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

Thx - I missed that bit.
But it does mean that I can stick to my bad-HSF theory (for the time being)...

 

yup, you were right. got my mobo replaced with a refurbished A04.

tech guy was pretty good. he didn't even look at possible problems, but went straight to replacing mobo. within 15mins everything as fixed and laptop ran as new... he said it was the experience of having replaced a few XPSs' mobo's before .

now everything is running smoothly again. tech guy said he has never been back with anyone to replace a replaced mobo...let's hope it stays like that.

 

i am another victim to the XPS M1330! Although I'm not sure my problem is exactly the same as havent experience the vertical lines, however up until now the laptop was freezing playing any games (and not exactly taxing games either - WOW!), until today that is, when the screen would change into what I can only describe as a mutlicoloured mess!

After that happened, it continued to do this when just sitting on desktop screen after about 5 mins of laptop being on. so finally i thought I would use the factory image restore option and now windows wont load at all!

the installation went find, but cant get past the progrees bar on windows loading... screen goes completely black with small flickering horizontal white lines in places.

Called xps support and engineer coming tomorrow to replace the motherboard! only had this latop since january 2008

 

Yep, about 6 months. Right on schedule. Personally, If I had a m1330 I would either ask for one with the Intel integrated graphics and a hundred dollars cash. Or a complete refund. There is no way I would settle for a m1330 with another 8400. IMO, the chassis is not roomy enough to keep the 8400m steam engine cool. Its a bad bad design/combo. I hope ATI globbers nvidia over this mess. Nvidia needs to stop pointing fingers and start listing the models affected or the bad PR is going to hugely hurt their market share for years to come.

 

i asked for this, they said they cant change the config

 

I am another victim as well :

http://forum.notebookreview.com/showthread.php?p=3591386#post3591386

Mine is about 4 months old

 

I had my motherboard replaced some weeks ago, and now I had some blinks again and the driver was reset by windows. I guess within next weeks I will get into trouble again.

Note: I'm using the new A11 bios, A04 board...

 

Got my mobo and fan replaced after the previous engineer stuffed up the first replacement.

A04 mobo, A11 bios, 62C idle temp in 24.5C ambient temp, latest official drivers.

 

Try using A12 and report differences to A11 regarding thermal control :

http://ftp.us.dell.com/bios/M1330A12.EXE

Fixes/Enhancements
------------------
1. Added enhancement for thermal control.
2. Update GM965 Graphics VBIOS from 1466 to 1588.
3. Update Intel CPU family name.

 

Will do now. 62C was for the GPU, CPU idles at 32C but it's undervolted and runs at 600MHz when there is no load.

A12 bios:

ambient: 24C, high humidity
CPU idle: 30C (T9500 @600MHz undervolted when at idle)
GPU idle: 61C (Stock clocks, low 3D at idle, Aero theme on, latest official drivers so giving correct temps)
Using HWMonitor which has always given me the highest temps

(updating as idle temps drop)

 

That sounds pretty nice actually, what about under load? Sorry for the lack of activity, been sidetracked with fruit phone.

 

Yup I saw that laptop. It looks really nice but unfortunately I'm in the US. The article below said it should be headed this way but I haven't heard of any updates since.

http://www.engadget.com/2008/01/07/lgs-p300-u-13-3-inch-laptop-headed-for-the-states/

I think I'll hold off on the M1330, at least until Montevina comes out.

 

What's the difference between speedstep (default = on) and your "undervolting"...?

thx,

 

I thought undervolting at the lower end is not really undervolting because it is already set at the lowest possible voltage value on the low multipliers. But it really helps out the upper end.

 

You could do even better:

. take lowest voltage you can set. This depends on your cpu and RMClock Registry hacking doesn't really allow you to go deeper if you read docs.
. Anyway: Let's assume 0,9 V is the lowest voltage.
. Then using Prime95, 2 Threads, [x] SmallFFT find the highest Frequency for this Voltage.
. For me that's 1200MHz in SuperLFM mode.
. And this is what I set for lowest frequency: SuperLFM@1200MHz @0,9V
. CPU power consumption is proportional to frequency, but @1200MHz it is twice as fast finished with it's job compared to @600MHz. That means spending more time in C3/C4 low power states, which results in less power consumption and heat.

 

Low power state is turned off to stop CPU whine and it runs at the lowest voltage possible. If 600mhz is too slow for a job it automatically bumps it up to 1200 anyway so what's the point?

My CPU's lowest temp is 30C now.

 

GPU peaks at 100C but once the fan is on full it runs at 92C running HL2: Lost coast. Will get CPU temps later today for you.

