hyc@math.lsa.umich.edu (Howard Chu) (11/20/90)
Sorry to stir up the pot with what seems to be a now-dead subject, but I don't recall the resolution of all the horror stories, and now have a (horrifying) 1040STe of my own to deal with. The main problem I'm having is that just about everything that loads up gets partially written onto video memory. I get vertical lines, I figure 16 pixels apart. I would have expected a plain-old overwrite problem to wipe out a couple rows, starting at the top of the screen. Instead I get 1 pixel wide vertical lines, evenly spaced out, and occasionally the bottom of the screen gets trashed. What's going on? The STe arrived with the STE_FIX program in the AUTO folder of its language disk. Can't boot into medium resolution without it. Yeehah. Along other lines... I just started using ICD's boot stuff with this new host adapter... Apparently, whenever the driver detects a write failure, the thing shifts the video into medium resolution. What I'm seeing is on my multisync monitor in monochrome mode, all of a sudden there'll be a glitch and my screen has been divided in half, vertically down the middle. Everything is drawn in both the left and right halves of the screen. The screen borders are resized, and look like the color rez borders, but everything stays in black and white. The preferences menu still says high rez. This is incredibly annoying, as well as making some stuff unreadable. (Although some of it is interesting, really. Anyone for a 4 pixel wide font? 160 columns of text. Oooo....) It says it's version 4.8.4, by the way. -- -- Howard Chu @ University of Michigan Mac// - adv., q.v. MacToo, e.g. McHave a McHappy McDay! McThanks, McYou MacToo!
bill@mwca.UUCP (Bill Sheppard) (11/22/90)
In article <1990Nov20.085041.25605@math.lsa.umich.edu> hyc@math.lsa.umich.edu (Howard Chu) writes: >Along other lines... I just started using ICD's boot stuff with this new >host adapter... Apparently, whenever the driver detects a write failure, the >thing shifts the video into medium resolution. What I'm seeing is on my >multisync monitor in monochrome mode, all of a sudden there'll be a glitch and >my screen has been divided in half, vertically down the middle. Everything is >drawn in both the left and right halves of the screen... >-- > -- Howard Chu @ University of Michigan I've seen the same thing (with the double mono screen - a cheap two-page monitor!) This happened when I was using a multisync in mono mode - overnight I turn off the monitor but leave the ST on - when I came in the next morning and turned on the monitor I saw what you describe. I suspect it has something to do with the synchronizing circuitry in the monitor becoming confused (in my case, when I turned on the monitor with the ST already on). Perhaps your ICD driver/STe is pulsing the horizontal sync in such a way that the monitor gets confused... -- ################################################################################ # Bill Sheppard -- bills@microware.com -- {uunet,sun}!mcrware!mwca!bill # # Microware Systems Corporation --- OS-9: Seven generations beyond __/_!! # #######Opinions expressed are my own, though you'd be wise to adopt them!#######
jvt@its.bt.co.uk (John Trickey) (11/23/90)
In article <1990Nov20.085041.25605@math.lsa.umich.edu> hyc@math.lsa.umich.edu (Howard Chu) writes: >Sorry to stir up the pot with what seems to be a now-dead subject, but I don't >recall the resolution of all the horror stories, and now have a (horrifying) >1040STe of my own to deal with. The main problem I'm having is that just >about everything that loads up gets partially written onto video memory. I get >vertical lines, I figure 16 pixels apart. I would have expected a plain-old >overwrite problem to wipe out a couple rows, starting at the top of the screen. >Instead I get 1 pixel wide vertical lines, evenly spaced out, and occasionally >the bottom of the screen gets trashed. What's going on? > >The STe arrived with the STE_FIX program in the AUTO folder of its language >disk. Can't boot into medium resolution without it. Yeehah. > >Along other lines... I just started using ICD's boot stuff with this new >host adapter... Apparently, whenever the driver detects a write failure, the >thing shifts the video into medium resolution. What I'm seeing