Getting Started, Documentation, Tips & Tricks

Graphics problem with Indigo2?

So I finally have my Indigo2 working fully with a proper full installation of IRIX on it, and normally I couldn't be happier; however, for some reason the display it gives me almost always has these distracting lines through it. You can see them quite clearly in the console screenshot attached to this post (which tells me it has something to do with software; if it was a mixed-up signal in the cable, I couldn't capture it in such a manner). It sometimes doesn't happen (seems to be random upon start-up, most commonly occurring). When it does happen, the only time it doesn't display exactly like this is at the login screen, where the display is still screwy, but in a different way - there are patterns of boxes and whatnot, I really can't describe them very well here.

Anyway, I was hoping people here might have some thoughts on how to get rid of these things.

_________________
Systems: :O200: :O200: + Image + :Indigo: + :Indigo2IMP: + :O2: +
- AdamMalkovich: P4 PC, LinuxMint Fluxbox CE 6
- Morningstar: Sun SS20, Solaris 9
- Guru: IBM RT/135, AIX 2.2.1
- Concord: Sun SPARCclassic, Solaris 2.6
- Lying Bastard: Apple iMac G3, Mac OS X 10.3.9
- Chunderspunk: NeXTstation, NeXTSTEP 3.3
I think a saw a post where this was related to bad trams or something. Might be worth searching.

_________________
Stuff.
yes, it appears to be a form of frame buffer memory fault. does the pattern of missing bits change when you select a different set of screen dimensions, such as 1024x768? different refresh rate? frame buffer depth?

_________________
I love my iPad!!!
The pattern doesn't change with a different screen resolution or refresh rate; I'm not sure how to change the frame buffer depth though. The one time it seemed to consistently go away was immediately upon installing IRIX (this happened a few times due to the nature of an error previously). However, after a few sessions, it returned. I shall search the forums for bad trams though.

_________________
Systems: :O200: :O200: + Image + :Indigo: + :Indigo2IMP: + :O2: +
- AdamMalkovich: P4 PC, LinuxMint Fluxbox CE 6
- Morningstar: Sun SS20, Solaris 9
- Guru: IBM RT/135, AIX 2.2.1
- Concord: Sun SPARCclassic, Solaris 2.6
- Lying Bastard: Apple iMac G3, Mac OS X 10.3.9
- Chunderspunk: NeXTstation, NeXTSTEP 3.3
If that's an Indigo2 with a graphics option made up from multiple PCBs (so HighImpact, MaxImpact), you could try to pull the PCBs apart and reseat them. It looks like either bad frame buffer memory, or a bad connection between frame buffer and display generator (DAC).

TRAM is used only for textured graphics and a terminal window isn't, so I wouldn't suspect TRAM.

Oh, and run the diagnostics.

_________________
Now this is a deep dark secret, so everybody keep it quiet :)
It turns out that when reset, the WD33C93 defaults to a SCSI ID of 0, and it was simpler to leave it that way... -- Dave Olson, in comp.sys.sgi

Currently in commercial service: Image :Octane2: :Onyx2: (2x) :0300:
In the museum: almost every MIPS/IRIX system.
It looks like the same problem this guy has. viewtopic.php?f=3&t=16718092&start=29

_________________
:Onyx: (Aldebaran) :Octane: (Chaos) :O2: (Machop)
:hp xw9300: (Aggrocrag) :hp dv8000: (Attack)
Welp, I took apart the graphics cards and got at the connections with my can of contact cleaner (which, incidentally, is how I got the machine to begin working properly originally); however, while I had them apart, I noticed something disconcerting - the chips on the underside of the boards, the ones with the little heat-sink-looking things attached to them, are not attached to their pads, or at least many of them are not. They are hanging on barely by perhaps another set of tenuously-connected pins towards the middle.

I shall attempt to solder these down later (the cleaning did not fix the problem); I personally do not have a good soldering station (or very steady hands for that matter ^^; ), but a housemate who is not presently in does. Report to follow.

