1271693 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種液晶驅動系統中的過度驅動電壓產生 方法，特別疋關於一種改善記憶體頻寬不足及色彩失直間題 的過度驅動電壓產生方法。 【先前技術】 液晶顯示器主要的優點在於容易達到高顯示 顯示器薄型化，因此廣為使用於筆記型電腦之中，近來隨著 大的持咖發，液晶顯示器也漸漸在桌上型電腦顯 席之地。然而’在許多廠商欲將液晶顯示技術 ^ =機產品，企圖以液晶電視取代傳統的陰極射線電視 守，液曰曰顯不技術在動態影像的表現上顯露出其技術瓶頸。 請參照圖-、，圖-為液晶顯示器10之基本圖。其中，液 j板12具有複數個像素單元❻㈣⑵，藉由對每一個 素单几121中驗晶分子辭特定的電壓 分巧旋轉角度，如此-來，液晶面板12下]方了的^^曰: 所提供之光線，在液晶面板12不_像素單元121中备有 =穿J率，而陣列的像素單元121則進*交織組成备H ，者觀看W晝面(frame)。若是更細微的區分，則每 早7L 121係可區分為三個次像素（sub_pixd) 121R、^加以 ^分別執掌每一個像素單元121中的紅色、綠色以 而如上所，液晶顯示技術在㈣影像的表現上 技術瓶頸’其主要原因正與液晶分子 = 為液晶分子接收到上述特料電壓之後，需要寺—:應· 1271693 間，以從一原先的角度旋轉至另一角度(與特定的電壓相對應 的角度）。而在顯示動態影像時，液晶分子的反應時間趕不^ 畫面的更新速率，造成顯示品質的延遲或失真缺點。因此， 可縮短液晶分子的反應時間，係為液晶顯示技術中的重要 課題。 〔液晶過度驅動技術（TFToverdrive)」係為目前經常使 用來縮短液晶反應時間的一種方法，其原理係藉由刻意地施 予液晶分子一過高（或過低)的電壓，使得液晶分子在預定的 時間間隔内旋轉至所需角度。以下將由圖二來說明液晶過度 驅=技術，圖二中縱軸係代表液晶分子旋轉之角度，橫軸係 為時間軸；如圖二所示，當施予液晶單元一控制電壓Vcl時， ，晶分子需要經過時間t2才得以由角度心旋轉至與控制電 壓Vcl相對應之角度，而液晶過度驅動技術係先行施予液 晶單元一過高之過度驅動電壓v〇D (v〇D係為角度62之控制 電^)，使得液晶分子提早在時間tl時即旋轉至角度Si，而 後續再將電壓調回為與角度心相對應之控制電壓Vcl，如此 一來則可縮短液晶之反應時間。 ， 然而，除了上述施予液晶分子一過高的電壓之外，液晶 過度驅動技術中亦存在有施予液晶分子一過低電壓的情況， 例如要使液晶分子由角度&反向旋轉至角度0。即為此狀 况，且圖一僅代表欲將液晶分子由角度θ。旋轉至角度θ !的 情形，實際操作液晶顯示器時，包含有更多的角度變化情況。 因此丄在液晶過度驅動技術中，需要不斷的將正要播放的晝 巧一個晝面進行訊號的比較，才能計算出適合的過度驅 ，，壓值(value of V0D);例如圖二中過度驅動電壓（v〇d)值的 言:算’必^將角度(屬於第二晝面中之一液晶分子)與角度 仏（屬於第-晝財之該液晶分子)進行比較，才能計算出合 7 1271693 理的過度驅動電壓值。如此說來，液晶過度驅動技術的原理 雖然簡單(利用過高或過低之電壓來縮短液晶反應時間），但 實施時卻有相當複雜的訊號比較與計算程序。 請參照圖三，圖三為習知液晶驅動系統示意圖，由圖三 ^以下之說明可進一步了解習知的液晶驅動系統與其中的液 晶過度驅動技術。液晶驅動系統20包括一處理器21以及一 記憶體26，液晶驅動系統20接收晝面資料流22之後，可在 内部進行訊號之處理與計算，而產生驅動訊號流24來驅動液 晶顯示器中的液晶分子。 。^由圖二中可見，驅動訊號流24之中，每一個畫面驅動訊 號係包括一過度驅動電壓v0D以及一控制電壓;例如， 第二晝面驅動訊號242係包括第二過度驅動電壓242V〇du 及第二控制電壓242Vcl。 ^當晝面資料流22之中的第二晝面資料222輸入液晶驅動 系統20時，處理器21會將第二晝面資料222分派至記憶體 26,由記憶體26將第二晝面資料222暫存起來，以成為^二 參考資料232。 ' 一 並且，處理器21係會將第二晝面資料222與先前已 於記憶體26中的第一參考資料231進行比較，並利用上述液 晶過度驅動技術的原理來計算出上述的第二過度驅動電壓 242VOD。 .至於上述的第二控制電壓242VCL則不需經過相鄰書 資料的比較程序’處理器21可根據第二晝面資料222 ; 計算出第二控制電壓242Va。 如此說來，接續地暫存於記憶體26中的各個查 (第-參考資料23卜第二參考資料说·· ·. ·)可被視J C 貪料流(reference data Stream)23，其用途係使液晶驅動^ 20在產生過度驅動電壓時有參考的依據。 7尔為 1271693 但’液晶過度驅動技術應用在高顯示解析度之影像 在液晶驅動系統20中出現了訊號寬度不足的問題。如圖二’ 示，記憶體26在暫存晝面資料時，一般而言最高只 位元(bit)做暫存的動作，換句話說，處理器21與記憶 之間的訊號寬度最寬只有32位元(bit)，雖然在 (640x480)或SVGA (800x600)等較低顯示解度的产 下，這樣的訊號寬度尚且足夠，但若是要顯示sjG (1280x1024)或是更高顯示解析度的晝面，則需要48位 或更寬的訊號寬度了。 ^ 面對此訊號寬度不足的問題，當然亦可直接發展或採用 訊號寬度更寬（例如64位元）的記憶體，然而其成本在目前 仍然偏高，且屬於記憶體相關之技術領域。而在關於本發= 所屬之液晶驅動系統技術領域之中，習知面對上述問題^做 法是將紅、藍、綠三原色的訊號各捨棄掉部分較後段的位元， 茲說明如下： 在低顯示解析度的情況下，處理器21在同一時序（cl〇ck) 接收並處理“一個”像素單元的像素資料，俗稱單通道G channel);舉例而言，VGA (640x480)之中一條掃瞄線需 要640個時序（ci〇ck)以處理完成。然而，高顯示析度時處 理器21在同一時序（cl〇ck)則接收並處理“二個，，像素單 元的像素資料，俗稱雙通道；以在例如SXGA (1280x1024) 的高顯示解析度時，可同樣在640個時序（ci〇ck)處理宗成 一條掃猫線的資料。 高顯示解析度48位元的訊號寬度係由雙通道(2 channels) 各24位元所組成’個別單通道的24位元係由紅、藍、綠各 8位元所組成，但是處理器21與記憶體26之間的訊號寬度 1271693 , /°’因此習知技術其巾之—種做法是將紅色訊號中 =後的2位讀棄，並且捨棄藍色以及綠色訊號中最後的3 而使得單通道紅、藍、、綠㈣+8)，位元的訊號縮 ^6+5+5)=16位元的訊號，如此—來在合併雙通道後才 可付合32位元的訊號寬度。 卷-ί彳于―提的是，上述將紅色訊號最後2位元捨棄，並捨 >、監，及綠色訊號最後3位元的習知作法僅是—種習知實施 Π其它亦有取紅、藍、綠各(5、5、5)位元的做法，或（5、 I、/ # 、5、6)······等，其目的僅在於捨棄部分較後段位元， 付=§fl錢度。以將原本8位元之紅色訊號捨棄最後的2 ，兀為例，原本8位元會有256色（紅色之256個灰階色）， 鈿減成為6位元後則只剩下64色（紅色之64個灰階色）， 因此習知作法很明顯的導致了色彩失真。 接著藉由圖四以加強說明上述習知技術的色彩失直 ，晝面資職為例，當記憶體26的訊=以 :，第一晝面資料222會被捨棄其中部份位元之資料，而形 、負料量較士的第二爹考資料232以暫存在記憶體％之 ,。如圖四所示，第二畫面資料222係包括複數個像素資料 (f xel data) 222p，而第二參考資料232係包括複數個來考像 素，料（referenCepixddata)232P。如上所述高顯示解析度時 矾號寬度為48位元，圖四中以像素資料222P中具有48個位 元點(標號25)來表示’而因為記憶體46的訊號寬度通常口有 32位元，所以如圖四所示，每一個參考像素資料中只 剩下32個位元點(標號25)。因此，由圖四中可以看出，由^ 記憶體26的訊號寬度不足，因此使得第二畫面資料222 j 個像素資料222ρ都被不完全地暫存，而成為不完整來者 像素資料232ρ ;其所帶出的影響擴及參考資料流23(圖^中 1271693 每一個晝面(frame)與每一個像素(pixe〇。 但習知技術中不甚重視此色彩失真缺點的理由在於—如 圖二所7F，色衫產生失真的資料是暫存於記憶體26中的第 一、第二參考資料231、232……等，即參考資料流23 (reference data stream)，參考資料流23係用來與正要播放的 晝面資料流22進行比較’所以，色彩失真的問題發生在參考 資料流23上;輸入液晶驅動系統2〇的晝面資料流22與輸出 之驅動訊舰24皆沒有峨寬度不足_題。·，^知 技術中並不甚重視此色彩失真缺點的嚴重性；麸事 明^考資料流的，失敍會表_液晶顯示^整體之顯 二嘈上Ξ太ίΪί影ί之中’影像之色彩失真係可被肉眼 戶^硪，原本希望能以高顯讀析度播放明_示品質的 晝面，反而產生了色彩失真的缺點。 ⑽Πί何改善域寬料足所帶麵影像色彩失真問 S解曰益增力的情況下，如何 人員所辆咐賴’輸_域的研發 【發明内容】 驅動ΐΐϊίϊΓ。的在於提供—種液晶驅動系統中的過度 度下動系統在高顯示解析 ^明之另—目的在於不增加記憶體硬 改〇液日日驅動系統的色彩失真缺點。 11 1271693 本發明係提供m轉系統巾之過度驅動電壓產生 j认液晶驅動系統係包括-處理器以及—記憶體，處理器 晶1f動系統之前—晝面資料暫存至記憶體，且處 雷Ϊ 3=!^晝面資料與前—晝面資料，而產0度驅動 中之液晶分子加速旋轉至所需角 ΐ液曰Ί Gif料皆包括了複數筆像素資料，以提 tr 佈之複數個像素單元進行顯像。本發 像ii 峰素單元與第二紐 象素早7G。本發明之過度驅動電壓產生方法包括下列步驟： ⑻將前一晝面資料中對應於該第一組像辛單亓之笸一 、。且像素資料完整地暫存至該記憶體。、、’、 料進麵巾之第—組像素資 r、々子±乂°十异出現恰晝面資料的過度驅動電壓。 後ιί將斯晝面㈣巾對應於第二組像素單元之第-έ 象素資料完整地暫存至該記憶體。 ’、 一、、且 (d)將下-晝面資料與暫存於記 貝料進行輯，崎料下—畫㈣料象素 =有多ίί』實iij早分j方 的像素單元能夠財其財完整的/元被暫存 灸右有更局規格的顯 1271693 示解析度的需求時’本發明可應用於相關的記憶體訊號寬度 不足情況。 【實施方式】 /請參照圖五，圖五為一液晶顯示器及其連接之一液晶驅 動系統示意圖。液晶驅動系統50係根據一晝面資料流52， 以提供一驅動訊號流54來驅動液晶顯示器60中的液晶分 子。驅動訊號流54之中，每一個畫面驅動訊號係包括一過度 驅動電壓v0D以及一控制電壓VcL。以第二晝面驅動訊號542 為例，其係包括第二過度驅動電壓542v〇d以及第二控制 壓 542VCL。 液晶驅動系統50包括一處理器51與一記憶體56。當晝 面資料流52之中的第二晝面資料522輸入液晶驅動系^ 犄，處理态56會將第二晝面資料522分派至記憶體56，由 =己憶體56將第二晝面資料522暫存起來，以作為第二參考資 料532。而處理器51係會將第二晝面資料522與先前^暫存 於記憶體56中的第一參考資料531進行比較，並計算出第二 過度驅動電壓542VOD。至於第二控制電壓542Vcl則不需經 過相鄰晝面資料的比較程序，處理器51可根據第二晝面資& 522而直接計算出第二控制電壓542Vcl。其中，接續地暫存 至記憶體56的複數個參考資料（例如第一、第二參考資料 53卜532···等）可視為一參考資料流53。 、 習知$術所遭遇的問題是：若欲播放高顯示解析度的影 像（處理器51同一時序接收並處理“二個，，像素單元的像 素資料一雙通道，48位元），則會因為記憶體56在暫存畫面 資料時只能以32位元(bit)做暫存的動作，因此一旦參考g料 々iL 53的況號i度大於32位元，將造成參考資料流％所代表 13 1271693 2彩產生ΐ真，而致使驅動訊號流54中的過度驅動電壓 弟一、弟—過度驅動電塵541 V〇D、542 V〇D...等）也產 生色彩失真的缺點。 ，繼績參照圖五，本發明所提供的過度驅動電壓產生方 ^，係用以產生過度驅動賴以軸液晶· ^⑹中之液晶 ^子加速旋轉至所需肖度，並改善習知技射色彩失真的缺 二。如圖五所不，液晶顯示器6〇的液晶面板泣中，係具有複 j個陣列分2之像素單元;本發明將該複數個像素單元分組為 弟-組像素單TC62A與第二組像素單元細。而本發明之過度 驅動電壓產生方法之步驟係敘述如下： 步驟6〇ι:暫存前-晝面資料（the previ〇us fr_如⑻中 對應於該第-組像素單元62A之第—組像素資料，其中每一 個像素資料係被完整地儲存。 步，603 ·將現時晝面資料（tJle current如⑽如拉）與暫 存之该第一組像素資料進行比對，以計算現時晝面資料的過度 驅動電壓。 一、 —步驟:暫存該j見時晝面資料中對應於該第二組像素單 几62B之第二組像素資料，其中每一個像素資料係被完整地 儲存。 > ^驟607 :將下一晝面資料（the next frame data)與暫存 之s亥第一組像素資料進行比對，以計算下一晝面資料的過度 驅動電壓。 一 ' 接著藉由圖六A'圖六B並配合以下文字以說明本發明 過度驅動電壓產生方法。以第二晝面資料522、第三晝面資 料523為例’第一晝面資料522係包括複數個像素資料（pixei dat=) 522p，第三晝面資料523係包括複數個像素資料523p。 而每一個像素資料522p、523p皆為48位元，圖中以48個位 14 1271693 不 元點(圖中標號55)來表 廳- 56訊號寬度不足的問題’本發明係先將液晶 ^不，6。中_分佈的複數個像素單元分組為第 f ^料522中對應於 像素身料:以作為第二參考資料532，且其中每一個像 =32ρ係被完整地儲存。如圖六B所示，每一個參考像素 貝料53办都具有μ個位元點（圖六八標號％)。’、 523中對應於第二組像素單元62糊五ΐ ”貝料533ρ都具有48個位元點（圖六Β標號55)。 料可侧的參考資 五液晶 曰曰 曰a 顯示第二 ====爾與圖五液 顯二考素中單 === 1271693 一口藉由相鄰二畫面的參考資料互相的補足，並利用液晶顯 示器60進行顯像時，畫面快速更新所帶來的視覺暫留效果。 本發明可，效改善習知技術中影像色彩失真的缺點。如圖六 C所示，每一個參考像素資料532p、533p、534p及535p皆 為f8 t元’相較於圖四之習知技術中每一個像素資料222p 都是不完全儲存，僅有32位元而言，本發明明顯地在記憶體 56未改㈣情況下，為習知液晶轉系統的色彩失直缺 供了解決方案。 ^ 關於實施本發明時，液晶顯示器6〇中複數個像辛單 何分組成為上述之第-組像素單元62A與第二峰=單元# 62B ,係有多種實施方法，將敘述如下·· <分組實施例A> ⑭請參照圖七A，實施例A係將液晶顯示器中，第奇數 第單元’以及第偶數條掃瞄線的第偶數 元係早Τ ’以念弟偶數條掃瞄線的第奇數個像素單 兀係疋義為弟一組像素皁元。 <分組實施例Β:> 請參照圖七Β，[實施例Β係將液晶顯示与中 第奇數個像素單元定義為第-組像素田線的 數個像素單元係定麟第二組像素單元。’ Μ線的第偶 <分組實施例C> ，照=C。由於液晶顯示器具有複數 像素早7L，其中母-像素單元係具有三個次!」刀佈之 別表示該像素單元之紅色、綠色與藍色組成。因^實3 16 1271693 C係將液晶顯示器中每一個像素單元進一步區分為一紅次像 素(R sub-pixel)、一綠次像素(G sub-pixel)及一藍次像素(B sub-pixel)。並且，在進行像素單元分組時以這些次像^為 單位，將第奇數條掃瞄線的第奇數個次像素單元了以及第^ 數條掃瞄線的第偶數個次像素定義為第一組次像素單元；而 將^奇數條掃瞄線的第偶數個次像素，以及第偶數條掃瞄線 的第奇數個次像素定義為第二組次像素單元。〃、 〈分組實施例D> 請參照圖七D，實施例D係將係將液晶顯示器中，第 =+1、4n+2條掃瞄線(n>0〜767)的第奇數個像素單^，以及籲 第4=3、4n+4條掃瞄線(11>0〜767)的第偶數個像素單元係定 第-組像素單元；而第4n+1、4n+2條掃目肖線(㈣〜 ，，數個像素單元，以及第4n+3、4n+4條掃聪線(n>〇〜767) 的弟可數個像素單元係定義為第二組像素單元。 <分組實施例E> 請參照圖七E，實施例E係將係將液晶顯示器中， 11 4n+2條掃瞄線(n>〇〜767)的第奇數個次像素，以及1271693 IX. Description of the Invention: [Technical Field] The present invention relates to a method for generating excessive driving voltage in a liquid crystal driving system, and more particularly to an excessive driving voltage generation for improving memory bandwidth shortage and color misalignment method. [Prior Art] The main advantage of the liquid crystal display is that it is easy to achieve a high display display, so it is widely used in notebook computers. Recently, with the large coffee, the liquid crystal display has gradually become more popular on desktop computers. Ground. However, in many manufacturers who want to replace the traditional cathode ray TV with liquid crystal display technology, liquid crystal display technology has revealed its technical bottleneck in the performance of dynamic images. Please refer to the figure -, - Figure - is the basic diagram of the liquid crystal display 10. Wherein, the liquid j plate 12 has a plurality of pixel units 四(4)(2), and the specific voltage is rotated by a specific angle for each of the crystals in each of the 121 sheets, so that the liquid crystal panel 12 is under the side of the liquid crystal panel 12 The light supplied is provided in the liquid crystal panel 12 in the non-pixel unit 121, and the pixel unit 121 of the array is interleaved to form a H, and the W frame is viewed. If it is a more subtle distinction, the 7L 121 system can be divided into three sub-pixels (sub_pixd) 121R, ^^^, respectively, to control the red and green colors in each pixel unit 121 as above, and