經濟部智慧財產局員工消費合作社印製 A7 44 1 8 "Printed by Employee Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs A7 44 1 8 "
4 4 3 5 t w f , d o c / 〇 〇 B ......—— - 五、發明説明（i‘） 本發明是有關於一種影像旋轉裝置與方法1且特別 是有關於一種以預取緩衝器（Prefetch buffer)及非顯示 記憶體（off-screen)達成影像旋轉之裝置與方法。 由於個人電腦的發展日新月異，許多的電腦週邊亦 相對的提供更爲人性化的功能’例如將顯示器由原來的 橫式改爲直立式，使得在顯示器上觀看文件就可以一次 看到更多的資料，減少上下文資料搜尋的麻煩，而更有 業者直接將液晶顯示螢幕製作成直立式，並且和個人電 腦結合成爲一體，而爲了要達成此功能，在硬體上就必 須將原來橫式顯示器上的資料轉換爲直式顯示器的資 料，來達成影像旋轉的目的。 習知影像旋轉有下列幾種方式： (1) 以軟體來達成影像旋轉，請參照第1A圖所 繪7K，此種方法利用所提供的驅動程式，當陰極射線管 60作影像旋轉時，以驅動程式載入電腦，利用中央處理 單元（CPU) 10來將橫式像素（pixel)顯示資料轉換爲 直式像素顯示資料，並將處理完的直式像素顯示資料經 由中央處理介面（Host Access) 20或者二維（2D)圖 形加速器30送至圖形記憶體（Display Memory ) 40，陰 極射線管控制器（Cathode-Ray Tube Controller，CRTC) 50則利用圖形記憶體4〇上的直式像素顯示資料，來控 制陰極射線管（CRT) 60顯示影像，進而達成影像旋轉 的目的。 (2) 以硬體來達成影像旋轉，請參照第1B圖所 3 E- m * nn ^ϋ( *一SJ c讀先閱讀背面之注意事項再填寫本頁) 本紙张尺度適用中國國家標準（CNS ) A4規格（210X297公釐） 4441 83 ^35twf.doc/0〇8 A7 B7 經濟』那智慧財產局R工消費合作杜印製 五、發明説明（了） 繪示，此種方法是將圖形顯示器17〇上的中央處理介面 Do以及二維圖形加速器〗30作硬體的修改，當中央處 理單元110將圖形資料經由中央處理介面120或者二維 圖形加速器130載入圖形記億體之前，必須經過中央處 理介面120修改的硬體電路180或者經過二維圖形加速' 器130修改的硬體電路190,此硬體電路會將橫式像素 顯示資料轉換爲直式像素顯示資料，並將直式像素顯示 資料載入圖形記憶體140,陰極射線管控制器150則利 用圖形記憶體140上的直式像素顯示資料，來控制陰極 射線管16〇顯示影像，進而達成影像旋轉的目的。 (3)以硬體來達成影像旋轉，請參照第1C圖所 繪示，此種方法是將圖形顯示器270上的陰極射線管控 制器250作硬體的修改，中央處理單元310將圖形資料 經由中央處理介面32〇或者二維圖形加速器330載入圖 形記憶體34〇,圖形記憶體34〇上的撗式像素顯示資料 再經過修改的硬體電路280,將橫式像素顯示資料轉換 爲直式像素顯示資料，並且控制陰極射線管360顯示影 像，進而達成影像旋轉的目的。 上述習知方法有以下的缺點： (1) 以軟體來達成影像旋轉，此方法簡單、具有 彈性且成本低，但是所有的運算必須全部仰賴中央處理 單元10來執行，造成系統的效能明顯的降低。 (2) 在寫入圖形記憶體11〇前，先經過中央處理 介面12〇修改的硬體電路18〇或者經過二維圖形加速器 n' HI ί ---- —·1· _·: m In ^^^1 n^i ^^^1^-SJ (請先閲讀背面之注意事項再填寫本1) 1 2 本紙張尺度適ϋϊΐ固家標準（CNS ) A4規格（210X297公釐) — 444183 4 35 .do c/0 0 8 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明（今） 130修改的硬體電路190來達成影像旋轉的目的，此方 法系統效能較以軟體達成影像旋轉佳，然而修改二維圖 形加速器的部分190更包括硬體以及軟體的修改，複雜 度非常的高。 (3)將陰極射線管控制器250加上硬體修改電路 280，此方法中央處理單元的負載不大而沒有系統效能 差的問題，但是顯示時因爲頻寬不足，以至於讀取速度 太慢，造成畫面品質太差_。 因此本發明係提出一種影像旋轉之裝置與方法，來 解決上述習知的缺點，在不減低系統效能下，同時維持 足夠大的顯示頻寬。 本發明提出一種影像旋轉之裝置，其包括一圖形記 憶體、一預取緩衝器、一非顯示記憶體以及一陰極射線 管控制器。圖形記憶體，用來接收像素顯示資料，並且 儲存像素顯示資料，以及送出像素顯示資料。預取緩衝 器，用來讀取圖形記憶體之連續位址，並將圖形記憶體 之連續位址轉換爲不連續位址，將這些不連續位址中所 儲存的像素顯示資料送出。非顯示記憶體，用來接收預 取緩衝器送出的像素顯示資料，並以非顯示記憶體之連 續位址來儲存這些像素顯示資料，並送出這些像素顯示 資料。陰極射線管控制器，用來讀取非顯示記憶體中的 像素顯示資料，將像素顯示資料轉換爲陰極射線管控制 訊號，傳至陰極射線管。 本發明提出一種以預取緩衝器及非顯示記憶體達成 5 本紙張尺度通用中ϋ家標準（CNS ) A4規格（210乂297公楚) " " ---------裝------訂 (請先閱讀背面之注意事項再填寫本頁) A7 B7 4435 t:wf.doc/008 五、發明説明（γ) 影像旋轉之方法，先檢查像素顯示資料是否經過更改， 若經過更改，則將像素顯示資料零新排列組合，將更新 後的像素顯示資料顯示出來。 爲讓¥發明之上述目的、特徵、和優點能更明顯易 懂，下文特舉較佳實施例，並配合所附圖式，作詳細說 明如下： 圖式之簡單說明： 第1A圖其所繪示爲習知以軟體作影像旋轉之方塊 圖； 第1B圖其所繪示爲習知更改中央處理介面以及二 維圖形加速器以達成影像旋轉之方塊圖； 第1C圖其所繪示爲習知更改陰極射線管控制器以 達成影像旋轉之方塊圖； 第2圖其所繪示爲本發明以預取緩衝器及非顯示記 憶體達成影像旋轉之裝置方塊圖.； 第3圖其所繪示爲預取緩衝器將圖形記憶體資料轉 換爲非顯示記憶體資料示意圖；以及 第4圖其所繪示爲本發明以預取緩衝器及非顯示記 憶體達成影像旋轉之方法流程圖。 標號說明： 10 中央處理單元 20 中央處理介面 30 二維圖形加速器 · 4〇 圖形記憶體 6 -裝------訂------1 (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 本紙張尺度適用中國國家標準（CNS > A4規格（2!