M1330 Display Issue (Vertical Lines on Reboot & Crash & Vertical Lines)

Well the good news is that this falls under std. operating procedures (see my RivaTuner values below for Portal).

Believe it or not, I think it's a tough call. The reason I say this is that my second, "refurbished" mobo has a problem with the firewire-port (see my Avatar... lol).

That's not to say that you can live with a dying GPU, but: what if it's simply a question of improper application of the HSF? Hard to imagine, seeing as we're talking about a thermal pad and not paste, but - what do I know - maybe the HSF is slightly warped, etc.

I would suggest the following:

1. Stop trying kill your GPU (i.e., don't take it out to the patio... )
   
   
2. Contact Dell immediately and have them ship a HSF along with the mobo.
   
   
3. Have the Rep try to analyze the current contact quality of the HSF to the GPU (you can also open up the rear panel now and try to see how large of a gap there is between GPU and HSF and verify that the pad is really nicely "squeezed" in between the two. You can maybe make use of my pics as a comparision.
   
   
4. Try to assess the quality of the replacement mobo (meaning is there a label on the shipping box with the word REFURBISHED in big, bold black letters) and then and only then, make the your decision (or accept the Rep's recommendation).

The only reason I'm suggesting you even bother with all of the above is that 104C is above average (for a stock GPU - see my results below for an OCed GPU...) and since theoretically, all M1330s are made alike (...), I think your HSF has even worse contact with the GPU then the avg M1330 (and I am assuming that your values are based on amb. temps of 24C or less...).

I mean your 105C is probably a ***forum record***. I stated that I once hit 104C myself but in retrospect, it was with my GPU overclocked (see below). Thus it would be more correct to go with 100C as my max recorded on stock speeds. That's the temp I managed to capture on a screenshot while running Half Life 2's Lost Coast tech demo.

---------------------------------------

Here's the max value during 3DMark06 ripped from a RivaTuner log (the first three values are core, shader & memory speeds, followed by the temps of the GPU and the two CPU cores. Ambient temp was 22C for all tests, using the 9-cell batt.).


New A04 mobo/GPU:
03-03-2008 01:37:44 432 864 602 97 52 54

Old A01 Mobo/GPU:
28-02-2008 00:14:06, 432 864 602 97 54 56

Old Mobo/GPU, OCed (haven't tried with the new one yet...):
15-01-2008 00:00:52 486 972 675 97 56 58

Old Mobo/GPU, OCed ( Portal):
15-01-2008 01:22:54, 486 972 675 103 59 63
15-01-2008 01:22:57, 297 594 301 104 59 62
15-01-2008 01:23:00, 297 594 301 94 60 63
...
15-01-2008 01:24:12, 486 972 675 87 59 63


Last but not least, I did a few A/B tests between the 6 & 9-cell batteries and the difference was exactly 2C. Not much of a difference, really, unless of course you're already at 100C+...

 

Of course... and they couldn't find the documentation on the WMDs either...
J/k - I just had to throw that one in...

Just one comment though on the 8400M GSes "Thermal Ceiling": Tom's as well as my own practical experiments show that it is set to 104C - this holding true for both the A04 and A01 mobos (which, as we've showed with your help too, are using anyways the same video BIOS revisions...).

Based on your last post, you're obviously in charge of repairing - no, make that "modding" - the jets in Okinawa*...! In any event, it's clear to me [now] that you have experience with electronic circuitry so I'm "buying it" now...

*fwiw, the closest I've come to Okinawa was through Beat Takeshi's films but my best friend was born there...

 

My m1330 that had CPU overheating issues after the motherboard replacement, was finally repaired today. They changed both motherboard (A04 again) and heatsink (A01).

The technician was here yesterday as well, but he had the wrong heatsink. It was smaller (thinner pipe) and it wasn’t covering all of the chips. I guess it was made for the one with the integrated graphics.

As far as I know he didn’t apply any kind of thermal paste but temperatures look normal again.

 

I found an old discussion of mine in regards to the temp issue. The card...meaning a non-integraded card which does not apply here...is rated up to 150C.

