Strange things happening at the depot

You're right I did confuse the northbridge with the southbridge. Sorry!

Embarassing

The temperature readings I was posting were for the SOUTH bridge. I did not measure the north bridge since it is covered up.

So the SOUTH bridge is running at a temperature of around 70C. I wouldn't know what the north bridge is running at, though it must be pretty hot since it's right next to the GPU and sharing the same heatsink.

Again, sorry for the confusion!!

Thanks for pointing it out.

 

By the way, dieselhorse, I notice in the picture that you posted, once again the silicone pad on the GPU/northbridge heatsink is shifted to the right, and doesn't quite line up with the heatsink top.

Why is this I wonder? Kinda strange! I wonder if there is a good reason for this... It doesn't seem right, look how the right side of the pad is just hanging off the edge, why don't they line the pad up better?

 

You're right, the pad being off by as much as it is points to poor manufacturing, however, I don't think that margin of error would cause much of a problem. That pad doesn't handle much of the thermal load from the GPU or north bridge. It seems to help keep the fan down in GPU idle conditions but does very little while the system is loaded.
That being said, I'm not saying the problems you're seeing are not being caused by thermal stress, but it seems to me you may being seeing a limitation in one of the voltage regulators powering the north bridge, not necessarily a problem with the north bridge itself.

 

Nurio I think your test results are very interesting, your shutdowns don't appear to be related to the temperature readings...strange...

Another strange thing, last night I was not able to get my system to crash doing the things I normally do. For some reason, after removing the palm rest and keyboard for the temperature testing, after I put everything back together, i couldn't seem to make it crash! It was strange. I ran the 3D graphics test several times and I even played Bioshock for over an hour, which is very unusual without crashing.

However, the first time I ran the 3D graphics test the GPU temperature got disturbingly high, then it came down again. Check this out:

http://jerry.cleedo.com/tempspike.jpg

It peaked at 97C. Isn't that a little high?

After this peak, it went down again to around 75C for the rest of the 3D graphics test, and when I played Bioshock later, it never got above about 76C. I don't get it...

 

That would be an interesting thing to look into! I remember a while back when I was experimenting with overclocking my Athlon 64 x2 I had a program that reported different system voltages, I wonder if it would be worthwhile to monitor some of the system voltage levels during 3D graphics tests and see if they stay within stable levels...

 

Hellbore,

In those pictures, I think I see the board battery in it's tradtional place which leaves wondering what the yellow thing is? Is it possible that's the accelerometer?

 

It only has 2 wires and I'm pretty sure I can see the shape of a battery under the shrink wrap. I could be wrong though.

Where exactly are you seeing the traditional board battery?

 

Hi Renee,
Please see page 81 of the HMM ( ftp://ftp.software.ibm.com/pc/pccbbs/mobiles/42x3546_01.pdf) which shows the device in question as the backup battery.

 

I didn't have a thermometer, I just used touch. The underside of the laptop right about in the middle of the board.

1 stick: I ran the 3d test many times and played WoW for a while to get the temp up to where it was when I had all the BSODs. I was unable to get an error w/ 1 stick.

2 stick: ran the 3d test twice (10 min ea) and both tests passed. Started up WoW and got a BSOD within a couple min.

 

This pad seems to be a thermal (and electric?) protection pad, probably to prevent a direct contact between the keyboard and the GPU heat sink, in case of a flex with the keyboard. In my laptop, that pad is perfectly aligned and I still have the shutdown problem.

Hmmm… have you had the chance to retest this after a complete shutdown? Is there a chance you might have moved something else on the board beside the pad?

Yes it seems quite high, I believe that 97C is very close to the red line. The bios probably lingered in kicking up the fan speed, now, it might have been just a hiccup, but you should probably keep an eye on that and see if it keeps happening.

 

use pc wizzard for monitoring your temps..it's free!

 

Hellbore, do you have a loaner system, new system, or your original system? I lost track, but thought you were getting a replacement system.

FWIW, I played BF2 last night for about an hour and while the back left corner of my machine got rather hot (had to shift it a few times to keep my leg from roasting), I had no shutdown issues at all, confirming the planar swap seemed to have done the trick.

