Content-Based Adaptive Refresh Schemes For Low-Power Displays

[0001] 1. Field

[0002] The present invention relates to a display system. More particularly, embodiments of the present invention relate to reducing the power consumption of a display panel.

[0003] 2. Background Information

[0004] Flat panel displays are gaining popularity in a wide range of electronic devices. Common types of flat panel displays include active matrix displays and passive matrix displays. Each pixel in an active matrix display is driven by active driving circuitry, while each pixel in a passive matrix display does not use such driving circuitry.

[0005] High-resolution color displays, such as modern computer displays, smart phones and televisions, typically use an active matrix display structure for better image quality and response time. An active matrix display of m×n pixel elements can be addressed with m row lines and n column lines. A set of thin-film transistors (TFTs) and capacitors are connected to each pixel element, allowing each column line to access one pixel. When a row line is selected, all of the column lines are connected to a row of pixels and voltages corresponding to the picture information are driven onto all of the column lines. The row line is then deactivated and the next row line is selected. All of the row lines are selected in sequence during a refresh operation. This refresh operation is repeated tens to hundreds of times per second. If the refresh rate is too low, the capacitors may leak current and unable to retain its state. This would appear as flickers and degrade the quality of the displayed image.

[0006] For portable devices, the power consumed by a display panel can be a major portion of the total power consumption of a portable device. To extend the battery life, it is necessary to reduce the power consumption of the display panel.

[0007] A method, system and apparatus for content-based adaptive refresh of a low-power display panel are described. In one embodiment, an active matrix display system includes a display panel having multiple rows of display elements arranged as a display matrix. The display panel is coupled to a scan driver and a data driver. The scan driver selects one row at a time to receive data signals, and the data driver provides the data signals. The active matrix display system also includes a timing controller operable to signal the scan driver to cause a first row of the display panel to be not refreshed in a current data frame and a second row of the display panel to be refreshed in the current data frame. In one embodiment, the active matrix display system also includes refresh logic operable to compare content of the first row in the current data frame with content of the first row in a previous data frame, and to determine whether the first row is to be refreshed in the current data frame based on a result of the comparison. Thus, the rows that display a dynamic content can be refreshed at a higher refresh rate than the rows that display a static content.

[0008] In another embodiment, a method for refreshing a display panel includes refreshing a first portion of the display panel at a first refresh rate, wherein the first portion displays a static content; and refreshing a second portion of the display panel at a second refresh rate higher than the first refresh rate, wherein the second portion displays a dynamic content. In one embodiment, the method further includes comparing content of the first portion of a current data frame with content of the first portion of a previous data frame to generate a comparison result, and determining whether the first portion displays a static content based on the comparison result. A portion of the display panel is refreshed when the portion of the display panel has a content change beyond a threshold or has reached a time out threshold.

[0009] In yet another embodiment, an apparatus includes a frame buffer to store a first portion of a previous data frame and a corresponding first portion of a current data frame, and refresh logic operable to compare content of the first portion of the current data frame with content of the corresponding first portion of the previous data frame to generate a comparison result, and to determine whether to refresh the first portion of the current data frame based on the comparison result. The refresh logic and the frame buffer may be located within a display system, a processor (such as a graphics processing unit (GPU) or a general purpose processing having graphics processing capabilities), an interface module coupled between the display system and the processor, or one or more hardware components coupled to a timing controller. The refresh logic is operable to assign different refresh rates to different portions of a display panel.

[0010] In one embodiment, the display panel of the above system, method and apparatus includes light emitting diode (LED) devices, micro LED devices, organic LED (OLED) devices, or liquid crystal display (LCD) elements. The term “portion” as used herein may be a row, a set of rows, a tile, a set of tiles, or a set of LED devices within a tile.