〇X297々f)_ 444183 4435twf.doc/008 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明（（-) 5〇 陰極射線管控制器 60 陰極射線管 70 圖形顯示器 Π0 中央處理單元 120 中央處理介面 130 二維圖形加速器 140 圖形記憶體 150 陰極射線管控制器 160 陰極射線管 170 圖形顯示器 180 中央處理介面硬體修改電路區塊 190 二維圖形加速器硬體修改電路區塊 210 中央處理單元 220 中央處理介面 230 二維二維圖形加速器 240 圖形記憶體 250 陰極射線管控制器 260 陰極射線管 270 圖形顯示器 280 陰極射線管控制器硬體修改電路區塊 310 中央處理單元 320 中央處理介面 330 二維二維圖形加速器 340 圖形記憶體 (請先閲讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準（CNS > A4規格（210X297公釐） 444 1 83 A7 B7 4 4 3 5 t w f . d 〇 c / 0 0 8 五、發明説明（G) •装------訂 (锖先閱讀背面之注意事項再填寫本頁) 350 陰極射線管控制器 360 陰極射線管 370 圖形顯示器 380 預取緩衝器 390 非顯示記憶體 410 圖形記憶體 420 圖形記憶體所對應的陰極射線管顯示之影像 位置 430 預取緩衝器 440 非顯示記憶體 450 非顯示記憶體所對應的陰極射線管顯示之影 像位置 第一實施例 第2圖爲依照本發明一較佳實施例，其繪示爲一種 影像旋轉之裝置方塊圖。該裝置至少包括圖形記憶體 34〇、預取緩衝器380、非顯示記憶體390以及陰極射線 管控制器350。 經濟部智慧財產局員工消费合作社印製 當中央處理單元3 10將圖形資料送入圖形顯示器 37〇要求顯示時，中央處理介面320會接收圖形資料並 作點對點的轉換，產生像素顯示資料，再將像素顯示資 料儲存至圖形記憶體340,而二維二維圖形加速器330 亦會接收圖形資料並作線對線的轉換，產生像素顯示資 料，再將像素顯示資料儲存至圖形記憶體34〇。當中央 處理介面320及二維二維圖形加速器330將像素顯示資 8 本紙張尺度適用中國國家標準（CNS ) Α4規格（2丨0Χ297公釐） ~ " 44叫 83 經濟部智慧財產局員工消費合作社印製 A7 B7 五、發明説明（?） 料儲存至圖形記憶體340後，在圖形記憶體340內的像 素顯示資料即爲橫式像素，顯示資料·當預取緩衝器380’ 檢查出圖形記憶體340有更新，則預取緩衝器380會讀 取多個圖形記憶體340的連續位址/並將此連續位址重 新計算並排列組合，成爲多個不連續位址，並將這些不 連續位址所儲存的像素顯示資料讀出，將這些像素顯示 資料以非顯示記憶體390的連續位址來存入，儲存完畢 後，則在非顯示記憶體39〇內的像素顯示資料即爲直式 像素顯示資料。陰極射線管控制器350則根據非顯示記 憶體390所儲存的像素顯示資料，轉換爲陰極射線管控 制訊號，送至陰極射線管，達成影像旋轉的目的。而預 取緩衝器380所作的轉換方式將在以下作進一步的描 述。 請參照第3圖其所繪示爲預取緩衝器430將圖形記 憶體42〇所儲存的像素顯示資料轉換爲非顯示記憶體440 所儲存的像素顯示資料示意圖。圖形記憶體410內的中 文數字代表該圖形記憶體410的位址，圖形記憶體41〇 內的阿拉伯數字代表該圖形記憶體410的像素顯示資 料’舉例來說，中文數字一十九代表圖形記憶體41〇的 位址，而阿拉伯數字I9代表第19筆像素顯示資料。陰 極射線管420由左至右由上到下依序顯示圖形記憶體中 的像素顯示資料，其爲橫式顯示。預取緩衝器430爲4 X 4緩衝器，可用以處理位址的轉換，預取緩衝器43 〇 一次會將4筆連續位址載入，並載入在同一列上，共要 9 本^張尺度適用中國國家樣準（CNS > A4規格（210X297公釐> —" — ----^------裝------訂 (請先閱讀背面之注意事項再填跨本頁) 444 1 83 4435t vjf.doc/008 五、發明説明（# ) (請先閱讀背面之注意事項再填寫本頁) 載入4次，亦即4列共16筆位址都在緩衝器上，而預 取緩衝器43 0上的位址是以送出一行爲一次，故原來以 連續位址載入的4筆位址則會變爲不連續，以每次送出 的不連續位址讀取該位址所儲存的像素顯示資料，寫入 非顯示記憶體440中的連續位址，共寫入4次。重複上 述的動作，直到所有在圖形記憶體410上的像素顯示資 料轉換至非顯示記憶體440上的像素顯示資料才算完 成。而陰極射線管450根據非顯示記憶體440上的像素 顯示資料所顯示的即爲直式顯示。 本發明之一種影像旋轉之動作過程可用第4圖之流 程圖來說明。首先，在步驟510中，由預取緩衝器檢查 圖形記憶體中的像素顯示資料是否更動，若未更動則跳 至步驟55〇，若有更動則跳至步驟52〇。接著步驟520， 將圖形記憶體中的連續位址取出並經轉換爲不連續位 址。在下來執行步驟53〇，取出不連續位址中所儲存的 像素顯示資料。在步驟540中，以非顯示記憶體的連續 位址儲存像素顯示資料。而後，根據步驟550，利用非 顯示記憶體的像素顯示資料來顯示影像。 經濟部智慧財產局員工消費合作社印製 因此，本發明的優點係大幅的提高了系統的效能， 並且圖形顯示器中的中央處理介面、二維二維圖形加速 器以及陰極射線管控制器的電路都不必經過修改。 本發明的另一優點係以預取緩衝器來作位址的轉 換，並且轉換後的像素顯示資料是以陰極射線管控制器 所要的格式儲存，解決了習知速度不夠快，頻寬不足的 本紙張尺度適用中國國家標準（CNS ) A4規格UlOxm公慶） 444183 4435twf-doc/008 A7 B7 五、發明説明（f) 問題。 綜上所述，雖然本發明已以較佳實施例揭露如上， 然其並非用以限定本發明，任何熟習此技藝者，在不脫 離本發明之精神和範圍內，當可作各種之更動與潤飾， 因此本發明之保護範圍當視後附之申請專利範圍所界定 者爲準。 --------裝------訂 (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員4消費合作社印製 本紙張尺度適用中國國家標準（CNS ) A4规格（2丨0父297公釐）4 4 3 5 twf, doc / 〇〇B ......——-5. Description of the Invention (i ') The present invention relates to an image rotation device and method 1 and in particular to a prefetch buffer Device and method for achieving image rotation with a Prefetch buffer and off-screen memory. Due to the rapid development of personal computers, many computer peripherals also provide more user-friendly functions, such as changing the display from the original horizontal to the vertical, so that you can see more information at a time when viewing files on the display , Reducing the trouble of searching for contextual data, and more professionals directly make the LCD screen into a vertical type and integrate it with a personal computer. In order to achieve this function, the original