home .. forth .. misc mail list archive ..

Offete_News


jfox@netcom.com (Jeff Fox) wrote:
>
>Dear MISC readers:
>
>Yesterday I got the following mail from Dr. C.H. Ting :
>
>>From TingCH@ccmail.apldbio.com Mon Oct  7 10:53:39 1996
>Subject: Forth Engines
>
>Dear Forth Engine User:
>
>More On Forth Engines, Volume 21 is now published.
[snip]

>He is pretty happy with the current 
>design of I21, and the prototype chips worked fairly well, it displayed some 
>very crispy pictures on a color monitor.

This is what we need 256+ colour pictures, I mean 384 pixels accross is
great in 16 colours, but if your going to make a micro-controller, you might
have well given it an LCD interface instead otherwise, higher resolutions
are better.

>
>I21 is being developed by iTv Corp.  Its goal is to produce a low cost Internet 
>ready computer for home use.  I21 connects TV, phone, and a keyboard together 
>to allow the user to surf Internet conveniently in a home entertainment 
>environment.  It attracted substantial investments and assembled a high power 
>team of Forth programmers to produce the system.  

Money is ready, I'm still waiting otherwise I'll have to buy the 160Mhz
Newton slate or the Webbook with the Sh-boom II chip.

Eugene I forgot to mention this one in reply to your post.
 
>
>At the beginning of this year, I visited Taiwan for a Forth Workshop.  There I 
>kind of committed myself to three projects: the first is to produce a system 
>for the MuP21H chip in the 44-pin PLCC package, the second was to help 
>marketing the CT100 microcontroller teaching/prototyping kit, and the third was 
>to replace the BIOS chip on 486 motherboard to build very cheap but very 
>powerful embedded processing systems.  I wanted to report the results of these 
>projects in this volume.  This is the reason why this volume is delayed for so 
>long.
>
>The new MuP21H kit uses a 82C51 UART chip to handle the serial communication 
>with the host PC.  There are many intriguing properties in 82C51 which ware not 
>fully understood when I first laid out the PCB for it.  Consequently the first 
>board didn't work.  After I ironed out all the bugs, the second layout worked 
>perfectly.  Now we can use this system to do serious programming, using a PC as 
>the terminal server and file server.  It is interesting to note that MuP21 is 
>very close to an Internet computer, because of its video coprocessor and the 
>very versatile I/O bus.  Hooking an UART to it is only a simple demonstration 
>of what it can do.

80 columns )

>
>As the DRAM price collapsed, as it should have years ago, the price of 
>assembled MuP21 kit is also lowered from  $350 to $250.  Hope this price 
>reduction will encourage you to try out this very exciting microprocessor.  It 
>is the only microprocessor which can generate video by itself.


