JPS63285589A - Image data transfer - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image data transfer method for a graph ink display device. More specifically, there is an The present invention relates to a transfer method when a condition that a transfer source address and an (n-1)th transfer destination address match is satisfied.

[Summary of the invention]

The present invention provides a graphic display device that reads data read from a pixel on a frame buffer specified by a destination address in a previous process.
Image data can be transferred at high speed by providing a step to transfer the contents of the first register to the second register so that it can be used as pixel data on the frame buffer specified by the source address. It is.

[Conventional technology]

Conventionally, image data transfer within a frame buffer has been achieved by repeating the steps described below. Incidentally, FIG. 2 shows a flowchart of a conventional transfer method.

First step: Read data from the pixel on the frame buffer specified by the source address and store it in the first register.

Second step: reading data from pixels on the frame buffer specified by the destination address,
Store in the second register.

Third step: the first step. The data in the second register is subjected to a logical operation 1, and the result is written to the pixel on the frame buffer specified by the destination address.

When processing image data, accessing the frame buffer takes time. In the conventional method, each of the above steps accesses the frame buffer, so three accesses are required for each word transfer. The processing time for one word is shown in FIG.

In order to increase the speed of the conventional method described above, a device has been devised to perform read, arithmetic, and write operations for one address in one operation as a read-modify-write operation. As a result, the second and third steps described above are performed in one operation as shown in FIG. Therefore, the transfer of one word requires two operations: a read operation and a read/modify/write operation.
Requires action.

[Problem that the invention seeks to solve]

However, for the purpose of displaying graphics, it has been desired to be able to transfer data at higher speeds. In particular, it has been desired that screen scrolling, which is essential for graph ink display devices and is frequently used, should be responsive to user operations, that is, should be scrolled quickly. Therefore, the present invention provides the source address of the n-th transfer and n
- The purpose is to perform transfer processing at high speed when the first destination addresses match.

[Means for solving problems]

In order to solve the above problems, the present invention provides a method comprising the steps described below.

First step: Store the data held by the pixels on the frame buffer specified by the source address. this is,
This is done by directly storing the data stored in the previous second step.

Second step: Store the data held by the pixels on the frame buffer specified by the destination address.

Third step: Perform logical operation 1 on each data stored in the first step and second step, and write the result to the pixel on the frame buffer specified by the destination address.

〔Example〕

Embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a flowchart of the image data transfer part within the frame buffer in the program. This is a transfer in which the source address in the nth transfer and the destination address in the (n-1)th transfer match. Each step will be explained assuming that the nth transfer is performed.

First step: Register 1 stores pixel data on the frame buffer designated by the destination address of the n-1th transfer. This pixel is the pixel whose source address should be specified in the nth transfer, so
Move to register 2.

Therefore, there is no need to specify the source address for the frame buffer and read out the pixel data; the desired data is stored in the register 2.

Second step: Specify the destination address for the nth transfer.

Third step: The data of the pixel on frame buffer 1 specified by the address specified in the second step is stored in register 1.

Fourth step: Register I stores data for 11 pixels in the frame buffer specified by the destination address, and register 2 stores data for pixels on the frame buffer specified by the source address. is in a state. A logical operation is performed between the data in register 1 and the data in register 2.

Fifth step: Write the calculation result of the fourth step to the pixel on the frame buffer specified by the destination address. This means that the nth transfer is completed.

6th step: If all transfers have not been completed, the 1st step
Return to step.

Access to the frame buffer occurs in the third and fifth steps, and as described above, the read/modify/write operation allows access to the frame buffer in the third and fifth steps. The 4th and 5th steps can be performed as one operation.

By adopting this method, if the n-th source address and the (n-1)th destination address match, each word can be transferred to the frame buffer at a speed of 1 operation. The processing time diagram shown in FIG. 5 shows this state.

Next, FIG. 6 shows an example of the configuration of an apparatus that specifically implements the present invention. The explanation of FIG. 6 will be given below along the flowchart of FIG. In FIG. 6, one control CPU controls the entire operation of the apparatus including image data transfer.