 

he currently has a loaner

 

Yes I have a loaner and I think they ordered me a new one but I'm not sure if they have put the order in yet.

For me there is no sense of urgency to get the new machine because the loaner gets me by for now, the only thing I am worried about is that I eventually get a machine that works fully with 4 gigs of RAM. If I get the new one and it has the same problem, we'll be right back where we started. I realize the system is stable on 2 gigs, but I shouldn't have to choose between 4 gigs of RAM or having a system that is stable, especially since 4 gigs of RAM is actually a configuration that is offered by Lenovo so it is supposed to be supported. If the machine can't be stable on 4 gigs or can't handle games then they should tell the customer that, then I would have known not to buy it lol...

I believe they are having an engineer look into the issue and see if they can determine exactly what is happening with these issues.

I hope they are able to find what is going on, to me the biggest clue that something is wrong is that various different programs have crashed for different people, but the common thread is that removing one stick of RAM makes the crashes stop happening. I can't see that being anything other than a hardware problem... I guess we'll see if they find it. I don't really know what else to try meanwhile.

 

It has to be load dependent Hellbore. I'm sorry you are having these problems. I have four gigs of ram and my system has never crashed, but I don't play games on the system.

 

Well, the hour of BF2 was with 4GB of RAM (2x2GB).... and no crashes. I did not dumb down the settings at all in BF2 -- I have everything set pretty high... Looked great, played as good as my desktop pc.

 

Don't know if it'll help you out any but I have a T61P with 2 1 gig sticks of memory and I just played through the entire Bioshock demo with no problems whatsoever. After quiting out and looking at the Nvidia monitor I peaked at about 85C on the video card near the end of the demo (it took playing for the entire demo to get up there) and 75c on the two processor cores. I never got any crashes or anything. I used the 169.01 Nvidia drivers.

 

Well my loaner T61P stopped having any crashes after I took it apart and put it back together... for a couple days at least.

I think the problem has come back. Today I took it to school with 4 gigs of RAM in it, and maybe 1 minute after it booted up, I was just browsing the notebookreview.com forums and the machine HARD LOCKED. No response to any keystrokes or trackpoint. Total hard lockup, I had to hold down the power button for 5 seconds to shut it down.

So yeah, I don't think the problems are solved

 

I just got my t61p last week, but only started running across your and other threads after running into the same issues. My specs arelighter than yours, 32 bit, 7500, and 2x 1 gig modules. The issues are similar though. (bios 1.26, all lenovo drivers up to date)

What I've found is that as with others, only 3d stuff civ 4 warlords, bf 2, even 3dmark (if I run it after the machine is hot) simply instantly shut down, typically after 10-30 minutes, give or take. Bf 2 I had to push pretty hard to get that to happen, civ it invariably does. If I try some burn in stuff thats non graphical, eg sandra, or super pi with a host of other things, i find it very difficult to force the shutdown.

However, with only one stick in, in the lower slot I haven'y managed to kill it yet ( on similar settings). What I have noticed is that the overall running temp is lower (at least 5-10 degrees both on and off load, measured and logged with speedfan beta 3.4, if mem serves). With both it easily creeps up to 75+, though I've seen it get 85-90+ (but not recently in my logs). With the one stick civ has been ruuning at ~70C core consistently ( feels noticeably cooler as well). I don't know, but it seems to me that it is running northbirdge significantly hotter under dual channel, and with the added load from the gpu in 3d, its enough to tip it over the edge.

The question is what to do about it. I'm seriously considering getting a 2 gig stick and running either it alone or 3 gigs ( though from what I've read this isn't always better). It doesn't help you with the 4 gigs as you need both slots, but I thought someone might find more information useful. I'm still trying to map out the problem using speedfan to log continuously while I try stuff, but this is what I've seen (thats relevant anyways)

 

@Macroscope,

I get the same problem with 3GB. Right now, the 1GB Lenovo stick is on top of my desk and the machine works fine with a 2gb stick in the bottom slot.

Now my problem has become intermittent wireless disconnects while gaming...

 

Is there anyone that has this problem with 2 1 gig sticks? I know that I played through Bioshock just fine with 2 1 gig sticks and have had no stability issues. Just asking out of curiosity.