horizontal display on the hardware must be The data is converted into data of a straight display to achieve the purpose of image rotation. There are several ways for image rotation: (1) Image rotation is achieved by software. Please refer to 7K as shown in Figure 1A. This method uses the provided driver program. When the cathode ray tube 60 performs image rotation, The driver is loaded into the computer, and the central processing unit (CPU) 10 is used to convert the horizontal pixel display data to the vertical pixel display data, and the processed vertical pixel display data is processed through the central access interface (Host Access). 20 or two-dimensional (2D) graphics accelerator 30 is sent to the graphic memory (Display Memory) 40, and the cathode-ray tube controller (CRTC) 50 uses the straight pixels on the graphics memory 40 to display data To control the cathode ray tube (CRT) 60 to display the image, thereby achieving the purpose of image rotation. (2) Use hardware to achieve image rotation. Please refer to Figure 1B. 3 E-m * nn ^ ϋ (* One SJ c read the precautions on the back before filling this page) This paper size applies Chinese national standards ( CNS) A4 specification (210X297 mm) 4441 83 ^ 35twf.doc / 0〇8 A7 B7 Economics ”The Intellectual Property Bureau R Industrial Consumption Cooperation Du Printed 5. Description of the Invention (The) This method is a graphic The central processing interface Do on the display 17 and the two-dimensional graphics accelerator 30 are modified in hardware. Before the central processing unit 110 loads the graphic data into the graphics memory through the central processing interface 120 or the two-dimensional graphics accelerator 130, it must be The hardware circuit 180 modified by the central processing interface 120 or the hardware circuit 190 modified by the two-dimensional graphics accelerator 130. This hardware circuit converts horizontal pixel display data into straight pixel display data and converts the straight pixel display data. The pixel display data is loaded into the graphic memory 140, and the cathode ray tube controller 150 uses the straight pixel display data on the graphic memory 140 to control the cathode ray tube 160 to display the image, thereby achieving the purpose of image rotation. . (3) Use hardware to achieve image rotation. Please refer to Figure 1C. This method is to modify the cathode-ray tube controller 250 on the graphics display 270 as a hardware modification. The central processing unit 310 passes the graphics data through The central processing interface 32o or the two-dimensional graphics accelerator 330 is loaded into the graphics memory 34o, and the pixel-type display data on the graphics memory 34o is then modified by the hardware circuit 280 to convert the horizontal pixel display data into a straight-type The pixels display data and control the cathode ray tube 360 to display the image, thereby achieving the purpose of image rotation. The above-mentioned conventional method has the following disadvantages: (1) The software is used to achieve image rotation. This method is simple, flexible and low cost, but