The stepdowns begin at 104C, stepping from high 3d to low 3d. The second stepdown occurs at 105C where it steps down to 2d...this essentially turns 3d rendering off. You can bypass both of these but I will not go into how to do this here, if you want to know, you can PM me. There is one more stepdown that occurs at 110C which is programmed outside of the drivers that drops it to 2d settings regardless of bypass settings. This one, I do not know how to bypass except by reprogramming the BIOS. I would not recommend doing that.

Since this notebook is going to be repaired by the Japanese, I'll have them give me their opinion on it.

You were close on my job. I used to be a maintainer, but I picked up a commission and now fly. I got my computer experience through Iomega a long while back in an emphasis in computer hardware. If you browse around a little, you'll probably find some of my other expertise. I ended up getting my degree in Aero Engineering and now this is just a hobby of mine.

Sadly, I may not be around computers too much more since I'm putting more time into my family life and work. I guess its age just catching up with me. I've already quit WoW and think and XBox is a good DVD player I'll let you know what the guys here think when they finally give me a call to schedule the fix. I just hope they don't have to ship the MB out here. Anyhow, I'm off for the night so take care.

 

Maybe the new heat sink had thermal pads on it, so applying thermal paste wasn't necessary.

 

I suggest not only to discuss the problem here, but also at Dell's community forums! I still can't believe Dell isn't aware of this problem!

Dell Community

 

Done deal

 

The technicians are here, working on my laptop right now .

 

A tech just called me and said he will be coming over later today to replace the motherboard. He verified that the part is refurbished (which means its probably a fixed DOA unit or something along those lines) He didn’t know what reversion the motherboard is however I will find out later today.

I did try messing around/tightening the heatsink just in case it wasn’t making contact however it didn’t seem to help. I did find out however that my CPU didn’t have very much thermal paste if at all. And after i did some stress tests on the CPU and monitored the temp and clock speed. I realized that it was declocking itself also when it was getting up to the high 90's

I really have a cooling problem here... I cleaned off the remaining thermal gunk off the cpu and the heatsink, but i still need to get some new thermal compound for it... Im going to buy some today after I get out of school. Just so it looks like there was thermal compound on it when the tech comes over later today.

I don't know if he’s bringing along another heatsink assembly but i doubt it. Maybe ill ask him to bring a spare if he has one…

 

First, I just want to say thank you for sharing all your experiences and knowledge in this thread.

Needless to say, I joined to say I am a member of the club now, hurrah!

Trying to decide what to do, as I have seen members post about how, even after a mobo replacement, the problem comes back 6 months later.

My husband and I have experience in computers, but really only in the software, would never consider opening the laptop up ourselves.

We're thinking that, if the XPS just needs the MOBO to be replaced every 6 months, and we'd have to buy the extended warranty just to cover a problem Dell should be acknowledging and fixing for everyone... Perhaps we should just buy a different laptop.

So, I've been looking at computers online and think, if we do this, I may go with an HP - I know in terms of software and computer stats what I need to function - and I would purchase an extended warranty from Best Buy for sure (it's just a quick drive from my office, so it would be easy to service)

So my question is... Is anyone else fixing their Dell, selling it, and moving on to another computer? I don't want to "give up" on it, I have never done something like this to a desktop before... but I don't want to keep fighting Dell for fixes either.

Any advice at all is appreciated, I watch this forum like a hawk

(If anyone could tell me what HSF means as well, I would be very grateful.)

 

HSF = Heat- Sink Fan assembly.

 

My vote's on extending the Warranty.

 

Thanks... I figured it was Heat, maybe Fan, but that's as far as I got.

Does installing a new HSF and a new MOBO make a difference in how long the fix lasts?

 

Ok, mother board and display replaced, my laptop works again . Now I have to wait to see if the problem appears again in ~6 months .

I guess Dell will continue replacing MOBOs till they find a "real" solution to the problem .

 

Well, the second motherboard replacement was installed yesterday and it's already failed. After discussing the problem with the tech (same guy), he thinks it could be the HSF. So...do I need a THIRD replacement mobo now or should I just get the HSF replaced? Does the problem somehow permanently damage the mobo? This is really maddening...

