot be optimized for adjacent processing stripes; we must not allow for a memory race condition!
+ if (yFirst > 0)
+ {
+ const int y = yFirst - 1;
+
+ const uint32_t* s_m1 = src + srcWidth * std::max(y - 1, 0);
+ const uint32_t* s_0 = src + srcWidth * y; //center line
+ const uint32_t* s_p1 = src + srcWidth * std::min(y + 1, srcHeight - 1);
+ const uint32_t* s_p2 = src + srcWidth * std::min(y + 2, srcHeight - 1);
+
+ for (int x = 0; x < srcWidth; ++x)
+ {
+ const int x_m1 = std::max(x - 1, 0);
+ const int x_p1 = std::min(x + 1, srcWidth - 1);
+ const int x_p2 = std::min(x + 2, srcWidth - 1);
+
+ Kernel_4x4 ker = {}; //perf: initialization is negligable
+ ker.a = s_m1[x_m1]; //read sequentially from memory as far as possible
+ ker.b = s_m1[x];
+ ker.c = s_m1[x_p1];
+ ker.d = s_m1[x_p2];
+
+ ker.e = s_0[x_m1];
+ ker.f = s_0[x];
+ ker.g = s_0[x_p1];
+ ker.h = s_0[x_p2];
+
+ ker.i = s_p1[x_m1];
+ ker.j = s_p1[x];
+ ker.k = s_p1[x_p1];
+ ker.l = s_p1[x_p2];
+
+ ker.m = s_p2[x_m1];
+ ker.n = s_p2[x];
+ ker.o = s_p2[x_p1];
+ ker.p = s_p2[x_p2];
+
+ const BlendResult res = preProcessCorners(ker, cfg);
+ /*
+ preprocessing blend result:
+ ---------
+ | F | G | //evalute corner between F, G, J, K
+ ----|---| //input pixel is at position F
+ | J | K |
+ ---------
+ */
+ setTopR(preProcBuffer[x], res.blend_j);
+
+ if (x + 1 < srcWidth)
+ setTopL(preProcBuffer[x + 1], res.blend_k);
+ }
+ }
+ //------------------------------------------------------------------------------------
+
+ for (int y = yFirst; y < yLast; ++y)
+ {
+ uint32_t* out = trg + Scaler::scale * y * trgWidth; //consider MT "striped" access
+
+ const uint32_t* s_m1 = src + srcWidth * std::max(y - 1, 0);
+ const uint32_t* s_0 = src + srcWidth * y; //center line
+ const uint32_t* s_p1 = src + srcWidth * std::min(y + 1, srcHeight - 1);
+ const uint32_t* s_p2 = src + srcWidth * std::min(y + 2, srcHeight - 1);
+
+ unsigned char blend_xy1 = 0; //corner blending for current (x, y + 1) position
+
+ for (int x = 0; x < srcWidth; ++x, out += Scaler::scale)
+ {
+#ifndef NDEBUG
+ breakIntoDebugger = debugPixelX == x && debugPixelY == y;
+#endif
+ //all those bounds checks have only insignificant impact on performance!
+ const int x_m1 = std::max(x - 1, 0); //perf: prefer array indexing to additional pointers!
+ const int x_p1 = std::min(x + 1, srcWidth - 1);
+ const int x_p2 = std::min(x + 2, srcWidth - 1);
+
+ //evaluate the four corners on bottom-right of current pixel
+ unsigned char blend_xy = 0; //for current (x, y) position
+ {
+ Kernel_4x4 ker = {}; //perf: initialization is negligable
+ ker.a = s_m1[x_m1]; //read sequentially from memory as far as possible
+ ker.b = s_m1[x];
+ ker.c = s_m1[x_p1];
+ ker.d = s_m1[x_p2];
+
+ ker.e = s_0[x_m1];
+ ker.f = s_0[x];
+ ker.g = s_0[x_p1];
+ ker.h = s_0[x_p2];
+
+ ker.i = s_p1[x_m1];
+ ker.j = s_p1[x];
+ ker.k = s_p1[x_p1];
+ ker.l = s_p1[x_p2];
+
+ ker.m = s_p2[x_m1];
+ ker.n = s_p2[x];
+ ker.o = s_p2[x_p1];
+ ker.p = s_p2[x_p2];
+
+ const BlendResult res = preProcessCorners(ker, cfg);
+ /*
+ preprocessing blend result:
+ ---------
+ | F | G | //evalute corner between F, G, J, K
+ ----|---| //current input pixel is at position F
+ | J | K |
+ ---------
+ */
+ blend_xy = preProcBuffer[x];
+ setBottomR(blend_xy, res.blend_f); //all four corners of (x, y) have been determined at this point due to processing sequence!
