mark@intek01.UUCP (Mark McWiggins) (04/01/89)
Any body have experience with these? Force makes real-time VME-based single-board machines running Uniflex, and we're considering them for a project. Thanks in advance for any insight.
larry@kitty.UUCP (Larry Lippman) (04/04/89)
In article <180@intek01.UUCP>, mark@intek01.UUCP (Mark McWiggins) writes: > Any body have experience with these? Force makes real-time VME-based > single-board machines running Uniflex, and we're considering them > for a project. I don't know anything about Force, but I do know something about UniFLEX, which is sold by Technical Systems Consultants, Inc. UniFLEX is the most awful "lookalike" to UNIX that I have ever seen. On page one of the UniFLEX manual it is claimed "The system's design has been influenced primarily by two operating systems, FLEX (tm) and UNIX. UniFLEX retains the flexibility and ease of use of FLEX while incorporating some of the widely accepted structures of UNIX." Don't believe it; in fact, any resemblence to UNIX is purely coincidental. Don't take my word for this; just ask to examine a UniFLEX manual for yourself. Want to bring the system to single-user mode? Forget about any UNIX init(1) commands - there ain't any; just use "shutup-1". I kid you not! What to create a zero length file? Well, forget about using something like "> foo". However, we do have the command "create-1". How about the "info-1" command, which will "display the contents of the information field associated with the specified binary file"? Hmmmm... features of UNIX, you say? How about "jobs-1" instead of ps(1)? It returns "No tasks active" if nothing is running other than the login shell. Ain't no cat(1), but there is "list-1". How's this to logout? No ^D, or logout. Nope, we have "log-1". Forget about chmod(1) or chown(1), we have "perms-1" and "owner-1" respectively. "Widely accepted structures of UNIX"?! Pardon me while I barf. <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP: {allegra|ames|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> VOICE: 716/688-1231, 716/773-1700 {att|hplabs|utzoo}!/ <> FAX: 716/741-9635, 716/773-2488 "Have you hugged your cat today?"
jac@penguin.UUCP (James Carter) (04/04/89)
In article <180@intek01.UUCP>, mark@intek01.UUCP (Mark McWiggins) writes: > Any body have experience with these? Force makes real-time VME-based > single-board machines running Uniflex, and we're considering them UniFLEX operating system was written (and copyrighted) by Technical Systems Consultants, currently of Chapel Hill, N.C. They started with a single user o/s called FLEX that was brought out back around 1976. It ran on Southwest Technical Products (Motorola) 6800's. They incorporated a subset of UNIX into it, and changed its name to UniFLEX(tm) in about 1978. It is currently in revision 4.14 +/-, and is a pretty solid o/s. As far as I know, they were running on several different makes of cpu, but were still pretty much tied to the Motorola chipset. I think they are running on the 68030 now. For a new user, their o/s is A LOT EASIER to learn. They don't have the standard "cd, pwd, ls" however, they tend to lean toward their early FLEX systems by using "chd, path, dir". I still support a couple of their systems that were installed back in '78. {Sometimes I wish they would break more often ;-)}. -- ========================================================================== Disclaimer: are you kidding? I own the place! James A. (JC) Carter Penguin Business Systems, Inc.
john@jwt.UUCP (John Temples) (04/06/89)
In article <3069@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes: >In article <180@intek01.UUCP>, mark@intek01.UUCP (Mark McWiggins) writes: >> Any body have experience with these? Force makes real-time VME-based >> single-board machines running Uniflex, and we're considering them >> for a project. > UniFLEX is the most awful "lookalike" to UNIX that I have ever seen. I'll second Larry's opinion. We evaluated UniFLEX on a Force VME board. The hardware itself seemed fine, but UniFLEX's claims of being "UNIX-like" were quite bogus. A couple additional problems we found not mentioned in Larry's posting were: The size of the code space had to be a power of two bytes. For larger programs, you jumped from 256K to 512K to 1024K, and if you had 1MB of RAM, you couldn't run a program that should have had a 600K executable. malloc(3C) was incredibly slow. A program we compiled which did a lot of memory allocation would take _several seconds_ to do the mallocs. Needles to say, UniFLEX went back to the vendor. -- John Temples - UUCP: {uiucuxc,hoptoad,petsd}!peora!rtmvax!bilver!jwt!john