Thinkpads Forum

While the 1st stick of additional RAM I tried (Archer brand, can't remember memory manufacturer) refused to work because of compatability problems with the stick which came with the notebook (T60 2007-72A), the Kingston stick I bought as a replacement worked without a problem. Unlike all the other users here who could at least boot into Windows before their second stick started giving problems, I couldn't even boot into Windows. I've yet to have a problem with my current 2x1Gb of RAM.

I've yet to change any memory setting in the BIOS and I was able to run emm386 with my current setup.
http://img.photobucket.com/albums/v85/r ... 000761.jpg

Conmee wrote:PLEASE PLEASE PLEASE!!!! If anyone can boot their ThinkPad T60 (using USB or floppy or CDROM) and have it load himem.sys and emm386.exe and see if you can get to a command prompt and run the mem command, I'd appreciate it. Especially those folks that have T60 machines with 2GB that don't lockup. I just want to know if EMM386 isn't working with the BIOS or the memory speed or worse, I have a defective memory controller. Since this problem occurs regardless of the type or amount of ram I have installed, I'm assuming EMM386 will lockup other T60s as well.

I did as you asked, with a floppy boot disk. Mem ran correctly, reporting 2GB of memory. I ran a Win98 version, so the display dropped the last digit, but that doesn't matter. Enabling extended memory test in the BIOS makes no difference here - that's just running the test, not changing the way that memory behaves.

I'm running BIOS version 1.05a (79ET60WW) dated 2006-04-18 and Embedded controller version 1.02.

Conmee wrote:One last bit: When I check the system info from the Catalyst Control Panel, where it lists all system components, it tells me that my memory is running at 0Hz. Just wondering if that is what other folks are seeing.
Daniel.

My Catalyst Control Center (CCC) shows
Core Click in MHz 398 MHz
Memory Clock in Mhz 324 MHz
I wonder if this is related to the pixelation problem?

mss1337 wrote:I am just about to receive a new t60 2ghz, x1400, 14'' sxga from my school (I don't know the model no.) It sounds like the t60 doesn't support 2gig of ram. Is this a big problem or has it only been happening in a select few computers?

I've had three T60p machines each with 2GB of memory (two 200793U and one 2323DDU). Two were shipped around March 16 2006, and the other (my main computer) was shipped on April 28 2006. I've never had a single lock-up, blue screen, or any other problem like you're seeing. I've run every 3DMark test, tested games, and have stressed the machines to the hilt. I think it's safe to say that the memory problem is limited to a few select (unlucky) machines. I ordered each of the 2GB machines with memory from Lenovo. On a fourth T60 machine (2623D6U), I ordered a Transcend 512M module from Newegg ($35) to bring its total memory to 1GB. That T60 has been in near-continuous use with no problems either.

Conmee wrote: So the net-net:

It looks as if Lenovo received a bad batch of early memory from Micron (at least in my case). As mentioned, I have two Micron parts purchased directly from Lenovo that were initially backordered. Both have a May 22, 2006 date stamp and are sequentially numbered. With both of these SODIMM installed, my T60 will simply lockup after a few moments or when initiating a 3D application. Either the SODIMM at 667MHz speeds are even more sensitive to timing and tolerances, and even the slightest variation causes problems, or there's someone along the way (Memory makers/Lenovo/etc) have incorrectly set specs. What do I know, though, since I'm not an engineer.

Daniel.

Very interesting as I ordered an extra 1GB RAM directly from Lenovo and it was backordered for a few weeks like yourself. I've been having problems with lock ups only when running both SODIMMs, but worked fine separately. I just looked at the date stamp and it's May 22, 2006. Now I'm starting to believe it's probably a bad SODIMM from Lenovo.

For myself, it always locked up on BIOS 1.06. I downgraded to 1.04 and it ran fine a few times then locked up. Now it continues to lock up even with 1.04

Conmee - What did you do with your Lenovo SODIMMs? Did you return them?

RonS wrote: I've had three T60p machines each with 2GB of memory (two 200793U and one 2323DDU). Two were shipped around March 16 2006, and the other (my main computer) was shipped on April 28 2006. I've never had a single lock-up, blue screen, or any other problem like you're seeing. I've run every 3DMark test, tested games, and have stressed the machines to the hilt. I think it's safe to say that the memory problem is limited to a few select (unlucky) machines. I ordered each of the 2GB machines with memory from Lenovo. On a fourth T60 machine (2623D6U), I ordered a Transcend 512M module from Newegg ($35) to bring its total memory to 1GB. That T60 has been in near-continuous use with no problems either.