 

I'm interested in hearing more about people's wireless disconnects. I just bought a T61p, and have been using a T43 for the past year. My T43 works great (except for a quirky issue where the DVD-ROM drive sporadically ejects, I suspect a short circuit or loose connection or something). Whenever I play Word of Warcraft on it, though, it heats up and causes the wireless to stop working altogether, usually in less than 15-20 minutes. I think the wifi components have much less heat tolerance, and wouldn't be surprised to see similar issues with the T61p. It's not a big deal for me to just do my gaming while on a wired connection, it's generally faster and more reliable that way anyway, it's just interesting to me.

 

Infinus, yes, as I said I have the problem with 2 x 1 gig sticks.

HayoB42, thanks. I'll keep that in mind. I havent't tried wireless very much I must admit.

I'm realizing that the temperatures are more complicated than I thought. The core temperatues do not seem wildly diff with one stick (under load), though I do find both the bottom of the computer and other sensors noticeably cooler (however they don't appear to vary with time which makes me more than a little suspicious).

Too confusing, just need to get some more ram and try it. I really don't want to have to go hellbore's route and have it sent back. This is partly because of the terribly unhelpful conversation with tech support I last had. With this vague a problem its hard to get them to commit to anything.

 

Huh, I'm confused then. I have 2x1gig sticks. I do see fairly warm temps with mine but I've had no issues of it freezing up yet. I ran the entire Bioshock demo with no issues at all. The video card at times hit 85c and the processor 75c, so it was definetly cooking, but it ran stable and smooth. I was fairly surprised at how well it performed in Bioshock.

 

To make things even more confusing, get this:

This loaner T61P has been working in games with 4 gigs of RAM for 3 days now, and the only change I made was I took it apart and put it back together. I know I haven't changed anything else, and I have windows update turned off. However, this was not the first time I had opened this computer up, I have swapped the RAM several times before this and the machine was still crashing with 4 gigs.

However, yesterday in class the computer completely froze up while I was opening a web page, which is not one of the problems I was having before, that's a new one.

So I dunno, the freeze-up could be unrelated, but how do you explain the computer suddenly behaving for no apparent reason, when all I did was take it apart and put it back together?

Is this the first Thinkpad to be ROHS compliant? I wonder if there are cold solder joints somewhere causing some kind of intermittent connection when the machine warms up or something.

 

I did not/do not have this problem on either my T41 or my T61p. Only wireless connection problems I ever had was when doing torrent type stuff on my T41.

 

Weird. Just out of curiosity, how did you get them to do anything with the laptop? I'm having a hell of a time convincing them to do anything, if you so much as mention the word games its like they aren't responsible anymore. They are really starting to irritate me.

 

Yes, tech support wasn't helpful at all when I talked about games, but when I told them about how I got a BSOD that says something about NMI Party Error they started listening. Also when I told them my problems disappeared when removing 1 stick of RAM that seemed to help the tech take it more seriously. Once my machine was in the depot the first thing they did was swap out the motherboard so I think the depot must have found something wrong.

 

Hey this is interesting, the Easyserv update page for my depot repair was updated today, here's the new status history:

OS 11/08/2007 14:37:00 M Machine being repaired
HP 10/30/2007 12:35:00 M Hold for Parts - Service call is on hold until parts arrive, estimated at: None provided
BO 10/30/2007 12:35:00 M The last status applied was backed off
OS 10/25/2007 14:36:00 M Machine being repaired
OS 10/25/2007 14:36:00 M Machine being repaired
HP 10/23/2007 08:54:00 M Hold for Parts - Service call is on hold until parts arrive, estimated at: None provided
HP 10/23/2007 08:50:00 M Hold for Parts - Service call is on hold until parts arrive, estimated at: None provided
BO 10/23/2007 08:50:00 M The last status applied was backed off
OS 10/22/2007 04:07:00 M Machine being repaired
AK 10/16/2007 23:31:00 M Airborne processing request to ship container
AK 10/16/2007 18:54:00 M Airborne processing request to ship container

I wonder if this is automatic or if that means the engineers are looking at this issue or what.