the liquid crystal display technology is in (4) images. The performance of the technical bottleneck 'the main reason is with the liquid crystal molecules = after the liquid crystal molecules receive the above-mentioned special material voltage, need the temple -: should be 1271693, to rotate from an original angle to another angle (with a specific voltage Corresponding angle). When displaying a moving image, the reaction time of the liquid crystal molecules does not catch up with the update rate of the picture, resulting in delay or distortion of the display quality. Therefore, the reaction time of liquid crystal molecules can be shortened, which is an important subject in liquid crystal display technology. [TFT overdrive technology (TFToverdrive) is a method currently used to shorten the reaction time of liquid crystals by deliberately applying a high (or too low) voltage to the liquid crystal molecules, so that the liquid crystal molecules are scheduled. Rotate to the desired angle within the time interval. The liquid crystal overdrive = technology will be described below with reference to Fig. 2. In Fig. 2, the vertical axis represents the angle of rotation of the liquid crystal molecules, and the horizontal axis is the time axis; as shown in Fig. 2, when the liquid crystal cell is supplied with a control voltage Vcl, The crystal molecule needs to pass the time t2 to be rotated from the angle center to the angle corresponding to the control voltage Vcl, and the liquid crystal overdrive technology first applies the liquid crystal cell to an excessively excessive driving voltage v〇D (v〇D is the angle The control circuit of 62 causes the liquid crystal molecules to rotate to the angle Si at time t1, and then adjusts the voltage back to the control voltage Vcl corresponding to the angle center, thereby shortening the reaction time of the liquid crystal. However, in addition to the excessively high voltage applied to the liquid crystal molecules, the liquid crystal overdrive technique also has a low voltage applied to the liquid crystal molecules, for example, to rotate the liquid crystal molecules from the angle & 0. That is, for this condition, and Fig. 1 only represents the liquid crystal molecules to be angled by θ. When rotating to the angle θ ! , there are more angle changes when actually operating the LCD. Therefore, in the liquid crystal overdrive technology, it is necessary to continuously compare the signals that are being played, and to calculate the appropriate overdrive, value of V0D; for example, excessive drive in Figure 2. The value of the voltage (v〇d) is calculated as: ' must ^ compare the angle (one of the liquid crystal molecules belonging to the second surface) with the angle 仏 (the liquid crystal molecule belonging to the first - 昼财), in order to calculate the combination 7 1271693 The excessive drive voltage value. In this way, the principle of liquid crystal overdrive technology is simple (using too high or too low voltage to shorten the liquid crystal reaction time), but there are quite complicated signal comparison and calculation procedures in implementation. Referring to FIG. 3, FIG. 3 is a schematic diagram of a conventional liquid crystal driving system. Further, the conventional liquid crystal driving system and the liquid crystal overdriving technology thereof can be further understood from the following description. The liquid crystal driving system 20 includes a processor 21 and a memory 26. After receiving the data stream 22, the liquid crystal driving system 20 can perform signal processing and calculation internally, and generate a driving signal stream 24 to drive the liquid crystal in the liquid crystal display. molecule. . As can be seen from FIG. 2, in the driving signal stream 24, each picture driving signal includes an overdrive voltage v0D and a control voltage; for example, the second kneading driver signal 242 includes a second overdrive voltage 242V〇du And a second control voltage 242Vcl. When the second page data 222 of the facet data stream 22 is input to the liquid crystal drive system 20, the processor 21 assigns the second facet data 222 to the memory 26, and the second face data is stored by the memory 26. 