[0021] In various embodiments, description is made with reference to figures. However, certain embodiments may be practiced without one or more of these specific details, or in combination with other known methods and configurations. In the following description, numerous specific details are set forth, such as specific configurations, dimensions and processes, etc., in order to provide a thorough understanding of the present invention. In other instances, well-known techniques and components have not been described in particular detail in order to not unnecessarily obscure the present invention. Reference throughout this specification to “one embodiment,” “an embodiment” or the like means that a particular feature, structure, configuration, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Thus, the appearances of the phrase “in one embodiment,” “in an embodiment” or the like in various places throughout this specification are not necessarily referring to the same embodiment of the invention. Furthermore, the particular features, structures, configurations, or characteristics may be combined in any suitable manner in one or more embodiments.

[0022] Embodiments of the present invention provide a method and system for reducing the power consumption of an active matrix display panel. For example, a major percentage of power consumed by an active matrix liquid crystal display (LCD) or organic LEDs (OLEDs) display panel is used on refreshing the display data to overcome panel parasitics that arise from switching transistors, storage capacitors, and row and column interconnects. In current display technologies the display data is typically refreshed at least 60 times per second (60 Hz) to avoid flickering in the displayed image, and power consumption required for display emission is as much as 5 to 10 times higher the amount of power consumption required for refreshing the display data. For example, for high resolution 10-inch diagonal displays, the power consumption for display emission may be 5-10 watts and the power consumption for refreshing the display data may be 1 watt. Accordingly, less emphasis has been placed on reducing power consumption during a refresh cycle than for reducing power consumption required for display emission.

[0023] In accordance with some embodiments of the invention, the active matrix display panel described herein is a micro LED display panel formed by micro LED devices. A micro LED display panel combines the performance, efficiency, and reliability of wafer-based LED devices with the high yield, low cost, mixed materials of thin film electronics used to form backplanes. The term “micro” in “micro LED device” refers to the scale of 1 to 100 μm. However, it is to be appreciated that the embodiments described herein may be applicable to larger, and possibly smaller size scales. In an embodiment, a display panel is similar to atypical OLED display panel, with a micro LED device having replaced the organic layer of the OLED display panel in each subpixel. Exemplary micro LED devices which may be utilized with some embodiments of the invention are described in U.S. patent application Ser. No. 13/372,222, U.S. patent application Ser. No. 13/436,260, U.S. patent application Ser. No. 13/458,932, and U.S. patent application Ser. No. 13/625,825, all of which are incorporated herein by reference. The micro LED devices are highly efficient at light emission and consume very little power (e.g., 250 mW) compared to 5-10 watts for LCD or OLED emission. Assuming, that the power consumption required for refreshing the display data remains the same (e.g. 1 watt), the percentage of power required for a refresh cycle to total power consumption for a micro LED display panel is more significant. Embodiments of the invention provide content-based adaptive refresh schemes for reducing the power consumption on display refresh. It should be appreciated, however, the adaptive refresh schemes are applicable to any active matrix displays, including but not limited to an LED display panel, a micro LED display panel, an OLED display panel, an LCD display panel, and other variations of display panels based on LED, OLED and LCD technologies.

[0024] One technique for reducing the refresh power consumption is to reduce the refresh rate of the display panel. With improved circuit elements, such as larger storage capacitors and switching transistors with lower OFF state leakage current, a display system may reduce its refresh rates without degrading the image quality beyond a predetermined design limit. Embodiments of the present invention provide a content-based adaptive refresh technique to reduce power consumption of a display system, which can be implemented in combination with or separately from the use of improved circuit elements. The content-based adaptive refresh technique refreshes different parts of a display panel at different rates based on data contents. A row of data that does not change its content, or changes less often than the other rows, is refreshed less often than these other rows. The refresh rate can be adjusted in real time during the data display process. When a row of data changes its content from static content (e.g., text or still image) to dynamic content (e.g., video), the refresh rate of that row can be increased to improve the image quality. When a row of data changes its content from dynamic content to static content, the refresh rate of that row can be decreased to reduce power consumption. That is, different rows of a display panel can be refreshed at different rates depending on their respective display contents.