+
+ setTopR(blend_xy1, res.blend_j); //set 2nd known corner for (x, y + 1)
+ preProcBuffer[x] = blend_xy1; //store on current buffer position for use on next row
+
+ blend_xy1 = 0;
+ setTopL(blend_xy1, res.blend_k); //set 1st known corner for (x + 1, y + 1) and buffer for use on next column
+
+ if (x + 1 < srcWidth) //set 3rd known corner for (x + 1, y)
+ setBottomL(preProcBuffer[x + 1], res.blend_g);
+ }
+
+ //fill block of size scale * scale with the given color
+ fillBlock(out, trgWidth * sizeof(uint32_t), s_0[x], Scaler::scale); //place *after* preprocessing step, to not overwrite the results while processing the the last pixel!
+
+ //blend four corners of current pixel
+ if (blendingNeeded(blend_xy)) //good 20% perf-improvement
+ {
+ Kernel_3x3 ker = {}; //perf: initialization is negligable
+
+ ker.a = s_m1[x_m1]; //read sequentially from memory as far as possible
+ ker.b = s_m1[x];
+ ker.c = s_m1[x_p1];
+
+ ker.d = s_0[x_m1];
+ ker.e = s_0[x];
+ ker.f = s_0[x_p1];
+
+ ker.g = s_p1[x_m1];
+ ker.h = s_p1[x];
+ ker.i = s_p1[x_p1];
+
+ scalePixel(ker, out, trgWidth, blend_xy, cfg);
+ scalePixel(ker, out, trgWidth, blend_xy, cfg);
+ scalePixel(ker, out, trgWidth, blend_xy, cfg);
+ scalePixel(ker, out, trgWidth, blend_xy, cfg);
+ }
+ }
+ }
+}
+
+
+struct Scaler2x
+{
+ static const int scale = 2;
+
+ template
+ static void blendLineShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<3, 4>(out.template ref(), col);
+ }
+
+ template
+ static void blendLineSteep(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<0, scale - 1>(), col);
+ alphaBlend<3, 4>(out.template ref<1, scale - 1>(), col);
+ }
+
+ template
+ static void blendLineSteepAndShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<1, 0>(), col);
+ alphaBlend<1, 4>(out.template ref<0, 1>(), col);
+ alphaBlend<5, 6>(out.template ref<1, 1>(), col); //[!] fixes 7/8 used in xBR
+ }
+
+ template
+ static void blendLineDiagonal(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 2>(out.template ref<1, 1>(), col);
+ }
+
+ template
+ static void blendCorner(uint32_t col, OutputMatrix& out)
+ {
+ //model a round corner
+ alphaBlend<21, 100>(out.template ref<1, 1>(), col); //exact: 1 - pi/4 = 0.2146018366
+ }
+};
+
+
+struct Scaler3x
+{
+ static const int scale = 3;
+
+ template
+ static void blendLineShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<1, 4>(out.template ref(), col);
+
+ alphaBlend<3, 4>(out.template ref(), col);
+ out.template ref() = col;
+ }
+
+ template
+ static void blendLineSteep(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<0, scale - 1>(), col);
+ alphaBlend<1, 4>(out.template ref<2, scale - 2>(), col);
+
+ alphaBlend<3, 4>(out.template ref<1, scale - 1>(), col);
+ out.template ref<2, scale - 1>() = col;
+ }
+
+ template
+ static void blendLineSteepAndShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<2, 0>(), col);
+ alphaBlend<1, 4>(out.template ref<0, 2>(), col);
+ alphaBlend<3, 4>(out.template ref<2, 1>(), col);
+ alphaBlend<3, 4>(out.template ref<1, 2>(), col);
+ out.template ref<2, 2>() = col;
+ }
+
+ template
+ static void blendLineDiagonal(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 8>(out.template ref<1, 2>(), col);
+ alphaBlend<1, 8>(out.template ref<2, 1>(), col);
+ alphaBlend<7, 8>(out.template ref<2, 2>(), col);
+ }
+
+ template
+ static void blendCorner(uint32_t col, OutputMatrix& out)
+ {
+ //model a round corner
+ alphaBlend<45, 100>(out.template ref<2, 2>(), col); //exact: 0.4545939598
