Why might a Macbook Pro be freezing on resume from sleep?

My Macbook Pro (17" 2.2 Ghz Intel Core 2 Duo, OS X 10.4.11) often locks up when it resumes from sleep. I usually know I'm in trouble when the display brightness doesn't adjust right away. Then it runs for a few seconds until it stops responding again. The pointer works, but that is about it.

I've spent the last few months stripping down all the background programs, to my inconvenience. I'm still tracking down a few that either didn't uninstall fully or don't have an obvious uninstaller, but I'm nearly down to a stock configuration and I'm running out of ideas. Has anyone else found a cause this behavior?

For a while I was getting kernel halts with an error on screen and a crash reporter on restart. Lately I'm just getting an unresponsive system. Here is the last entry from panic.log, back when it was begin detected:

Mon Aug  3 07:50:07 2009
panic(cpu 0 caller 0x001A49CB): Unresolved kernel trap (CPU 0, Type 14=page fault), registers:
CR0: 0x8001003b, CR2: 0x00000000, CR3: 0x018ba000, CR4: 0x000006e0
EAX: 0x00000428, EBX: 0x00000000, ECX: 0x050ef820, EDX: 0x00000002
CR2: 0x00000000, EBP: 0x25e7b918, ESI: 0x000000d8, EDI: 0x050ef804
EFL: 0x00010206, EIP: 0x3ce5ceff, CS:  0x00000008, DS:  0x050e0010

Backtrace, Format - Frame : Return Address (4 potential args on stack) 
0x25e7b6b8 : 0x128d0d (0x3cc65c 0x25e7b6dc 0x131f95 0x0) 
0x25e7b6f8 : 0x1a49cb (0x3d2a94 0x0 0xe 0x3d22b8) 
0x25e7b808 : 0x19b3a4 (0x25e7b820 0x1 0x25e7b838 0x3cc2c67f) 
0x25e7b918 : 0x3ce57723 (0x50ef804 0x4751804 0x3 0x3) 
0x25e7ba48 : 0x3cd2b60c (0x50ef804 0x4751804 0x25e7bb30 0x25e7bb30) 
0x25e7bb58 : 0x3cd2b33a (0x50ef804 0x45ae004 0x1 0x1c) 
0x25e7bbb8 : 0x3ce5210a (0x50ef804 0x45ae004 0x1c 0x3ce97160) 
0x25e7bc38 : 0x3ce524f2 (0x50ef804 0x4751804 0x100000 0x25e7bd4c) 
0x25e7bd98 : 0x3ce784cd (0x50ef804 0x4751804 0x0 0x19a4f1) 
0x25e7be78 : 0x3cd222ac (0x50ef804 0x50b4004 0x0 0x50ef804) 
0x25e7bec8 : 0x3cbf240b (0x50ef804 0x50b4004 0x0 0x4e60270) 
0x25e7bf08 : 0x39b96f (0x24352004 0x4da4c80 0x1 0x2) 
0x25e7bf58 : 0x39ab41 (0x4da4c80 0x0 0x0 0x19e23a) 
0x25e7bf88 : 0x39a877 (0x4b7dcc0 0x25e7bfd4 0x134db9 0x13d8a2) 
0x25e7bfc8 : 0x19b21c (0x4b7dcc0 0x0 0x19e0b5 0x442cb04) Backtrace terminated-invalid frame pointer 0x0
      Kernel loadable modules in backtrace (with dependencies):
         com.apple.nvidia.nv50hal(5.0.8)@0x3ce07000
            dependency: com.apple.iokit.IOPCIFamily(2.2)@0x35ea5000
            dependency: com.apple.NVDAResman(5.0.8)@0x3cbde000
         com.apple.NVDAResman(5.0.8)@0x3cbde000
            dependency: com.apple.iokit.IOPCIFamily(2.2)@0x35ea5000
            dependency: com.apple.iokit.IOGraphicsFamily(1.4.8)@0x3cadb000
            dependency: com.apple.iokit.IONDRVSupport(1.4.8)@0x3caf6000

Kernel version:
Darwin Kernel Version 8.11.1: Wed Oct 10 18:23:28 PDT 2007; root:xnu-792.25.20~1/RELEASE_I386

System.log from a later instance, after having system board replaced:

Sep 11 18:07:44 rocky kernel[0]: System SafeSleep
Sep 11 18:07:44 rocky kernel[0]: hibernate_page_list_setall start
Sep 11 18:07:44 rocky kernel[0]: hibernate_page_list_setall time: 245 ms
Sep 11 18:07:44 rocky kernel[0]: pages 334939, wire 60443, act 67784, inact 24030, zf 7800, could discard act 66117 inact 108765
Sep 11 18:07:44 rocky kernel[0]: hibernate_page_list_setall found pageCount 334939
Sep 11 18:07:44 rocky kernel[0]: IOHibernatePollerOpen, ml_get_interrupts_enabled 0
Sep 11 18:07:44 rocky kernel[0]: IOHibernatePollerOpen(0)
Sep 11 18:07:44 rocky kernel[0]: writing 333180 pages
Sep 11 18:07:44 rocky kernel[0]: image1Size 110096896
Sep 11 18:07:44 rocky kernel[0]: all time: 7872 ms, comp time: 1617 ms, deco time: 0 ms, 
Sep 11 18:07:44 rocky kernel[0]: image 317386752, uncompressed 648388608 (158298), compressed 310973092 (47%), sum1 1396d187, sum2 75ecef4d
Sep 11 18:07:44 rocky kernel[0]: hibernate_write_image done(0)
Sep 11 18:07:44 rocky kernel[0]: sleep
Sep 11 18:07:44 rocky kernel[0]: Enabling XMM register save/restore and SSE/SSE2 opcodes
Sep 11 18:07:44 rocky kernel[0]: Started CPU 01
Sep 11 18:07:44 rocky kernel[0]: IOBluetoothHCIController::restartShutdownWL this is a wake from sleep
Sep 11 18:07:44 rocky kernel[0]: System Wake
Sep 11 18:07:44 rocky kernel[0]: IOUSBWorkLoop::closeGate - interrupt Thread being held off
Sep 11 18:07:46 rocky configd[38]: setting hostname to "rocky.local"
Sep 11 18:07:46 rocky kernel[0]: AppleYukon2 - en0 link active, 100-Mbit, full duplex, symmetric flow control enabled port 0
Sep 11 18:07:46 rocky lookupd[4558]: lookupd (version 369.8) starting - Fri Sep 11 18:07:46 2009
Sep 11 18:07:50 rocky configd[38]: posting notification com.apple.system.config.network_change
Sep 11 18:07:50 rocky configd[38]: setting hostname to "rocky"
Sep 11 18:07:54 rocky usbmuxd[3745]: stopping.
Sep 11 18:07:54 rocky lookupd[4560]: lookupd (version 369.8) starting - Fri Sep 11 18:07:54 2009
Sep 11 18:07:54 rocky usbmuxd[4559]: usbmuxd-167.1 built for iTunesEightTwo on Jul  9 2009 at 14:02:00, running 32 bit
Sep 11 18:08:00 rocky kernel[0]: NVChannel(GL): Graphics channel timeout!

From your kernel panic log:

Kernel loadable modules in backtrace (with dependencies):
     com.apple.nvidia.nv50hal(5.0.8)@0x3ce07000
        dependency: com.apple.iokit.IOPCIFamily(2.2)@0x35ea5000
        dependency: com.apple.NVDAResman(5.0.8)@0x3cbde000
     com.apple.NVDAResman(5.0.8)@0x3cbde000
        dependency: com.apple.iokit.IOPCIFamily(2.2)@0x35ea5000
        dependency: com.apple.iokit.IOGraphicsFamily(1.4.8)@0x3cadb000
        dependency: com.apple.iokit.IONDRVSupport(1.4.8)@0x3caf6000

That says that either Apple's Nvidia drivers or the Nvidia card itself is causing the computer to kernel panic so it may be a bad graphics card or simply bad drivers.

Now that Snow Leopard is out and you're planning to upgrade you can try an upgrade (which has newer drivers) and see if that fixes your issue. If the issue remains then it sounds like something is faulty with the graphics card and you'll want to contact Apple Tech Support for a repair under warranty (if you're still covered - 1 year with the computer, 3 years if you bought AppleCare)

Is it a GeForce 8600M? The warranty has been boosted to 3 years because of the card having a design fault - see Apple's technote.