Memtest86 on Laptop Question (OT)
Hi,
I've got a Dell C400 laptop with 768MB of RAM, that obviously I can't overclock. The reason I'm posting this question here is that I think this ng is about as knowledgeable about Memtest86 as any other ng. I've had some blue screen of death (BSOD) problems with the notebook when trying to use wireless networking cards in the PC card slot; this is a not uncommon problem from what I can tell, reading various Dell forums and other ng postings. I believe it is software related. Anyway, I spent yet another 45 mins on the phone with Dell laptop tech support yesterday, speaking to a fellow who was obviously pretty knowledgeable about computers (much moreso than your usual telephone support person; he sounded like a computer nerd). Anyway, he told me that the only other thing he could think of to do that I hadn't done was to remove the one removeable Sodimm from the notebook and see if the problems are eliminated. That makes sense, however I've run 3 or 4 passes of Memtest86 on the laptop and not had any errors; my sense is that (1) it is very unlikely that Memtest86 would miss a memory error; (2) if there are memory errors, there are 3 possible sources, e.g. 2 sticks of ram only one of which is user removable plus the cpu which is also not user accessible, and (3) given #1 and #2, we are talking about such a very small possible yield that it isn't worth doing, e.g. removing the memory module as a test procedure. Alternatively, I guess I could run another memory testing utility but I question if there is one out there that is any better than Memtest86. If anyone has a reaction or suggestion, please let me have it! Best, ken |
Ken Fox wrote:
Hi, I've got a Dell C400 laptop with 768MB of RAM, that obviously I can't overclock. The reason I'm posting this question here is that I think this ng is about as knowledgeable about Memtest86 as any other ng. I've had some blue screen of death (BSOD) problems with the notebook when trying to use wireless networking cards in the PC card slot; this is a not uncommon problem from what I can tell, reading various Dell forums and other ng postings. I believe it is software related. Anyway, I spent yet another 45 mins on the phone with Dell laptop tech support yesterday, speaking to a fellow who was obviously pretty knowledgeable about computers (much moreso than your usual telephone support person; he sounded like a computer nerd). Anyway, he told me that the only other thing he could think of to do that I hadn't done was to remove the one removeable Sodimm from the notebook and see if the problems are eliminated. That makes sense, however I've run 3 or 4 passes of Memtest86 on the laptop and not had any errors; my sense is that (1) it is very unlikely that Memtest86 would miss a memory error; (2) if there are memory errors, there are 3 possible sources, e.g. 2 sticks of ram only one of which is user removable plus the cpu which is also not user accessible, and (3) given #1 and #2, we are talking about such a very small possible yield that it isn't worth doing, e.g. removing the memory module as a test procedure. Alternatively, I guess I could run another memory testing utility but I question if there is one out there that is any better than Memtest86. After booting into Memtest, press "c" "2" "3" and "Enter" to do all tests. Test #11 is very thorough. Some will say GoldMem 5.07 is better. http://www.goldmemory.cz/ |
"S.Heenan" wrote in message news:LgbSb.319487$JQ1.7855@pd7tw1no... Ken Fox wrote: Hi, I've got a Dell C400 laptop with 768MB of RAM, that obviously I can't overclock. The reason I'm posting this question here is that I think this ng is about as knowledgeable about Memtest86 as any other ng. I've had some blue screen of death (BSOD) problems with the notebook when trying to use wireless networking cards in the PC card slot; this is a not uncommon problem from what I can tell, reading various Dell forums and other ng postings. I believe it is software related. Anyway, I spent yet another 45 mins on the phone with Dell laptop tech support yesterday, speaking to a fellow who was obviously pretty knowledgeable about computers (much moreso than your usual telephone support person; he sounded like a computer nerd). Anyway, he told me that the only other thing he could think of to do that I hadn't done was to remove the one removeable Sodimm from the notebook and see if the problems are eliminated. That makes sense, however I've run 3 or 4 passes of Memtest86 on the laptop and not had any errors; my sense is that (1) it is very unlikely that Memtest86 would miss a memory error; (2) if there are memory errors, there are 3 possible sources, e.g. 2 sticks of ram only one of which is user removable plus the cpu which is also not user accessible, and (3) given #1 and #2, we are talking about such a very small possible yield that it isn't worth doing, e.g. removing the memory module as a test procedure. Alternatively, I guess I could run another memory testing utility but I question if there is one out there that is any better than Memtest86. After booting into Memtest, press "c" "2" "3" and "Enter" to do all tests. Test #11 is very thorough. Some will say GoldMem 5.07 is better. http://www.goldmemory.cz/ thanks so much for this response; after 5 hours we haven't quite gotten to test#11 (we're at 10), but obviously it is testing something and if it comes back with no errors I have little reason to question the RAM. Thanks, ken |
Ken Fox wrote:
"S.Heenan" wrote in message news:LgbSb.319487$JQ1.7855@pd7tw1no... Ken Fox wrote: Hi, I've got a Dell C400 laptop with 768MB of RAM, that obviously I can't overclock. The reason I'm posting this question here is that I think this ng is about as knowledgeable about Memtest86 as any other ng. I've had some blue screen of death (BSOD) problems with the notebook when trying to use wireless networking cards in the PC card slot; this is a not uncommon problem from what I can tell, reading various Dell forums and other ng postings. I believe it is software related. Anyway, I spent yet another 45 mins on the phone with Dell laptop tech support yesterday, speaking to a fellow who was obviously pretty knowledgeable about computers (much moreso than your usual telephone support person; he sounded like a computer nerd). Anyway, he told me that the only other thing he could think of to do that I hadn't done was to remove the one removeable Sodimm from the notebook and see if the problems are eliminated. That makes sense, however I've run 3 or 4 passes of Memtest86 on the laptop and not had any errors; my sense is that (1) it is very unlikely that Memtest86 would miss a memory error; (2) if there are memory errors, there are 3 possible sources, e.g. 2 sticks of ram only one of which is user removable plus the cpu which is also not user accessible, and (3) given #1 and #2, we are talking about such a very small possible yield that it isn't worth doing, e.g. removing the memory module as a test procedure. Alternatively, I guess I could run another memory testing utility but I question if there is one out there that is any better than Memtest86. After booting into Memtest, press "c" "2" "3" and "Enter" to do all tests. Test #11 is very thorough. Some will say GoldMem 5.07 is better. http://www.goldmemory.cz/ thanks so much for this response; after 5 hours we haven't quite gotten to test#11 (we're at 10), but obviously it is testing something and if it comes back with no errors I have little reason to question the RAM. Just a thought, laptops are quite different to desktops, maybe the extra RAM module competes for resources with the PC card slot? Or uses power from the same PCB trace or something? There may be a reason the tech suggested you remove it. It shouldn't take long, then you'd know if it was going to make a difference, you can always put it back in. -- ~misfit~ |
"~misfit~" wrote in message
... Ken Fox wrote: "S.Heenan" wrote in message news:LgbSb.319487$JQ1.7855@pd7tw1no... Ken Fox wrote: Hi, I've got a Dell C400 laptop with 768MB of RAM, that obviously I can't overclock. The reason I'm posting this question here is that I think this ng is about as knowledgeable about Memtest86 as any other ng. I've had some blue screen of death (BSOD) problems with the notebook when trying to use wireless networking cards in the PC card slot; this is a not uncommon problem from what I can tell, reading various Dell forums and other ng postings. I believe it is software related. Anyway, I spent yet another 45 mins on the phone with Dell laptop tech support yesterday, speaking to a fellow who was obviously pretty knowledgeable about computers (much moreso than your usual telephone support person; he sounded like a computer nerd). Anyway, he told me that the only other thing he could think of to do that I hadn't done was to remove the one removeable Sodimm from the notebook and see if the problems are eliminated. That makes sense, however I've run 3 or 4 passes of Memtest86 on the laptop and not had any errors; my sense is that (1) it is very unlikely that Memtest86 would miss a memory error; (2) if there are memory errors, there are 3 possible sources, e.g. 2 sticks of ram only one of which is user removable plus the cpu which is also not user accessible, and (3) given #1 and #2, we are talking about such a very small possible yield that it isn't worth doing, e.g. removing the memory module as a test procedure. Alternatively, I guess I could run another memory testing utility but I question if there is one out there that is any better than Memtest86. After booting into Memtest, press "c" "2" "3" and "Enter" to do all tests. Test #11 is very thorough. Some will say GoldMem 5.07 is better. http://www.goldmemory.cz/ thanks so much for this response; after 5 hours we haven't quite gotten to test#11 (we're at 10), but obviously it is testing something and if it comes back with no errors I have little reason to question the RAM. Just a thought, laptops are quite different to desktops, maybe the extra RAM module competes for resources with the PC card slot? Or uses power from the same PCB trace or something? There may be a reason the tech suggested you remove it. It shouldn't take long, then you'd know if it was going to make a difference, you can always put it back in. -- ~misfit~ The tech specifically said that there was a small possibility that Memtest could miss a RAM error; that's what he was addressing. I did run the "all tests" sequence of Memtest86 which literally took 42 hours; there were zero errors. The reason why I doubt bad RAM is that this notebook functions flawlessly except for that one problem, dealing with that wireless card. I've had bad ram before in other systems, and the hallmark of bad ram in my experience is lots of problems you have trouble identifying, with little or no pattern to it, generally with scads of blue screens; in essence, a very unstable and unreliable system. My laptop does not have these symptoms. I am going to try a couple of more things; one is, that a driver that keeps showing up on the blue screens is "ndis.sys," which is a networking driver of some sort. I'm going to replace the one(s) on the laptop with the same file from my desktop to rule out corruption of that file. I'm also going to try a "repair install" of Windows 2000, sp4 version. If these actions fail to solve the problem I think it is unsolvable until or unless MS, Dell, or the wireless card makers alter their software, since this problem is clearly affecting quite a few users. Thanks all, ken |
