• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

PowerWave weirdness | picky about video card

jessenator

Well-known member
So when I first acquired my PowerWave, it had some posting issues which were due to RAM, which were resolved quickly and the machine ran well. However, after decided to upgrade from the "fine" Mach64 with a Rage128 I had sitting around, things got weird: get a chime and then the screen gets a signal briefly, then nothing. not even a black screen, my monitor just says there's no signal.

The machine wouldn't post at all, even after a board reset before powering back on. Tried pulling RAM, reseating the ROM, reseating and changing out CPU daughtercards. Nothing. I put the Mach64 and it immediately posts and boots right up. I upgrade/reinstall the ATi drivers, shutdown, replace with the Rage128, reset the board, power on, and after a few seconds it finally posts. Okay, fine, no harm done I guess.

Time comes where I have to reset the PRAM. Upon doing so, I'm back to a black screen, no posting at all, nothing beyond the bong. 

Would the ROM on the PowerWave just see the Rage128 and not know what to make of it because it's too new? This all seems very odd. Just wondering if anyone has any advice or has had a similar experience.

Edit: as an addendum, if it really is something about a contemporary (to a PowerMac 85/9500) that it needs as far as a graphics card, what would be a decent replacement for the Mach64? VRAM modules for them are nearly double what you can get a standalone card for.

 
Last edited by a moderator:

MOS8_030

Well-known member
Interesting. I used in a PCI Rage128. probably from a B&W G3, in my 8500 with no issues.

 

jessenator

Well-known member
Now that I'm thinking more about it, again, not familiar with the particulars of the classic OS loading order, I wonder if the OS would really have much bearing on the card's compatibility. I'm thinking no, but someone correct me.

So when I boot I get: 1) chime, 2) the monitor going from standby to actual black (getting signal?), 3) then head seek on the HDD, 4) then back to standby, and it hangs. Is it loading parts of the OS at that time, or is it just reading the ROM and prepping everything to load the OS?

I'm loading my questions with a lot of supposition… barring any software snafus, I wonder what other hardware considerations there are to think about.

 

ppcoutlaw

Active member
Have you tried placing the video card in a different slot? There may be an issue with the newer card running in that slot. I ask as I am reminded of posts on another site regarding the design of the Umax SuperMac 900 models regarding card placement being an issue. Some folk had issues with card hierarchy, kind of like the scsi addressing system. It had something to do with the bus controller I believe.

"When PowerPC was outlawed, only outlaws had PowerPC. I am a PowerPC outlaw."

 

jessenator

Well-known member
Have you tried placing the video card in a different slot?
I did try that, and thank you for your thoughts.

Turns out now that I can't zap the PRAM with alt+cmd+p+r ...that is very odd. When I try it just makes the machine hang on a black screen, no chime, nothing. Maybe that's a bigger issue here :lol:  

So so far the only method that works is pull the PRAM battery, put the Mach64 in, boot it, get everything set up, shut down, replace the Mach64 with the Rage128 (no board reset, no PRAM reset). Then it works. But there's something fishy going on here...

I'm really starting to wonder if there's something wrong with the foundational aspects, maybe the ROM? Something else?

 
Last edited by a moderator:

jessenator

Well-known member
I'm really starting to wonder if there's something wrong with the foundational aspects, maybe the ROM? Something else?
Edit: this first test was with the 225/45 PowerComputing CPU card

well, I ran TechTool 3 and got some interesting results... it failed on some aspects of the disc, but I can't get the TT CD to boot, so I can't do full repairs on the HDD. No surprises that there would be some after all these years. What was strange was it failed the CPU math test. Don't know what to make of that, to be honest.

r1AEaui.jpg.33a9bdc8d245a18472ca789a9cea3208.jpg


The repair advice mentioned a few things, but basically said a qualified technician would be able to give a proper diagnosis. One thing that was mentioned was bad solder joints... on the CPU daughtercard connector?

 
Last edited by a moderator:

jessenator

Well-known member
I thought I was making progress with my other CPU card (132), and I got the PRAM to zap just fine and booted up. Then I put the Rage128 in and it won't post again. Square one...

 

ppcoutlaw

Active member
Do you have another PCI mac to test the

rage 128 on? Aside from the test showing a possible issue on the board, from what you have written, the machine works with the original card, but goes toes up with the rage.

It seems likely that it may be that the card

is bad, or it is not a mac card. It is possible that the clone cant see it as it is not there

for it to see.

 

jessenator

Well-known member
Do you have another PCI mac to test the rage 128 on?
Runs just dandy on my 4400 and my beige PMG3. It's a Mac-specific card as well—came from a B&W G3. And, I mean, up until there's a board reset or the PRAM battery is taken out, it runs just fine. 

I'm kind of tempted to try it again now that I have my PowerLogix G3 daughtercard working. Part of me thinks something funky was going on with the two 604 cards, as they were not the originals that came with the machine (when I got it). It came with the G3, but was running OS 8.6. Right now it's running very stably, and with 0 issues. When I was doing the diagnoses with the other two cards I would get random freezes every now and again. I think I'll try that next just for kicks.