[0025] As used herein, the term “data frame” refers to the display data that fills the entire display area of a display panel. Each row of a data frame is referred to as a “data line.” The term “frame time” or “frame cycle” refers to the time for selecting and displaying all of the rows in a data frame. The term “line time” or “line cycle” refers to the time for selecting and displaying one of the rows in a data frame. Further, the terms “display panel” and “display” are used interchangeably.

[0026] Moreover, the term “ON” as used herein in connection with a device state refers to an activated state of the device, and the term “OFF” used in this connection refers to a de-activated state of the device. The term “ON” as used herein in connection with a signal received by a device refers to a signal that activates the device, and the term “OFF” used in this connection refers to a signal that de-activates the device. A device may be activated by a high voltage or a low voltage, depending on the underlying electronics implementing the device. For example, a PMOS transistor device is activated by a low voltage while a NMOS transistor device is activated by a high voltage. Thus, it should be understood that an “ON” voltage for a PMOS transistor device and a NMOS transistor device correspond to opposite (low vs. high) voltage levels.

[0027] FIG. 1 illustrates a display system 100 in accordance with an embodiment. The display system 100 includes a display panel 120 driven by display driver integrated circuitry, which includes a data driver 122 and a scan driver 124. The scan driver 124 selects a row of the display panel 120 at a time by providing an ON voltage to the selected row. The selected row is activated to receive display data from the data driver 122.

[0028] The display panel 120 is an active matrix display that includes a two-dimensional matrix of display elements. In some embodiments, the display panel 120 may include multiple stacked layers of two-dimensional matrix of display elements. In one embodiment, each display element is an emissive device, which, for example, may include LEDs, OLEDs, micro LEDs, or other light-emissive elements. However, it is to be appreciated that embodiments of the present invention are not necessarily so limited. In alternative embodiments, each display element may be an LCD element, or other types of display elements.

[0029] The display panel 120 includes a matrix of pixels. Each pixel may include multiple subpixels that emit different colors of lights. In a red-green-blue (RGB) subpixel arrangement, each pixel includes three subpixels that emit red, green and blue lights, respectively. It is to be appreciated that the RGB arrangement is exemplary and that embodiments are not so limited. Examples of other subpixel arrangements that can be utilized include, but are not limited to, red-green-blue-yellow (RGBY), red-green-blue-yellow-cyan (RGBYC), or red-green-blue-white (RGBW), or other subpixel matrix schemes where the pixels may have different number of subpixels, such as the displays manufactured under the trademark name PenTile®.

[0030] In one embodiment, the data driver 122 and the scan driver 124 are controlled by a timing controller 118. The timing controller 118 provides the scan driver 124 a select signal indicating which row is to be selected next for refresh. The timing controller 118 also presents a data line to the data driver 122 in the form of a row of data voltages. Each data voltage drives a corresponding subpixel in the selected row to emit a colored light at a specified intensity level. The rows may be refreshed sequentially. Alternatively, the rows may be addressed directly and refreshed one at a time in any given order. In yet another embodiment, the rows are partitioned into several non-overlapping (or overlapping) row segments. Each row segment can be addressed directly and the rows within a segment can be refreshed sequentially or in any given order. The illustrated embodiment shows that the timing controller 118, the data driver 122 and the scan driver 124 as separate components. In an alternative embodiment, the timing controller 118 and the data driver 122 may be one integrated component. In yet another embodiment, the timing controller 118, the data driver 122 and the scan driver 124 may be one integrated component.

[0031] The display system 100 includes a receiver 116 to receive display data from outside of the display system 100. The receiver 116 may be configured to receive data wirelessly, by a wire connection, or by an optical interconnect. Wireless may be implemented in any of a number of wireless standards or protocols including, but not limited to, Wi-Fi (IEEE 802.11 family), WiMAX (IEEE 802.16 family), IEEE 802.20, long term evolution (LTE), Ev-DO, HSPA+, HSDPA+, HSUPA+, EDGE, GSM, GPRS, CDMA, TDMA, DECT, Bluetooth, derivatives thereof, as well as any other wireless protocols that are designated as 3G, 4G, 5G, and beyond.

