[Home] [This version is outdated, a new version is here]
*Title...
*Search:...
*Read Me/FAQ/General Info...
*_IBM...
*ACC Micro...
*ALD...
*ALi...
*AMD . . . . . . . [no datasheets, some info]...
*Chips & Technologies...
*Contaq  . . . . . [no datasheets, some info]...
*Efar Microsystems [no datasheets, some info]...
*ETEQ...
*Faraday...
*Forex . . . . . . [List only, no datasheets found]...
*Intel...
**82498/493   Cache Controller / Cache RAM (for P54 Pentium)    <Nov94
***Notes:...
***Info:...
***Configurations:...
***Features:...
**
**Later chipsets (basic spec):
**440 series:...
**450NX  (?)            06/29/98:...
**?????  (Profusion)    c:99...
**800 series...
*Headland/G2...
*HMC (Hulon Microelectronics)...
*Logicstar...
*Motorola...
*OPTi...
*PC CHIPS/Amptron/Atrend/ECS/Elpina/etc...
*SIS...
**5591/5592/5595 (David)   Pentium PCI A.G.P. Chipset        <01/09/98
***Info:...
***Configurations:...
***Features:
o   Support Intel/AMD/Cyrix Pentium CPU and Other Compatible CPU 
    Host Bus at 60/66 MHz and 3.3V Bus Interface
    − Support the Pipelined Address of Pentium compatible CPU
    − Support the Linear Address Mode of Cyrix CPU
o   Support the Pipelined Address Mode of Pentium CPU
o   Fully Compliant to A.G.P. Revision 1.0 Specification
o   Meet PC97 Requirements
o   Supports PCI Revision 2.1 Specification
o   Integrated Second Level (L2) Cache Controller
    - Write Back Cache Mode
    - Support L2 Cache Flushing for entire L2 cache or specific 
      4K page
    - 8 bits or 7 bits Tag with Direct Mapped Cache Organization
    - Integrated 32K bits Dirty SRAM
    - Integrated 32K bits Invalid SRAM
    - Support Pipelined Burst SRAM
    - Support 256K/512K/1MBytes Cache Sizes
    - Cache Hit Read/Write Cycle of 3-1-1-1
    - Cache Back-to-Back Read/Write Cycle of 3-1-1-1-1-1-1-1
    - Support Single Read Allocation for L2 Cache
    - Support Concurrency of CPU to L2 cache and A.G.P. master to 
      DRAM accesses
o   Integrated DRAM Controller
    - Support 6 RAS Lines for FPM/EDO/SDRAM DIMMs/SIMMs
    - Support 2Mbytes to 768Mbytes of main memory
    - Support Cacheable DRAM Sizes up to 256 MBytes.
    - Support 256K/512K/1M/2M/4M/8M/16Mx N FPM/EDO/SDRAM DRAM
    - Support 64 Mb DRAM Technology
    - Support Parity Checker or ECC Function
    - Support 3.3V or 5V DRAM
    - Supports Symmetrical and Asymmetrical DRAM
    - Support Concurrent Write Back
    - Support CAS before RAS Refresh, Self Refresh
    - Support Relocation of System Management Memory
    - Programmable CAS#, RAS#, RAMWE# and MA Driving Current
    - Fully Configurable for the Characteristic of Shadow RAM (640 
      KBytes to 1 MBytes)
    - Support FPM DRAM 5/6-3-3-3(-3-3-3-3) Burst Read Cycles
    - Support EDO DRAM 5/6-2-2-2(-2-2-2-2) Burst Read Cycles
    - Support SDRAM 5/6/7-1-1-1(-2/3-1-1-1) Burst Read Cycles
    - Support X-1-1-1/X-2-2-2/X-3-3-3 Burst Write Cycles
    - Two Programmable Non-cacheable Regions
    - Option to Disable Local Memory in Non-cacheable Regions
    - Shadow RAM in Increments of 16 Kbytes
    - Pseudo Directory/Page Scheme for Mapping Graphical Texture 
      Access to Physical Memory Address
    - Built-in 8 Way Associative/16 Entries GART cache to Minimize the 
      Number of Memory Bus Cycles Required for Accessing Graphical 
      Texture Memory
    - Programmable Counters to Ensure Guaranteed Minimum Access Time 
      for A.G.P., CPU, and PCI accesses
o   Provides High Performance PCI Arbiter.
    - Support up to 5  PCI Masters
    - Support Rotating Priority Mechanism
    - Hidden Arbitration Scheme Minimizes Arbitration Overhead.
    - Support Concurrency between CPU to Memory and PCI to PCI
    - Support Concurrency between CPU to 33Mhz PCI Access and 33Mhz 
      PCI to A.G.P. Access
    - Support Concurrency between CPU to 66Mhz PCI Access and A.G.P. 
      to 33Mhz PCI Access
    - Programmable Timers Ensure Guaranteed Minimum Access Time for 
      PCI Bus Masters, and CPU
o   Integrated Host-to-PCI Bridge
    - Support Asynchronous and Synchronous PCI Clock
    - Translates the CPU Cycles into the PCI Bus Cycles
    - Zero Wait State Burst Cycles
    - Support IDE Posted Write
    - Support Pipelined Process in CPU-to-PCI Access
    - Support Advance Snooping for PCI Master Bursting
    - Maximum PCI Burst Transfer from 256 Bytes to 4 Kbytes
    - Support Memory Remapping Function for PCI master accessing 
      Graphical Window
o   Integrated A.G.P. Compliant Target/66Mhz Host-to-PCI Bridge
    - Support Asynchronous and Synchronous A.G.P. Clock
    - Support 1X, and 2X Mode for A.G.P. 66/133 MHz 3.3V device
    - Support Graphic Window Size from 4Mbytes to 256Mbytes
    - Different arbitration policy for A.G.P. devices and 66Mhz PCI 
      devices.
    - Translates Sequential CPU-to-A.G.P. Memory Write Cycles into 
      A.G.P. Bus (PCI66) Burst Cycles
    - Zero Wait State Burst Cycles
    - Support Pipelined Process in CPU-to-A.G.P. Access
    - Support Advance Snooping for A.G.P. Master initiate system 
      memory access with PCI Cycles
    - Support 8 Way, 16 Entries Page Table Cache to enhance A.G.P. 
      Read/Write Performance
    - Support Both 1-Level and 2-Level GART (Graphic Address Re-
      Mapping Table)
    - Maximum PCI Burst Transfer from 256 Bytes to 4 Kbytes
    - Programmable Counters to Ensure Guaranteed Minimum Access Time 
      for Low Priority Request, CPU to A.G.P./and A.G.P. Master 
      Transaction
    - Support PCI-to-PCI bridge function for memory write from 33Mhz 
      PCI bus to A.G.P. bus
o   Integrated Posted Write Buffers and Read Prefetch Buffers to 
    Increase System Performance
    - CPU-to-Memory Posted Write Buffer (CTMFF) with 8 QW Deep, Always 
      Sustains 0 Wait Performance on CPU-to-Memory
    - CPU-to-Memory Read Buffer with 4 QW Deep
    - CPU-to-PCI Posted Write Buffer(CTPFF) with 8 DW Deep
    - PCI-to-Memory Posted Write Buffer(PTHFF) with 8 QW Deep, Always 
      Streams 0 Wait Performance on PCI-to/from-Memory Access
    - PCI-to-Memory Read Prefetch Buffer(CTPFF) with 8 QW Deep
    - CPU-to-PCI66 Posted Write Buffer(CTAFF) with 8 DW Deep
    - PCI66-to-Memory Posted Write Buffer(ATHFF) with 8 QW Deep
    - A.G.P. Request Queue With the Depth of 32
    - A.G.P. High Priority Write Queue with 64 QW Deep
    - A.G.P. Low Priority Write Queue with 64 QW Deep
    - A.G.P. High Priority Read Return Queue with 64 QW Deep
    - A.G.P. Low Priority Read Return Queue with 64 QW Deep
o   Fast PCI IDE Master/Slave Controller
    - Bus Master Programming Interface for ATA Windows 95 Compliant 
      Controller
    - Plug and Play Compatible
    - Support Scatter and Gather
    - Support Dual Mode Operation - Native Mode and Compatibility 
      Mode
    - Support IDE PIO Timing Mode 0, 1, 2 ,3 and 4
    - Support Multiword DMA Mode 0, 1, 2
    - Support Ultra DMA/33
    - Two Separate IDE Bus
    - Two 16 DW FIFO for PCI Burst Transfers.
o   Support NAND Tree for Ball Connectivity Testing
o   553-Balls BGA Package
o   0.35μm 3.3V CMOS Technology

