jam

Registered Users
  • Content count

    4,832
  • Joined

  • Last visited

  • Days Won

    3

jam last won the day on July 12

jam had the most liked content!

Community Reputation

91 Vet

About jam

  • Rank
    Member
  • Birthday
  1. Research is needed to get at the facts of the matter, I reckon. One of the many things we can say about Internet posting is that one encounters a lot of people who think they know something - and while it is true that everyone knows something that someone else doesn't know, there are many who think they know something in particular and don't know that they don't really know it - that something that they think they know, that is. ,s Joe Rumsfeld
  2. Still no joy, but . . . I found a potential explanation here: http://forums.wz2100.net/viewtopic.php?f=4&t=3000&sid=bf57d3dcc60fffed77aac56f671915b5&start=15 And it goes like this: While this post is from a Warzone 2100 forum, it seems to explain something that points back to the suggestion Oldzeke made earlier on - and may cause me to eventually bite the bullet and upgrade my system to Windows 7 64-bit, if the underlying argument is true. Player (WZ) Kaldera asserts that a 32-bit OS can only address something less than 4 GB of memory - including the memory of the peripheral devices. If he is correct, then, in my case, I have 4 GB of cache RAM in my system, plus another 2 GB. on the Quadro 3800M board - 6 GB. in all. So the memory address space violation issue may be due to the newer nvidia driver treating my system like a 64-bit OS system attempting to address 64-bit OS memory locations (on the video card) that do not exist. I can: Try the low-mem option that OZ suggested. Continue to find an nvidia driver that works. As an experiment, pull out 2 GB of memory from my system, leaving a total of 4 GB counting what is in the vidcard. This may still be an issue because the total is over the 3.5 GB. Long-term, the best thing to do to access the full power of my system is to go to Windows 7 64-bit. It also should go without saying that I ought to verify the assertions of Kaldera even if his suggestions do make sense, logically. Cheers, Joe
  3. I am going to keep an eye on this thread. I started a thread here where I CTD, usually after playing from minutes to an hour or so since I upgraded to 260.99 on a laptop using a Quadro FX3800M. For a few months, on the laptop, I had no such problem. The driver that worked well was the one that came with the Dell M6500 from early 2010. I uninstalled 260.99 and replaced it with 197.16 with no change. In my case, the error comes when a hit a particular terrain view. It is repeatable. I.E. if I have a CTD while spawning in to a particular FB and I re-spawn in the same place, I get another CTD immediately. Go somewhere else and I am fine. Not meaning to high-jack the thread, just adding another OpenGL driver related issue. My next step wil be to find a good nVidia driver cleaner and try an older driver a second time. Cheers, Joe
  4. Small update: I noticed in the Nvidia control panel that: OpenGL Rendering GPU was set to "auto-select," which, I suppose, ought to be fine. But the laptop only has the one GPU and so I set it for the "Quadro 3800M," the only other choice in the drop-down box. Afterwards, I had five missions without a CTD, 2 were under 1 minute and the other three were 10 minutes, 5 minutes and 18 minutes for a total time w/o CTD of about 33 or 34 minutes. I have "threaded optimizations" off and the above setting for the chosen GPU for OpenGL. The game seems fine to my eye - with good frame rates and fluidity. I may not have fixed anything, but so far, there seems to be some improvement. Cheers, Joe
  5. The rollback option on the video drivers AFAIK only provides one step back to the previous driver, which was also problematic. I frequently create restore points, but for some reason I did not do it with this recent round of video driver updates. Anyway . . . I had one eventful mission tonight that lasted 9 minutes and then, I respawned in a new location, found my Flak-30 on its side and as I was getting ready to despawn, I CTD'd Prior to playing, I set "threaded optimizations" to off for wwiiol.exe, but even so, I crashed. As we can clearly see, the CTD cause is the same. I am not about to give up on some experimentation. Cheers, Joe
  6. Intersting thread here: http://www.opengl.org/discussion_boards/ubbthreads.php?ubb=showflat&Number=232426 I may try turning off "threaded optimizations" and see what happens. Apart from a mild gamma adjustment, I have not tweaked any of the three nvidia driver settings to be game-specific with any of the three drivers that have been used on this laptop. ===================================== There is a lesson to be learned here that I am sure I will fail to learn - and that many before me have failed to learn: "If it aint broke, don't fix it." If you have good or acceptable frame rates If you have completely fluid game play with no stutters If you can max out most of the settings If you are, overall, quite satisfied with the performance of the game and of the other applications you run, then why the hell would you update your video driver? I can think of a few off-color analogies here, but I'll leave them for OT. Cheers, Joe