[0032] The receiver 116 may receive display data from a processor 110 via an interface controller 114. In one embodiment, the processor 110 may be a graphics processing unit (GPU), a general-purpose processor having a GPU located therein, or a general-purpose processor with graphics processing capabilities. The interface controller 114 provides display data and synchronization signals to the receiver 116, which in turn provides the display data to the timing controller 118. The display data may be generated in real time by the processor 110 executing one or more instructions in a software program, or retrieved from a system memory 112.

[0033] Depending on its applications, the display system 100 may include other components. These other components include, but are not limited to, memory, a touch-screen controller, and a battery. In various implementations, the display system 100 may be a television, tablet, phone, laptop, computer monitor, automotive heads-up display, automotive navigation display, kiosk, digital camera, handheld game console, media display, ebook display, or large area signage display.

[0034] According to one embodiment of the invention, the display system 100 also includes a refresh logic 115 and a frame buffer 130. The refresh logic 115 determines which rows should be refreshed and which rows should be skipped. Thus, the refresh logic 115 is responsible for refreshing different rows of the display panel 120 at different refresh rates based on the data content of the rows. The refresh logic 115 may base its decision on the data content stored in the frame buffer 130 and/or the mode in which the display panel 120 operates.

[0035] In the illustrated embodiment, the refresh logic 115 and the frame buffer 130 are located in the display system 100 and are couple to the timing controller 118. In an alternative embodiment, the refresh logic 115 and the frame buffer 130 may be located within the timing controller 118, the receiver 116, the interface controller 114, the processor 110, or another component within or coupled to the display system 100. The refresh logic 115 may be implemented by hardware circuitry (such as integrated circuitry) or firmware components. The frame buffer 130 may be a memory device, such as a dynamic random access memory (DRAM), or other types of volatile or non-volatile random access memory devices.

[0036] The refresh logic 115 determines which rows to refresh and when to refresh the rows. In one embodiment, the refresh logic 115 compares a data line of a current data frame with the corresponding data line of a previous data frame to determine whether or not this data line should be refreshed. The data lines used for the comparison are stored in the frame buffer 130. A number of alternative approaches for performing the comparison are described below; however, it is to be appreciated that the described approaches are exemplary and that embodiments are not so limited.

[0037] For example, the frame buffer 130 may store a previous data frame and one current data line to be displayed. If this current data line is the same as or within a predefined variation from the corresponding data line of the previous data frame, no refresh will be performed on the data line (unless the data line is going to be timed out, as will be described below). If the current data line is different or exceed a predefined variation from the corresponding data line of the previous data frame, the refresh logic 115 will signal the downstream component (i.e., a hardware component between the refresh logic 115 and the display panel 120) to cause the data line to be refreshed. The downstream component may send acknowledgements to the refresh logic 115 or other upstream components (i.e., components between the processor 110 and the downstream component). The current data line can then overwrite the corresponding data line of the previous data frame in the frame buffer 130, and a next data line can be loaded into the frame buffer 130 for comparison. As another example, the frame buffer 130 may store the previous data frame and a current segment (which is a set of rows) to be displayed. The refresh logic 115 may compare the current segment with a corresponding segment of the previous data frame to determine whether the segment or any of the data lines in the segment needs to be refreshed. In one embodiment, after comparing a line/segment, the line/segment may overwrite the corresponding line/segment in the previous data frame.