Ken Fox wrote:
"~misfit~" wrote in message ... Ken Fox wrote: "S.Heenan" wrote in message news:LgbSb.319487$JQ1.7855@pd7tw1no... Ken Fox wrote: Hi, I've got a Dell C400 laptop with 768MB of RAM, that obviously I can't overclock. The reason I'm posting this question here is that I think this ng is about as knowledgeable about Memtest86 as any other ng. I've had some blue screen of death (BSOD) problems with the notebook when trying to use wireless networking cards in the PC card slot; this is a not uncommon problem from what I can tell, reading various Dell forums and other ng postings. I believe it is software related. Anyway, I spent yet another 45 mins on the phone with Dell laptop tech support yesterday, speaking to a fellow who was obviously pretty knowledgeable about computers (much moreso than your usual telephone support person; he sounded like a computer nerd). Anyway, he told me that the only other thing he could think of to do that I hadn't done was to remove the one removeable Sodimm from the notebook and see if the problems are eliminated. That makes sense, however I've run 3 or 4 passes of Memtest86 on the laptop and not had any errors; my sense is that (1) it is very unlikely that Memtest86 would miss a memory error; (2) if there are memory errors, there are 3 possible sources, e.g. 2 sticks of ram only one of which is user removable plus the cpu which is also not user accessible, and (3) given #1 and #2, we are talking about such a very small possible yield that it isn't worth doing, e.g. removing the memory module as a test procedure. Alternatively, I guess I could run another memory testing utility but I question if there is one out there that is any better than Memtest86. After booting into Memtest, press "c" "2" "3" and "Enter" to do all tests. Test #11 is very thorough. Some will say GoldMem 5.07 is better. http://www.goldmemory.cz/ thanks so much for this response; after 5 hours we haven't quite gotten to test#11 (we're at 10), but obviously it is testing something and if it comes back with no errors I have little reason to question the RAM. Just a thought, laptops are quite different to desktops, maybe the extra RAM module competes for resources with the PC card slot? Or uses power from the same PCB trace or something? There may be a reason the tech suggested you remove it. It shouldn't take long, then you'd know if it was going to make a difference, you can always put it back in. -- ~misfit~ The tech specifically said that there was a small possibility that Memtest could miss a RAM error; that's what he was addressing. I did run the "all tests" sequence of Memtest86 which literally took 42 hours; there were zero errors. The reason why I doubt bad RAM is that this notebook functions flawlessly except for that one problem, dealing with that wireless card. I've had bad ram before in other systems, and the hallmark of bad ram in my experience is lots of problems you have trouble identifying, with little or no pattern to it, generally with scads of blue screens; in essence, a very unstable and unreliable system. My laptop does not have these symptoms. I am going to try a couple of more things; one is, that a driver that keeps showing up on the blue screens is "ndis.sys," which is a networking driver of some sort. I'm going to replace the one(s) on the laptop with the same file from my desktop to rule out corruption of that file. I'm also going to try a "repair install" of Windows 2000, sp4 version. If these actions fail to solve the problem I think it is unsolvable until or unless MS, Dell, or the wireless card makers alter their software, since this problem is clearly affecting quite a few users. Thanks all, ken If your wireless card has a power setting try turning transmit power to a lower level. They pull a fair amount of current, for a PCMCIA card, and maybe the notebook is having trouble providing it. |
"David Maynard" wrote in message
