Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!watmath!clyde!burl!ulysses!bellcore!decvax!decwrl!pyramid!hplabs!tektronix!reed!psu-cs!omepd!davec From: davec@omepd.UUCP Newsgroups: net.info-terms Subject: Re: Wyse 60 review + termcap Message-ID: <105@omepd> Date: Fri, 8-Aug-86 14:24:04 EDT Article-I.D.: omepd.105 Posted: Fri Aug 8 14:24:04 1986 Date-Received: Mon, 11-Aug-86 04:52:34 EDT Reply-To: davec@omepd.UUCP (Dave Cobbley) Followup-To: net.info-terms Distribution: net Organization: Intel Corp. Hillsboro, Oregon Lines: 126 Keywords: Wyse 60 Summary: WY-60 problems, more termcaps The group I work with ordered 30 Wyse 60 terminals the day they were officially announced back in April. We specified the white phosphor versions with the native keyboards. Most of them are used in native mode, since it is very similar to the WY-50 that was being used here by some previously. We have found them to be quite usable terminals, although we have run into several problems. We selected the WY-60 because of number of lines on the screen, hidden attributes, keyboard layout, user interface, and of course, price. It had more features we wanted/needed than almost any other terminal in its price range [nearest competitor was the Falco 500; rejected for price (we didn't need full ANSI), keyboard layout, "rogue-like" setup interface("you are in a maze of twisted passages that all look alike")]. The 9600 baud handshaking requirement has been a bother for us also, especially the emacs users. One user has gone so far as to rlogin to another machine and then back again in order to slow things down enough to not require flow-control. Following is a complaint list from local users that I compiled recently at the request of Wyse factory support: - The keyboard is often ignored when a lot of output is being sent to the terminal at the higher baud rates, or when the terminal is doing screen clears, etc. Wyse has admitted that the keyboard scan rate is extremely slow and at low priority. You cannot, for example, invoke the "vi" editor and then type ahead as the screen is being painted, as many input characters may be lost (often the beginning "/" of a search pattern, and then the following pattern characters are interpreted as vi commands!). Another user has written a program to exercise the graphics capabilities of the terminal by having it draw various geometric shapes continuously on the screen. The slow scan rate often prevents interrupts from being received, requiring the process to be killed from another terminal. - Tab stops get set incorrectly when the screen is in 80 column mode and you try to set tops off the edge of the screen. Also, you cannot reliably switch back and forth between 80- and 132- column mode without the tab stops being messed up. Escape characters also often are lost during this time. - Command sequences intended for the terminal show up on the screen and the intended operation doesn't happen. In other words, the sequences aren't recognized as commands. Similarly, tabs are sometimes set at the wrong columns. One guess as to the cause of the problem is that there are holes in input flow control and that the input buffer is being overrun, causing characters to be lost. It may be spending too much time at interrupt level. - The terminal clock loses time. If you load the terminal, it loses a lot of time. Sending continuous clear rectangle commands at 9600 baud will make