Ron, is the memory in each T60p matched/identical? I'm wondering based on your report and others whether the problem lies in the ram.

Also, my datapoint on service update - I was scheduled to have a tech out today to "replace the system board and memory", as I was told last night. Unfortunately that wasn't true - no tech showed up and when I finally called towards the end of the service window (8am -5 pm ) I was told that no tech was ever sent since they don't have the parts. Apparently, the replacement memory is out of stock and they are waiting for a shipment from Taiwan to come in and they estimate it will be 5-10 days before it arrives. I did ask about depot service, but apparently the depot is out of replacement memory as well.

i have the same laptop.

i have noticed my laptop only freezes like that if i hold my laptop in one hand from the front right hand sidewhere the finger print scanner is and move the laptop around.

my laptop freezes instantly. it mightnot be the same case for you guys.

I got my pair of Crucial 1GB SODIMMs this morning from Newegg, installed them together in my T60p (2007-92U) and they've been running without problem all day. I tried running the video stress test and memory stress tests from Dr. DOS with out probelm for about 15 minutes each.

I could, however, crash the system by running the memory stress test and a complex 3d game at the same time. Just playing the game for half an hour--only in the spirit of testing, of course -- didn't freeze the system at all.

Since I don't know how sophisticated the memory stress test is. I suppose it's possible that it may have confused the cache or TLB while the graphics game wa srunning. (I'm no Intel maven, but I'm assuming they have something at least similar to a TLB).

I also tried letting the T60p cool all the way down before sustained tests warmed it up just to see there were any thermal problems.

The new SODIMMs were both from Crucial. Each was marked on the back with its Micron part numer: MT16HTF12864HY-667B3 -- the same Micron part number as on my "official" Lenovo SODIMM. the new ones both show a date code from a couple of weeks ago and are "Made in USA," whereas the other Micron board was produced about four weeks ago in in Singapore.

Incidentally, all of this happened under BIOS 1.06.

This thing is going to have to work without problems for several days before I trust it, though.

When I found this topic,I predicted this issue could happen in my new T60P,
because I ordered a T60P(200784U) which has 1GB Lenovo RAM,and I already had a another vendor 1GB RAM for extension.
But actually new machine has been working correctly with no freeze in 2GB(2 x 1GB).
I checked it by 3DMark05 several times.Still need to verify.But maybe I am lucky!

SoDimm #1 - Samsung - 1GB 2Rx8 PC2-5300S-555-12-E3
SoDimm #2 - ACP MemoryUpgrades - 1GB PC2-5300(AA667D2S5/1GB)(Lenovo 40Y7734-AA?? by Buy.com)
Bios 1.05a(79ET60WW)

ibmthink wrote:i have the same laptop.

i have noticed my laptop only freezes like that if i hold my laptop in one hand from the front right hand sidewhere the finger print scanner is and move the laptop around.

my laptop freezes instantly. it mightnot be the same case for you guys.

That is a different problem; usually related to loose internal connections and mobo issues. It isn't good to hold it like that either.

Update guys:
I think the problem (at least for myself and conmee) was associated with 1GB SODIMM (dated May 22, 2006) from Lenovo. I've tested both SODIMMs and they always worked independently but locked up when together (tested both RAM slots too).

I ended up buying a transcend 1GB SODIMM (from this thread http://forum.thinkpads.com/viewtopic.php?t=26650) and installed it yesterday. Have not had any problems (running BIOS 1.06) and ran memtest all night with no errors.

I am now returning my SODIMM from Lenovo and keeping the transcend. This may not solve everyone's problems, but hopefully it will help some.

My Lenovo SODIMMs were also 5-22-2006 and I can't return them because EPP purchases on Accessories are no return. So I've asked to get two NEW SODIMM (backordered) and will sell them at a great discount when I get them!

