Today, microprocessor systems with a local display device such as a liquid-crystal display may use a discrete graphics chip or an integrated graphics engine (commonly called a GPU) built into a core-logic chip such as a north bridge or south bridge.
The second bus interface 26 communicates with a graphics engine external 42 to the south bridge 14, which can be coupled to a display 44 and a second memory 46.
The integrated solution of
In UMA systems that have the main memory interface, integrated into the microprocessor, the integrated-graphics solution requires display refresh data to be transferred from the memory 24 to the processor core 15 via the memory bus, then from the processor 12 to the graphics engine 42 in the south bridge 14 via a front-side bus. This extra transfer wastes energy and reduces system performance by consuming some of the bandwidth of the front-side bus.
In some systems, the graphics engine 42 is integrated into the microprocessor chip.
A fully integrated UMA solution such as that above is acceptable for low-end systems, but not for mid-range or high-end systems. Modern processor cores are already large and expensive to manufacture, and a good graphics engine is just as large and complex as a processor core. Combining both a processor core and a graphics engine on one chip imposes an unacceptably high cost penalty. Accordingly what is desired is a microcomputer that overcomes the above-mentioned issues. The present invention addresses such a need.
A processing system is disclosed. The processing system comprises a first integrated circuit. The first integrated circuit includes a processor core, a display interface and memory controller coupled to a first bus interface. The display interface is adapted to display graphical information generated by a graphics engine. The graphics engine is not on the first integrated circuit. The processing system includes a second bus interface for allowing communication with the first integrated circuit via the first bus interface. The second bus interface is adapted to allow for communication to the graphics engine.
The present invention relates generally to microprocessors and more particularly to an integrated display interface. The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements. Various modifications to the preferred embodiments and the generic principles and features described herein will be readily apparent to those skilled in the art. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features described herein.
As known by one of ordinary skill in the art, this invention, including any logic circuit or transistor circuit, may be modeled, generated, or both by computer based on a description of the hardware expressed in the syntax and the semantics of a hardware description language (HDL). Such HDL descriptions are often stored on a computer readable medium. Applicable HDLs include those at the layout, circuit netlist, register transfer, and/or schematic capture levels. Examples of HDLs include, but are not limited to: GDS II and OASIS (layout level); various SPICE languages, and IBIS (circuit netlist level); Verilog and VHDL (register transfer level); and Virtuoso custom design language and Design Architecture-IC custom design language (schematic capture level). HDL descriptions may also be used for a variety of purposes, including but not limited to layout, behavior, logic and circuit design verification, modeling, and/or simulation.
It is desirable to be able to offer both high-end and low-end microprocessors using a common package pinout. Such commonality would improve the chances of commercial success for both products, since they could use a common motherboard design. A microprocessor system in accordance with the present invention addresses the above identified features of the present invention in more detail, refer to the following.
A system and method in accordance with the present invention, a microprocessor includes memory interface and a display interface as well as the logic necessary to connect these two interfaces. The microprocessor package pinout could be made the same as a fully integrated product that includes a microprocessor, a memory interface, a graphics engine, and a display interface.
In this embodiment, the bus interface 126 communicates with an internal graphics engine 142, which can be coupled to a display 144.
The microprocessor system 100 could operate with a graphics engine 142 that has an interface to a display 144 or the system 100 could operate with a graphics engine 142 that does not have an interface to a display.
As before mentioned, the microprocessor system 100 includes one or more processor cores 115, the display interface 148 which includes the necessary logic to support display-refresh operations and to display graphical information by a graphics engine, and the DRAM controller 116 including a memory interface. The logic to support display-refresh operations is often called a CRT controller function. The CRT controller function may be used with any type of display. Accordingly the CRT controller function is not limited to controlling a CRT but could control a LCD, OLED, LCOS, laser scanning or any other type of display. In conventional microprocessor systems, the CRT controller function is combined with the graphics engine 142. In the present invention, the CRT controller function is separate from the graphics engine 142 typically as part of the display interface 148.
The CRT controller function may also perform compositing operations in which multiple graphical images are combined to generate an image for the display. Compositing is a known technique in computer graphics and is used in Microsoft's Windows Vista operating system.
Alternately, processor 112 could include a CRT controller function but not a graphics engine, or could include both a CRT controller function and a graphics engine but the graphics engine is disabled because a discrete graphics chip is present in the south bridge 114.
Optionally, the processor or other component containing the display interface and CRT controller, but not the graphics engine, could also be equipped with all or part of a digital video decoding engine. Such a video engine can be much less complex than a high-end graphics engine, although in some implementations a graphics engine can perform video-decoding functions in software. When the video-decoding functions are divided among the graphics engine and dedicated logic blocks, for example, when the graphics engine performs transport and stream demultiplexing plus video and audio decoding and dedicated logic performs color-space conversion and video scaling, the functions implemented in dedicated logic blocks would preferably be integrated into the processor or other component containing the display interface and CRT controller function.
Separating the CRT controller function from the graphics engine and putting these two functions on separate chips allows the processor core 115 and graphics engine 142 to be designed and built separately, yet function together. In an embodiment, the CRT controller function is configured according to the requirements of the display device and the operating system.
The graphics engine 142 performs graphics operations as required by the operating system and applications, and sets pointers in the CRT controller according to the location of the frame buffer(s) located in system main memory, graphics memory, or both. Alternately these pointers may be set by software running on the processor core 115, especially for 2D-only display operations.