First step: Transfer controller 2 instructs selector 5 so that the input to register 2 in 4 comes from register 1 in 3, and transfers the contents of register 1 to register 2. Control of registers 2 and 5 of 1°4, AI of 8, and U is controlled by 2 through control &i10.
This is done by the transfer controller.

Second step: The address generator at 6 informs the frame buffer at 7 of the destination address through the address bus 9.

Third step: Data is read from the pixel specified by the address notified in the second step and stored in register 1 of No. 3.

Fourth step: Set the data in register 1 and register 2 to 8
The ALU performs logical operations.

Fifth step 2: Write the calculation result of the fourth step to the pixel designated by the address notified to the frame buffer in the second step.

Sixth step: If it is determined that the control m c p u of 1 continues the transfer, the process returns to the first step.

As described above, with this configuration, the flowchart of the embodiment of the present invention shown in FIG. 1 can be realized by a specific device. In addition, in this configuration example, if the transfer controller 2 instructs the selector 5 to input to the register 2 in 4 from the frame buffer 7, image data can be transferred using the conventional method. You can also do that.

When the n-th source address and the (n-1)th destination match, screen scrolling, which is an essential function in a graphic display device, is transferred under this condition. Figure 7 shows an example of screen scrolling. Below, in the nth transfer, data 1 read from the pixel specified by address 1 on frame buffer 7 is transferred to the pixel specified by address 2. Explain when to do so. The description will proceed with reference to the flowchart shown in FIG. 1 and the configuration shown in FIG. 6. Since the destination address is address 1 in the third step of the n-1th transfer, data l is stored in register 1 (7-2), and is not processed until the sixth step 7-3), n Return to the first step of the second transfer.

First step: Data 1 is moved to register 2 (7
-4).

Second step: Address 2 is designated as the destination address.

Third. 4th. Fifth step: Data 1 stored in register 2 as data of the pixel specified by the source address is transferred to address 2, which is the destination address.
(5-.6.5-7).

According to this method, in the nth transfer, data 1
Since there is no need to read out from the frame buffer,
High-speed screen scrolling can be achieved.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to transfer image data within a frame buffer, which is frequently performed in which the source address of the nth transfer and the destination address of the (n-1)th transfer match. Can be processed at high speed. Since one read/modify/write operation is required to transfer one word, it is possible to achieve an unprecedented speedup. In particular, since it can be realized under conditions where the present invention can be applied,
This has the effect of speeding up screen scrolling, which is one of the functions most desired to speed up in a graphic display device.

[Brief explanation of drawings]

FIG. 1 shows a flowchart of image data transfer according to an embodiment of the present invention, and FIG. 2 shows a flowchart of image data transfer according to a conventional method. Fig. 3 is a transfer processing time diagram according to the conventional method, Fig. 4 is a transfer processing time diagram according to the conventional method when read/modify/write operations can be performed in one operation, and Fig. 5 is a transfer processing time diagram according to the present invention. , FIG. 6 is a configuration diagram showing an example of specifically configuring the present invention as a device, and FIG.
The figure is a diagram showing the movement of data during screen scrolling according to an embodiment of the present invention. The above picture 1 elephant day 7/'170-Nat Figure 1 Figure 2

Claims (1)

[Claims] When transferring image data, in which data is read from a pixel on the frame buffer specified by the source address and written to a pixel on the frame buffer specified by the destination address, the pixel on the frame buffer specified by the source address is The first step is to store the data held by the pixel on the frame buffer specified by the destination address, and the second step is to store the data held by the pixel on the frame buffer specified by the destination address. The third step consists of performing a logical operation on each of the data and writing the result to the pixel on the frame buffer specified by the destination address, and the first step is the same as that in the previous second step. , an image data transfer method is performed by directly storing the stored data, and the data is transferred at high speed when the nth source address and the n-1th destination address match. .