[0038] As yet another example, the frame buffer 130 may store the previous data frame and an entire current data frame to be displayed. The refresh logic 115 may compare the corresponding data lines in the two frames line by line, or segment by segment, to identify the data lines or segments that need to be refreshed. In one embodiment, after comparing a line/segment, the line/segment in the current data frame may overwrite the corresponding line/segment in the previous data frame. Alternatively, after comparing the two frames, the current data frame is set to the “new” previous data frame and a next data frame can be loaded into the frame buffer 130 as the “new” current data frame. In addition or alternative to the comparison described above, the refresh logic 115 may determine whether to refresh a data line based on the operation mode of the display panel 120. In one embodiment, the refresh logic 115 may receive signals from the processor 110 or another source indicating a current display mode of the display panel 120, such as a static mode (e.g., when displaying still images, texts or e-books), a dynamic mode (e.g., when displaying videos), a mixed mode (e.g., when displaying mixed text and video contents). For the mixed mode display, the refresh logic 115 may receive signals from the processor 110 or another source indicating the coordinates or row indices of the region or regions that need a different refresh rate from the rest of the display. For example, an online news page may display text, with a small window showing a video. The coordinates of the window, or the row indices of the window, may be sent to the refresh logic 115 such that the refresh logic 115 may pass this information to the downstream component to selectively refresh the data lines.

[0039] According to embodiments of the invention, a data line is refreshed when its data content has changed beyond a threshold. Additionally, a data line is refreshed when its content is going to be timed out. Otherwise the data line does not need to be refreshed. The content of a data line is timed out when the elapsed time from its last refresh has reached a threshold. This threshold is the maximum amount of time that a display element (e.g., a subpixel) can hold its electric charge in the storage capacitor without causing a change to the display image that exceeds a predetermined design limit. The refresh logic 115 may keep one or more timers to track the elapsed time from the last refresh for the rows; e.g., one timer for each data line or each segment. When the elapsed time reaches the threshold, the refresh logic 115 signals the downstream component to refresh the data line/segment, even if the data line/segment has no content change.

[0040] As described above, one approach to reducing power consumption on data refresh is by selectively lowering the refresh rates, such as skipping a data line/segment during data refresh. In addition, the data transfer rate into the display panel 120 can be reduced to realized further power saving. As described above, the refresh logic 115 may be located in the display system 100 or outside the display system 100. For example, the refresh logic 115 may be located within any of the following components: the timing controller 118, the receiver 116, the interface controller 114, the processor 110, or coupled to any of the above components. The location of the refresh logic 115 and the signal it sends to its downstream component can affect the data transfer rate along the data path into the display panel 120. In one embodiment, the refresh logic 115 is located within the processor 110. If the processor 110 only outputs the line/segment content that needs to be refreshed, the data transfer rate along the entire path from the processor 110 to the display panel 120 can be reduced. Alternatively, the processor 110 may output all of the line/segment content regardless whether a refresh is needed, and for each line/segment the processor 110 also outputs an indicator (e.g., an activate/ignore signal) indicating whether that line/segment needs to be refreshed. If the entire data content is forwarded all the way to the data driver 122, which selectively refreshes the line/segment according to the indicator, then no reduction in the data transfer rate can be realized along the path from the processor 110 to the display panel 120. If the data content is forwarded to a component downstream from the processor 110 before reaching the display panel 120 (e.g., the timing controller 118 or the interface component 114), at which point only the data lines that need refresh is forwarded to the display panel 120, then the data transfer rate can be reduced along the path from this component to the display panel 120.

[0041] In an embodiment where the refresh logic 115 is within the timing controller 118 or is coupled to the timing controller 118 as shown in the illustrated embodiment of FIG. 1, the data transfer rate along the path from the timing controller 118 to the display panel 120 can be reduced if the timing controller 118 only sends the data lines that need refresh to the display panel 120. If the timing controller 118 sends all of the data lines to the data driver 122, it may also send an “ignore” signal for the line/segment that is not to be refreshed and an “activate” signal for the line/segment that is to be refreshed. However, this latter approach only reduces the display refresh power and does not reduce the data transfer rate on the path to the display panel 120. Alternative scenarios may exist for the location of the refresh logic 115 and the component that starts the reduction in data transfer rate. It is to be appreciated that the above described scenarios are exemplary and that embodiments are not so limited.