**5596/5513      (Genesis) Pentium PCI Chipset               <03/26/96...
**5597/5598      (Jedi)    Pentium PCI/ISA Chipset           <04/15/97...
**530/5595       (Sinbad) Host, PCI, 3D Graphics & Mem. Ctrl.<11/10/98...
**540            (Spartan) Super7 2D/3D Ultra-AGP Single C.S.<11/30/99...
**55x            SoC (System-on-chip)                        <03/14/02...
**
**Support chips:
**85C206     Integrated Peripheral Controller [no datasheet]         ?...
**5595       Pentium PCI System I/O                          <12/24/97...
**950        LPC I/O                                         <07/16/99...
**Other:...
**PII/III/Pro...
**Athlon etc...
*Symphony...
*TI (Texas Instruments)...
*UMC...
*Unresearched:...
*VIA...
**VT8501       Apollo MVP4,Single-Chip 66-100MHz & AGP       <11/04/98
***Notes:...
***info:
The  Apollo MVP4  is  a PC  Socket-7  system logic  North Bridge  with
integrated 2D  / 3D Graphics  accelerator.  The core logic  portion of
the chip is  based on the popular 100MHz VIA  Apollo MVP3 chipset with
enhanced features  and graphics accelerator based  on the Cyber9398DVD
from Trident  Microsystems, Inc.  The  combination of the  two leading
edge  technologies   provides  a  stable,   cost-effective,  and  high
performance solution for personal computers, embedded systems, set-top
boxes and  others.  As  shown in Figure  1 [see datasheet]  below, the
Apollo MVP4 will interface to:

o Socket 7 CPU (66 – 100 MHz)
o L2 Cache RAM & Tag
o SDRAM Memory Interface
o PCI Bus (30 - 33 MHz)
o Analog RGB Monitor with DDC
o DFP / Digital Monitor Interface (TMDS)
o Video Capture / Playback CODECs

Apollo MVP4 Core Logic Overview
The Apollo  MVP4 –  System Media Accelerated  North Bridge (SMA)  is a
high performance, cost-effective and energy efficient solution for the
implementation  of Integrated  2D/3D  Graphics -  PCI  - ISA  personal
computer  systems from  66 MHz  to 100  MHz based  on  64-bit Socket-7
(Intel Pentium and Pentium MMX; AMD K6 and K6-2; Cyrix / National 6x86
/ 6x86MX, IDT / Centaur C6/WinChip), and Rise MP6 processors.

The Apollo  MVP4 controller provides superior  performance between the
integrated  2D/3D Graphics  Engine, CPU,  optional  synchronous cache,
DRAM,  and PCI bus  with pipelined,  burst, and  concurrent operation.
For  L2-Cache  solutions  using  pipelined  burst  synchronous  SRAMs,
3-1-1-1-1-1-1-1  timing  can  be  achieved  for both  read  and  write
transactions at 100 MHz.  Tag timing is specially optimized internally
(less  than 4 nsec  setup time)  to allow  implementation of  L2 cache
using an external tag for t  he most flexible cache organization (0K /
256K / 512K / 1M /  2M).  Four cache lines (16 quadwords) of CPU/cache
to  DRAM  write  buffers  with concurrent  write-back  capability  are
included on chip to speed up cache read and write miss cycles.

The Apollo  MVP4 supports six  banks of DRAMs  up to 768MB.   The DRAM
controller  supports  standard Fast  Page  Mode  (FP) DRAM,  EDO-DRAM,
Synchronous DRAM  (SDRAM), and Virtual  Channel Synchronous DRAM  in a
flexible mix  / match manner.   The Synchronous DRAM  interface allows
zero wait state bursting between the  DRAM and the data buffers at 100
MHz.  The six banks of DRAM can be composed of an arbitrary mixture of
1M / 2M  / 4M / 8M  / 16MxN DRAMs.  The DRAM  controller also supports
optional ECC (single-bit error  correction and multi-bit detection) or
EC (error checking) capability separately selectable on a bank-by-bank
basis.   The  DRAM Controller  can  run at  either  the  host CPU  bus
frequency (66  / 100 MHz) or  at the PC100 memory  frequency (100 MHz)
with  built-in deskew  PLL  timing control.   With  the advanced  DRAM
controller,  the  Apollo  MVP4   allows  implementation  of  the  most
flexible, reliable, and high-performance DRAM interface.

The  Apollo MVP4  also  supports  full AGP  v2.0  capability with  the
internal 2D/3D Graphics Engine for maximum software compatibility.  An
eight level request  queue plus a four level  post-write request queue
with thirty-two  and sixteen quadwords  of read and write  data FIFO’s
respectively   are  included   for  deep   pipelined  and   split  AGP
transactions.   A  single-level  GART  TLB with  16  full  associative
entries and  flexible CPU/AGP/PCI  remapping control is  also provided
for  operation  under  protected  mode operating  environments.   Both
Windows-95 VXD and Windows-98 / NT5 miniport drivers are supported.

The Apollo MVP4 supports one 32-bit  3.3 / 5V system bus (PCI) that is
synchronous  /  pseudo-synchronous to  the  CPU  bus.   The chip  also
contains a built-in AGP bus  -to- PCI bus bridge to allow simultaneous
concurrent  operations  on each  bus.   Five  levels (doublewords)  of
posted write buffers are included  to allow for concurrent CPU and PCI
operation.  For PCI master operation, forty-eight levels (doublewords)
of posted  write buffers and sixteen levels  (doublewords) of prefetch
buffers are  included for concurrent PCI bus  and DRAM/cache accesses.
The   chip  also   supports  enhanced   PCI  bus   commands   such  as
Memory-Read-Line,   Memory-Read-Multiple,   and   Memory-Write-Invalid
commands to  minimize snoop overhead.  In  addition, advanced features
are  supported such  as snoop  ahead, snoop  filtering,  L1 write-back
forward to  PCI master, and L1  write-back merged with  PCI post write
buffers  to minimize  PCI master  read latency  and  DRAM utilization.
Delayed transaction  and read caching mechanisms  are also implemented
for further improvement of overall system performance.

The Apollo MVP4 provides independent  clock stop control for the CPU /
SDRAM, PCI, and AGP buses and Dynamic CKE control for powering down of
the SDRAM.  A separate suspend-well plane is implemented for the SDRAM
control  signals  for  Suspend-to-DRAM  operation.  Coupled  with  the
324-pin Ball Grid Array VIA VT82C596B south bridge chip, a complete PC
main board can be implemented with no external TTLs.

The Apollo MVP4 controller  coupled with VIA’s highly integrated south
bridge,  the   VT82C686A,  is  ideal  for   high  performance,  energy
efficient,  and  highly integrated  computer  systems.  The  VT82C686A
supports a PCI-to-ISA bus  controller, four USB ports, dual bus-master
IDE  with UltraDMA33/66,  AC97  basic digital  audio, system  hardware
monitoring, and integrated "Super-I/O" functionality.
***Configurations:...
***Features:...
**VT82C680     Apollo P6, Pentium-Pro Chip Set               <08/30/96...
**Support chips:
**VT82C505     Pentium/486 VL to PCI Bridge                  <05/30/94...
**VT82C586/A/B PCI Integrated Peripheral Controller          <10/13/96...
**VT82C596/A   Mobile PCI Integrated Peripheral Controller   <11/05/97...
**VT82C686A/B  PCI Super-I/O Integrated Peripheral Ctrl.     <02/10/98...
**Later P-Pro/II/III/Celeron...
**Later AMD...
**Other...
*VLSI...
*Western Digital...
*Winbond...
*ZyMOS...
*General Sources:...

(c) Copyright mR_Slugs Warehouse - All rights Reserved