+ //alphaBlend<14, 1000>(out.template ref<2, 1>(), col); //0.01413008627 -> negligable
+ //alphaBlend<14, 1000>(out.template ref<1, 2>(), col); //0.01413008627
+ }
+};
+
+
+struct Scaler4x
+{
+ static const int scale = 4;
+
+ template
+ static void blendLineShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<1, 4>(out.template ref(), col);
+
+ alphaBlend<3, 4>(out.template ref(), col);
+ alphaBlend<3, 4>(out.template ref(), col);
+
+ out.template ref() = col;
+ out.template ref() = col;
+ }
+
+ template
+ static void blendLineSteep(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<0, scale - 1>(), col);
+ alphaBlend<1, 4>(out.template ref<2, scale - 2>(), col);
+
+ alphaBlend<3, 4>(out.template ref<1, scale - 1>(), col);
+ alphaBlend<3, 4>(out.template ref<3, scale - 2>(), col);
+
+ out.template ref<2, scale - 1>() = col;
+ out.template ref<3, scale - 1>() = col;
+ }
+
+ template
+ static void blendLineSteepAndShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<3, 4>(out.template ref<3, 1>(), col);
+ alphaBlend<3, 4>(out.template ref<1, 3>(), col);
+ alphaBlend<1, 4>(out.template ref<3, 0>(), col);
+ alphaBlend<1, 4>(out.template ref<0, 3>(), col);
+ alphaBlend<1, 3>(out.template ref<2, 2>(), col); //[!] fixes 1/4 used in xBR
+ out.template ref<3, 3>() = out.template ref<3, 2>() = out.template ref<2, 3>() = col;
+ }
+
+ template
+ static void blendLineDiagonal(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 2>(out.template ref(), col);
+ alphaBlend<1, 2>(out.template ref(), col);
+ out.template ref() = col;
+ }
+
+ template
+ static void blendCorner(uint32_t col, OutputMatrix& out)
+ {
+ //model a round corner
+ alphaBlend<68, 100>(out.template ref<3, 3>(), col); //exact: 0.6848532563
+ alphaBlend< 9, 100>(out.template ref<3, 2>(), col); //0.08677704501
+ alphaBlend< 9, 100>(out.template ref<2, 3>(), col); //0.08677704501
+ }
+};
+
+
+struct Scaler5x
+{
+ static const int scale = 5;
+
+ template
+ static void blendLineShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<1, 4>(out.template ref(), col);
+
+ alphaBlend<3, 4>(out.template ref(), col);
+ alphaBlend<3, 4>(out.template ref(), col);
+
+ out.template ref() = col;
+ out.template ref() = col;
+ out.template ref() = col;
+ out.template ref() = col;
+ }
+
+ template
+ static void blendLineSteep(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<0, scale - 1>(), col);
+ alphaBlend<1, 4>(out.template ref<2, scale - 2>(), col);
+ alphaBlend<1, 4>(out.template ref<4, scale - 3>(), col);
+
+ alphaBlend<3, 4>(out.template ref<1, scale - 1>(), col);
+ alphaBlend<3, 4>(out.template ref<3, scale - 2>(), col);
+
+ out.template ref<2, scale - 1>() = col;
+ out.template ref<3, scale - 1>() = col;
+ out.template ref<4, scale - 1>() = col;
+ out.template ref<4, scale - 2>() = col;
+ }
+
+ template
+ static void blendLineSteepAndShallow(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 4>(out.template ref<0, scale - 1>(), col);
+ alphaBlend<1, 4>(out.template ref<2, scale - 2>(), col);
+ alphaBlend<3, 4>(out.template ref<1, scale - 1>(), col);
+
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<1, 4>(out.template ref(), col);
+ alphaBlend<3, 4>(out.template ref(), col);
+
+ out.template ref<2, scale - 1>() = col;
+ out.template ref<3, scale - 1>() = col;
+
+ out.template ref() = col;
+ out.template ref() = col;
+
+ out.template ref<4, scale - 1>() = col;
+