So far my Elpida SODIMMs are working flawlessly. After some more testing, I've found that EMM386.EXE will not work if I boot from a USB Key or if the T60 is docked AND there is a SATA drive in the dock bay. EMM386.EXE works fine if booting from a floppy. I've tried excluding high memory addresses based on what Lenovo says are reserved for PXE, BIOS, Video, and USB support, but there is just something about the T60 that doesn't like a memory manager fussing with memory in these scenarios. The good news is that I don't necessarily need EMM386.EXE to install/run in those scenarios. But... if anyone can create a bootable USB Key using MS-DOS 6.22 or Win95/Win98 and use the EMM386.EXE /NOEMS line in the config sys file, and see if you can get to a command prompt and do a dir listing, I'd appreciate it. I'm 99% convinced I don't have any problem with my T60 or memory controller, and that my new SODIMMs are flawless, but I just would like a few data points to see if others encounter lockups using EMM386 while docked (with a SATA drive in the dock bay) or when booting with a USB key.

Current setup is using BIOS 1.04, with virtualization set to disable and Extended Memory test set to disable. The one test I need to try is upgrading to BIOS 1.06 with my new memory and seeing if there is any video pixelation. I have a feeling it was the memory that was causing all my problems, not the BIOS update. I'll report on that later.

Daniel.

Mtrack wrote:Update guys:
I think the problem (at least for myself and conmee) was associated with 1GB SODIMM (dated May 22, 2006) from Lenovo.

I'm (somewhat) relieved to see I'm not the only one having problems with a T60p.


I just called Lenovo (or IBM, whatever ), and after asking me for the memory DIMM part number the tech support person told me that replacing the DIMMs would absolutely solve the problem.

Chances are the DIMMs we got are defective and they know about it; I'm getting replacement DIMMs on tuesday, so I'll let you know how the new DIMMs work -- but this is the 4th thinkpad I've own and after previous experiences with thinkpad support I'm sure things will be smooth after tuesday!

Just so you know, the guy asked me to look at the DIMM and look for a number of the form VJ16U, and told me this part was definitely causing the lockups. If you have these DIMMs and you haven't called Lenovo yet call them up and they'll replace the DIMMs for you. If you don't know how to tell, download the hardware maintenance guide (http://www-307.ibm.com/pc/support/site. ... MIGR-62733) for instructions on how to open the laptop and take the DIMMs out. HTH

My two SODIMMs also have VJ16U in the Serial Number... so hopefully, it was just a case of a bad batch of SODIMMs...

Daniel.

Conmee wrote:My two SODIMMs also have VJ16U in the Serial Number... so hopefully, it was just a case of a bad batch of SODIMMs...

Daniel.

Just looked at my SODIMM from Lenovo and it has the same VJ16U.

I just discovered this forum in a desperate search for background information on the problem I was having with my T60p.

I am amazed at the level of tolerance some people have. The ThinPad is billed as a business machine. To me, that means reliability. Experimenting with various memory parts is fine if you have the time and money, but even if you are lucky and find something that works, the reliability of the machine is undermined.

The person who defended their support line is too kind. It took me four phone calls to get a reasonable analysis. The first rep offered outright to replace the disk drive, which I declined until I ran diagnostics. It took two calls to locate the correct version of PC Dr. After that was run, the fourth rep assured me there was a memory problem which PC Dr. could not detect.

The rep asked me to unscrew the case so I could tell him the serial no. of the memory part, but I declined. Obviously, IBM has a list of defective memory parts, but IBM does not seem to have a list of which computers those parts are installed on.

Conclusion: someone in some part of the world is popping memory into a $3000 business machine, but not keeping track of what goes in; or, if they are keeping track, they are not telling anyone; or, if they are telling, the people who need to know (IBM) are not listening.

I am planning to send this computer back for a replacement. The question in my mind is, if Lenovo/IBM did not know a defective part was in my first machine, how will they know about the replacement?

robbi wrote: I am planning to send this computer back for a replacement. The question in my mind is, if Lenovo/IBM did not know a defective part was in my first machine, how will they know about the replacement?

They know now and that is enough; I imagine that they are irate to some extent. IBM, now IBM/Lenovo, has always had to juggle and mix and match and jump through hoops just to get the parts they spec'd from the suppliers. The suppliers are all to keen to get business from companies like IBM, but after they acquire them as a customer they move on to other things and are no longer all that eager to cut deals. The same thing has happened to other large companies, even the ones that tested their memory; the biggest, in recent memory, affected HP and their notebooks. HP ended up doing a system-wide recall of just about all the memory they thought was bad. Lucky that this got caught as soon as it did.