[0042] FIG. 2 illustrates an example of the content-based adaptive refresh in further detail according to one embodiment of the invention. In this example, the display panel 120 is in a mixed display mode, in which a dynamic content 230 such as video is shown in a window while the rest of the display shows a static content 220 such as a page of text. The timing controller 118 of FIG. 1 sends a select signal to the scan driver 124 indicating a next row to be displayed, and sends a data signal to the data driver 122 containing the data line to be displayed. The select signal may specify a row index to directly address a row in the display panel 120 for refresh, or may prompt the scan driver 124 to select the next row in the sequential order. In an alternative embodiment, the timing controller 118 may also send an “activate/ignore” signal indicating whether the data signal presented to the data driver 122 is to be loaded into the display panel 120. In this alternative embodiment, the activate/ignore signal may be sent to the scan driver 124 alone, or to both the scan driver 124 and the data driver 122.

[0043] To select a row for refresh, the scan driver 124 may use a row select circuit 213 as shown in the illustrated embodiment. The row select circuit 213 may be a demultiplexer, which, based on an input row index, outputs an ON voltage to directly select one of the rows of the display panel 120. The row that is turned on is refreshed with the data line from the data driver 122. Alternatively, the row select circuit 213 may include a shift register. The shift register may contain at least the same number of bits as the number of rows in the display panel 120. The shift register may include one shift token of a bit value of one, and the rest of the bits in the shift register are zeros. It is understood that the bit values of one and zero are exemplary and may be reversed in certain embodiments. In this alternative embodiment, the select signal from the timing controller 118 prompts the shift register to shift the token to a next row in the sequential order to thereby sequentially select a row at a time. A high-definition (HD) display panel typically has thousands of rows and thousands of columns. A single shift register may not have enough number of bits for the entire row dimension. In one embodiment, the scan driver 124 may include multiple shift registers (e.g., S1, S2, S3, . . . , Sn), each for selecting one of the row segments (i.e., a subset of rows) in the display panel 120. The shift registers may be coupled to a demultiplexer, which directly selects one of the shift registers at a time. It is appreciated that other known hardware implementations for row selection may also be used and the arrangement described above is exemplary and not limiting.

[0044] In the example of FIG. 2, the scan driver 124 can refresh the rows at a higher rate for the dynamic content 230 and a lower rate for the static content 220. The scan driver 124 may directly select the rows of the dynamic content 230, directly select the segments that contain the dynamic content 230, sequentially clock through the rows until upper boundary of the dynamic content is reached, or otherwise select the dynamic content 230 by a combination of direct and sequential selection schemes using the row select circuit 213 as described above. In other scenarios where a row has changed content or reached a time out threshold, the scan driver 124 may select that row for refresh directly or by sequential selection using row select circuit 213 as described above.

[0045] FIG. 3A, FIG. 3B, FIG. 4A, FIG. 4B and FIG. 5 illustrate examples of timing diagrams of the content-based adaptive refresh according to embodiments of the invention. In these examples, it is assumed that the display panel has 1080 rows, and the default frame cycle is 16.67 ms, which corresponds to a default refresh rate of 60 Hz. It is to be appreciated that these numbers are chosen for illustration purposes; alternative embodiments may have different default frame cycles and default refresh rates. Each of the timing diagrams may be generated by the system, components and signals shown in FIG. 1 and FIG. 2. Row selection may be made using a demultiplexer and/or one or more shift registers, as described above with reference to the row select circuit 213.

[0046] In the example of FIG. 3A and FIG. 3B, it is assumed that rows R₃ and R₄ have no content change and have not reached a time out threshold in the current data frame that is to be displayed. As a result, no operations are performed in the line cycles of R₃ and R₄ (shown in dotted lines). That is, the line cycles of R₃ and R₄ are idle as R₃ and R₄ are not selected in these cycles for receiving data signals. FIG. 3B shows that the corresponding frame time stays the same from one frame cycle to the next. However, not all the rows are refreshed in every frame cycle. Thus, even though the default refresh rate is 60 Hz, not all of the rows are refreshed 60 times per second. The rows that do not have content change or have slow content changes can be refreshed at a slower rate, such as 40 Hz, 30 Hz or even less, depending on the electric properties of other circuit components in the display panel. The reduction in the refresh rate for a subset of the rows in the display panel can reduce power consumption of the display panel.

