corpenin@kodak.UUCP (darrell corpening) (10/14/86)
< line eater > Has anyone done a comparison of the various FORTH's available for the CA? o JForth is about $100 and claims blazing speed and says it will run sieve in 10 sec. o Mach1 was suppose to be out in Sept., but haven't seen anything yet -- However, they claim to run sieve in 2-3 sec. on a MAC. o Creative Solutions Forth is about $180 and I have seen some good words in the reviews, but nothing on speed. o UBZ Forth has been around awhile, but I have seen no reports. Any comparisons/recommendations would be appreciated!! Darrell Corpening Eastman Chemicals Division Kingsport Tennessee (615)-229-3298
keithe@tekgvs.UUCP (Keith Ericson) (10/15/86)
In article <594@kodak.UUCP> corpenin@kodak.UUCP (darrell corpening) writes: >< line eater > > >Has anyone done a comparison of the various FORTH's available for the CA? >o UBZ Forth has been around awhile, but I have seen no reports. > >Any comparisons/recommendations would be appreciated!! > I've been trying to order UBZ Forth for months! The Tek purchasing department required that they be able to talk to someone at UBZ, but they could never get an answer at their Mableton GA phone number. I had called them earlier and HAD talked to someone, but couldn't later on. I called FIG since they were still running advertisements in Forth Dimensions to find out - not much help. To date I haven't placed an order since I'm not sure that UBZ Software really exists. (My strong suspicion is that UBZ is their "night job." But what else is new?) I'd like to know about this if anyone HAS had any good luck with UBZ. I'm waiting to send in my order! keith
kim@amdahl.UUCP (Kim DeVaughn) (10/18/86)
[ "Send lawyers, guns, and money ..." ]
Warning: If you don't like seeing Product Announcements on the
net, hit your 'n' key now ... flames to /dev/null!
The Delta Research people were at the BADGE meeting last night (10/16),
and were demonstrating their "JForth" product.
I'm not a "Forth person", so I can't really evaluate their product, but
since there has been some discussion on the availability, etc. of various
Forth's for the Amiga, I thought I'd go ahead and pass this along.
The following is extracted from Delta Research's flyer on their JForth
product. I've edited out most of the superlative adjectives that are
typical in product flyers, and tried to extract the facts (as stated
by Delta). Please do not consider this to be an endorsement of their
product by me, or my employer. (Whew!)
With that said ...
vvvvvvvvvvvvvvvvvvvvvvvvvvvvv JForth vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
developed for the Amiga
Completely interactive programming environment.
JForth contains both an interpreter and a compiler in one language,
providing an interactive environment. You may compile programs
directly from the keyboard.
Programs are interactively debugged immediately following compilation.
Whereas Manx 'C' requires about 3.5 minutes to compile the Sieve of
Erastothenes, JForth completely compiles the standard Forth version
in less than 2 seconds.
JForth utilizes a technique known as "JSR-Threading" to tie together
compiled programs. This essentially means that all code produced by
the compiler is directly executable.
The following chart compares JForth with 2 other Forth's, one of which
is similarly JSR-Threaded. The operations are common Forth primatives
used in virtually all Forth programs. The sieve mentioned is the
"Standard Unmodified" Sieve of Erastothenes benchmark.