The CRT controller then fetches display data from the DRAM 124 and transmits it to the display 150.
When software running on the processor core 114 or the graphics engine 142, or hardware in the graphics engine 142, determine that the contents of the display are not changing, such as during idle periods between keystrokes or when no applications are running, etc., the display interface 148 can be configured to refresh the display device from a compressed or otherwise static frame buffer without further graphics engine 142 intervention. To enter this mode of operation, the contents of the currently active frame buffer are copied from the dedicated graphics memory (if present) to main memory, shown in
Or similarly, if the graphics processing that would otherwise be performed by the graphics engine 142 is simple enough to be performed efficiently by the processor core 115, the graphics engine 142 can be put into a low-power configuration or powered off entirely and its functions can be taken over by the processor core 115. This situation might arise if no application is currently performing any 3D rendering, since software running on a processor core 115 can perform 2D-only functions such as windows management. The basis for choosing between graphics engine 142-based operation and processor core 115-based operation could include whether the user will detect significant performance degradation due to processor core 115-based operation and whether the processor core 115 can perform the necessary functions using less total power than the graphics engine 142.
Several different system configurations become possible using this design.
1. Fully integrated solutions for low-end systems, with the microprocessor including fewer or smaller processor cores 15, an integrated graphics engine 42, and a display interface as shown in
2. An improved UMA implementation for mid-range systems in which the microprocessor includes the processor cores 115 and display interface 148, a separate chip includes the graphics engine 142, and system memory 124 is used for both shown in
3. An improved non-UMA implementation for high-end mobile systems in which the microprocessor includes the processor core 115 and display interface 148, a discrete graphics chip includes the graphics engine 142 and a local graphics-memory interface, and the display 150 is connected to the display interface 148 as shown in
4. An improved non-UMA implementation for high-end mobile systems in which the microprocessor includes the processor core 115 and display interface 148; a discrete graphics chip includes the graphics engine 142, a local graphics-memory interface, and a local display interface; and displays 150 and 144 are connected to the display interface 148 as well as the interface on the graphics engine 142 shown in
5. The traditional non-UMA implementation in which the microprocessor includes the processor core 115 and a display interface 148 that is not used; a discrete graphics chip includes the processor core 115, a local graphics-memory interface, and a local display interface; and the display is connected to the display interface on the graphics chip of
A system and method in accordance with the present invention thus allows the microprocessor systems to be built that have the lowest possible cost for a given level of performance because they have no unneeded display interfaces and a more economical separation of the processor core and the graphics engine. They also have the lowest possible energy consumption for display refresh options.
Although the present invention has been described in accordance with the embodiments shown, one of ordinary skill in the art will readily recognize that there could be variations to the embodiments and those variations would be within the spirit and scope of the present invention. Accordingly, many modifications may be made by one of ordinary skill in the art without departing from the spirit and scope of the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
5649173 | Lentz | Jul 1997 | A |
5790134 | Lentz | Aug 1998 | A |
5918057 | Chou et al. | Jun 1999 | A |
6184906 | Wang et al. | Feb 2001 | B1 |
RE37944 | Fielder et al. | Dec 2002 | E |
6518970 | Glen et al. | Feb 2003 | B1 |
6608625 | Chin et al. | Aug 2003 | B1 |
6801207 | Tischler et al. | Oct 2004 | B1 |
6822654 | Iwamoto et al. | Nov 2004 | B1 |
6850240 | Jones, Jr. | Feb 2005 | B1 |
6975324 | Valmiki et al. | Dec 2005 | B1 |
7051139 | Peleg et al. | May 2006 | B2 |
7139849 | Tojima et al. | Nov 2006 | B2 |
7391409 | Zalewski et al. | Jun 2008 | B2 |
7472301 | Ginggen et al. | Dec 2008 | B2 |
7617403 | Capps et al. | Nov 2009 | B2 |
20030001856 | Sawada et al. | Jan 2003 | A1 |
20050049729 | Culbert et al. | Mar 2005 | A1 |
20050156933 | Lee et al. | Jul 2005 | A1 |
20050160251 | Zur et al. | Jul 2005 | A1 |
20050268141 | Alben et al. | Dec 2005 | A1 |
20050285863 | Diamond | Dec 2005 | A1 |
20060066907 | Nakata et al. | Mar 2006 | A1 |
20060206627 | Kardach et al. | Sep 2006 | A1 |
20060267987 | Litchmanov | Nov 2006 | A1 |
20060271797 | Ginggen et al. | Nov 2006 | A1 |
20070219644 | Sonobe | Sep 2007 | A1 |
20070283175 | Marinkovic et al. | Dec 2007 | A1 |
20080028236 | Capps et al. | Jan 2008 | A1 |
20080143731 | Cheng et al. | Jun 2008 | A1 |
20080168286 | Tupman | Jul 2008 | A1 |
20080204460 | Marinkovic et al. | Aug 2008 | A1 |
20090109230 | Miller et al. | Apr 2009 | A1 |
20100302567 | Ozaki | Dec 2010 | A1 |
20110169840 | Bakalash | Jul 2011 | A1 |
20110212761 | Paulsen et al. | Sep 2011 | A1 |
20120042190 | Azar et al. | Feb 2012 | A1 |