[0047] In the example of FIG. 4A and FIG. 4B, it is also assumed that rows R₃ and R₄ have no content change and have not reached a time out threshold. Thus, R₃ and R₄ are skipped from being refreshed for the current frame cycle. Instead of being idle in the line cycles of R₃ and R₄, the scan driver 124 proceeds to select the next rows following R₃ and R₄. As shown in FIG. 4A, the line cycle that was scheduled for refreshing R₃ is now used for refreshing R₅, which is the next row following R₃ and R₄ that has a content change. FIG. 4B shows that the corresponding frame time stays the same from one frame cycle to the next. However, similar to the example of FIG. 3B, the entire data frame is not refreshed in every frame cycle. For example, in the frame cycles shown in FIG. 4B, each frame time includes an active refresh period in which rows (but not necessarily all of the rows in the display panel) are refreshed in each line cycle, and an idle period in which no refresh operation is performed. This idle period enables power saving. In the example of FIG. 5, a data line is refreshed whenever it has a content change or reached a time out threshold, without following a fixed schedule. The timing controller 118 dynamically schedules the refresh on demand. If none of the rows have content change or have reached the time out threshold, no refresh operations are performed (shown in FIG. 5 as horizontal lines indicating idle periods). If a data line reaches its time out threshold when another data line is up for refresh due to changed content, the timed out data line may be serviced before the other data line.

[0048] Although not shown in the figures, in alternative embodiments, the time out refresh may be performed at fixed intervals (e.g., every five frame cycles). Between the two time out refreshes, only the data lines having content change beyond a threshold are refreshed. As some data lines may not be refreshed at all between the two consecutive time out refreshes, the elapsed time between the two time out refreshes for each data line cannot be greater than the time out threshold.

[0049] FIG. 6 is a circuit diagram of a pixel in an active matrix display 600 in accordance with an embodiment of the invention. The active matrix display 600 is meant to be one example of the display panel 120 shown in FIG. 1 and FIG. 2, and other types of active matrix displays are contemplated in accordance with embodiments of the invention. As illustrated, the scan lines V_select are oriented horizontally and a driven by one or more scan drivers, and data lines V_data are oriented vertically and are driven by one or more data drivers. In one embodiment, red, green, and blue light emitting micro LED devices 610 are placed between the scan lines and data lines. It is to be appreciated that in alternative embodiments, the micro LED devices 610 may be an LED device or an OLED device, and the active matrix display 600 may be based on light emission devices such as LED, OLED and LCD devices.

[0050] FIG. 7 is a circuit diagram of a subpixel with 2T1C circuitry in an active matrix display (e.g., the active matrix display 600 of FIG. 6) in accordance with an embodiment of the invention. In such an embodiment, the circuit includes a switching transistor T1, a driving transistor T2, a storage capacitor Cs and a micro LED device 610. The transistors T1, T2 can be any type of transistor such as a thin film transistor. For example, the switching transistor T1 can be an n-type metal-oxide semiconductor (NMOS) transistor, and the driving transistor T2 can be a p-type metal-oxide semiconductor (PMOS) transistor. The switching transistor T1 has a gate electrode connected to a scan line V_select and a first source/drain electrode connected to a data line V_data. The driving transistor T2 has a gate electrode connected to a second source/drain electrode of the switching transistor T1 and a first source/drain electrode connected to a power source V_dd. The storage capacitor Cs is connected between the gate electrode of the driving transistor T2 and the first source/drain electrode of the driving transistor T2. Alternatively, instead of connecting to the first source/drain electrode of T2, Cs may be connected to its own ground line, the anode electrode of the LED device 610, a previous/next row of the scan line, or other alternative locations in the circuitry. The micro LED device 610 has an anode electrode connected to a second source/drain electrode of the driving transistor T2 and a cathode electrode connected to a ground V_ss. Alternatively, the micro LED device 610 may have the cathode electrode connected to the second source/drain electrode of T2 and the anode electrode connected to V_dd, where V_dd in FIG. 7 is replaced by V_ss in this alternative embodiment. It is to be appreciated that the embodiment of FIG. 7 is exemplary and alternative embodiments may exist.