 

WOW!!! This is unbelievable! I applied a bunch of high quality (arctic silver) thermal paste to the cpu after cleaning it. I reassembled the HFS assembly with the stock pads for the NB and GPU intact, I did put a bit of the thermal paste to the HFS's CPU pad.

I booted it up and the fan wouldn't turn on! I though i killed the fan! It booted into vista, still no fan... I opened up Intel's Thermal Analysis Tool (TAT) and i turned on the stress test, and the fan finally came on!

At max load my CPU hits 77-78c max now where as before it would go into the high 90s and give me a message that the CPU is overheating! (but only when i was doing a stress test) What a shame the last tech that replaced the motherboard didn't even consider replacing the thermal paste on the CPU!!

Whats even more amazing to me is that my GPU doesn't go over 101C now! It dosent hit 104C like it used to!

Could it be that the increased heat transfer from the CPU helps the heatsink work better overall??? It would be cool if someone could find out!

I did absolutely nothing to the GPU chip and the thermal pad making contact with it!

I'm in the same room with a similar ambient temp as the last few days (maybe a tad cooler) I'm using the same flat desk, same stress/benchmarking software!

This truly made my day!

And to top it off the thermal paste is only Arctic silver ceramique which isn't fully cured! (i have AS 5 which i will use after the tech replaces the motherboard later today or tomorrow)

Tech should be here in about 2 hours, or tomorrow if he doesn't make it

 

Ok maybe i spoke too soon... It (GPU) still hits 104/105c at times... But it did it after i turned off the CPU stress test. (i was running both the GPU and CPU stress test at once)

 

Clearly the HSF was never properly mounted (on any of the chips). Maybe the HSF's left-side screw was a bit to loose and when you re-mounted it, the pad achieved better contact with the GPU! In any event, it certainly wasn't the improved contact with CPU which can only do the opposite - that is, introduce more heat onto the "pipeline"... .

 

Was that a typo?

 

Dont think that was a typo, you could put AS5 on all contact patches on the HSF (GPU, CPU and Northbridge)

 

There's no thermal pad between the CPU and the Heatsink.

 

Hey all, I have the same GPU overheating problem with the m1330 and 8400m gs...
I did find out that is you use the 167.45 32-bit Vista driver from lv2go (with decent performance as well), the gpu idles ten degrees cooler and it doesn't rise much over 97°C. I know this is a crappy fix and not acceptable for many of you (not even me... running official 174.31 drivers here).

Is there anything Dell can/will do to fix this? It's been bothering me a long time and they even sent a new notebook because they couldn't fix my old one... I don't want to be tampering with thermal paste and the like in my not-even-six-months-old laptop, but I really need this notebook running properly (ie I play games on it ).
Hope this helps...

 

Thats interesting that a driver can lower the temp... Do you know how it impacts performance & clock/memory speed

As for my laptop... the tech didn't make it last night. He's coming out today for sure. And yes, I did put thermal paste on the thermal pad that contacts the GPU (probably a bad idea but i thought i would try it)

I can't wait to see what reversion this motherboard is...

 

I just discovered this thread, add me to the list.
It happened to me today; the screen is messed up with purple vertical lines. Rebooted several times and its fine now, I think.

Btw, bios version is A05

In summary, the problem is overheating GPU? Or drivers related? I know I should just read the thread but theres 48 pages :S

 

Yeah, it's pretty much an overcooked GPU due to less than acceptable thermal issues with the M1330 and although as mentioned above, using the 167.45 drivers will allow you to run the M1330 10 C. cooler, once you have the symptoms as you have, the only remedy is a replacement motherboard. I had a false sense of hope on my system when the lines started one day, and then went away. I thought it was only a glitch......then it got worse and eventually it wouldn't boot at all. Wish I had something more positive to contribute. You seem to be joining this club of failed GPU's and boy is it growing. I hope you don't get an ignorant tech support person at Dell when you contact them about this problem.

 

Oh no....
So they have to replace the mobo? or just send a new laptop?