time stand still. (It will also severely starve the keyboard.) The clock also stops when the terminal is powered off, instead of being powered via shutdown power or batteries. - ^S/^Q stuff doesn't seem quite right. When working under VMS (telnet'd through a 4.2 Vax), the host VAX often gets a ^S and stops output, but not the corresponding ^Q. (This may be a telnet/UNIX/VMS problem, but has never been seen with the WY50, AAA, or other terminal here.) We were told that new firmware is soon forthcoming. I'm not sure if firmware can solve all the problems; it appears that in going from the WY-50 to the WY-60, they added a lot of features without proportionately increasing the processing power to implement those features. A lot of the problems we've seen, though, are indeed the kind often seen in a freshly-released product, and hopefully will be rectified soon. Here are several termcaps for the WY-60 that have been in use here for ~3 months. Of course the vendor never tells you how much delay you need to add for each function, and we haven't had time to sit down and time each action, so the paddings are likely to be way off in several cases. We depend more on flow-control to keep things slowed down enough, and will continue to do so until the firmware stabilizes. w6|wy60|wy60-43|Wyse 60 w/ 43 lines:\ :is=\E~4\Ee6\Ee1\Ec21\E`\72\Ee+:\ :vs=\EZ1+\E[A\177\EZ1,\E[B\177\EZ1-\E[D\177\EZ1.\E[C\177\EZ1*\E[H\177:\ :ve=\EZ0+^K\177\EZ0,^J\177\EZ0-^H\177\EZ0.^L\177\EZ0*^^\177:\ :al=\EE:am:bs:bt=\EI:pt:ct=\E0:st=\E1:ce=\ET:cm=\Ea%i%2R%3C:\ :cl=150\E*:cd=150\EY:co#80:dc=\EW:dl=\ER:ei=\Er:ho=\E{:\ :LC:im=\Eq:ku=\E[A:kd=\E[B:kr=\E[C:kl=\E[D:kh=\E[H:k0=\EQ:\ :k1=\EW:k2=\ET:k3=\Er:k4=\EE:k5=\ER:k6=\EY:k7=\Eq:k8=\EK:\ :k9=\EJ:li#43:mi:mb=\EG2:md=\EG4:mh=\EGp:mr=\EG4:me=\EG0:\ :so=\EG4:se=\EG0:us=\EG8:ue=\EG0:nd=^L:up=^K:\ :hs:ts=\EF:fs=\r:ds=\EF\r: ws|wy60-s|wy60-42|Wyse 60 w/ status line:\ :is=\E~4\Ee6\Ee1\Ec21\E`\72\Ee*:li#42:i2=\Ez\177\EA1t\Ez(\r:\ :hs:ts=\Ez(:fs=\r:ds=\EA10\Ez(\r:tc=wy60: wa|wy60-w|wy60-43-w|Wyse 60 wide - 132 columns:\ :is=\E~4\Ee6\Ee1\Ec21\E`;\Ee+:co#132:tc=wy60: wd|wy60-w-s|wy60-42-w|Wyse 60 wide - 132 columns w/ sysline:\ :is=\E~4\Ee6\Ee1\Ec21\E`;\Ee*:co#132:tc=wy60-s: wb|wy60-24|wy60-24-s|Wyse 60 with 24 lines on the screen w/ sysline:\ :is=\E~4\Ee6\Ee1\Ec21\E`\72\Ee(:li#24:tc=wy60-s: wc|wy60-24-w|Wyse 60 with 24 lines by 132 columns:\ :is=\E~4\Ee6\Ee1\Ec21\E`;\Ee(:li#24:tc=wy60-s: # # Reverse video versions of Wyse 60 # wr|wy60-rv|wy60-43-rv|Wyse wy-60 w/ 43 lines:\ :is=\E~4\Ee6\Ee1\Ec21\E`\72\Ee+\EA04\EA14\EA2|\EA3|:tc=wy60: wS|wy60-s-rv|wy60-rv-s|wy60-42-rv|Wyse 60 w/ status line:\ :is=\E~4\Ee6\Ee1\Ec21\E`\72\Ee*\EA04\EA14\EA2|\EA3|:li#42:\ :i2=\Ez\177\EA1t\Ez(\r:hs:ts=\Ez(:fs=\r:ds=\EA14\Ez(\r:tc=wy60: wA|wy60-w-rv|wy60-43-w-rv|Wyse 60 132 columns reverse video:\ :is=\E~4\Ee6\Ee1\Ec21\E`;\Ee+\EA04\EA14\EA2|\EA3|:co#132:tc=wy60: wD|wy60-w-s-rv|wy60-42-w-rv|Wyse 60 132 columns reverse video w/ sysline:\ :is=\E~4\Ee6\Ee1\Ec21\E`;\Ee*\EA04\EA14\EA2|\EA3|:co#132:tc=wy60-s-rv: wB|wy60-24-rv|wy60-24-s-rv|Wyse 60 with 24 lines on the screen w/ sysline:\ :is=\E~4\Ee6\Ee1\Ec21\E`\72\Ee(\EA04\EA14\EA2|\EA3|:li#24:tc=wy60-s-rv: wC|wy60-24-w-rv|Wyse 60 with 24 lines by 132 columns:\ :is=\E~4\Ee6\Ee1\Ec21\E`;\Ee(\EA04\EA14\EA2|\EA3|:li#24:tc=wy60-s-rv: Dave Cobbley Intel Corporation Hillsboro, Oregon Phone: (503) 681-4157 Usenet: {decvax,ucbvax}!tektronix!ogcvax!inteloa!omepd!davec