Something must have fallen through the cracks and, given the nature of this problem and that IBM/Lenovo generally sells machines with less than 2GB and the client upgrades it, they probably didn't get enough feedback and/or a chance on the models in question to perform highly detailed testing as people in this thread have done to pin down the problem.

Talked to Lenovo/IBM today. I'm sending back my "defective" SODIMM from them and they're sending me a new one.

I had also the same problem with my 2 GB T60p purchased in Germany (1 GB chipped on board and one as an Lenovo add-on).

An IBM technician told me they already know the problem and you can identify the faulty ram by the serial number (the number starting with 11S printed on the memory).

Beside the already known "VJ 16U" 1GB chips marked with "JV 172" have the same problem not suitable to work together with other 1 GB chips.

My local dealer replaced my chips within 1 day with new ones: both from samsung. Now everthing is running fine.

is anyone out there seeing this issue with 512mb sticks? I started a thread after encountering numerous lock-ups on my 2623-D6U w/ 2 x 512:

http://forum.thinkpads.com/viewtopic.php?t=26992

Well I was called yesterday by the tech for the onsite service. Apparently I'm still in the system as needing a new system board and a new memory stick. He doesn't have the memory stick, but wants to replace the system board.

After reviewing the Hardware Maintenance Manual on the IBM site for the T60, the steps involved in replacing the system board seem extensive, and quite frankly I'm worried about the skill level required for the system board replacement and all the steps required. If he doesn't show up with the proper tools and parts I'm not going to allow him to work on the machine.

Has anyone had a field tech replace their system board for the memory lock up problem?

dr72 wrote: Has anyone had a field tech replace their system board for the memory lock up problem?

I'm not a field tech but I've replaced the system board on my T42. It's a piece of cake, you just need a screw driver and the hardware maintenance manual.

The tech visited today. Thankfully he had replacement ram. He replaced the Micron ram with another stick of Samsung ram (matching the original one) and it seems to be working fine. I've run 3DMark05 in a loop for some time and everything seems to work great. Same for firefox, office apps, etc.

Is there a more specific memory test I should run?

As a side note, the palmrest wasn't re-installed properly, so I had to re-install it. Thankfully, we have ESD workstations here at work and used that to put the palmrest back on correctly.

dr72 wrote:The tech visited today. Thankfully he had replacement ram. He replaced the Micron ram with another stick of Samsung ram (matching the original one) and it seems to be working fine. I've run 3DMark05 in a loop for some time and everything seems to work great. Same for firefox, office apps, etc.

Is there a more specific memory test I should run?

As a side note, the palmrest wasn't re-installed properly, so I had to re-install it. Thankfully, we have ESD workstations here at work and used that to put the palmrest back on correctly.

Install one of the game demos mentioned in the other game crashing thread, and see if they work.

jenss wrote:Just thought my input might help tracking down the problem. The fact that 2 GB running on low clock speed (533 MHz) do not cause lockups might give some support for the 'overheating hypothesis' as a cause for the lcokups.

Sounds to me like Lenovo or ATI (or both) have a poor implementation that can't handle dual-channel mode correctly. Running two different speed SO-DIMMs keeps it from going into dual-channel mode. If it's a hardware issue, eveyrone is probably screwed out of the 10% or so that dual-channel mode gets you (possibly a bit better than that, when you're using enough VRAM to kick HyperMemory into action).

Bummer.

I'm running out of time to order while the rebate is still in effect, and all these problems I'm reading about are making it *really* hard to pull the trigger on a T60p, FlexView sweetness or not.

My TP60p was locking up when I installed a second Lenovo-purchased 1GB SODIMM (from Micron, and with the infamous VJ16U stamp) along with the Hynix module supplied with the laptop. I called Lenovo support and they ordered a new SODIMM to replace the Micron SODIMM, but told me that it was backordered until mid-July.

Since I have an overseas trip coming up I bought two Crucial SODIMMs from Newegg and tried them. The TP60p locked up after a few minutes

Amazingly, the Lenovo replacement SODIMM, a Samsung, showed up last Tuesday. I installed it with the Hynix and they've been working together with no problem for the last three days of hard use on the TP60p.