Is it confirmed a new mobo will solve the problem or will it just overheat again.

 

People on here have said that there has been no change in the design on the cooling method on their new motherboards and so it will most likely lead to the same gradual cooking of the GPU until it no longer works.

 

...sure, they'll send you a brand new unit...
...and for your troubles, they'll even throw in a Penryn and a 64GB Solid State Drive!

Yeah, that's the ticket... not.

lol, rofl, lmao, etc.

1. Sorry (really)
2. Join the club
3. and try to smile when they replace your otherwise-fine mobo with a refurbished job...
4. sorry (really)

 

That doesn't help

Refurbished mobo? What gives when its a design fault. There should be a recall.

 

About the 167.45 driver... it's performance is actually very good, and battery life as well. Clock speeds go from 168/100 to 300/200 to 600/400 like they should, don't know about overclocking though.
The only (big) problem is that there are some issues relating to specific games in the 16x.xx driver branches that are fixed in the 174.31... but at the cost of a coocked gpu (although I haven't had this happen to me yet).

As far as I know, a motherboard replacement doesn't help at all, if it doesn't worsen it(!)
Guess we'll just have to keep peskering Dell about this issue untill they decide it's worth looking at...

 

If I am not mistaken, it’s been discussed in other relevant threads that drivers REPORT different temperatures. This doesn’t mean that the temps are actually different. In this case I prefer seeing higher values.

 

In conjunction with psychrometric's comments above, I find it hard to digest because only a handful of SW changes could make a GPU run cooler:

1. An immense increase in efficiency at the API level (you'd need a firmware update to get anything out at the microcode level...).
2. Making better use of the GPU's sub-systems (i.e. shader units, etc)
3. Passing certain rendering functions [back] to the CPU.
4. Making an "adjustment" to the reported clock and/or temp speeds.

While I could maybe buy the first two arguments, I would only do so if the newer drivers were more efficient (cooler) than the older but the claims are the 16x run cooler than the 17x which btw, run just as hot as the original Dell-supplied drivers (15x?)

The third is also an option, but hardly likely since it probably involves the DirectX API too, and that hasn't changed in this scenario.

So if anything, I'm more inclined to go with the 4th option, in which case nVidia's reporting the wrong clock speeds or the wrong GPU temp, as psychrometric's post suggests.

My take on this is if you're a gamer*, go with drivers that provide the best performance w/o any side-effects (not waking from sleep mode and what not) and pray that nothing "cooks".

If you're not a gamer, install RivaTuner and downclock the GPU and video memory by about 10-15%, for example: 360 core, 540 memory. I doubt this will alter the speed of VISTA's Aero or the effects of HW-supported motion compensation (of your favorite SW DVD player).

*Addendum: If your a gamer with big ones, mod the existing HSF assembly (as documented elsewhere) and be sure to buy some Hubba-Bubba-purple thermal-pads beforehand and store them in the event you need to present your 1330 to the local authorities (read: your local Rep-guy).

 

Hi, i write from Sweden and i just bought this XPS 1330, i am not a gamer and will only surf Internet and do some easy stuff fore the computer ( music, video ). I am wery worried what i can get : ). I have the HP HDX 9050 and for me its a great machine, and have never been a troubel for me, only the touchbottoms have failed. I have own two Dells before 1710 and a celeron ( Dell ) notebook and have never have any problems before with Dell. In Swedish forum i have not read so many post about overheating yet, and there are lot of people who have buy this XPS ( cooking? ) machine

Stefan

 

The Dell guy just got through replacing my motherboard on my M1330 with a brand new one. I hope this fixes the problem for me. (ie. Purple frozen screen then crash and lines on re-boot).

 

I have an M1330 which died with the vertical lines issue 10 days ago.

I use a Zalman cooling pad underneath it 95% of the time. I play games on it rarely (maybe once a month for a few hours).

My machine dates from November 2007 and has the nvidia gfx.

Dell techs *rapidly* diagnosed a motherboard fault on the phone. But after four scheduled visits and three replacement motherboards, the laptop is still dead.

 

They are out of parts because my M1330 is now on its fourth motherboard and it still doesn't work!