 

Thanks Kreeeee, you're the best. I assume that's pretty decent... or at least better than what you've gotten in the past. Battery life isn't sacrificed, is it? I may update the bios, but i'm sorta holding off on it right now if battery life is affected.

 

All Nvidia G84 and G86s are bad

Means all GeForce 8600M GT/GS und GeForce 8400M GT/GS/G are officially affected on the mobile side. Some desktop GPUs seem to be too ... (but the cooling there's better, so nothing to worry about)

It's stated that HP, Dell and Asus already confirmed mentionable quantities of their mobile computers failing. HP is already offering an extended repair service on their notebooks. The XPS1330 gets explicitly mentioned by heise (Germany, didn't find the article on their English sites).

Both, The Inquirer and heise.de, state that the cash amount for the correction of the problems will heavily exceed the $200 million mentioned by Nvidia.

Let's hope there will be a long term solution coming.

Edit: The Inquirer also mentioned that "Dell is ... replacing them with ATI parts" — Anyone knows about this?

 

well, dell used ATI for their Studio Line

 

Excellent article! Seems that German tech magazines (hardcopy / online) always go deeper into the underlying technology, something a bit harder to find in their English counterparts...

 

It's still far from ideal but a lot better than A10. I get 12 minutes less battery life on my 9 cell.

 

Got my board + gpu replaced recently... casing still remains quite warm, and base is still much hotter than other notebooks I've owned.

So far so good, but not sure how long this will hold up.

 

Let me rephrase: You can do much better!

CPU whining is due to C4 low power state on every notebook that I've seen (MacBook, Thinkpad, Dell, ...)
German:
http://think-wiki.org/index.php?title=Ger%C3%A4uschentwicklung_reduzieren#C4-Fiepen
( google translated english)

Dell thinks it's due to C3:
http://support.dell.com/support/topics/global.aspx/support/dsn/en/document?c=us&cs=19&dl=false&l=en&s=dhs&docid=0A7D5CD2E17F5125E0401E0A55176204&doclang=en

That means you can still use C3, C2, C1 including enhanced states without whining and gain less power consumption:
http://laptoplogic.com/resources/detail.php?id=48&page=4

Disabling the entire C-states is the worst solution of all. After that going >= 1200 MHz SuperLFM or similar things for lowest P-State does make sense.

Some advanced readings if you're interested:
http://mjg59.livejournal.com/88608.html

 

I had just disabled popup to stop whine, I was incorrect in my previous post.

Just done most of your suggestions with zero decrease in temperature, thanks anyway though

 

I think there's more to say about RMClcok than a single setting. RMClock doesn't work perfectly on Santa Rosa and especially Dell XPS M1330.

One observation is that if one doesn't adjust the energy profile processor load settings minimum == maximum, then one gets FID/VID spikes:
http://forum.rightmark.org/topic.cgi?id=6:1591

Another observation is that if one changes any advanced chipset settings in RMClock, it breaks Thermal Management on Dell XPS M1330. I think this applies to [x] Disable C4 as well as to [ ] Enable Popup Mode:
http://forum.rightmark.org/topic.cgi?id=6:1533#5

 

I don't get those spikes

 

Okay, that's cool. I haven't anylzyed in detail, whether there are other conditions and settings, which influence this. However I've seen it more than on half a dozen of machines.

I have a couple of notes regarding this:

. I never used [X] Disable C4 mode nor [ ] enable popup mode. One reason for me is that it draws more power. Second reason is it breaks thermal management on my dell. The latter point isn't an issue, since I'm undervolting anyway and can't reach thermal throttle temperature. However if it breaks thermal management, I don't know what else it breaks. Furthermore I never investigated which of the two modes (if both stop whining for you) results in less power consumption.

. If your CPU runs 99% idle, how much cooler will it run with 99,5% idle? Your measurement method most likely will not reveal that and that's exactly the case with the 600 MHz vs. 1200 MHz SuperLFM mode. Don't expect a lower idle temp from it. But let's assume you're surfing the web, doing some office work at 40% workload @600MHz. It definitely makes a difference running this at 1200 MHz with resulting 20% workload.

. The measurement of temperatures is difficult. Ambient temps vary. Workloads may vary. I'm testing with different workloads and I'm watching temps as well as power consumption. This is not an easy test that is done in 5 min. On the other hand: If you don't notice a difference during 5min of testing, then the differences don't matter to you and you shouldn't be bothered. The biggest gain is due to undervolting - all the remainder is minor stuff. I agree.