222 is temporarily stored to become the reference material 232. '1, the processor 21 compares the second side data 222 with the first reference material 231 previously in the memory 26, and uses the principle of the above liquid crystal overdrive technology to calculate the second excessive The driving voltage is 242VOD. As for the second control voltage 242VCL described above, the comparison program ‘the processor 21 can calculate the second control voltage 242Va according to the second page data 222 without going through the adjacent book data. In this way, each of the checks temporarily stored in the memory 26 (the first reference material 23, the second reference material said · · · ·) can be regarded as the JC reference data stream 23, its use The liquid crystal driver 20 has a reference for generating an excessive driving voltage. 7 is 1271693 However, the application of the liquid crystal overdrive technology to a high display resolution has caused a problem of insufficient signal width in the liquid crystal drive system 20. As shown in Fig. 2', when the memory 26 temporarily stores the data, the highest bit is generally stored as a temporary memory. In other words, the signal width between the processor 21 and the memory is the widest. 32-bit (bit), although the signal width is sufficient for low display resolutions such as (640x480) or SVGA (800x600), if you want to display sjG (1280x1024) or higher display resolution Inside, you need a signal width of 48 bits or more. ^ The problem of insufficient width of this signal can of course directly develop or adopt a memory with a wider signal width (for example, 64 bits). However, the cost is still high at present and belongs to the technical field related to memory. In the technical field of the liquid crystal drive system to which the present invention belongs, it is conventional to face the above problem. The practice is to discard the signals of the three primary colors of red, blue and green, and to discard some of the bits of the latter stage, as explained below: In the case of displaying the resolution, the processor 21 receives and processes the pixel data of the "one" pixel unit at the same timing (cl〇ck), commonly known as a single channel G channel); for example, one scan of VGA (640x480) The line requires 640 timings (ci〇ck) to complete the processing. However, at the same timing (cl〇ck), the processor 21 receives and processes "two, pixel units of pixel data, commonly known as dual channels; at high display resolution, for example, when SXGA (1280x1024) has a high display resolution. The data of the sweeping cat line can also be processed in 640 timings (ci〇ck). The signal width of the 48-bit high display resolution is composed of two channels (2 channels) of 24 bits each. The 24-bit system consists of 8 bits each of red, blue and green, but the signal width between the processor 21 and the memory 26 is 1271693, /°'. Therefore, the conventional technique is to use a red signal. The middle 2 digits are discarded, and the last 3 of the blue and green signals are discarded, so that the single channel red, blue, and green (four) +8), the bit signal is reduced by ^6+5+5)=16 bits. The signal of the yuan, so - to merge the dual channel can only pay the 32-bit signal width. Volume - 彳 ― ―, that is, the above two bits of the red signal are discarded, and And the traditional practice of the last three digits of the green signal is only a kind of practice implementation. Others also have red, blue and green colors ( 5, 5, 5) the practice of the bit, or (5, I, / #, 5, 6) ······, etc., the purpose is only to discard some of the later bits, pay = § fl money. For example, the original 8-bit red signal is discarded by the last 2, for example, the original 8-bit will have 256 colors (red 256 gray-scale colors), and after subtracting to 6-bit, only 64 colors will be left. Red 64 grayscale colors), so the conventional practice obviously leads to color distortion. Next, by using Figure 4 to enhance the color loss of the above-mentioned prior art, the face of the job is taken as an example, when the memory 26 News = by:, the first side of the information 222 will be discarded some of the information of the bits, and the second and second data of the shape and the negative amount of the material is