Q: display anomalies

So, I'm curious. People who have had new motherboards have had the same versions of MacOS installed, apparently without problems. If completely wiping the PRAM is the fix that has been achieved with a new motherboard, does that mean that nearly any version of OS10 will be stable after that??? Or is it only the latest 10.6.8 ATI driver that has been fixed???

My impression is that there is some dodgy code in newer versions of Safari that badly interacts with the ATI driver to express a small bug in the ATI driver. That Safari is the only application that crashes regularly in an otherwise stable 10.5.8 minus ATIRadeonX1000.kext is evidence of this.

>>>> Half-___ed measures won't cut it!

did you read my post? 

I'm not suggesting anyone follow what I said over what marco said.  I just can't find my media as of this moment to do the entire list of tasks.

loupalladino, because I find it a pain to keep doing achive/installs, I am going to try unplugging the PRAM battery then my usual software fix recipe to see what happens. I think the effect on cleaning out the system and not exposing the bad ATI drivers is similar. The key to that is that one has to delete the ATI drivers *and* rebuld the kextcache to prevent the bad drivers from reloading. You could try that too if you can't find disks....

It's possible that it's just a matter of time before the same problem rears its ugly head after a system board replacement (or a PRAM erasure, for that matter). There may be a particular set of conditions that occurs infrequently to create the ATIRadeon-related problems.

It's as if the ATIRadeon resources become corrupted or altered in a way that eventually corrupts the PRAM. Then, it becomes an ongoing problem. The version of the ATIRadeon resources may not be the deciding factor, but whether or not they've become corrupted in some way, or whether the "particular set of conditions" happens again.

Safari's relationship with the ATIRadeon resources could be a factor, as well.

We're working with a soup recipe with a lot of ingredients!

All I know for sure is that my MacBook Pro 1,1 seems like a new machine compared to how it's been behaving. If the problem reccurs, at least I'll know how to go about solving it.

I wasn't directing that phrase at you, although it did seem that some people in the thread were talking about implementing Marco's recommendations half-heartedly.

I'd like to add my thank you to the list. Marco, fantastic job! I don't have the X1600, so I won't need this particular fix, but I really admire what some people come up with in response to issues. It seems like with almost every major hardware revision, there is some common issue, and someone like Marco comes up with a fix that leaves me wondering how on earth they ever thought of it. This is another of those fixes. Well done Marco.

Now something for thinkers of his type to consider. Not long after the MBPs with the X1600 (maybe even the next revision?) PRAM batts were done away with, and replaced with capacitors. Relatively common issues similar to those in this thread continue to occur. Unfortunately for many users, however, they cannot follow Marco's lead in applying the idea of his fix to their machines because said capacitor is going to hold onto the very information that apparently needs to be zapped. I can't help but wondering if Marco's process would work with at least some of the many other display/GPU issues (and who knows what else), if this info became more widespread, and a process for discharging said capacitor were found (along with finding the proper capacitor itself).

For industrious individuals, perhaps someone with a newer MBP with this type of issue, I hope that you will consider sharing, as Marco has, your efforts, both those that work, and those that don't, toward applying a process similar to this, to other machines. That's what I enjoy most about these forums. Good luck to all.

If PRAM needs to be wiped in a machine with a capacitor, it is relatively simple to short out capacitor terminals to discharge as long as the capacitor and its terminals can be identified.

Hay marco...POLLO or anyone, I am looking for measurements of the PRAM battery (diameter and thickness), since if I have to pull my machine apart to unplug the battery, I figure I may as well replace the battery. PCB mount lithium button cells are widely available and should be easy to solder to the wires from the old battery.

The candidates are from Panasonic's extended reliability range:

http://www.panasonic.com/industrial/batteries-oem/oem/primary-coin-cylindrical/b r-cr.aspx

PS - just found a picture on iFixit that shows the battery markings - a CR2025.

Now I'm hoping I can change it to a BR series, which is more temperature resistant, but there isn't a BR2025, so to do that I have to go a little larger 3mm (BR2325) or a little thicker 0.7mm (BR2032). Any comments on what might be more likely to fit??