+ alphaBlend<2, 3>(out.template ref<3, 3>(), col);
+ }
+
+ template
+ static void blendLineDiagonal(uint32_t col, OutputMatrix& out)
+ {
+ alphaBlend<1, 8>(out.template ref(), col);
+ alphaBlend<1, 8>(out.template ref(), col);
+ alphaBlend<1, 8>(out.template ref(), col);
+
+ alphaBlend<7, 8>(out.template ref<4, 3>(), col);
+ alphaBlend<7, 8>(out.template ref<3, 4>(), col);
+
+ out.template ref<4, 4>() = col;
+ }
+
+ template
+ static void blendCorner(uint32_t col, OutputMatrix& out)
+ {
+ //model a round corner
+ alphaBlend<86, 100>(out.template ref<4, 4>(), col); //exact: 0.8631434088
+ alphaBlend<23, 100>(out.template ref<4, 3>(), col); //0.2306749731
+ alphaBlend<23, 100>(out.template ref<3, 4>(), col); //0.2306749731
+ //alphaBlend<8, 1000>(out.template ref<4, 2>(), col); //0.008384061834 -> negligable
+ //alphaBlend<8, 1000>(out.template ref<2, 4>(), col); //0.008384061834
+ }
+};
+}
+
+
+void xbrz::scale(size_t factor, const uint32_t* src, uint32_t* trg, int srcWidth, int srcHeight, const xbrz::ScalerCfg& cfg, int yFirst, int yLast)
+{
+ switch (factor)
+ {
+ case 2:
+ return scaleImage(src, trg, srcWidth, srcHeight, cfg, yFirst, yLast);
+ case 3:
+ return scaleImage(src, trg, srcWidth, srcHeight, cfg, yFirst, yLast);
+ case 4:
+ return scaleImage(src, trg, srcWidth, srcHeight, cfg, yFirst, yLast);
+ case 5:
+ return scaleImage(src, trg, srcWidth, srcHeight, cfg, yFirst, yLast);
+ }
+ assert(false);
+}
+
+
+bool xbrz::equalColor(uint32_t col1, uint32_t col2, double luminanceWeight, double equalColorTolerance)
+{
+ return colorDist(col1, col2, luminanceWeight) < equalColorTolerance;
+}
+
+
+void xbrz::nearestNeighborScale(const uint32_t* src, int srcWidth, int srcHeight, int srcPitch,
+ uint32_t* trg, int trgWidth, int trgHeight, int trgPitch,
+ SliceType st, int yFirst, int yLast)
+{
+ if (srcPitch < srcWidth * static_cast(sizeof(uint32_t)) ||
+ trgPitch < trgWidth * static_cast(sizeof(uint32_t)))
+ {
+ assert(false);
+ return;
+ }
+
+ switch (st)
+ {
+ case NN_SCALE_SLICE_SOURCE:
+ //nearest-neighbor (going over source image - fast for upscaling, since source is read only once
+ yFirst = std::max(yFirst, 0);
+ yLast = std::min(yLast, srcHeight);
+ if (yFirst >= yLast || trgWidth <= 0 || trgHeight <= 0) return;
+
+ for (int y = yFirst; y < yLast; ++y)
+ {
+ //mathematically: ySrc = floor(srcHeight * yTrg / trgHeight)
+ // => search for integers in: [ySrc, ySrc + 1) * trgHeight / srcHeight
+
+ //keep within for loop to support MT input slices!
+ const int yTrg_first = ( y * trgHeight + srcHeight - 1) / srcHeight; //=ceil(y * trgHeight / srcHeight)
+ const int yTrg_last = ((y + 1) * trgHeight + srcHeight - 1) / srcHeight; //=ceil(((y + 1) * trgHeight) / srcHeight)
+ const int blockHeight = yTrg_last - yTrg_first;
+
+ if (blockHeight > 0)
+ {
+ const uint32_t* srcLine = byteAdvance(src, y * srcPitch);
+ uint32_t* trgLine = byteAdvance(trg, yTrg_first * trgPitch);
+ int xTrg_first = 0;
+
+ for (int x = 0; x < srcWidth; ++x)
+ {
+ int xTrg_last = ((x + 1) * trgWidth + srcWidth - 1) / srcWidth;
+ const int blockWidth = xTrg_last - xTrg_first;
+ if (blockWidth > 0)
+ {
+ xTrg_first = xTrg_last;
+ fillBlock(trgLine, trgPitch, srcLine[x], blockWidth, blockHeight);
+ trgLine += blockWidth;
+ }
+ }
+ }
+ }
+ break;
+
+ case NN_SCALE_SLICE_TARGET:
+ //nearest-neighbor (going over target image - slow for upscaling, since source is read multiple times missing out on cache! Fast for similar image sizes!)