temporarily stored in the memory, as shown in Figure 4. The second picture material 222 includes a plurality of pixel data (f xel data) 222p, and the second reference material 232 includes a plurality of reference pixels (referenCepixddata) 232P. The high display resolution is nickname as described above. The width is 48 bits, and in FIG. 4, there are 48 bit points (reference numeral 25) in the pixel data 222P to represent 'because the signal of the memory 46 The degree usually has 32 bits, so as shown in Figure 4, there are only 32 bit points (reference 25) left in each reference pixel data. Therefore, as can be seen from Figure 4, the memory 26 is The signal width is insufficient, so that the second picture data 222 j pixel data 222 ρ are incompletely temporarily stored, and become incomplete source pixel data 232 ρ ; the influence brought by the expansion is extended to the reference data stream 23 (Fig. 1271693 Each frame and every pixel (pixe〇. But the reason why the color distortion is not seriously valued in the prior art is that - as shown in Figure 2, 7F, the color distortion of the color shirt is temporarily stored in the memory. The first and second reference materials 231, 232, etc. in 26, that is, the reference data stream 23, the reference data stream 23 is used to compare with the kneading data stream 22 to be played. The problem of color distortion occurs on the reference stream 23; the input data stream 22 of the input liquid crystal drive system 2 and the output of the drive ship 24 are not insufficiently wide. ·, knowing the technology does not pay much attention to the seriousness of the shortcomings of this color distortion; the bran matter ^ test data flow, the misunderstanding table _ liquid crystal display ^ overall display two 嘈 Ξ Ϊ Ϊ Ϊ Ϊ Ϊ ' ' The color distortion can be seen by the naked eye, and it is hoped that the high-definition reading will play the surface of the quality, which will cause the color distortion. (10) Π 何 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善 改善It is provided that the excessive degree of the down-motion system in the liquid crystal drive system is reflected in the high display. The purpose is to not increase the color distortion of the memory hardening system. 11 1271693 The present invention provides an overdrive voltage for the m-turn system towel. The liquid crystal drive system includes a processor and a memory. Before the processor crystal 1f moves, the data is temporarily stored in the memory, and the mine is Ϊ 3=!^昼面资料与前—昼面资料, and the liquid crystal molecules in the 0 degree drive are accelerated to the desired angle ΐ liquid 曰Ί Gif material includes a plurality of pixel data to increase the number of tr The pixel units are developed. The image of the ii peak element is 7G earlier than the second pixel. The overdrive voltage generating method of the present invention comprises the following steps: (8) Having a corresponding one of the first group of images corresponding to the first group of images. And the pixel data is completely temporarily stored in the memory. ,, ', the material into the face of the towel - group of pixel resources r, scorpion ± 乂 ° ten different appearance of the excessive driving voltage of the data. After the ιί, the first-pixel data corresponding to the second group of pixel units is completely temporarily stored in the memory. ', one, and (d) will be the next - 昼 surface data and temporarily stored in the note material for the series, under the raw materials - painting (four) material pixels = how many ίί" real iij early points j square pixel unit can be wealth When the financial integrity/yuan is temporarily stored in the moxibustion, there is a need for a more detailed specification of the 1271693 resolution. The invention can be applied to the related memory signal width deficiency. [Embodiment] / Please refer to FIG. 5, which is a schematic diagram of a liquid crystal display and a liquid crystal driving system thereof. The liquid crystal drive system 50 is based on a facet data stream 52 to provide a drive signal stream 54 for driving liquid crystal molecules in the liquid crystal display 60. Each of the drive signal streams 54 includes an overdrive voltage v0D and a control voltage VcL. Taking the second