Interesting what Apple generally says about the function of its PRAM: http://support.apple.com/kb/ht1242

I am curious as to what ever happened to the theory on this discussion board that our early MacBook Pros were subject to a video firmware update that immediately corrupted the PRAM; did anyone ever manage to "undo" the firmware change and revert to ATI's original code? It seems going back to earlier OS X versions did nothing except lessen the heat of the machine when running. The SMC fix etc, solved heat but not stability of video.

I am relatively stable on 10.5.8, but do get the random kernel panic. It says PRAM retains info on those too. Wonder what connection might be. I just want to be sure before I take out all those little screws that MARCO's fix is the last one on this brick of a MBP!

I am glad there has been some success on my potential solution. I am not a hardcore Mac tech anymore. I did not know why my method worked and may work on others. It is just the steps I took that worked, so I HAD to post and get word out for others to try!

As for those that wonder about sending boards in for replacment ATI chips. I am not sure if I mentioned it here, but my theory is STILL that its both hardware and software related. I can not be sure. But my guess is that ATI had a bad batch of chips. Possiby hit by a high voltage static charge. Just like with RAM, the chips begins a very slow death that is not detectable right away. Hence the fact that machines act up longer after. The same way RAM can go bad. But I think ATI told Apple, or not, that there may be an issue on the ATI chips already soldering to many motherboards and that it may or may not come up as an issue later. So what might have been done was a software "solution" to skirt around any issue. Problem is that the patch only worked so well. Later drivers may work better to resolve any erouneous data the chips spew out. Again, just a guess. But given that the chips fail later and this solution seems to help, if not cure, then it can not be just hardware that is the issue. May it come up again in a few months time on the same machine? Maybe. But for now I am still playing Spore (sad game really) and using the MacBook to watch shows in iTunes, etc. All without issue. My GPU hits temps of about 145 and has yet to exceed 147 degrees.

I encourage those who can try the fix (I understand the capacitor issue) please do. Its worth a shot, but I really hope everyone tries the instructions to the letter. Maybe, if this works well, a software solution can be created to clear out the PRAM completely, remove the old drivers and any potential support files that may be an issue and then install new drivers. But software I do not write! Good luck folks!

I think the big difference here is that while NVIDIA publicly admitted they had a bad batch of chips (they were also supplying Dell and HP at the time), ATI did not.  I can only surmise that NVIDIA coming clean as it were backed Apple into a corner, and they had no choice but to recitfy the situation for NVIDIA users at Apple's expense.  As such, Apple has not publicly acknowledged this particular issue.

That being said, I took my MBP to the nearest Apple store, and long story short, they finally offered to repair it at no charge.  It took me 2 visits to get this accomplished.

My first visit, I was told I was out of luck, that they had never heard of this issue, and that my only option was to pay for a logic board replacement.  I showed them pages such as this one but received the equivalent of a blank stare.

When I received the email asking for feedback, I was firm, polite, yet completely scathing.  I laid out my [our] case - similar issue as the NVIDIA problem, etc, and the complete lack of support from Apple on this one.  This prompted a phone call from the local store manager, who then extended an offer to repair my laptop at no cost.  Apparently, someone at Apple reads those letters, and feedback works.

Just wanted to add my experience with the solution devised by Marco...POLLO! The system seems to be running well without any kext removed; I'm on 10.7.1. When using Safari in non-maximized view, a single-lined graphic anomaly appears. Using Chrome or Firefox in non-maximized view, no artifacts appear. When using Safari and Firefox in maximized view or full-screen mode; single-lined graphic anomaly reappears. Using Chrome in maximized view or full-screen mode shows no anomalies.

I was able to perform XBench successfully. However, when using TechTool Pro 6 to test the video RAM, the program just crashes; this could perhaps just be a compatibility issue with Lion?

I have not run any graphics heavy games...Angry Birds however runs smoothly. Aside from the issue mentioned above, I have not yet noticed any other graphic anomalies.

Thanks again Marco...POLLO!

How are you running 10.7.1?  I thought the requirements for Lion were a minimum of Core 2 Duo... The first gen MBP was only a a Core Duo.  Is that a recommendation and it will actually install??

I'm on a MacBook Pro 15" 2.16GHz Core 2 Duo (MacBook Pro2,2).

Gotcha.  Do you have the ATI x1600 also?