Edit; nope. same story... this is almost to the comical stage for me.

 
Last edited by a moderator:

CC_333

Well-known member
Perhaps there's some kind of weirdly-specific fault in the logic board?

c

 
Last edited by a moderator:

EvilCapitalist

Well-known member
Turns out now that I can't zap the PRAM with alt+cmd+p+r ...that is very odd. When I try it just makes the machine hang on a black screen, no chime, nothing. Maybe that's a bigger issue here :lol:  

So so far the only method that works is pull the PRAM battery, put the Mach64 in, boot it, get everything set up, shut down, replace the Mach64 with the Rage128 (no board reset, no PRAM reset). Then it works. But there's something fishy going on here...

I'm really starting to wonder if there's something wrong with the foundational aspects, maybe the ROM? Something else?
Seemingly silly question, but are you using a stock Apple keyboard (or a PowerWave branded one) or just some generic ADB keyboard?  I've only got my experiences on it but for whatever reason every single non-Apple keyboard I've used has been crap in terms of resetting PRAM or other startup key combinations.  I'd be holding the keys down and nothing would happen, then I'd switch to an Apple keyboard (using the same ADB cable) and *boom*, all of a sudden the computer is recognizing all the startup key combos.

 

jessenator

Well-known member
Seemingly silly question, but are you using a stock Apple keyboard (or a PowerWave branded one) or just some generic ADB keyboard?  
Y'know, that's something that crossed my brain and immediately dismissed out of stubbornness/pride :lol: I had talked up the keyboard in my head...

I have been using the PCC adb keyboard ever since I got it, and noticed some anomalies, mostly while playing DOOM— not being able to maneuver like I'm accustomed to. I'll test it out with my AEKII. Thanks for the thought.

 

jeremywork

Well-known member
I would verify whether this behavior exists when running in your 4400, but here are a couple of my observations in a 9600:

I've noticed many of my ATI cards from that era (Rage 128, Radeon PCI, 7000, 8500) will send a short (1-3 second) dark signal to the display before the system is ready to output video. This usually results in:

Power on keyboard (num/caps/scroll light momentarily, as the fans/drives spin up)

Startup chime just following startup

RAM test (longer for more memory)

Monitor awake - black screen

1-3 seconds later monitor returns to sleep

2-5 seconds later monitor awakens again, this time displaying the Mac's checkerboard pattern nearly immediately.

I've found in cases where something inhibits startup, the ATI card will still wake the monitor the first time, but will remain inactive after the initial 1-3 seconds.

For instance, if I install a 3dfx card alongside the ATI card, the 3dfx card will only allow the system to boot if it detects a monitor itself.

With two screens: The ATI card does its initial flash, the 3dfx screen turns on, then the ATI screen turns on.

With only ATI screen (3dfx card still installed): The monitor turns on, displays black for a few seconds, then goes back to sleep. No hard drive activity in the background.

With only 3dfx screen (ATI card still installed): The monitor turns on normally and the system boots. The ATI control panel does not detect any hardware.

If your Rage 128 also does the double-wake in your 4400, then I would ignore the symptom of the monitor turning on and just treat it as a normal no-video symptom.

Beyond that I don't think I have any more specific advice, but I would see if those math tests still fail immediately following a PRAM reset. If the presence of the Rage 128 causes the PRAM to become corrupt that would be interesting to know/examine.

 

Bolle

Well-known member
That behavior you’re experiencing is a common issue with more modern ATI cards in old world PowerMacs as far as I remember.

I will see if I can dig up the information from back then somewhere.

Solution was basically something along the lines of don’t reset PRAM once you got it working or things will get messed up for no reason.

So just like you already figured out.

 
Last edited by a moderator:

ppcoutlaw

Active member
I forgot to add some more info on the s900.

it could pertain to the power clone machines as well.

There was a card population issue. This cars in slot one, that one in slot two, etc. Sometimes the cards would be recognized and run fine, other times the operator had to move the cards around until a workable order was discovered. I.e. video card in the first slot, SCSI controller in the third slot. Might that be something to examine?

 

jessenator

Well-known member
Solution was basically something along the lines of don’t reset PRAM once you got it working or things will get messed up for no reason. 
That would make me feel less inane :lol:

see if those math tests still fail immediately following a PRAM reset.
That only happened with the 225 MHz PCC daughtercard, actually. Not with the 132 or with the PoweForce G3 (currently installed), and hasn't repeated itself. It was incredibly weird though. 

As far as the other machines I've used it in, I have 0 issues with the card, pre-, post-, or with a PRAM reset. The only OWR machine that's given me issues is the PowerWave :/  

I did experience that sort of monitor wake thing, but I leave it for minutes on end (in the PowerWave) and I get nothing. Part of me wants to get another NWR graphics card and see if it's repeated.

 
Last edited by a moderator:
Top