_________________
Systems: :O200: :O200: + Image + :Indigo: + :Indigo2IMP: + :O2: +
- AdamMalkovich: P4 PC, LinuxMint Fluxbox CE 6
- Morningstar: Sun SS20, Solaris 9
- Guru: IBM RT/135, AIX 2.2.1
- Concord: Sun SPARCclassic, Solaris 2.6
- Lying Bastard: Apple iMac G3, Mac OS X 10.3.9
- Chunderspunk: NeXTstation, NeXTSTEP 3.3
Interesting - please post some photos if you can - this problem usually surfaces with TRAM chips but it sounds like you are referring to something else?
Your problem is different than mine was but I recently fixed a problem with my MaxImpact that I originally suspected to be TRAMs. It turned out to be main memory instead. My symptom was corrupted textures in a few applications (SkyRocket and distort). Reseating all of the DIMMs fixed the problem (I actually re-arranged them, too). Results from internal diagnostics is what led me to suspect the DIMMs.

Anyway, as I said your problem is different but if your diagnostics report main board errors then it could be something apparently unrelated like main memory.

_________________
:Indigo2IMP: :1600SW: R10K Indigo2 MaxIMPACT, 4 TRAMS, 768MB RAM, 2x9GB HD, CD-ROM, Phobos G160
Black Cardinal
Baphijmm wrote:
Welp, I took apart the graphics cards and got at the connections with my can of contact cleaner (which, incidentally, is how I got the machine to begin working properly originally); however, while I had them apart, I noticed something disconcerting - the chips on the underside of the boards, the ones with the little heat-sink-looking things attached to them, are not attached to their pads, or at least many of them are not. They are hanging on barely by perhaps another set of tenuously-connected pins towards the middle.

I shall attempt to solder these down later (the cleaning did not fix the problem); I personally do not have a good soldering station (or very steady hands for that matter ^^; ), but a housemate who is not presently in does. Report to follow.




Yikes! I've rarely seen components pull loose from PCBs. A bit of a concern, makes you wonder how well the machine was maintained.


Regards


Ryan


_________________
MAYA, nut-
:Octane2: :Octane2: Octane 2 R14k 600 V12 4GB, Octane2 R14K 600 V10 1GB ,
:Onyx2: :Onyx2: Onyx2 IR3 4GB Quad R14K 500 DIVO, Onyx2 IR Quad R12K 400 2GB,
:Indigo2: SGI Indigo 2 R8K75 TEAL Extreme 256MB,
:Indigo2IMP: SGI Indigo 2 R10K 195 Solid Impact 256MB, MAX Impact Pending
,
Apple G5 Quad, NV Quadro 4500 + 7800GT, 12GB RAM
Sun Blade 1000 Dual 900 XVR 1000
Ryan: It was obtained through a local federal surplus property auction, so, not at all.

The boards still don't seem to be working properly; unfortunately, one of the ICs had a loose contact immediately against another and we could not get the soldering iron in to fix it. I don't think this can be repaired past this point.

For those of you wondering which chips I mean, I'm talking about the ones on the underside of the top two boards; there are six each, and they each have a small what appears to be a heatsink glued to the top of them.

_________________
Systems: :O200: :O200: + Image + :Indigo: + :Indigo2IMP: + :O2: +
- AdamMalkovich: P4 PC, LinuxMint Fluxbox CE 6
- Morningstar: Sun SS20, Solaris 9
- Guru: IBM RT/135, AIX 2.2.1
- Concord: Sun SPARCclassic, Solaris 2.6
- Lying Bastard: Apple iMac G3, Mac OS X 10.3.9
- Chunderspunk: NeXTstation, NeXTSTEP 3.3
maybe you should try the heat gun and tinfoil approach? That's worked great for me on several of those ill-designed radeon thinkpads.

_________________
:Onyx: (Aldebaran) :Octane: (Chaos) :O2: (Machop)
:hp xw9300: (Aggrocrag) :hp dv8000: (Attack)
Can you post a picture of the components that have lifted? would be worth seeing what we're up against here.

_________________
:Fuel: :Indigo: :Indigo: :Indigo: :Octane2: :O2: :O2: :Indigo2: :Indigo2IMP: :Indy: :Indy: