DEC & ALPHA CPU's  

LSI 11/2 CPU Module
This module was used as CPU of the PDP-11/3 System. The LSI-11/2
offered
83 standard PDP11 instructions, 8 optional EIS/FIS instructions,
32 KW adressing range, MicroODT (Octal Debugging Technique on the console)
and one interrupt level.
source: http://hampage.hu


DEC M8186 11/23 CPU MODULE KDF11-A

 

The KDF11-A module has four 40-pin sockets for the chipset. One is for the DCF11 two-chip hybrid (21-15541AB data path and 23-001C7 control chip), one's for the KTF11 MMU option and one's for the KEF11 floating-point option (another two-chip carrier; the MMU must be installed to use the FPP option). The remaining socket can be used with the CIS option, which spans two sockets wide (so you loose the FPP). The KDF11-B has five sockets, so there are no problems
with this.

In floating-point-heavy applications, the FPF11 can be used. This is a quad-height module that is connected to the KEF11-socket on CPU board with a cable. It's based on bit slice processors, the performance is six times of the performance of the KEF11 option. (64-bit data path, 17-digit accuracy)

source: http://hampage.hu

 


DEC M8190-AE KDJ11-BF CPU PDP 11/84

This PDP-11/84 CPU, M8190 or KDJ11-BF (1985), microprocessor-based J-11, was a high-end PDP, which supported advanced solutions such as memory cache (mounted on same card), the FPU on board and memory interconnect system, a communication channel between high-speed CPU and memory. The J-11 (1983) was essentially an implementation of the microprocessor in the form of PDP 11/70, with support for SMP and cache memory. The execution of floating point instructions can be accelerated by adding an optional FPU. Like its predecessor F-11. The J-11 also supported the instruction set "commercial" CIS (by adding a special microcode ROM). The J-11 is a multi-chip microprocessor, consists of two separate base unit (Control Chip and Data Chip) and a possible third unit, the FPU (FPA Floating-Point Accelerator).
 
Chip Identification
  DC334: data chip (DC), 40K transistors, Harris "D4-6900-5"
DC335: control chip (CC), 80K transistors, Harris "D4-6901-5"
DC334 + DC335 = dual-chip ceramic module DEC "DCJ11-AC 57-19400-09"
DC321: floating-point unit (FPA), 34K transistors, DEC "21-21858-08"
DC351: DMA controller gate array, LSI Logic L1A0305, "21-22800-01"
DC350/DC394: system and cache controller gate array TC176080AY-0001

Source: http://ummr.altervista.org



DEC KA820-AA CPU


The DEC KA820-AA (1985), based on the bus VAX-BI, contains the CPU V-11
"Scorpio" clocked at 5 MHz. This processor was used in the VAX minicomputer
 8250/8350 (1986). The V-11 was from a historical point of view the first implementation of the architecture in the form of microprocessor VLSI VAX.
The V-11 CPU has four distinct units: the chip ROM / RAM, the integer
execution unit, a unit of memory management, cache and address translation,
 which also acts as a I/O controller, and an accelerator for floating point operations


 
Chip Identification
  DC327: RAM / ROM 208K transistors
DC327: Execution Unit (IE), 60K transistors
DC329: System Control Chip (M chip), 54K transistors
DC330: Floating Point Unit (F chip), 29,6K transistors



Source: http://ummr.altervista.org

 


DEC KA680-AA CPU

This is a processor for VAX machines 4000/M500. The board hosts a 75 MHz CPU
 NVAX (left, with the gold heat spreader) . The NVAX was the last step evolutionary
VAX architecture, and the fastest CISC CPU available (1991). Developed in parallel
with the Alpha 21064 RISC processor with similar production process, technological
 and human resources is widely regarded as the maximum achievable within CISC. Contains approximately 1.3 million transistors in CMOS4 process with 0.75 um, triple
 metallization, reduced to 0.50 um in the "Plus" version (CMOS5). As for the complexity is comparable NVAX at Intel 80486, but the performance and
architectural solutions are definitely higher! This is a single processor chip (which
 contains both the floating point unit as the cache controller) based on macro-pipelined architecture type with branch prediction, highly optimized implementation
of VAX instructions with the use offewer possible clock cycles.
 
Chip Identification
  DC246: NVAX CPU
DC243: NDAL to CDAL I/O interface chip (NCA): "DEC 243C 21-33499-02"
DC244: Main Memory Controller (NMC): "DEC 244B 21-33500-01"
DC527: Q22 bus interface controller (CQBIC): "21-25972-02"
DC541: Ethernet Interface (SGEC): "21-28059-03"
DC542: DSSI interface (SHAC): "21-27667-02"
DC511: System support chip (SSC): "DEC 78332-GA 511H 21-24942-03"
DC509: clock chip (CC): "21-24673-02"

Source: http://ummr.altervista.org




Processor Card AlphaServer 1000 4/266



This is a processor card for an AlphaServer 1000 4/266, DEC 21064 CPU
clocked at 266 MHz. The AlphaServer 1000 midrange system was a single CPU
based architecture EISA / PCI. One could expand the RAM up to 1 GB. The
card contains the DEC chipset 21072.

Source: http://ummr.altervista.org
 

 
Processor Card AlphaServer 1200 5/533

CPU module Alpha 21164A 533 MHz. Contains  4 MB of L2 cache
 

 

Processor Card AlphaServer 21264/ 733