Creative-Solutions | Palo Alto Shipping | Delta Research
Multi-Forth | Mach2 | JForth
(Amiga 7.16MHz) | (Mac-Plus 8MHz) | (Amiga 7.16MHz)
Indirect-Threading | JSR-Threading | JSR-Threading
| |
---Sequence-----------Time (in sec.) for 1,000,000 sequences-----------
| |
Empty DO LOOP 14.14 | 3.75 | 2.86
SWAP 11.22 | 9.68 | 4.64
DUP DROP 11.14 | 5.15 | 3.54
OVER DROP 11.72 | 5.98 | 5.96
23 45 + DROP 26.06 | 15.66 | 8.62
>R R> 12.84 | 11.61 | 7.16
ROT 15.44 | 23.61 | 9.92
23 2* DROP 19.38 | 11.60 | 5.34
Sieve (10 iterations) 21.48 | 11.61 | 8.82
Much more besides speed:
+ Object-Oriented dialect (similar to NEON)
+ ASCII or 'SCREEN files (SCR editor provided)
+ 1983 Standard (1979 and FIG loadable)
+ Complete DEBUG package
+ Complete Floating Point package [sorry, I didn't ask if it is IEEE]
+ JForth newsletter (free)
+ Local Stack variables
+ Assembler and Disassembler
+ Complete 32-bit implementation
+ Call ANY Amiga function by NAME
+ Optimizing target compiler ... no royalties
+ Many utilities provided
+ Updates available for postage, handling, and media
+ Elegant Amiga STRUCTURE interface
Price: $99.95 (CA residents add 7% sales tax)
Contact: Delta Research
4054 Wilkie Way
Palo Alto, CA 94306
415-856-3669
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
I did ask a few of the more obvious questions:
Q: What's the availability of JForth?
A: We're shipping it this month (October).
Q: Is it copy protected?
A: No! [ Much applause from the onlookers! ]
Q: Does it support ALL the Amiga functions?
A: Yes. (Everything that's in any Amiga Library.)
Q: Does it work with 1.1 and 1.2?
A: Yes, it was developed under 1.2.
Q: Does it support extended (Fast) memory?
A: We're running it with Pacific Cypress's memory board/expansion box
right now.
My own impressions are that it is indeed a fast programming system, and
the product looked stable. The person giving the demo had 6 or 7
JForth windows open at once, showing code, debug screens, program
execution that was using graphics calls, etc.
Hmmmm ... I've always wanted to learn about Forth ... maybe now is
the time. (Time ... who has enough of it :-) )
Just for the record: I'm not associated with Delta Research in any
way ... just passing along the information.
/kim
--
UUCP: {sun,decwrl,hplabs,pyramid,ihnp4,seismo,oliveb}!amdahl!kim
DDD: 408-746-8462
USPS: Amdahl Corp. M/S 249, 1250 E. Arques Av, Sunnyvale, CA 94086
CIS: 76535,25
[ Any thoughts or opinions which may or may not have been expressed ]
[ herein are my own. They are not necessarily those of my employer. ]tenney@well.UUCP (Glenn S. Tenney) (10/18/86)
There is a public domain version of MVP FORTH that has been
available for a year. It is on one of the early FISH disks
(with complete source -- remember, public domain) with
documentation available from Mountain View Press (415) 961-4103.
Performance was not a prime criteria for this 79 STANDARD system,
so you might get a much better performer elsewhere.
MVP FORTH for the Amiga was the first FORTH available on the Amiga.
-- Glenn Tenney
UUCP: {hplabs,glacier,lll-crg,ihnp4!ptsfa}!well!tenney
ARPA: well!tenney@LLL-CRG.ARPA Delphi and MCI Mail: TENNEY
As Alphonso Bodoya would say... (tnx boulton)
Disclaimers? DISCLAIMERS!? I don' gotta show you no stinking DISCLAIMERS!jesup@steinmetz.UUCP (Jesup) (10/19/86)
In article <594@kodak.UUCP> corpenin@kodak.UUCP (darrell corpening) writes: >Has anyone done a comparison of the various FORTH's available for the CA? >o JForth is about $100 and claims blazing speed and says it will >run sieve in 10 sec. I believe their numbers, because I wrote a subroutine threaded Forth for the amiga that does it in 9 sec (for 10 iterations.) I may release it to the net, if I ever have time to clean it up a little. >o Mach1 was suppose to be out in Sept., but haven't seen >anything yet -- However, they claim to run sieve in 2-3 sec. on a MAC. I don't believe that, if it's for 10 iterations, unless they did it entirely in assembler. More probably one iteration, which fits with a very fast threaded version, or a slow subroutine threaded version. Randell Jesup jesup@ge-crd.ARPA seismo!rochester!steinmetz!jesup