欢迎访问ic37.com |
会员登录 免费注册
发布采购

IC42S16400F-6TL 参数 Datasheet PDF下载

IC42S16400F-6TL图片预览
型号: IC42S16400F-6TL
PDF下载: 下载PDF文件 查看货源
内容描述: 1梅格位×16位× 4银行( 64兆位)同步动态RAM [1 Meg Bits x 16 Bits x 4 Banks (64-MBIT) SYNCHRONOUS DYNAMIC RAM]
分类和应用: 存储内存集成电路光电二极管动态存储器时钟
文件页数/大小: 55 页 / 822 K
品牌: ISSI [ INTEGRATED SILICON SOLUTION, INC ]
 浏览型号IC42S16400F-6TL的Datasheet PDF文件第16页浏览型号IC42S16400F-6TL的Datasheet PDF文件第17页浏览型号IC42S16400F-6TL的Datasheet PDF文件第18页浏览型号IC42S16400F-6TL的Datasheet PDF文件第19页浏览型号IC42S16400F-6TL的Datasheet PDF文件第21页浏览型号IC42S16400F-6TL的Datasheet PDF文件第22页浏览型号IC42S16400F-6TL的Datasheet PDF文件第23页浏览型号IC42S16400F-6TL的Datasheet PDF文件第24页  
samebank.ThePRECHARGEcommandshouldbeissuedꢀ  
                                                            