 

I asked for this and it was refused point blank.

When I managed to escalate it -- don't ask how, long story -- I was told that Dell's auditors forbade them from offering existing customers warranty extensions. Not sure I believe it, might be possible given the New York state legal action against Dell.

 

Fourth motherboard!?.....and you going for a 5? cant you get the money back and some compesation?...i have give up and take tham to the coart !

 

Well i got my 3rd motherboard now... (SO HAPPY!...) NOT...

The tech that came out this time did a more thorough job it seems... He was more careful with the laptop and the parts. He did however forget to take out my express card cover out of the old mobo :-( but i have my remote in there for the meanwhile until the tech brings the old one back or they send me a new one...

I ordered a zalman cooling pad on newegg.com, I should be getting it on Monday hopefully it will help

Since my warranty expires in September, i fell for Dells extended warranty since it only set me back like $350+/- for 3 years... and they give me $75 dell credit so i can buy one of 9 cell batteries, or an auto/air power adapter.

I did this because I don't trust the design of these motherboards and their cooling solution. And if any of you read my earlier post with the link to dells online store, the refurbished motherboard for the m1330 w/ 8400gs graphics costs over $500.

So in my mind that warranty is more than worth it. I decided to keep this sad excuse of a laptop because i like the overall design and when its not having any video/cooling related issues its still a great laptop to me!

And if i wanted to sell this laptop, obviously i wouldn't get very much for it. Therefore i wouldn't have much for a replacement.

Even though this laptop is somewhat unreliable at the moment, its unreliability doesn't affect me too much since this isn't my only computer or laptop for that matter. I have 3 older laptops that i can use as i backup if i really wanted! And then i have my desktops.

Sadly I'm tempted to buy another laptop :-( i feel guilty for being wasteful

LOL

But yea this is whats tempting me http://www.newegg.com/Product/Product.aspx?Item=N82E16834220266

Its so cheap and small... perfect for carrying around in my bookbag to school for note taking and such... And then there is the novelty/coolness gadget factor

 

If anyone is curious... the replacement board i got was a A01

 

Does that mean it's the original version? How many versions are there, and why don't you just change to the integrated graphics option if you don't game. Better battery life

 

Three replacements and still nada...

what did the rep say or suggest the cause might be: I mean, I'd hate to believe you ended up with three faulty refurbs...

As for the initial failure, I know how you feel. I played a few games on mine, just to test out the 1330's capabilities. But... I was doing nothing more than surfing on it in a room that had 22C when it died... .

 

Well, before you get too bent out of shape, my "A04" is an A01 with an A04 sticker stuck to the original... . Secondly, I went over my pics of both mobos and couldn't find anything larger than a resistor that was different between the two (i.e., no new GPU or audio chip, or anything that drastic).

The only "rumor" I read was that the A04 was supposed to (somehow) improve the noise issue on the audio path (extra shielding).

And who knows... all they have to do is forget to pop the A04 sticker over the A01 label and you have a mis-labled A04...

 

Guess so... But oh well... they can just keep on replacing my motherboards for 3 more years (my extended warranty) I'm gonna squeeze every last penny out of that warranty!!! Lets see what dell thinks about that!
At least i wont be paying $580 per motherboard with this warranty

As for the integrated graphics... LOL well, i paid like $1,800 for this laptop and i want what what a paid for I want a high maintenance laptop with 8400gs graphics (The Benz of computers, since most new high end Benz's are quite unreliable also from what iv read. Mainly because of the new unproven technology)

OK well i don't want the high maintenance part, but i do want 8400gs graphics... I mean i chose to have them after all right? I got the 8400 because i did plan on playing a game or two occasionally on it... Such as Grand Theft Auto SA... But i haven't had the time for it yet, and from the cooling problem i don't know if i will be playing it really.

Besides that, I though it would be good to get the best configuration I could afford so the laptop wont become obsolete as fast as it would with basic integrated graphics. (Then again I heard that Intels X3100 gpu is really revolutionary for being integrated)

Yea your probably right... I don't care what reversion it is anymore... As long as it works. They can keep sending us replacement motherboards until they figure out a long term fix...