As you might know, I currently have a loaner T61P which is good since I have been without my laptop for 25 days now. I was told I was going to be given a brand new T61P to replace my old one, but I haven't been given any proof yet, i.e. I don't have an order number or any ETA on when I might expect it.

It's kind of annoying because even though this loaner T61P is pretty good, it's not as fast as the one I PAID for. I paid a lot of extra money for the T7800 CPU (fastest CPU they offered) to use for CPU-intensive applications and I'm not getting the benefit of this option, the loaner T61P only has a T7700 CPU. You might not think it's a big difference, but it just rankles that I paid a premium for the fastest machine they offer but now for 25 days I have lost the use of this machine that I paid so much for, and my only consolation is I get to use a slower machine while I'm waiting.

For those of you who are new to this thread, if you have had this issue with crashes where the crashes stop if you remove 1 stick of RAM, you should probably email [email protected] and tell him about your issues.

Those of us having this problem need to stick together and stay on Lenovo's case to make sure that solving this issue is a priority and they don't put it on the back burner. They are a large company and I'm sure they prioritize things, if we don't make ourselves heard, we can't expect this to ever be resolved. It's the old addage "the squeaky hinge gets the grease"!

 

Well, if anyone is still following this thread... Here's today's update.

The original laptop came back from the depot today (!!!)

However, I was told I would be getting a new one. As far as I know that is still true, I am guessing I will just need to send this one back or something. We'll see what they say. It's probably nothing to worry about but we'll see.

 

Did they fix it?

 

Update:

False alarm, there's no problem, they are still replacing my laptop with a new one, they just sent me the old one when the depot was done with it apparently, no big deal. They are going to ship it back to Lenovo.

I don't know whether the old one is fixed, I haven't had time to even look at it yet. I've been so busy lately, fortunately they are not rushing me to get it shipped back to them because I probaby won't have time to mess with it for a few days. Lenovo, or I should say Mark, the guy I've been dealing with, has been very helpful.

 

Well, I got an order number for the new laptop. Naturally there is going to be a wait because it's a configured system, but the loaner laptop they sent me is working great so I don't mind.

This has been an unfortunate situation to be in, but Lenovo is taking care of things and the net result is that I only was without a laptop for about a week. That's not so bad really. I don't regret buying a Thinkpad. Situations like this occasionally happen with any brand, I just got unlucky I guess, but they are taking care of things and that's what really matters.

 

Just to throw in my experience:

My problems were not, as I had thought, solved by the BIOS update, and recurred with a vengeance in Ubuntu (merely enabling Compiz can cause the crash) and now are cropping up in XP again. I called Lenovo today and they are sending out a box to send the notebook in for repairs. We;ll see how that goes; I'll keep everyone posted.

 

I'm new to here, but found everything described above to be very similar to my situation. I wish I discovered this forum earlier.

I got my T61p (nvidia quadro, two 2GB hynix modules, XP) few weeks ago and my laments started. The thing was BSODing several of times a day, not particularly in 3D applications but in casual web-browsing or like.

So I called support. Then it was gone through entire cycle of: updating bios -> pcdoctor -> clean install -> repair depot -> clean install -> ... with hours of helpless (mostly) talks with the support team in between. Once I was recommended to call Microsoft to figure out the problem. Finally, one guy told me to take out one of the memory modules and see how it goes. Surprisingly for me it went well and they sent the replacement. Now I have two 2Gb modules: hynix and samsung, and the system has been running for 12 hours without a problem. Both 3DMark and BurnInTest did good, so I'm downloading Bioshock to ensure that my laments are finally over. Wish me a good luck.

p.s Hellbore, your story is even more exciting and I am definitely looking forward to hear the happy end.

 

I know this is an "old" topic, but I said I'd keep you posted, so here goes.

I got my machine back from the depot today. Here's a rundown of what they did:
-Reimage the hard drive - with the WRONG operating system. (I ordered the machine with Vista Business 64-bit, but it came back from the depot with 32-bit.)
-Update BIOS/Drivers - which they didn't actually do, the GPU drivers it came back with were 101.55 (latest being 169.12).

They didn't touch any of the hardware at all. Best I can tell, they didn't even take the damn thing apart. I of course specifically stated everything I did myself to try to fix the problem, among which was reformatting about 15 times with 5 different OSes. But why should logic apply to anything Lenovo does? It hasn't in the past, why start now? But hey, I plugged it in and booted it, just in case. And it bluescreened before even successfully booting up once.