all operations must be performed by the central processing unit 10, which results in a significant decrease in system performance. . (2) Before writing to the graphics memory 110, the hardware circuit 18 modified by the central processing interface 12 or the two-dimensional graphics accelerator n 'HI ί ---- — · 1 · _ :: m In ^^^ 1 n ^ i ^^^ 1 ^ -SJ (Please read the precautions on the back before filling in this 1) 1 2 This paper is suitable for the standard of the household (CNS) A4 (210X297 mm) — 444183 4 35 .do c / 0 0 8 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. 5. Description of the invention (today) 130 Modified hardware circuit 190 to achieve the purpose of image rotation. The rotation is good, however, the part 190 for modifying the two-dimensional graphics accelerator further includes hardware and software modifications, and the complexity is very high. (3) The cathode-ray tube controller 250 is added with a hardware modification circuit 280. In this method, the load of the central processing unit is not large and there is no problem of poor system performance. However, the display speed is too slow due to insufficient bandwidth. , Resulting in poor picture quality. Therefore, the present invention proposes a device and method for image rotation to solve the above-mentioned shortcomings, while maintaining sufficient display bandwidth without reducing system performance. The invention provides an image rotation device, which includes a graphics memory, a pre-fetch buffer, a non-display memory, and a cathode ray tube controller. Graphics memory is used to receive pixel display data, store pixel display data, and send pixel display data. The prefetch buffer is used to read the continuous addresses of the graphic memory, convert the continuous addresses of the graphic memory into discrete addresses, and send the pixel display data stored in these discrete addresses. Non-display memory is used to receive the pixel display data sent by the pre-fetch buffer, and store these pixel display data with the continuous address of the non-display memory, and send these pixel display data. The cathode ray tube controller is used to read the pixel display data in the non-display memory, convert the pixel display data into a cathode ray tube control signal, and transmit it to the cathode ray tube. The present invention proposes a universal paper standard (CNS) A4 specification (210 乂 297 cm) with a prefetch buffer and non-display memory to achieve 5 paper sizes. &Quot; " --------- install ------ Order (Please read the precautions on the back before filling this page) A7 B7 4435 t: wf.doc / 008 V. Description of the invention (γ) For the method of image rotation, first check whether the pixel display data has been changed If it is changed, the pixel display data is arranged in a new arrangement, and the updated pixel display data is displayed. In order to make the above-mentioned objects, features, and advantages of the invention more obvious and easy to understand, the following specific examples are given in conjunction with the accompanying drawings to make a detailed description as follows: Brief