forth, GForth EC on PSC1000/ShBoom and others

GForth EC on PSC1000/ShBoom and others

Author	Message
Jens #1 / 6	GForth EC on PSC1000/ShBoom and others Hi! You will find some status report about the GForth EC development together with some demo files at: www.jwdt.com/gforth/ec Current targets are: 6502 C165 PSC1000 8086 bye, Jens Wilke
Wed, 19 Jan 2000 03:00:00 GMT

Keith Wootte #2 / 6	GForth EC on PSC1000/ShBoom and others [snipped] Quote: >Hi! >You will find some status report about the GForth EC development >together with some demo files at: >www.jwdt.com/gforth/ec >PSC1000 >Jens Wilke Thanks! -- Keith Wootten
Thu, 27 Jan 2000 03:00:00 GMT

Keith Wootte #3 / 6	GForth EC on PSC1000/ShBoom and others Quote: >Hi! >You will find some status report about the GForth EC development >together with some demo files at: >www.jwdt.com/gforth/ec >Current targets are: >6502 >C165 >PSC1000 >8086 >bye, Jens Wilke I bought my PSC1000 (ShBoom) demo board a couple of months ago, with a view to evaluating the processor for a (not-too-distant) future embedded project. I was a bit disappointed to find no mention of Forth in the documentation, even though it's a stack machine - the assembler mnemonics are documented with stack comments but no mention of the dreaded F-word. I tried the supplied C compiler. The example ran and said "Hello, World". I modified this to say "Hello, Keith". My first real C program and already I was hating it! :-) Then I downloaded Jens Wilke's Forth demo. Within five minutes I had made an LED on the output port turn on an off using only a 32bit 40MHz stack machine and a 200MHz Pentium Pro terminal!!! This would have taken several man-years in C. All embedded systems start with an LED being turned on and off. This is an inviolable law of nature. Tomorrow, the world. Thanks for the kick-start. I look forward to future developments. It's notable that while the ShBoom is without doubt a Forth machine, it's being touted as a Java engine (and supplied with a C compiler). If this means that it's going to become a large volume established part, then I'm all for it. Makes you wonder if the marketeers feel that mentioning Forth would have a negative impact. [ BTW PSC1000 demo boards are available in the UK from Thame Components for about UKP200. The processor itself is ridiculously small in a 100 pin PQFP. It's made by NatSemi - this and the packaging imply that either it's a serious part or somebody's going to lose a lot of money. The former, IMO. ] -- Keith Wootten
Thu, 03 Feb 2000 03:00:00 GMT

Marcel Hendr #4 / 6	GForth EC on PSC1000/ShBoom and others Quote: > I bought my PSC1000 (ShBoom) demo board a couple of months ago, with a > view to evaluating the processor for a (not-too-distant) future embedded > project. I was a bit disappointed to find no mention of Forth in the > documentation, even though it's a stack machine - the assembler > mnemonics are documented with stack comments but no mention of the > dreaded F-word. I tried the supplied C compiler. The example ran and > said "Hello, World". I modified this to say "Hello, Keith". My first > real C program and already I was hating it! :-) Who wrote the C-compiler? And how many bugs are still in it :-) I suppose it's quite an investment somebody has made there, especially as for UKP200 for a full kit (?) they seem to be giving it away. That would hardly cover the price of the manuals, unless they plan to sell an enormous quantity. [..] Quote: > [ BTW PSC1000 demo boards are available in the UK from Thame Components > for about UKP200. The processor itself is ridiculously small in a 100 > pin PQFP. It's made by NatSemi - this and the packaging imply that > either it's a serious part or somebody's going to lose a lot of money. > The former, IMO. ] I thought the ShBoom ran at 200 MHz? And why advocate Java for a controller? Did I miss something again :-) -marcel
Tue, 08 Feb 2000 03:00:00 GMT

Eric Dorm #5 / 6	GForth EC on PSC1000/ShBoom and others Quote: >I thought the ShBoom ran at 200 MHz? 'Fraid not. The current uP docs (which came yesterday, yay :) ) say 100Mhz internal clock is the max on the current stuff. Their blurb says they have room to go faster with the smaller processes... Quote: >And why advocate Java for a controller? Did I miss something again :-) Because they want to sell more psc1000s :) If the '1000 executes java bytecodes _at least_ as fast as x86 hardware (with just-in-time compilation) it'll be cheaper to use '1000s and run JavaOS in PDAs, set-top boxes and NCs than to carry around enough Intel/Microsoft junk to run Win95 or whatever (they hope :) ). I've always been very interested in pen computers, PDAs, and what they now call 'NCs'; I'm still searching for the perfect device for each task. The psc1000 looks like a really cool unit for these applications. Quote: >-marcel Regards, Eric Dorman
Tue, 08 Feb 2000 03:00:00 GMT

Keith Wootte #6 / 6	GForth EC on PSC1000/ShBoom and others writes [snipped] Quote: >>I thought the ShBoom ran at 200 MHz? >'Fraid not. The current uP docs (which came yesterday, yay :) ) say 100Mhz >internal clock is the max on the current stuff. Their blurb says they >have room to go faster with the smaller processes... The evaluation board runs at 40MHz with promises of a 100MHz PSC1005 soon. Maybe the 200MHz is confusion with Chuck Moore's MISC chips? Quote: >>And why advocate Java for a controller? Did I miss something again :-) >Because they want to sell more psc1000s :) If the '1000 executes >java bytecodes _at least_ as fast as x86 hardware (with just-in-time >compilation) it'll be cheaper to use '1000s and run JavaOS in PDAs, set-top >boxes and NCs than to carry around enough Intel/Microsoft junk to run Win95 >or whatever (they hope :) ). I think the Java angle is for a 'Web Book' - obviously this is where Patriot see the volume market. -- Keith Wootten
Wed, 09 Feb 2000 03:00:00 GMT

Page 1 of 1

[ 6 post ]