Specifications.
[pintos-anon] / specs / freevga / hardrec.htm
diff --git a/specs/freevga/hardrec.htm b/specs/freevga/hardrec.htm
new file mode 100644 (file)
index 0000000..0f865b3
--- /dev/null
@@ -0,0 +1,127 @@
+<HTML>
+<HEAD>
+   <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
+   <META NAME="Author" CONTENT="Joshua Neal">
+   <META NAME="Description" CONTENT="Pure VGA/SVGA hardware programming (registers, identification, and otherlow-level stuff.)">
+   <META NAME="KeyWords" CONTENT="VGA SVGA hardware video programming">
+   <TITLE>FreeVGA - Product Recommendations for Video Developers</TITLE>
+</HEAD>
+<BODY>
+
+<CENTER><A HREF="home.htm">Home</A> <A HREF="#intro">Introduction</A> <A HREF="#monitors">Recommended</A>
+<A HREF="#failures">Failed</A> <A HREF="#test">Test</A> <A HREF="home.htm#product">Back</A>&nbsp;
+<HR WIDTH="100%"><B>Hardware Level VGA and SVGA Video Programming Information
+Page</B></CENTER>
+
+<CENTER>Product Recommendations for Video Developers&nbsp;
+<HR WIDTH="100%"></CENTER>
+<A NAME="intro"></A><B>Introduction</B>
+<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; This page is to provide
+hardware recommendations for those implementing the information on this
+site.&nbsp; There are no recommendations for video cards, as the goal is
+to increase programmer support for all cards, existing or otherwise, rather
+than try to influence people to buy a specific video card implementation.&nbsp;
+I will, however recommend hardware, other than the video cards themselves
+that are helpful in the development of software for video cards in general.
+Monitors are a strong issue to me, both for safety concerns, and financial
+concerns, as it is usually advantageous to buy an new, indestructable monitor
+than to burn through many cheap, expendable monitors.
+
+<P><B>&nbsp;<A NAME="monitors"></A>Monitors Recommended</B>
+<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; For a monitor to be recommended
+it must meet all of the following criteria:
+<UL>
+<LI>
+It should be able to tolerate improperly formed video inputs for an extended
+period of time without the possibility of being damaged.&nbsp; It should
+also be tolerant to extremely frequent mode changes.&nbsp; Damage due to
+this kind of operation should not be excluded by the standard manufacturer's
+warranty.&nbsp; This is critical due to the replacement cost of high-performance
+monitors, and due to the possible safety and fire hazards failing monitors
+may cause.</LI>
+
+<LI>
+It should be able to synchronize to a wide variety of properly formed signals,
+including both standard and custom video timings.&nbsp; This important
+for developing the modes required for special applications.</LI>
+
+<LI>
+It should handle the maximum frequencies/resolutions that can be generated
+by current and future (to a reasonable extent) video chipsets.</LI>
+
+<LI>
+It should be compliant to all levels of display to host communication and
+power mamagement so code can be developed that implements these features
+of the video hardware.</LI>
+
+<LI>
+If the monitor allows the picture controls to be saved/restored on changes
+in mode, it should be allowed to defeat this feature so that the generated
+video timings can be adjusted to minimize the visible effects of mode change.</LI>
+
+<LI>
+It should be currently available on the market and covered by manufacturers
+warranty for the period of time required to develop the desired application.</LI>
+
+<LI>
+It has been put through my own personal monitor torture tests, as well
+as operated for an extended period of time under conditions related to
+video software development.</LI>
+</UL>
+The following monitors have been evaluated by myself personally, and have
+been determined to meet all these criteria.
+<UL>
+<LI>
+At this time, there are no monitors that I have determined meet all these
+criteria.</LI>
+</UL>
+<A NAME="failures"></A><B>Monitor Failures</B>
+<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; This section lists monitors
+that have either died for me while testing, or have died for others in
+a fashion that would imply that the programmer was responsible for their
+failure.&nbsp; This does not imply that the programmer was at fault, as
+these things naturally happen when developing drivers.&nbsp; I strongly
+recommend the purchase of one of the recommended monitors, to avoid damaging
+a valuable monitor
+
+<P><B>Compaq VGA (not SVGA)</B> -- I do not know the model specifically.&nbsp;
+It was a fixed frequency model, and the horizontal circuitry was damaged.&nbsp;
+The problem was repeatable after repairs were made, so I believe that the
+monitor can be damaged by normal mode testing. I have met others who claim
+to have experienced this same problem.&nbsp; Not recommended.
+
+<P><B>CTX CMS-1561LR</B> -- The problem with this monitor occured when
+driving the monitor at the high end of its frequency envelope.&nbsp; The
+monitor synced to the frequency, but may have been slightly overdriven.&nbsp;
+The horizontal output transistor and some capacitors were replaced and
+the monitor was restored to working order.&nbsp; The problem has not been
+repeated, so ordinary failure is likely.
+
+<P><B>NCR MBR 2321</B> -- This one comes from a friend in Fayetteville,
+AR, whose monitor blew caps while writing a svgalib video driver.&nbsp;
+The explosion from the capacitors shattered the rear of the picture tube,
+damaging the monitor beyond repair.&nbsp; Not recommended due to the catastrophic
+nature of the failure.&nbsp; The operation being performed when the failure
+occurred was frequent mode changing.
+
+<P><A NAME="test"></A><B>Test Equipment Recommended</B>
+<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; There are certain pieces
+of test equipment that can come in handy when working with video cards.&nbsp;
+This can be especially important when verifying that the video signal being
+generated is, in fact the one intended by the programmer.&nbsp; Failure
+to do this can cause catastrophic failure when the driver is used in conjunction
+with a fixed-frequency or other monitor that can be damaged by improper
+inputs.
+<UL>
+<LI>
+At this time, I cannot recommend any test equipment other than a good frequency
+counter, as this is really not in my area of expertise.&nbsp; If you can
+help me in my research into this, I would be greatly appreciative.</LI>
+</UL>
+Notice: All trademarks used or referred to on this page are the property
+of their respective owners.
+<BR>All pages are Copyright &copy; 1997, 1998, J. D. Neal, except where
+noted. Permission for utilization and distribution is subject to the terms
+of the <A HREF="license.htm">FreeVGA Project Copyright License</A>.&nbsp;<IMG SRC="http://www.goodnet.com/~tinara/cgi-bin/imgserv.cgi?logo3.gif" >
+</BODY>
+</HTML>