description of the drawings: Figure 1A Shown as a block diagram of image rotation using software; Figure 1B is shown as a block diagram of changing the central processing interface and a two-dimensional graphics accelerator to achieve image rotation; Figure 1C is shown as a conventional block diagram. Change the block diagram of the cathode ray tube controller to achieve the image rotation; Figure 2 shows the block diagram of the device for pre-fetch buffer and non-display memory to achieve the image rotation; Figure 3 shows the block diagram Schematic diagram of converting graphics memory data into non-display memory data for the pre-fetch buffer; and FIG. 4 shows a flowchart of a method for image rotation using the pre-fetch buffer and non-display memory according to the present invention. Explanation of symbols: 10 central processing unit 20 central processing interface 30 two-dimensional graphics accelerator 4 graphics memory 6 -installation -------- order ----- 1 (please read the precautions on the back before filling in this Page) Printed by the Intellectual Property Bureau's Employees' Cooperatives of the Ministry of Economic Affairs This paper is printed in accordance with Chinese national standards (CNS > A4 size (2! 〇X297々f) _ 444183 4435twf.doc / 008 A7 B7 Printed 5. Description of the invention (-) 50 cathode ray tube controller 60 cathode ray tube 70 graphics display Π0 central processing unit 120 central processing interface 130 two-dimensional graphics accelerator 140 graphics memory 150 cathode ray tube controller 160 cathode ray Tube 170 Graphic display 180 Central processing interface hardware modification circuit block 190 Two-dimensional graphics accelerator hardware modification circuit block 210 Central processing unit 220 Central processing interface 230 Two-dimensional two-dimensional graphics accelerator 240 Graphic memory 250 Cathode ray tube controller 260 cathode ray tube 270 graphic display 280 cathode ray tube controller hardware modification circuit block 310 central processing unit 320 central Processing interface 330 2D 2D graphics accelerator 340 graphics memory (please read the precautions on the back before filling this page) This paper size applies to Chinese national standards (CNS > A4 size (210X297 mm) 444 1 83 A7 B7 4 4 3 5 twf. D occ / 0 0 8 V. Description of the Invention (G) • Equipment ---- Order (锖 Please read the notes on the back before filling this page) 350 Cathode Ray Tube Controller 360 Cathode Ray Tube 370 Graphic display 380 Prefetch buffer 390 Non-display memory 410 Graphic memory 420 Image position of cathode ray tube corresponding to graphic memory 430 Prefetch buffer 440 Non-display memory 450 Non-display memory The first embodiment of the position of the image displayed by the cathode-ray tube. Figure 2 is a block diagram of a device for image rotation according to a preferred embodiment of the present invention. The device includes at least a graphics memory 34 and a prefetch buffer. 