 

You know what... i figured it out! I think these GPUs are dieing from boredom/lack of real graphics processing needs! :laugh:
I say that because it seems almost every one had theirs die while doing some light menial task that doesn't require anything more than a basic 2d graphics card... My first one died while i was doing nothing more than watching a family guy episode! The second time it started acting up, I was chatting on Facebook.

 

System fixed....mobo replaced....Bios A08 upgraded to A10 on new mobo...like new again.

 

Hello everyone,

I have researched the problem a bit and I think I have found some stuff which could help.

The problem you are having is often caused by overheating of, usually; your GPU (Graphics Card), which in turn damages your motherboard. There are several reasons for this and the same issues have been seen in other high powered compact laptops.

The problem often happens when someone on a small, high powered laptop is running Vista, has “Vista Extras”, or other graphics intensive programs running, such as "DremaScene" which causes your graphics card to run HOT, which then overheats your GPU and causes the system to crash and/or destroys the motherboard. So, first, please make sure such extras are turned off.

Overheating is also caused by laying your laptop on your bed. The covers restrict the airflow through the fan which sucks in air at the bottom of your laptop. Having your laptop on a hard surface will help.

Keeping your laptop angled will help as well. The "chimney effect" will reduce your laptop’s temp by at least 2%.

If you are using a program/game which is graphics intensive, turn on the AC. A hot room can cause your laptop to overheat.

Lastly, you can download a system temperature monitor off the web. By using these small utilities you can keep an eye on the temperature of your laptops components.

I am typing this email on a 1330 right now and have tested each of the above recommendations with a temp monitor. Each one of the above suggestions makes a BIG difference in the core temp (especially the GPU temp). I was able to make this laptop crash to the vertical colored lines by placing the laptop on my bed with “dreamscene” turned on and the Air Conditioner off. It took less than 15 mins to overheat and crash. After taking the above steps the core temp dropped from almost 100C to an average of 55C. Even the GPU is down under 65C and I have not had a crash or vertical line screen since or before I conducted this test.

If your laptop no longer boots, and especially if your laptop only boots to a screen with many vertical colored lines (even after letting it cool for an hour), it is more than likely that your motherboard is fried. Replace the motherboard and then use the steps above to stop it from happening again.

I have seen the vertical colored line issue on many small, high powered laptops which run Vista (which is graphics intensive and causes computers to run at higher temperatures than past operating systems.) If you have had the above issue, make sure to do what has been recommended above. If you simply replace your motherboard but continue to utilize your computer as before, without taking the above precautions, your motherboard will be destined to fry again. Small laptops make it difficult for air to flow through it. Laptops with high powered components such as the 1330 make things even worse.

I see the vertical colored line crash issue become rampant during summer months, as temperatures rise and put an already overheated laptop over the edge.

I hope this helps and good luck.

Edit: I can not stress enough... Make sure any "Vista Extras" such as "DreamScene" are turned OFF. They may look nice, but they, on average, increased the temp of this test 1330 by over 25%! And that was just while sitting there looking at the background with no programs running. Turning them off INSTANTLY caused the core temp to drop significantly.

 

My laptop died today, 2 days after the crazy lines appeared
My GPU temp never went near 100C. Max I saw was 80C when gaming (which I rarely do anyway) and normally hovers around 40-56C during normal use. Not sure how accurate i8kfan is though.

 

Mine died when it was running XP.

Anything that stresses out the GPU will lead to premature failure.

Its been already shown that the HSF design is leading to inefficient heat dissipation leading to possible GPU failure.

Very early on the thread, I have included pictures and hypothesized that the HSF is suspect.

I do not believe it is software or even M/B revision related.

I figure that Dell would have clued in by now and released a revised HSF to dissipate the heat better. I wonder what their product teams are doing.

 

probably a m1330 refresh/rebuild/new product coming out in the next 6 months

and by the time your warranty runs out (get a 4-yr warranty), dell would have replaced so many mobos that hopefully they'll upgrade u to the new line, though its extremely inconvenient....