>
>The CT100 Lab Board is an 8051 based microcontroller laboratory which contains 
>about 20 different circuit groups for experiments on microcontroller 
>programming and interfacing.  You connect the circuit groups with jumper cables 
>and are ready to write software to exercise the hardware.  It is a great tool 
>to teach microcontrollers.  As the 8051 is driven by a F83 system, Forth is the 
>natural language to develop software on this system.  I was so impressed by 
>this system that I promised to develop a teaching course for it.  It took a 
>while, but I finished building the course and also wrote a manual for it.  I 
>like to market it in the US.  I think it is a great tool to train people in 
>microcontroller programming and Forth, of course.
>
>I had more troubles than what I asked for in trying to replace BIOS with Forth.  
>The vision was that 486 motherboards are become so cheap that they are now the 
>cheapest and most powerful embedded system money can buy.  For $100, you can 
>get a 486 board with 1 MB of DRAM, 2 serial ports, 1 parallel port, a floppy 
>controller, and a IDE hard disk controller, on an a single piece of PCB.  My 
>goal was to put an eForth in the BIOS socket and I could have a powerful 
>computing element, which can be used as a node in a parallel computing system.  
>486 microprocessor is interesting because of the 32-bit linear addressing space 
>in the protected mode, and the powerful floating point math coprocessor.  If I 
>can link 10 or 100 of these beasts together, I could have a super computer on 
>my desk.
>
>The progress was slow and tedious.  I threw away the DOS extender I used in the 
>386 eForth, V1.01, because the DOS extender itself is too complicated and not 
>suitable for ROMming.  I used Chuck's OK as the protected mode loader and get 
>the eForth to work well in the 32-bit protected mode, entered from DOS.  
>However, all attempts to replace BIOS ROM failed.  The logic analyzer showed 
>that the first subroutine call failed to return to the caller.  Investigating 
>code in BIOS made me feel hopeless because different motherboards use different 
>chip sets which are initialized differently by different BIOS chips.  It is 
>apparently impossible to build a Forth which can deal with all the motherboard 
>in a uniform way.  
>
>The next best approach was to replace the Video BIOS ROM on the video display 
>card.  This can be done uniformly, because the motherboard BIOS calls the Video 
>BIOS after the motherboard is completely initialized, and that the Video BIOS 
>is always located at C0000H.  This strategy worked, and I was able to boot up 
>eForth on a bare-bone 486 motherboard with an Award BIOS ROM.  We still need to 
>test it on other motherboards.
>
>You will find reports on these three project in this volume.  In addition I 
>included two papers I presented at the 1996 Rochester Forth Conference held in 
>Toronto in June.  Laws of Computing tries to summarize the principles of 
>programming and prove that Forth is the best programming language.  We all know 
>that, but it is nice to prove it from the first principles.  Now, we have the 
>first principles and the proof.  Traffic Controller and Zen of State Machine is 
>my answer to the Programming challenger presented at the Forth Day Conference 
>at the Silicon Valley FIG Chapter.  The most amazing discovery was that Forth 
>is the perfect language to program state machines.  No new words or new concept 
>outside the F-PC are necessary.
>
>Remember ShBoom?  It is now being developed by Patriot Scientific Corporation 
>in Poway, CA.  George Shaw was invited to SVFIG to talk about it and he 
>distributed data sheets, which are reproduced here.  Klaus Schleisiek also 
>visited SVFIG and gave a talk on his new Forth chip  IX.  His handout is also 
>included.
>
>Several users requested RTX processor and it reminded me what a nice 
>microprocessor it is.  I am still distributing the RTX kit.  If you have an 
>application which requires fast and precisely timed reponse from the computer, 
>RTX with cmForth is still the best answer.
>
>Just before the Volume 21 goes to the printer, I worked out the logic design of 
>a 16 bit microprocessor P16, modeled after Chuck Moore's MuP21.  A paper is 
>appended to the end of this volume.  P16 design is in standard schematics and 
>had passed logic simulation.  It demonstrates my understanding of the P21 
>architecture and its logic operations.
>
>Dr. C. H. Ting                        Offete Enterprise
>San Mateo, California                     1306 South B Street
>October, 1996                     San Mateo, CA 94402
>                              (415)574-8250, Fax (415)571-5004
>
>New Products from Offete Enterprises:
>
>4121  More on Forth Engines, Volume 21, $20
>Editor's Notes          
>New products from Offete Enterprises 
>MuP21H Kit
>My Troubles with This Darn 82C51
>CT100 Lab Board
>Born to be Free  
>Laws of Computing
>Traffic Controller and Zen of State Machines      
>ShBoom Microprocessor
>Programmable Fieldbus Controller IX1
>Logic Design of a 16-Bit Microprocessor P16
>
>4018  MuP21H Chip in PLCC, $40
>MuP21 is now available in 44-pin PLCC packages.  It has much smaller (1/3) 
>footprint compared to DIP, and is suitable for surface mount..
>4019  MuP21H Evaluation Kit, $100
>Include PLCC MuP21H, a printed circuit board, a 128KB EPROM with eForth V2.08, 
>instructions and assembler diskette.  Build your own stand-alone MuP21 
>development system with 82C51 UART.  
>4020  Assembled MuP21H Evaluation Kit, $250
>MuP21 system with 1Mx20 DRAM, 128KB EPROM eForth V2.08, and 82C51 UART to talk 
>to PC host.  Apply 5V power and it generates NTSC signals to drive a color TV 
>monitor.  Manual, MuP21 assembler and sample code included to guide application 
>development.
>4012  Assembled MuP21 Evaluation Kit, $350, $250
>MuP21 system with 1Mx20 DRAM, 128KB EPROM, and 8-bit parallel I/O ports.  Apply 
>5V power and it generates NTSC signals to drive a color TV monitor.  Manual, 
>MuP21 assembler and sample code included to guide application development.  
>Note price reduction due to DRAM price drop.
>1014  MuP21 Programming Manual, C. H. Ting-- $25.00
>Second Edition with MuP21H update. Primary reference for MuP21 microprocessor.  
>Architecture, instruction set, video coprocessor programming, assembler, 
>bootstrap code, Chuck Moore's OK demonstration system, and his lectures on P21 
>and OK.
>9008  CT-100 Lab Board, $400.00
>This board is manufactured in Taiwan to train students in microcontroller 
>interfacing.  It has 8051 as the CPU, surrounded by interesting devices like 
>LED and LCD displays, ADC, DAC, stepper motor controller, key array and 
>keyboard, buzzer, etc.  It provides enough materials for a one semester 
>microcontroller course at college level.  Students can get on to the 
>experiments directly without having to worry about hardware assembly.  Manual 
>is available separately as Item #2132.
>2132  CT100 Lab Board User Manual, C. H. Ting--$20.00
>CT-100 is a 8051 based microcontroller training system, with integrated 
>peripherals like LED and LCD displays, keys and keyboard input, ADC. DAC, 
>stepper motor controller, relays, RS232, 8255, and 8279 peripherals.  This 
>manual details experiments using 8051 Forth to controller these devices.
>2133  32-Bit 386 eForth V.4.02, C. H. Ting--$25.00
>This is a 32 bit implementation of eForth for 80386 in the protected mode.  
>  It enters into protected mode directly without a DOS extender.  Only COM1 is 
>active at 9600 baud.  Keyboard, screen and disks are all silenced. You need 
>another PC to serve as a host through the serial port.
>2134  32-Bit 386 eForth V.4.03, C. H. Ting--$25.00
>This is a 32 bit implementation of eForth for 80386 in the protected mode.  
>  It replaces the Video BIOS at C0000H and booted by system BIOS.  It converts 
>a 386/486 motherboard to an embedded Forth processor, communicating with a PC 
>through the COM1 port..
>
>


--