... If your wireless card has a power setting try turning transmit power to a lower level. They pull a fair amount of current, for a PCMCIA card, and maybe the notebook is having trouble providing it. Hi David, The wireless card does have such a setting; I was (at least initially, haven't pursued it much since but may go back to it) unable to get a decent connection on the lower power settings last night, unfortunately. I had removed the wireless card software and drivers from my laptop so I wasn't able to try the power setting initially, by itself. Because these blue screen errors have been blamed on out of date bioses in the past, I'd been reluctant to see if the bios was related being as I already had the most up to date bios flash on my notebook (it dates from May 2003 but they have not updated it since my flash which was probably in June). I decided to reflash the bios. The Dell utility offers an option to flash without using a floppy; I tried that, but all I got was an error message saying I already had that level of bios so the program defaulted to not flashing the bios. The next time I refused the floppy-less flash, and forced it to make a boot floppy with the bios on it, then rebooted. The bios flash (per floppy) program apparently doesn't check what level you were already at, so it just flashed the bios without asking. I rebooted, loaded the software and drivers for the wireless card, and did not get any blue screens. I had trouble getting a decent connection, and as it was late decided not to bother with the access point last night and deferred dealing further with this until this afternoon. After rebooting the wireless AP, I'm now getting good connections on full power to the card, but did not no go back to the power setting on the driver. Since I only use the laptop on wireless when it is plugged in anyway, I wonder how much the power setting really matters anyway, but if I get blue screens again I'll try the power setting and see if that resolves them. For now, the bios reflash appears to have done the trick. I should add, that since I was able to use the wireless card for about 6 months without any problems and without setting it to low power, I think this makes a pretty good case that my bios got corrupted sometime over the last few months, for some reason, and that this was the source of all my aggravations with my DLink card! Thanks again for your input. ken |
This is completely off-topic to this forum, but I'm posting a brief further
followup on the chance that someone will pull this thread up on a google search investigating wireless card-caused blue screen errors on laptops. In the prior post I mentioned that reflashing the bios seemed to solve the problem; it did for a day or so but the problems have returned, e.g. repetitive blue screen errors with "DRIVER_IRQL_NOT_LESS_OR_EQUAL, blah blah blah. I've run google searches on the exact set of hexidecimal codes and have determined this is a very common problem afflicting all sorts of wireless cards, apparently mostly based on a TI chip, and it is found in some PCI wireless cards as well as laptop PC cards. No one has found a solution that works, unfortunately. The driver is bad and conflicts with certain sorts of hardware in certain notebooks and primarily laptops. Unless the drivers are rewritten and or windows is patched, I don't think there is a solution that an actual enduser can utilize. I did retry the lower power settings in the laptop for the wireless card and they don't help. Thanks, Ken "Ken Fox" wrote in message ... "David Maynard" wrote in message ... If your wireless card has a power setting try turning transmit power to a lower level. They pull a fair amount of current, for a PCMCIA card, and maybe the notebook is having trouble providing it. Hi David, The wireless card does have such a setting; I was (at least initially, haven't pursued it much since but may go back to it) unable to get a decent connection on the lower power settings last night, unfortunately. I had removed the wireless card software and drivers from my laptop so I wasn't able to try the power setting initially, by itself. Because these blue screen errors have been blamed on out of date bioses in the past, I'd been reluctant to see if the bios was related being as I already had the most up to date bios flash on my notebook (it dates from May 2003 but they have not updated it since my flash which was probably in June). I decided to reflash the bios. The Dell utility offers an option to flash without using a floppy; I tried that, but all I got was an error message saying I already had that level of bios so the program defaulted to not flashing the bios. The next time I refused the floppy-less flash, and forced it to make a boot floppy with the bios on it, then rebooted. The bios flash (per floppy) program apparently doesn't check what level you were already at, so it just flashed the bios without asking. I rebooted, loaded the software and drivers for the wireless card, and did not get any blue screens. I had trouble getting a decent connection, and as it was late decided not to bother with the access point last night and deferred dealing further with this until this afternoon. After rebooting the wireless AP, I'm now getting good connections on full power to the card, but did not no go back to the power setting on the driver. Since I only use the laptop on wireless when it is plugged in anyway, I wonder how much the power setting really matters anyway, but if I get blue screens again I'll try the power setting and see if that resolves them. For now, the bios reflash appears to have done the trick. I should add, that since I was able to use the wireless card for about 6 months without any problems and without setting it to low power, I think this makes a pretty good case that my bios got corrupted sometime over the last few months, for some reason, and that this was the source of all my aggravations with my DLink card! Thanks again for your input. ken |
All times are GMT +1. The time now is 03:00 PM. |
Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
HardwareBanter.com