  7. Again, The laptop was working flawlessly until I updated the driver, so I think the drop-out or CTD is driver related. I see as confirmation, the wording of the WWIIOL/BE error log: There are a handful of clues above that I have not yet investigated. If one Googles "nvoglnt", there is lots of info . . . I'll work on this some and eventually report my results. It appears the CTD is being caused by an access violation by one of the driver's OpenGL support elements, specifically, nvoglnt.dll. Other games have had this issue in the past couple of years. There are a number of discussion threads for NWN or Never Winter Nights, KOTOR and Q3, to name a couple. So this is not a "never before been seen" issue. Here is another clue from an Nvidia developer's forum: Someone else had a similar issue with his app, but there was no nVidia response in that thread. =============================== I am definitely not interested in pursuing using anything other than an NVidia Quadro driver. I am still betting that, I may find a tweak for the present driver, or else I may roll back my driver to the last one that worked fine, or else, I may just wait it out for a new driver. As far as the hardware monitoring goes . . . The software is reading the sensors on the dies, so accuracy of readings is not an issue. There seems to be little difference (as in no discernible difference) between SpeedFan and HWMonitor. I am convinced that this is not a temperature related issue because the CTD for the game does not correlate to observed temperature. Before using the current driver, I had the game up for hours on this laptop and although it really did get warm to the touch, it never quit. The fans would periodically speed up and then slow down again, indicating that no uncontrollable terminal temperature or runaway had occurred - no artifacts; no BSOD; no CTD - all OK. I am still very open to suggestions though. Thanks to those who have made them, Joe
  8. I'll have a look at HWmonitor; thanks for the suggestion. Even so, there is no indication that the problem is heat-related. Yes, I do know its actually a quad-core, though this is not germain to the issue. No, there is nothing wrong with using a Quadro vidcard for gaming. This is my third Quadro and I have never had any problems with the other two - or with this one, until I updated the driver recently. Yes, if you read the first posts, you'd know I used the correct driver. Although I haven't tried installing the *wrong* driver on purpose, I doubt that Nvidia would fail to warn the installer - or let one install the wrong driver without some compatibility checking. In any case, I am not going to attempt to install a GF driver. Does anyone think it could be a 32-bit OS versus 64-bit OS issue? I don't think so. In the end, this will likely turn out to be either a tweak needed or a replacement driver. It's fair to say that Quadro drivers are not written with gaming in mind and are probably not regression tested for the gaming environment. Regards, Joe
  9. Nah, I usually play the game in my underwear, Bill, so I stay cool. ,s J/K Actually, I installed a neat little applet called SpeedFan which monitors and indicates the temperatures of the GPU and each of the 8 cores and there is no excessive heat. No artifacts, no problems with any other apps. Just for the heck of it a bought a neat Rosewill black anodized aluminum laptop cooler with a pair of quiet 60mm fans in it. When the GPU is at 54 C at idle, and I turn on the cooler, within 2 minutes the temperature goes up to 59C, lol. Turn off the cooler and it goes back down to 54 C. I haven't quite figured out yet why the cooler increases the temperature. It draws air from underneath and pushes it up, but I suspect that it may fight the air flow in the M6500. A little further investigation is in order. The GPU and the cores go up to nearly 70 C while playing the game, but the CTD does not seem to correspond to any spike in temperature. The internal fan speeds are CPU controlled and when I was having no trouble, they would only increase in speed when the game was running for half an hour or more and after awhile they would slow again. So the laptop does not *think* it is overheating. BTW, the M6500 costs a lot of scratch, but I am enjoying it. Had I to do it all over, I would have specified Win 7 64, but I usually wait for an OS to be out for awhile before I embrace it. Win 7 seems a real winner, but it is a lot of trouble - PIA - to go from XP to 7 when one has many apps installed. Regards, Joe
  10. Ran all of the DirectX tests and there were no issues detected. Cheers, Joe
  11. After this, my laptop and I feel pretty exposed. Time to put some clothes back on, lol: I will run the tests, but frankly, I'd be surprised if it fails any of them. Cheers, Joe
  12. I am fairly sure this is more than anyone wants to see, lol:
  13. I am overrunning the 10,000 character limit with each copy & paste
  14. This uses a lot of 1s and 0s, lol . . .