kneading drive signal 542 as an example, the second overdrive voltage 542v〇d and the second control voltage 542VCL are included. The liquid crystal drive system 50 includes a processor 51 and a memory 56. When the second side data 522 of the facet data stream 52 is input to the liquid crystal driving system, the processing state 56 dispatches the second side data 522 to the memory 56, and the second side is the second side. The data 522 is temporarily stored as the second reference material 532. The processor 51 compares the second page data 522 with the first reference material 531 previously stored in the memory 56, and calculates a second overdrive voltage 542VOD. As for the second control voltage 542Vcl, the comparison program of the adjacent data is not required, and the processor 51 can directly calculate the second control voltage 542Vcl according to the second surface & 522. The plurality of reference materials (for example, the first and second reference materials 53 532···) successively stored in the memory 56 can be regarded as a reference data stream 53. The problem encountered by the conventional $ technique is that if the image with high display resolution is to be played (the processor 51 receives and processes "two pixels at the same timing, the pixel data of the pixel unit is a dual channel, 48 bits"), Because the memory 56 can only temporarily store 32-bit (bit) in the temporary storage of the screen data, once the reference number i degree of the reference material 々iL 53 is greater than 32 bits, the reference data stream% will be caused. The representative 13 1271693 2 color produces the true, and the excessive driving voltage in the driving signal stream 54, the younger brother - excessive driving of the dust 541 V 〇 D, 542 V 〇 D ..., etc.) also produces the drawback of color distortion. According to FIG. 5, the overdrive voltage generating method provided by the present invention is used to generate an overdrive to drive the liquid crystal in the liquid crystal (^) to accelerate to a desired degree, and to improve the conventional technique. There is a lack of color distortion. As shown in Figure 5, the LCD panel of the liquid crystal display has a pixel array of 2 pixels. The present invention groups the plurality of pixel units into a group-pixel pixel TC62A. Fine with the second group of pixel units. And the present invention is excessive The steps of the driving voltage generating method are as follows: Step 6: 暂: pre-stored data (the previ〇us fr_ as in (8) corresponds to the first group of pixel units 62A of the first group of pixel data, each of which A pixel data is stored completely. Step 603 · Compare the current face data (tJle current (10) such as pull) with the temporarily stored first set of pixel data to calculate the overdrive voltage of the current face data. 1. Step: temporarily store the second set of pixel data corresponding to the second group of pixels 62B in the face data, wherein each pixel data is completely stored. > ^Step 607: The next frame data is compared with the first set of pixel data of the temporary storage to calculate the overdrive voltage of the next data. A 'throw by Figure 6A' Figure 6 B and the following text is used to explain the excessive driving voltage generating method of the present invention. Taking the second side data 522 and the third side data 523 as an example, the first side data 522 includes a plurality of pixel data (pixei dat=) 522p. , the third face data 523 series including The pixel data is 523p. Each pixel data 522p and 523p are 48 bits. In the figure, 48 bits 14 1271693 are not points (55 in the figure) to the table - 56 signal width is insufficient. First, the plurality of pixel units of the liquid crystals are not grouped into 6. The corresponding pixel units in the first material 522 are corresponding to the pixel material: as the second reference material 532, and each of the image=32ρ is completely stored. As shown in FIG. 6B, each of the reference pixel materials 53 has μ bit points (Fig. 6-8). ', 523 corresponds to the second group of pixel units 62 paste five ΐ 贝 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 533 === 尔和图五液显二素素中单=== 1271693 The visual effect brought by the rapid update of the screen when the