+ yFirst = std::max(yFirst, 0);
+ yLast = std::min(yLast, trgHeight);
+ if (yFirst >= yLast || srcHeight <= 0 || srcWidth <= 0) return;
+
+ for (int y = yFirst; y < yLast; ++y)
+ {
+ uint32_t* trgLine = byteAdvance(trg, y * trgPitch);
+ const int ySrc = srcHeight * y / trgHeight;
+ const uint32_t* srcLine = byteAdvance(src, ySrc * srcPitch);
+ for (int x = 0; x < trgWidth; ++x)
+ {
+ const int xSrc = srcWidth * x / trgWidth;
+ trgLine[x] = srcLine[xSrc];
+ }
+ }
+ break;
+ }
+}
diff --git a/filter/xbrz.h b/filter/xbrz.h
new file mode 100644
index 00000000..ccbb299e
--- /dev/null
+++ b/filter/xbrz.h
@@ -0,0 +1,88 @@
+// ****************************************************************************
+// * This file is part of the HqMAME project. It is distributed under *
+// * GNU General Public License: http://www.gnu.org/licenses/gpl.html *
+// * Copyright (C) Zenju (zenju AT gmx DOT de) - All Rights Reserved *
+// * *
+// * Additionally and as a special exception, the author gives permission *
+// * to link the code of this program with the MAME library (or with modified *
+// * versions of MAME that use the same license as MAME), and distribute *
+// * linked combinations including the two. You must obey the GNU General *
+// * Public License in all respects for all of the code used other than MAME. *
+// * If you modify this file, you may extend this exception to your version *
+// * of the file, but you are not obligated to do so. If you do not wish to *
+// * do so, delete this exception statement from your version. *
+// ****************************************************************************
+
+#ifndef XBRZ_HEADER_3847894708239054
+#define XBRZ_HEADER_3847894708239054
+
+#include "port.h"
+//#include //size_t
+//#include //uint32_t
+#include
+#include "xbrz-config.h"
+
+namespace xbrz
+{
+/*
+-------------------------------------------------------------------------
+| xBRZ: "Scale by rules" - high quality image upscaling filter by Zenju |
+-------------------------------------------------------------------------
+using a modified approach of xBR:
+http://board.byuu.org/viewtopic.php?f=10&t=2248
+- new rule set preserving small image features
+- support multithreading
+- support 64 bit architectures
+- support processing image slices
+*/
+
+/*
+-> map source (srcWidth * srcHeight) to target (scale * width x scale * height) image, optionally processing a half-open slice of rows [yFirst, yLast) only
+-> color format: ARGB (BGRA byte order), alpha channel unused
+-> support for source/target pitch in bytes!
+-> if your emulator changes only a few image slices during each cycle (e.g. Dosbox) then there's no need to run xBRZ on the complete image:
+ Just make sure you enlarge the source image slice by 2 rows on top and 2 on bottom (this is the additional range the xBRZ algorithm is using during analysis)
+ Caveat: If there are multiple changed slices, make sure they do not overlap after adding these additional rows in order to avoid a memory race condition
+ if you are using multiple threads for processing each enlarged slice!
+
+THREAD-SAFETY: - parts of the same image may be scaled by multiple threads as long as the [yFirst, yLast) ranges do not overlap!
+ - there is a minor inefficiency for the first row of a slice, so avoid processing single rows only
+
+
+*/
+void scale(size_t factor, //valid range: 2 - 5
+ const uint32_t* src, uint32_t* trg, int srcWidth, int srcHeight,
+ const ScalerCfg& cfg = ScalerCfg(),
+ int yFirst = 0, int yLast = std::numeric_limits::max()); //slice of source image
+
+void nearestNeighborScale(const uint32_t* src, int srcWidth, int srcHeight,
+ uint32_t* trg, int trgWidth, int trgHeight);
+
+enum SliceType
+{
+ NN_SCALE_SLICE_SOURCE,
+ NN_SCALE_SLICE_TARGET,
+};
+void nearestNeighborScale(const uint32_t* src, int srcWidth, int srcHeight, int srcPitch, //pitch in bytes!