Since Crucial SODIMMS are really thinly-disguised Microns, I have to assume that the TP has some issue with Micron SODIMMS. At 667MHz, clocking data on both edges of the 1.5ns pulse, even small differences in the circuit board traces on the laptop's mother board, the SODIMM socket and the memory stick can have a big effect.

I don't believe that the Micron sticks are defective--just that they don't work right in TP60s.

Oh my! I'm expecting a T60 (200764U) with 2GB RAM from Lenovo this week and I'm worried that there might be problems.

I didn't get the T60p anymore and I hope it will be fine.

I have this issue on my T60P with 1gb factory installed and an extra 1gb Lenovo installed (Micron manufactured 23/05/2006) I get the lockup when playing Counterstrike Source after about 30 mins. I phoned IBM support in the UK and they seem to want me to prove that it's a defective memory chip by just using the extra 1gb Micron chip - I asked if they knew anything about defective Micron chips and they said not. Bit of a catch22 this as so far the problem only appears with both chips installed - I mentioned this to the Tech support person who said it may indicate another problem with the system such as systemboard, cpu fan e.t.c.

My T60p, only a few weeks old, is experiencing lockups too . It does have two 1GB RAM Micron cards with the "VJ16U" code. When calling IBM tech support, however, the person I talked to did not know of this issue. He wanted me to do re-imaging and run the IBM/Thinkvantage diagnostic tests.

Was anyone able to have the IBM diagnostics test fail with these Micron memory cards? What tech support did you call to have them resend different memory cards?

I think IBM/Lenovo should automatically replace all these defective memory cards when a user requests that.

Thanks.

rfm wrote:

Mtrack wrote:Update guys:
I think the problem (at least for myself and conmee) was associated with 1GB SODIMM (dated May 22, 2006) from Lenovo.

I'm (somewhat) relieved to see I'm not the only one having problems with a T60p.


I just called Lenovo (or IBM, whatever ), and after asking me for the memory DIMM part number the tech support person told me that replacing the DIMMs would absolutely solve the problem.

Chances are the DIMMs we got are defective and they know about it; I'm getting replacement DIMMs on tuesday, so I'll let you know how the new DIMMs work -- but this is the 4th thinkpad I've own and after previous experiences with thinkpad support I'm sure things will be smooth after tuesday!

Just so you know, the guy asked me to look at the DIMM and look for a number of the form VJ16U, and told me this part was definitely causing the lockups. If you have these DIMMs and you haven't called Lenovo yet call them up and they'll replace the DIMMs for you. If you don't know how to tell, download the hardware maintenance guide (http://www-307.ibm.com/pc/support/site. ... MIGR-62733) for instructions on how to open the laptop and take the DIMMs out. HTH

The Hynix and Samsung pair of 1GB sticks have been playing in my T60p now for over a week of hard use with not so much as a hiccup.

It seems as though my T60p has a timing issue with the Micron SODIMMs.

I'll have to try my remaining pair of brand-new Crucial/Micron SODIMMs in another system--maybe a Mac mini or Macbook.

Well after another call with Thinkpad support here in the UK, I'm still not much further forward. The hardware support rep I spoke to after trying out their suggestion of installing one memeory card at a time (they both work fine singly but as soon as I install them both together the system locks up after 20 mins or so and interestingly will not run 3dmark05 without crashing), suggested that I send the machine in for analysis and repair. I pressed about defective memory chips particularly ones from Micron with the VJ16U stamp which I have, but she indicated they did'nt know of any problems.
Being reluctant to return and lose my machine for 7 -10 days, she eventually spoke to a supervisor who suggested I revert to BIOS 1.04 - I looked at doing this but noticed that BIOS 1.07 was available. I've upgraded the system to 1.07 and things have improved - the system will now run 3dmark05 without crashing - I get a score of 3541 which seems very good. The system still locks up but it seems to take much longer on BIOS 1.07 sometimes it can take hours before it locks up.
Clearly, there is a problem here which seems related to BIOS, Dual channel memory usage, Graphics card and memory brand but getting Thinkpad support to recognise this and do something is proving difficult - maybe this problem needs to be escalated?

I have a 1 GB Elpida and a 1 GB Micron... both installed by Lenovo. I would get dead hangs multiple times daily with BIOS 1.07 and 1.06. However, BIOS 1.04 would hang only about once a day.

Lenovo Tech Support is sending me another 1GB memory stick to replace the VJ16U Micron.