reference data of the adjacent two screens complement each other and the liquid crystal display 60 is used for development The present invention can improve the shortcomings of image color distortion in the prior art. As shown in FIG. 6C, each of the reference pixel data 532p, 533p, 534p and 535p is f8 t-e' compared to FIG. In the prior art, each pixel data 222p is incompletely stored, and only 32 bits, the present invention clearly shows that the color of the conventional liquid crystal transfer system is lacking in the case where the memory 56 is not changed (4). Solution. ^ In the implementation of the present invention, a plurality of image groups in the liquid crystal display 6 are grouped into the above-described first-group pixel unit 62A and second peak-level unit 62B, which are described in the following. ·· <Grouping Example A> 14 Referring to Figure 7A, In the case of the liquid crystal display, the odd-numbered unit and the even-numbered scanning lines of the even-numbered scanning lines are earlier than the first-numbered pixels of the even-numbered scanning lines. Group Pixel Soap Element. <Grouping Example &:> Referring to FIG. 7A, [Embodiment 将 is to define a liquid crystal display and an odd number of pixel units in the first group of pixel rows. The second group of pixel units of the lining. The first even < grouping embodiment C> of the Μ line, photo = C. Since the liquid crystal display has a complex pixel 7L early, wherein the mother-pixel unit has three times! Indicates the red, green, and blue components of the pixel unit. Because the real 3 16 1271693 C system further distinguishes each pixel unit in the liquid crystal display into a red sub-pixel, a green sub-pixel, and a blue sub-pixel (B sub-pixel). ). And, in the pixel unit grouping, the odd-numbered sub-pixel units of the odd-numbered scan lines and the even-numbered sub-pixels of the first-numbered scan lines are defined as the first group in units of these sub-images The sub-pixel unit; and the even-numbered sub-pixels of the odd-numbered scan lines and the odd-numbered sub-pixels of the even-numbered scan lines are defined as the second set of sub-pixel units.分组, <Group Embodiment D> Referring to FIG. 7D, Embodiment D will be the odd-numbered pixel list of the +1st, 4n+2th scan lines (n>0~767) in the liquid crystal display. ^, and the even number of pixel units of the 4=3, 4n+4 scan lines (11 > 0~767) are the first group of pixel units; and the 4n+1, 4n+2 The line ((4)~,, a few pixel units, and the 4th+3, 4n+4 Sweep line (n> 〇~767) are defined as the second group of pixel units. <Grouping Embodiment E> Referring to FIG. 7E, Embodiment E is an odd-numbered sub-pixel of 11 4n+2 scanning lines (n>〇~767) in a liquid crystal display, and
4Γ4條掃崎㈣〜767)的第偶數個次像素係定義 ： 4η+1 ' 4η+2 ^8-^(π>〇~767) 第卉—Ίίί，以及第4η+3、4η+4條掃瞄線(η>0〜767) 弟可數個:欠像素係定⑽第二次組像素單元。 有像本發明之實施例’係皆屬於將液晶顯示器之所 組:象素二與第二組!f素單”或第 素單元中的料單丨組像素單70與第二組像 #寬声48 #素^不重覆的情況。以高顯示解析度的訊 又 兀而言，例如實施例Α中，第一組像素單元與 17 1271693 度24位元的資料 ，因此可以 例A〜有32位元的記憶體中。就上述實施 度沒有被使^體有6位元的訊號寬 顯示品質，改善習知的“真；:ί供較f知技術更好的 -組數加以细，則可財暫存第 ί驟 動電壓的產生方法之步驟係包含下列 第—組像素資料’—組 資料’及對應於該第二組像素單 步驟803 :將現時晝面資料與暫存之 行比對，以計算_晝面㈣的财轉輕像素貝科進 步驟805 :暫存-第二組像素資料 包括現時晝面資料中，對應於該第二組 像常負料。 步驟807 :將下-晝面·與暫存 行比對，以計算下-晝面資料的過度驅動電壓像常貝科進 綜合以上所述可知，關於形成第„組像素單元與第二纪 像素單几的分組方法可有多種變形的實施方式。其 在液晶驅動系統記憶體的習知限制之下，利用^ 組，以使抑個別被暫存的像素單元能夠保有其原^完整的位 18 1271693 元資料之相關技術皆應屬本發明之延伸。 使得液晶驅動系統突破液晶顯示器高顯示^二日:^， 技術瓶頸’並且使得顯示影像之色彩維持其斤= 發明可“=更:趙=顯度：=的需求時’本 本發明雖以較佳實例闡明如上然其^ ^ ^ :精=^實體，僅止於上述實施例疒 ί。^以用其它元件或方式來產生相同的功 應包含在下述之申請專利範圍内。 特之修改，均 【圖式簡單說明】 ，由以下詳細之描述結合所附圖示，將可輕易的了 内谷及此項發明之諸多優點，其中： 解上达 圖=顯示液晶顯示器基本圖； 圖亡為液晶分子旋轉角度與反應時間關係圖； 圖二為習知液晶驅動系統示意圖； 圖四為圖三之第二晝面資料暫存入記憶體示意圖； 圖f為液晶顯示器及其連接之液晶驅動系統示意圖； 圖為圖五之第二晝面資料暫存入記憶體示意圖； 圖為圖五之第三晝面資料暫存入記憶體示意圖； =六C為暫存至記憶體的第二〜第五晝面資料示意圖； 圖七Α表示像素單元分組實施例a· 圖七B表示像素單元分組實施例B 圖七C表示像素單元分組實施例C 圖七D表示像素單元分組實施例D;以及 圖七E表示像素單元分組實施例E。 