is on my >multisync monitor in monochrome mode, all of a sudden there'll be a glitch and >my screen has been divided in half, vertically down the middle. Everything is >drawn in both the left and right halves of the screen. The screen borders are >resized, and look like the color rez borders, but everything stays in black and >white. The preferences menu still says high rez. This is incredibly annoying, >as well as making some stuff unreadable. (Although some of it is interesting, >really. Anyone for a 4 pixel wide font? 160 columns of text. Oooo....) It says >it's version 4.8.4, by the way. OUCH! Firstly apologies for including most of the original but I felt deletions would destroy the sense of Howard's message. I too have an ICD host adaptor with the same software. However I do not have STE_FIX and I am certainly not getting most of these problems. The only one I get is not booting Med-Res if the desktop is saved in Lo or Med. One *simple* fix I've found is to save the desktop in Hi-res then when you boot with a colour monitor it comes up in Med. I believe there is a hardware bug in the latest ICD board. I have contacted them & they are sending me a daughter board "which will cure it". Cannot confirm yet as Snail Mail is still doing its worst!! The problem is around noise on the DMA bus. To see if you have this problem, turn on write verify. If it reports error, you probably have read errors also. A short term fix is to find a DMA lead which allows the disk to write without error but it will vary from drive to drive. This may be the cause of your other problems - no promises though. John. -- John Trickey <jvt@its.bt.co.uk> || ..!mcsun!ukc!axion!its G4REV @ GB7SUT Voice: +44 21 333 3369 #include <std/disclaimer>
hyc@math.lsa.umich.edu (Howard Chu) (11/25/90)
In article <1751@mwca.UUCP> bill@mwca.UUCP (Bill Sheppard) writes: >I've seen the same thing (with the double mono screen - a cheap two-page >monitor!) This happened when I was using a multisync in mono mode - overnight >I turn off the monitor but leave the ST on - when I came in the next morning >and turned on the monitor I saw what you describe. I suspect it has something >to do with the synchronizing circuitry in the monitor becoming confused (in >my case, when I turned on the monitor with the ST already on). Perhaps your >ICD driver/STe is pulsing the horizontal sync in such a way that the monitor >gets confused... Well, I've seen it in low-rez mode now, too. In mono, what happens is the video shifter gets set to medium rez, but lineA, VDI, AES, etc., aren't notified of any change. From low-rez mode, you get the same vertical division of the screen, only this time, the right half is black. It's really strange, I didn't think the way memory was mapped would allow something coherent to be displayed when modes shifted like this, but there's no scrambling at all. -- -- Howard Chu @ University of Michigan Mac// - adv., q.v. MacToo, e.g. McHave a McHappy McDay! McThanks, McYou MacToo!
bill@mwca.UUCP (Bill Sheppard) (11/28/90)
In article <1990Nov25.005225.20937@math.lsa.umich.edu> hyc@math.lsa.umich.edu (Howard Chu) writes: >In article <1751@mwca.UUCP> bill@mwca.UUCP (Bill Sheppard) writes: >>I've seen the same thing (with the double mono screen - a cheap two-page >>monitor!) This happened when I was using a multisync in mono mode... >Well, I've seen it in low-rez mode now, too. In mono, what happens is the >video shifter gets set to medium rez, but lineA, VDI, AES, etc., aren't >notified of any change... > -- Howard Chu @ University of Michigan Are you sure that it is in the ST? I'm assuming that the weirdness is being caused by the multi-sync circuitry of the monitor, since this seems never to have shown up on a non-multisync, and I find it very unlikely that the video circuitry in the ST could 'accidentally' send out each scan line _twice_ on the same line...seems that this would require a doubling of the horizontal scan rate since no information is being lost. -- ################################################################################ # Bill Sheppard -- bills@microware.com -- {uunet,sun}!mcrware!mwca!bill # # Microware Systems Corporation --- OS-9: Seven generations beyond __/_!! # #######Opinions expressed are my own, though you'd be wise to adopt them!#######