380, non-display memory 390 and cathode ray tube controller 350. Printed by the Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs when the central processing unit 3 10 sends the graphic data to the graphic display 37. When displaying is requested, the central processing interface 320 will receive graphic data and perform point-to-point conversion to generate pixel display data, and then store the pixel display data in the graphics memory 340, and the two-dimensional and two-dimensional graphics accelerator 330 will also receive the graphic data. The line-to-line conversion is performed to generate pixel display data, and the pixel display data is stored in the graphics memory 34. When the central processing interface 320 and the two-dimensional and two-dimensional graphics accelerator 330 display pixels, the paper size is applicable to the Chinese National Standard (CNS) A4 specification (2 丨 0 × 297 mm) ~ " 44 called 83 employees of the Intellectual Property Bureau of the Ministry of Economic Affairs for employee consumption Printed by the cooperative A7 B7 V. Description of the invention (?) After the material is stored in the graphic memory 340, the pixel display data in the graphic memory 340 is the horizontal pixel. The display data. When the prefetch buffer 380 'checks the graphic If the memory 340 is updated, the prefetch buffer 380 will read the consecutive addresses of multiple graphics memories 340 / and recalculate and arrange the consecutive addresses to form a plurality of discontinuous addresses. The pixel display data stored in the continuous address is read out, and these pixel display data are stored in the continuous address of the non-display memory 390. After the storage is completed, the pixel display data in the non-display memory 390 is Vertical pixel display data. The cathode-ray tube controller 350 converts the pixel-display data stored in the non-display memory 390 into a cathode-ray tube control signal and sends it to the cathode-ray tube to achieve the purpose of image rotation. The conversion method performed by the prefetch buffer 380 will be described further below. Please refer to FIG. 3, which is a schematic diagram of converting the pixel display data stored in the graphics memory 42 into the pixel display data stored in the non-display memory 440 as the prefetch buffer 430. The Chinese numbers in the graphic memory 410 represent the address of the graphic memory 410, and the Arabic numbers in the graphic memory 410 represent the pixel display data of the graphic memory 410. For example, the Chinese number 19 represents the graphic memory The address of the body 41〇, and the Arabic numeral I9 represents the 19th pixel display data. The cathode ray tube 420 sequentially displays the pixel display data in the graphic memory from left to right and from top to bottom, which is a horizontal display. The prefetch buffer 430 is a 4 X 4 buffer, which can be used to handle address conversion. The prefetch buffer 43 will load 4 consecutive addresses at a time and load them on the same row. A total of 9 copies are required. ^ The scale is applicable to the Chinese national standard (CNS > A4 specification (210X297 mm > — " — ---- ^ ------ installation ------ order) (please read the note on the back first) Matters are refilled across