+ uint32_t* trg, int trgWidth, int trgHeight, int trgPitch,
+ SliceType st, int yFirst, int yLast);
+
+//parameter tuning
+bool equalColor(uint32_t col1, uint32_t col2, double luminanceWeight, double equalColorTolerance);
+
+
+
+
+
+//########################### implementation ###########################
+inline
+void nearestNeighborScale(const uint32_t* src, int srcWidth, int srcHeight,
+ uint32_t* trg, int trgWidth, int trgHeight)
+{
+ nearestNeighborScale(src, srcWidth, srcHeight, srcWidth * sizeof(uint32_t),
+ trg, trgWidth, trgHeight, trgWidth * sizeof(uint32_t),
+ NN_SCALE_SLICE_TARGET, 0, trgHeight);
+}
+}
+
+#endif
diff --git a/win32/render.cpp b/win32/render.cpp
index bb252a27..a2a99d7e 100644
--- a/win32/render.cpp
+++ b/win32/render.cpp
@@ -195,6 +195,9 @@
#include "../filter/2xsai.h"
#include "../filter/hq2x.h"
#include "snes_ntsc.h"
+#include "../filter/xbrz.h"
+#include
+#include
// Private Prototypes, should not be called directly
void RenderPlain (SSurface Src, SSurface Dst, RECT *);
@@ -209,6 +212,7 @@ template void RenderHQ2X (SSurface Src, SSurface Dst, RECT *rect);
template void RenderHQ3X (SSurface Src, SSurface Dst, RECT *rect);
void RenderLQ3XB (SSurface Src, SSurface Dst, RECT *rect);
void RenderHQ4X (SSurface Src, SSurface Dst, RECT *rect);
+void Render4xBRZ(SSurface Src, SSurface Dst, RECT* rect);
void RenderEPXA (SSurface Src, SSurface Dst, RECT *);
void RenderEPXB (SSurface Src, SSurface Dst, RECT *);
void RenderEPXC (SSurface Src, SSurface Dst, RECT *);
@@ -249,6 +253,30 @@ enum BlarggMode { UNINITIALIZED,BLARGGCOMPOSITE,BLARGGSVIDEO,BLARGGRGB };
snes_ntsc_t *ntsc = NULL;
BlarggMode blarggMode = UNINITIALIZED;
+int num_xbrz_threads = 4;
+
+struct xbrz_thread_data {
+ HANDLE xbrz_start_event;
+ HANDLE xbrz_sync_event;
+ HANDLE thread_handle;
+ static int scalingFactor;
+ static SSurface *src;
+ static SSurface *dst;
+ static uint8* dstPtr;
+ int yFirst;
+ int yLast;
+};
+xbrz_thread_data *xbrz_thread_sync_data;
+
+int xbrz_thread_data::scalingFactor = 4;
+SSurface *xbrz_thread_data::src = NULL;
+SSurface *xbrz_thread_data::dst = NULL;
+uint8 *xbrz_thread_data::dstPtr = NULL;
+
+HANDLE *xbrz_sync_handles;
+
+DWORD WINAPI ThreadProc_XBRZ(VOID * pParam);
+
TRenderMethod FilterToMethod(RenderFilter filterID)
{
switch(filterID)
@@ -281,6 +309,7 @@ TRenderMethod FilterToMethod(RenderFilter filterID)
case FILTER_BLARGGRGB: return RenderBlarggNTSCRgb;
case FILTER_SIMPLE4X: return RenderSimple4X;
case FILTER_HQ4X: return RenderHQ4X;
+ case FILTER_4XBRZ: return Render4xBRZ;
}
}
@@ -316,6 +345,7 @@ const char* GetFilterName(RenderFilter filterID)
case FILTER_EPX3: return "EPX3";
case FILTER_SIMPLE4X: return "Simple 4X";
case FILTER_HQ4X: return "hq4x";
+ case FILTER_4XBRZ: return "4xBRZ";
}
}
@@ -344,6 +374,7 @@ int GetFilterScale(RenderFilter filterID)
return 3;
case FILTER_SIMPLE4X:
case FILTER_HQ4X:
+ case FILTER_4XBRZ:
return 4;
}
}
@@ -363,6 +394,7 @@ bool GetFilterHiResSupport(RenderFilter filterID)
case FILTER_SIMPLE3X:
case FILTER_SIMPLE4X:
case FILTER_HQ4X:
+ case FILTER_4XBRZ:
return true;
default:
@@ -453,6 +485,20 @@ void InitRenderFilters(void)
BlendBuffer = BlendBuf + EXT_OFFSET;
memset(BlendBuf, 0, EXT_PITCH * EXT_HEIGHT);
}
+
+ SYSTEM_INFO sysinfo;
+ GetSystemInfo( &sysinfo );
+ num_xbrz_threads = sysinfo.dwNumberOfProcessors * 2;
+ if(!xbrz_thread_sync_data) {
+ xbrz_thread_sync_data = new xbrz_thread_data[num_xbrz_threads];
+ xbrz_sync_handles = new HANDLE[num_xbrz_threads];
+ for(int i = 0; i < num_xbrz_threads; i++) {
+ xbrz_thread_sync_data[i].xbrz_start_event = CreateEvent(NULL, FALSE, FALSE, NULL);
+ xbrz_thread_sync_data[i].xbrz_sync_event = CreateEvent(NULL, FALSE, FALSE, NULL);
+ xbrz_thread_sync_data[i].thread_handle = CreateThread(NULL, 0, ThreadProc_XBRZ, &xbrz_thread_sync_data[i], 0, NULL);
+ xbrz_sync_handles[i] = xbrz_thread_sync_data[i].xbrz_sync_event;
+ }
+ }
}
#define R5G6B5 // windows port uses RGB565
@@ -467,6 +513,10 @@ void InitRenderFilters(void)
((((pixel) >> 5) & 0x3f) << /*GreenShift+2*/10) | \
(((pixel) & 0x1f) << /*BlueShift+3*/ 3))
#define NUMBITS (16)
+ #define CONVERT_32_TO_16(pixel) \
+ (((((pixel) & 0xf80000) >> 8) | \
+ (((pixel) & 0xfc00) >> 5) | \
+ (((pixel) & 0xf8) >> 3)) & 0xffff)
#else
#define Mask_2 0x03E0 // 00000 11111 00000
#define Mask13 0x7C1F // 11111 00000 11111
@@ -477,6 +527,10 @@ void InitRenderFilters(void)
((((pixel) >> 5) & 0x1f) << /*GreenShift+3*/11) | \
(((pixel) & 0x1f) << /*BlueShift+3*/ 3))
#define NUMBITS (15)
+ #define CONVERT_32_TO_16(pixel) \
+ (((((pixel) & 0xf80000) >> 9) | \
+ (((pixel) & 0xf800) >> 6) | \
+ (((pixel) & 0xf8) >> 3)) & 0xffff)
#endif
static int RGBtoYUV[1<= yLast || height <= 0 || width <= 0) return;
+
+ for (int y = yFirst; y < yLast; ++y)
+ {
+ uint32_t* trgLine = trg + y * width;
+ const uint16_t* srcLine = reinterpret_cast(reinterpret_cast(src) + y * srcPitch);
+
+ for (int x = 0; x < width; ++x)
+ trgLine[x] = CONVERT_16_TO_32(srcLine[x]);
+ }
+}
+
+
+//stretch image and convert from ARGB to RGB565/555
+inline
+void stretchImage32To16(const uint32_t* src, int srcWidth, int srcHeight,
+ uint16_t* trg, int trgWidth, int trgHeight, int trgPitch,
+ int yFirst, int yLast)
+{
+ yFirst = std::max(yFirst, 0);
+ yLast = std::min(yLast, trgHeight);
+ if (yFirst >= yLast || srcHeight <= 0 || srcWidth <= 0) return;
+
+ for (int y = yFirst; y < yLast; ++y)
+ {
+ uint16_t* trgLine = reinterpret_cast(reinterpret_cast(trg) + y * trgPitch);
+ const int ySrc = srcHeight * y / trgHeight;
+ const uint32_t* srcLine = src + ySrc * srcWidth;
+ for (int x = 0; x < trgWidth; ++x)
+ {
+ const int xSrc = srcWidth * x / trgWidth;
+ trgLine[x] = CONVERT_32_TO_16(srcLine[xSrc]);
+ }
+ }
+}
+
+std::vector renderBuffer; //raw image
+std::vector xbrzBuffer; //scaled image
+
+DWORD WINAPI ThreadProc_XBRZ(VOID * pParam)
+{
+ xbrz_thread_data *thread_data = (xbrz_thread_data *)pParam;
+ while(true) {
+ WaitForSingleObject(thread_data->xbrz_start_event, INFINITE);
+ int trgWidth = xbrz_thread_data::src->Width * xbrz_thread_data::scalingFactor;