[0051] In operation, a voltage level scan signal turns on the switching transistor T1, which enables the data signal to charge the storage capacitor Cs. The voltage potential that stores within the storage capacitor Cs determines the magnitude of the current flowing through the driving transistor T2, so that the micro LED device 610 can emit light based on the current. It is to be appreciated that the 2T1C circuitry is meant to be exemplary, and that other types of circuitry or modifications of the traditional 2T1C circuitry are contemplated in accordance with embodiments of the invention. For example, more complicated circuits can be used to compensate for variations in the electrical properties of the driver transistor and the micro LED device, or for their instabilities.

[0052] FIG. 8 illustrates an alternative embodiment of a system 800 in which the adaptive refresh schemes described above may apply. The system 800 includes a processor (such as the processor 110 of FIG. 1) and a display controller 820. The display controller 820 controls a set of tile controllers 830 and each tile control 830 controls a tile 840. The tiles 840, which form the display panel 120, can be arranged into any geometric shapes, such as a line, a rectangle, a circle, and the like. An example of the tile 840, shown in FIG. 9A, includes an array of LED devices 950 (or micro LED devices). FIG. 9B illustrates an embodiment in which the LED devices 840 form a passive matrix which are arranged as rows and columns, where the rows are driven by a scan driver 920 and the columns are driven by a data driver 910. The scan driver 920 and the data driver 910 may be part of the tile controller 830 or may be coupled to the tile controller 830. FIG. 9C illustrates another embodiment in which each LED device 840 is directly driven by the tile controller 830 via a direct connection to the tile controller 830.

[0053] In the embodiment of FIG. 8, a tile that has content change beyond a threshold or has reached a time out threshold will be refreshed. As described above with reference to FIG. 1 and FIG. 2, the determination of whether a tile should be refreshed is made by the refresh logic 115. The refresh logic 115 and the frame buffer 130 may be located within the processor 110 or the display controller 820, or otherwise coupled to the processor 110 or the display controller 820. The refresh logic 115 may compare a tile of data in a current frame with a corresponding tile in a previous frame to determine whether the tile is to be refreshed in the current frame. The frame buffer 130 holds the tiles that are being compared. In alternative embodiments, instead comparing and refreshing one tile at a time, multiple tiles (“tile segment”) may be compared and refreshed at a time.

[0054] As described above, location of the refresh logic 115 and the signal it sends to its downstream component can affect the data transfer rate along the data path into the display panel 120. For example, if the refresh logic 115 is located within the processor 110, the processor 110 may send only the data that needs to be refreshed downstream, thereby reducing the data transfer rate along the path from the processor 110 to the display panel 120. If the refresh logic 115 is located with the display controller 820, the display controller 820 may send only the data that needs to be refreshed downstream, thereby reducing the data transfer rate along the path from the display controller 820 to the display panel 120. It is understood that for simplicity of the description, not all of the possible scenarios are described herein.

[0055] In some embodiments as shown in FIG. 9C, adaptive refresh may be performed on the individual LED devices 950 such that only those LED devices 950 that need a refresh are refreshed. Thus, some of the LED devices 950 may be refreshed while some of the other LED devices 950 within the same tile are not refreshed. The details of the refresh operations are similar to, or the same as, the operations described above with respect to the display system 100 of FIG. 1 and FIG. 2 and therefore are not repeated herein.

[0056] In utilizing the various aspects of this invention, it would become apparent to one skilled in the art that combinations or variations of the above embodiments are possible for performing adaptive refresh. Although the present invention has been described in language specific to structural features and/or methodological acts, it is to be understood that the invention defined in the appended claims is not necessarily limited to the specific features or acts described. The specific features and acts disclosed are instead to be understood as particularly graceful implementations of the claimed invention useful for illustrating the present invention.