this page) 444 1 83 4435t vjf.doc / 008 V. Description of the invention (#) (Please read the notes on the back before filling this page) Loaded 4 times, that is, 4 columns with a total of 16 addresses Are all on the buffer, and the address on the prefetch buffer 43 0 is sent once, so the 4 addresses originally loaded as consecutive addresses will become discontinuous, Consecutive addresses read the pixel display data stored at this address and write the consecutive addresses in the non-display memory 440 for a total of 4 times. Repeat the above operation until all the pixels on the graphics memory 410 are displayed. Data conversion to the pixel display data on the non-display memory 440 is completed. The cathode ray tube 450 displays the data based on the pixels on the non-display memory 440. The display is a straight display. An image rotation operation process according to the present invention can be illustrated by the flowchart in FIG. 4. First, in step 510, the prefetch buffer is used to check whether the pixel display data in the graphics memory is displayed. If there is no change, go to step 55. If there is a change, go to step 52. Then, in step 520, take out the consecutive addresses in the graphics memory and convert them into discontinuous addresses. Then go to step 53. 〇, take out the pixel display data stored in the discontinuous address. In step 540, store the pixel display data with the continuous address of the non-display memory. Then, according to step 550, use the pixel display data of the non-display memory to Printed image. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economics. Therefore, the advantages of the present invention are greatly improved system performance, and the central processing interface in the graphic display, the two-dimensional two-dimensional graphics accelerator, and the cathode-ray tube controller. The circuit does not need to be modified. Another advantage of the present invention is that the prefetch buffer is used for address conversion, and the converted image The plain display data is stored in the format required by the cathode ray tube controller, which solves the problem that the learning speed is not fast enough and the bandwidth is insufficient. The paper size is applicable to the Chinese National Standard (CNS) A4 specification UlOxm official celebration 444183 4435twf-doc / 008 A7 B7 V. Explanation of the Invention (f) Problem. In summary, although the present invention has been disclosed as above with the preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art will not depart from the spirit and scope of the present invention. Within the scope, various modifications and retouching can be made, so the scope of protection of the present invention shall be determined by the scope of the appended patent scope. -------- Installation ------ Order (please Please read the notes on the back before filling this page) Member of the Intellectual Property Bureau of the Ministry of Economic Affairs 4 Consumption Cooperatives Printed on this paper The size of the paper applies to the Chinese National Standard (CNS) A4 specification (2 丨 0 parent 297 mm)