19 12716934Γ4 扫崎(四)~767) The even sub-pixel definition: 4η+1 ' 4η+2 ^8-^(π>〇~767) The first plant—Ίίί, and the 4η+3, 4η+4 Scan lines (η > 0~767) can be counted: under the pixel system (10) second group of pixel units. There are embodiments like the present invention, which belong to the group of liquid crystal displays: pixel two and the second group! f single, or the unit cell group 70 and the second group image # in the first unit The sound 48 is not repeated. In the case of high display resolution, for example, in the first embodiment, the first group of pixel units and 17 1271693 degrees 24 bits of data, so can be a In the 32-bit memory, the above implementation degree is not caused by the 6-bit signal width display quality of the body, and the improvement of the conventional "true;: ί is better than the knowledge-known technology - the number of groups is fine, The step of generating the ί 骤 电压 电压 包含 包含 包含 包含 包含 包含 ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί ί The comparison, to calculate the _ 昼 ( (4) of the fortune light pixel Becco into the step 805: temporary storage - the second group of pixel data including the current face data, corresponding to the second group of image constants. Step 807: Comparing the lower-side surface with the temporary storage line to calculate the excessive driving voltage of the lower-surface data. As described above, the above-mentioned group of pixel units and second-order pixels are formed. A single grouping method can have various modified embodiments. Under the conventional limitation of the memory of the liquid crystal driving system, the group is used to enable the individual temporarily stored pixel units to retain their original bits 18 1271693 The related technology of the metadata should be an extension of the present invention. The liquid crystal drive system breaks through the high display of the liquid crystal display ^ 2 day: ^, the technical bottleneck 'and the color of the displayed image is maintained by its weight = invention can be "= more: Zhao = Significantness: When the demand is = 'The present invention, while exemplifying the above, is a ^ ^ ^ : fine = ^ entity, only the above embodiment 疒 ί. The use of other elements or means to produce the same function is included in the scope of the following claims. In particular, the following is a brief description of the drawings. The following detailed description, together with the accompanying drawings, will be able to easily overcome the advantages of the inner valley and the invention, wherein: the upper map = the basic display of the liquid crystal display Figure 2 is a schematic diagram of a conventional liquid crystal driving system; Figure 4 is a schematic view of the second surface of Figure 3 temporarily stored in a memory; Figure f is a liquid crystal display and its connection The schematic diagram of the liquid crystal driving system; the picture is the second side of the figure 5 is temporarily stored in the memory; the picture is the third side of the figure 5 is temporarily stored in the memory; = six C is temporarily stored in the memory FIG. 7B shows a pixel unit grouping embodiment a. FIG. 7B shows a pixel unit grouping embodiment B. FIG. 7C shows a pixel unit grouping embodiment C. FIG. 7D shows a pixel unit grouping embodiment. D; and FIG. 7E shows a pixel unit grouping embodiment E. 19 1271693
【主要元件符號說明】 液晶顯示器10、60 液晶面板12、62 次像素 121R、121G 及 121B 像素單元121 背光源14 液晶驅動系統20、50 處理器21、51 晝面資料流22、52 第二晝面資料222 像素資料222p、522p、523p 參考資料流23、53 第一參考資料231、531 第二參考資料232、532 參考像素資料232p 驅動訊號流24、54 弟一晝面驅動訊號242、542 第二控制電壓242VCL、542VC]L —-----—-—__ 第二過度驅動電壓242V〇D、 542V0D 位元點25、55 —~~-—____ 記憶體26、56 第二晝面資料522 第三晝面資料523 參考像素資料532p、533p、 534p > 535p ~—---------- 第四參考資料534 ~~ --- 第三參考資料533 ^^-------- 第五參考資料535 第一過度驅動電壓541VciD ^ — 第—組像素單元62Α 第二組像素單元62B 20[Main component symbol description] Liquid crystal display 10, 60 Liquid crystal panel 12, 62 sub-pixels 121R, 121G and 121B Pixel unit 121 Backlight 14 Liquid crystal drive system 20, 50 Processor 21, 51 Face data stream 22, 52 Second Surface data 222 Pixel data 222p, 522p, 523p Reference data stream 23, 53 First reference data 231, 531 Second reference data 232, 532 Reference pixel data 232p Drive signal stream 24, 54 Brother one face drive signal 242, 542 Two control voltages 242VCL, 542VC]L —--------__ Second overdrive voltage 242V〇D, 542V0D Bits 25, 55 —~~-—____ Memory 26, 56 Second surface data 522 Third page data 523 Reference pixel data 532p, 533p, 534p > 535p ~----------- Fourth reference material 534 ~~ --- Third reference material 533 ^^--- ----- fifth reference material 535 first overdrive voltage 541VciD ^ - first group of pixel units 62 Α second group of pixel units 62B 20