One more trip to the depot, in which I sense nothing will happen, and it's time to take them up on the Florida lemon law.

 

wow..seems as if lenovo doesn't listen at all....why would they reimage your HD if you have formated 15 times and it didn't do any good!!

 

For any of you who are experiencing depot issues or just ThinkPad hardware issues in general I'd suggest calling Lenovo and getting your case escalated to a manager.

Normally it expedites the entire process instead of arguing with a tech over something you know is broken.

 

Very much agreed. Don't deal with tech support if they're not helping, ask straight for the supervisor. You bought a system worth over $1,000 and you shouldn't have to settle with defective anything.

 

Um.... Aren't the 101.55 the approved Lenovo drivers? The 169 drivers are all modded inf drivers from laptop video 2 go aren't they? I may be wrong here but if they are you won't get any support on those. Don't take me wrong, I'm not hinting anything regarding your experience with Lenovo, just commenting on the video drivers.

 

The latest official driver is 156.66 I believe. At least that's what it is on their website for the nVidia Quadro's.

 

Ah, ok. I hadn't actually looked on the site, but I figured it was a safe assumption they weren't up to 169.... they usually lag a bit.

 

Another update from the sagas of the depot.

I got my machine back again today. Much to my amazement, they replaced the planar card (motherboard) - I know they actually did something, because I had to re-register the machine on my uni's network, meaning the MAC address was different. And try as I might, I was unable to reproduce the crash I had before. That's not to say the machine is problem-free, but the show-stopping video issues are gone. Sadly, the processor whine is still there, and I swear the fan is making a slight thumping noise now. And they didn't upgrade the 64-bit VRAM bus, either.

 

Awesome...

Yeah I am really becoming a believer that this issue is hardware and that only SOME motherboards have it, because Lenovo replaced my whole darn laptop, and the new one... well.. it's great! I can't make it crash so far! I'm trying...I really am...lol... I want to make sure it's fixed but I think it is!

 

Hey anybody want to look into an interesting possibility?

Now this is a long shot but... hear me out...

OK I have had 2 T61P's in my hands that had this graphics issue.

Both of them have a little rubbery pad under the keyboard that was off center. Here is a picture showing what I mean:

Ignore the red circle, look where the pink arrow is pointing:



OK see how the purple rubbery pad is the same shape as the metal object it sits on, but it's not centered? It's shifted to the right.

Well, both the "bad" laptops I have seen had this pad shifted just like that. However, when I got a NEW T61P that appears to NOT have this problem, I took off the keyboard and guess what? The pad is not shifted, it is perfectly centered!

So, if anyone is not afraid to take off the keyboard and check this (it's pretty easy, if you have accessed your RAM before, to take the keyboard off only requires unscrewing 1 extra screw on the bottom, the one with the keyboard picture by it, near the middle of the bottom).

I would be very curious to hear from those of you who have this issue, whether your pad is offset or not, and those of you who don't have this problem as well.

Those who had the problem fixed at the depot, you don't count, because the depot guys might have straightened the pad, which means nothing.

The only reason I'm curious about this is because I just wonder if there was some issue with the assembly line that has been since corrected, or something like that, just wondering if there is any correlation between the shifted pad and the graphics issues. If there did happen to be a correlation, it could serve as an easy way to identify what laptops have this issue.

EDIT: Removing the palm rest and / or keyboard does NOT void your warranty. I am 100% sure of that.

 

There's an update: There has been a new bios update addressing this issue. Check out this thread http://forum.notebookreview.com/showthread.php?p=2787378#post2787378

here's the link to the bios update http://www-307.ibm.com/pc/support/si...cid=MIGR-67988

Version 7LETA7WW (2.07-1.08)

* (New) Added the new error code 1830 to indicate that a single DIMM is not installed in the memory slot 0. Solution for the error code: With a single DIMM, install the DIMM into the memory slot 0 (a slot different from having used it).
* (Fix) Unexpected interrupts from the USB controller may occur.
* (Fix) The system with NVIDIA graphics chip may cause hang.