+ int trgHeight = xbrz_thread_data::src->Height * xbrz_thread_data::scalingFactor;
+ copyImage16To32(reinterpret_cast(thread_data->src->Surface), xbrz_thread_data::src->Width, xbrz_thread_data::src->Height, xbrz_thread_data::src->Pitch,
+ &renderBuffer[0], thread_data->yFirst, thread_data->yLast);
+ SetEvent(thread_data->xbrz_sync_event);
+ WaitForSingleObject(thread_data->xbrz_start_event, INFINITE);
+
+ xbrz::scale(thread_data->scalingFactor, &renderBuffer[0], &xbrzBuffer[0], xbrz_thread_data::src->Width, xbrz_thread_data::src->Height, xbrz::ScalerCfg(), thread_data->yFirst, thread_data->yLast);
+ SetEvent(thread_data->xbrz_sync_event);
+ WaitForSingleObject(thread_data->xbrz_start_event, INFINITE);
+
+ if (xbrz_thread_data::src->Height > SNES_HEIGHT_EXTENDED)
+ trgHeight /= 2;
+ if (xbrz_thread_data::src->Width == 512)
+ trgWidth /= 2;
+ stretchImage32To16(&xbrzBuffer[0], xbrz_thread_data::src->Width * xbrz_thread_data::scalingFactor, xbrz_thread_data::src->Height * xbrz_thread_data::scalingFactor,
+ reinterpret_cast(xbrz_thread_data::dstPtr), trgWidth, trgHeight, xbrz_thread_data::dst->Pitch, thread_data->yFirst * xbrz_thread_data::scalingFactor, thread_data->yLast * xbrz_thread_data::scalingFactor);
+ SetEvent(thread_data->xbrz_sync_event);
+ }
+ return 0;
+}
+
+void Render4xBRZ(SSurface Src, SSurface Dst, RECT* rect)
+{
+ xbrz_thread_data::scalingFactor = 4;
+
+ xbrz_thread_data::dstPtr = Dst.Surface;
+ SetRect(rect, SNES_WIDTH, SNES_HEIGHT_EXTENDED, xbrz_thread_data::scalingFactor);
+ xbrz_thread_data::dstPtr += rect->top * Dst.Pitch + rect->left * sizeof(uint16_t);
+
+ if (Src.Width <= 0 || Src.Height <= 0)
+ return;
+
+ renderBuffer.resize(Src.Width * Src.Height);
+ xbrzBuffer.resize(renderBuffer.size() * xbrz_thread_data::scalingFactor * xbrz_thread_data::scalingFactor);
+
+ xbrz_thread_data::src = &Src;
+ xbrz_thread_data::dst = &Dst;
+
+ // init + convert run
+ int ySlice = Src.Height / num_xbrz_threads;
+ for(int i = 0; i < num_xbrz_threads; i++) {
+ xbrz_thread_sync_data[i].yFirst = ySlice * i;
+ xbrz_thread_sync_data[i].yLast = (i == num_xbrz_threads - 1) ? Src.Height : ySlice * i + ySlice;
+ SetEvent(xbrz_thread_sync_data[i].xbrz_start_event);
+ }
+ WaitForMultipleObjects(num_xbrz_threads, xbrz_sync_handles, TRUE, INFINITE);
+
+ // xbrz run
+ for(int i = 0; i < num_xbrz_threads; i++) {
+ SetEvent(xbrz_thread_sync_data[i].xbrz_start_event);
+ }
+ WaitForMultipleObjects(num_xbrz_threads, xbrz_sync_handles, TRUE, INFINITE);
+
+ // convert run
+ for(int i = 0; i < num_xbrz_threads; i++) {
+ SetEvent(xbrz_thread_sync_data[i].xbrz_start_event);
+ }
+ WaitForMultipleObjects(num_xbrz_threads, xbrz_sync_handles, TRUE, INFINITE);
+}
+/*#################### /XBRZ support ####################*/
+
void RenderBlarggNTSCComposite( SSurface Src, SSurface Dst, RECT *rect)
{
if(blarggMode!=BLARGGCOMPOSITE) {
diff --git a/win32/snes9xw.vcproj b/win32/snes9xw.vcproj
index 7ddb4586..c0f74e75 100644
--- a/win32/snes9xw.vcproj
+++ b/win32/snes9xw.vcproj
@@ -4698,6 +4698,18 @@
RelativePath=".\snes_ntsc_impl.h"
>
+
+
+
+
+
+