IS42S16400F  
IC42S16400F  
READS  
READ COMMAND  
READꢀ burstsꢀ areꢀ initiatedꢀ withꢀ aꢀ READꢀ command,ꢀ asꢀ  
shownꢀinꢀtheꢀREADꢀCOMMANDꢀdiagram.  
CLK  
Theꢀstartingꢀcolumnꢀandꢀbankꢀaddressesꢀareꢀprovidedꢀwithꢀ  
theREADcommand,andautoprechargeiseitherenabledorꢀ  
disabled for that burst access. If auto precharge is enabled,  
the row being accessed is precharged at the completion of  
theꢀburst.ꢀForꢀtheꢀgenericꢀREADꢀcommandsꢀusedꢀinꢀtheꢀfol-  
lowing illustrations, auto precharge is disabled.  
HIGH-Z  
CKE  
CS  
RAS  
DuringꢀREADꢀbursts,ꢀtheꢀvalidꢀdata-outꢀelementꢀfromꢀtheꢀ  
starting column address will be available following the  
CASꢀlatencyꢀafterꢀtheꢀREADꢀcommand.ꢀEachꢀsubsequentꢀ  
data-outꢀelementꢀwillꢀbeꢀvalidꢀbyꢀtheꢀnextꢀpositiveꢀclockꢀ  
edge.TheCASLatencydiagramshowsgeneraltiming  
for each possible CAS latency setting.  
CAS  
WE  
COLUMN ADDRESS  
AUTO PRECHARGE  
Uponꢀcompletionꢀofꢀaꢀburst,ꢀassumingꢀnoꢀotherꢀcommandsꢀ  
havebeeninitiated,theDQswillgoHigh-Z.Afull-pageburstꢀ  
will continue until terminated. (At the end of the page, it will  
wrap to column 0 and continue.)  
A0-A7  
A8, A9, A11  
A10  
DataꢀfromꢀanyꢀREADꢀburstꢀmayꢀbeꢀtruncatedꢀwithꢀaꢀsub-  
sequentREADcommand,anddatafromaxed-lengthꢀ  
READꢀburstꢀmayꢀbeꢀimmediatelyꢀfollowedꢀbyꢀdataꢀfromꢀaꢀ  
READꢀcommand.ꢀInꢀeitherꢀcase,ꢀaꢀcontinuousꢀflowꢀofꢀdataꢀ  
canꢀbeꢀmaintained.ꢀTheꢀfirstꢀdataꢀelementꢀfromꢀtheꢀnewꢀ  
burst follows either the last element of a completed burst  
or the last desired data element of a longer burst which  
is being truncated.  
NO PRECHARGE  
BANK ADDRESS  
BA0, BA1  
TheꢀDQMꢀinputꢀisꢀusedꢀtoꢀavoidꢀI/Oꢀcontention,ꢀasꢀshownꢀ  
inꢀFiguresꢀRW1ꢀandꢀRW2.ꢀTheꢀDQMꢀsignalꢀmustꢀbeꢀas-  
serted(HIGH)atleastthreeclockspriortotheꢀWRITEꢀ  
commandꢀ(DQMꢀlatencyꢀisꢀtwoꢀclocksꢀforꢀoutputꢀbuffers)ꢀ  
tosuppressdata-outfromtheREAD.OncetheꢀWRITEꢀ  
commandꢀisꢀregistered,ꢀtheꢀDQsꢀwillꢀgoꢀHigh-Zꢀ(orꢀremainꢀ  
High-Z),regardlessofthestateoftheDQMsignal,providedꢀ  
theꢀDQMꢀwasꢀactiveꢀonꢀtheꢀclockꢀjustꢀpriorꢀtoꢀtheꢀWRITEꢀ  
commandꢀthatꢀtruncatedꢀtheꢀREADꢀcommand.ꢀIfꢀnot,ꢀtheꢀ  
secondꢀWRITEꢀwillꢀbeꢀanꢀinvalidꢀWRITE.ꢀForꢀexample,ꢀifꢀ  
DQMwasLOWduringT4inFigureRW2,thentheWRITEsꢀ  
atT5ꢀandT7ꢀwouldꢀbeꢀvalid,ꢀwhileꢀtheꢀWRITEꢀatꢀT6ꢀwouldꢀ  
be invalid.  
ThenewREADcommandshouldbeissuedxcyclesbefore  
the clock edge at which the last desired data element is  
valid, where x equalsꢀtheꢀCASꢀlatencyꢀminusꢀone.ꢀThisꢀisꢀ  
shownꢀinꢀConsecutiveꢀREADꢀBurstsꢀforꢀCASꢀlatenciesꢀofꢀ  
twoꢀandꢀthree;ꢀdataꢀelementꢀn +ꢀ3ꢀisꢀeitherꢀtheꢀlastꢀofꢀaꢀ  
burstꢀofꢀfourꢀorꢀtheꢀlastꢀdesiredꢀofꢀaꢀlongerꢀburst.Theꢀ64Mbꢀ  
SDRAMꢀusesꢀaꢀpipelinedꢀarchitectureꢀandꢀthereforeꢀdoesꢀ  
not require the 2n rule associated with a prefetch architec-  
ture.AREADcommandcanbeinitiatedonanyclockcycleꢀ  
followingꢀaꢀpreviousꢀREADꢀcommand.ꢀFull-speedꢀrandomꢀ  
read accesses can be performed to the same bank, as  
shownꢀinꢀRandomꢀREADꢀAccesses,ꢀorꢀeachꢀsubsequentꢀ  
READꢀmayꢀbeꢀperformedꢀtoꢀaꢀdifferentꢀbank.  
TheꢀDQMꢀsignalꢀmustꢀbeꢀde-assertedꢀpriorꢀtoꢀtheꢀWRITEꢀ  
commandꢀ(DQMꢀlatencyꢀisꢀzeroꢀclocksꢀforꢀinputꢀbuffers)ꢀ  
to ensure that the written data is not masked.  
DataꢀfromꢀanyꢀREADꢀburstꢀmayꢀbeꢀtruncatedꢀwithꢀaꢀsub-  
sequentWRITEcommand,anddatafromaxed-lengthꢀ  
READꢀburstꢀmayꢀbeꢀimmediatelyꢀfollowedꢀbyꢀdataꢀfromꢀaꢀ  
WRITEꢀcommandꢀ(subjectꢀtoꢀbusꢀturnaroundꢀlimitations).ꢀ  
TheꢀWRITEꢀburstꢀmayꢀbeꢀinitiatedꢀonꢀtheꢀclockꢀedgeꢀim-  
mediately following the last (or last desired) data element  
fromꢀtheꢀREADꢀburst,ꢀprovidedꢀthatꢀI/Oꢀcontentionꢀcanꢀbeꢀ  
avoided. In a given system design, there may be a pos-  
sibilityꢀthatꢀtheꢀdeviceꢀdrivingꢀtheꢀinputꢀdataꢀwillꢀgoꢀLow-Zꢀ  
beforeꢀtheꢀSDRAMꢀDQsꢀgoꢀHigh-Z.ꢀInꢀthisꢀcase,ꢀatꢀleastꢀ  
a single-cycle delay should occur between the last read  
dataꢀandꢀtheꢀWRITEꢀcommand.  
Axed-lengthREADburstmaybefollowedby,ortruncatedꢀ  
with, aPRECHARGE commandtothesamebank(provided  
that auto precharge was not activated), and a full-page burst  
mayꢀbeꢀtruncatedꢀwithꢀaꢀPRECHARGEꢀcommandꢀtoꢀtheꢀ  
x cycles before the clock edge at which the last desired  
data element is valid, where x equals the CAS latency  
minusꢀone.ꢀThisꢀisꢀshownꢀinꢀtheꢀREADꢀtoꢀPRECHARGEꢀ  
20  
Integrated Silicon Solution, Inc. — www.issi.com  
Rev. A  
03/19/08  
 复制成功!