 

You seriously think I can do MUCH better than 30C in a 24.5C-25.5C ambient environment? No offense but that's just daft.

Web browsing never uses more than 10% at 600Mhz, averaging 1-2% and when opening multiple tabs it jumps to 2.5GHz for half a second then straight back to 600MHz with zero effect on temperature.

 

My quote "Let me rephrase: You can do much better!" was refering to power consumption regarding this statement of yours:

Given that, my words hold. However you corrected yourself by saying you meant something completely different. By doing that and also mentioning you didn't notice a temperature drop, I stated even more precisely under which conditions you will see a very noticable power consumption (and temperature) drop. That's the section of my previous posting with "99% idle" in it. Please reread this section and try to understand it.

Regarding your 30°C:
You're using Zalmann NC-1000. There's nothing special about that:
http://forum.notebookreview.com/showpost.php?p=3602871&postcount=40

You can do whatever benchmark you want to do, but your benchmarks seem mostly meaningless.

I'd be happy to look at a processor load screenshot (task manager) while you open google maps and scroll the image. You haven't yet recognized Web 2.0 applications I guess.

If you really think you can surf the internet with 200 MHz (this equals to 33% workload on your 600 MHz SuperLFM configuration), I'd like to invite you to perform an experiment:
. open RMClock, set Performance profile to use just 6x multiplier in SuperLFM Mode (= 600MHz)
. open performance profile [x] Use throttling and set to [x] Index 2 : 37,5%
. This configuration corresponds to 284 MHz on my machine (dual core!)

Now surf the web! Or try to post in this forum and view some threads.

We'd be glad if you share with us your insight!

 

No, that temp is without the Zalman. I have not tested idle temps with the device, be wary of your assumptions.

Why do you have an obscure fascination with proving that a processor running at 4.5C-5.5C above ambient in a laptop with poor thermal design could be MUCH better?

 

Hi All, I just came across this and thought it might be interesting to some.

"I am the Dell Community Liaison for laptops. I wanted to weigh in regarding refurbished parts. I would be taken aback, to say the least, if I was given what I thought were used major components to replace defective ones in my brand new laptop.
We do not replace motherboards on m1330’s with refurbished parts. In fact we do not refurbish motherboards at all. The reason our replacement parts go out with a refurbished sticker attached when they are new is due to packaging laws. We do not manufacture motherboards, but in fact purchase them from a number of manufacturers. When we stock our shelves with replacement motherboards that we purchase from the manufacturer, we are required to place a “Refurbished” sticker on it.
I hope this explanation helps ease your mind about your motherboard, and that you are able to enjoy the m1330."

 

Where did you get this statement? Link? thanks.

 

Here's the link if you want it...

http://www.petertan.com/blog/2008/07/02/my-broken-dell-xps-m1330-repaired/

 

bump, any updates on this from dell?

 

Wouldn't be.. If at all, any improvement to the heatsink design would come with the montevina refresh.. I am not too hopeful, though.. I guess at this point dell should just ship out a copper piece to all existing 1330 owners, and legalize the mod..

 

Dell will not provide modified heatsinks to existing M1330 owners. Why would they want to lose money when they can bill motherboard and other replacement costs to nVidia? Further, it would be a big setback to XPS brand, image and sales, if Dell claimed that the existing heatsinks are not adequate.

I personally feel that the existing heatsink is more than adequate with a non-faulty nVidia chip.

 

Hi all, unfortunately i also am yet another victim with this issue with M1330 machine.

great forum here, but unfortunately to the widespread rpoblem is very long

my experience:
Bought M1330 in Dec 07 in Aust, purchase the extended next day warranty for 3 years.
Am business traveller in Middle East, and git initial symptops, whcih screen would change with wird charatcer block(as if the was another layer of windows screen in back, overlap, and hen would freeze and hang, did not get the vortical color lines issue as originally described here) that is becasue the GPU was dying but noit dead, could recover from this by hard reset.

after 1 week, then also received the wonderful vertical color lines, it basically woudl start qith a solid balck screen and these color loines, gradually fade in. Then , the whole screen would fade out to a solid White screen. Buy this tiume when we get this as unfortunatley as we all know the GPU is dead.

Havugn read this thread i had intaled I8Kfan and was monitoring GPU temos and i dont play games (typical bussiness apss etc, and i was getting max about 52 degree C. when it died.

anyways, heres my experince and tips to others epecially buisness travellers:
called Dell support i home country, and they cant help , so they refer you to the closest Dell call centre, which we you get through (on averge 20min wait every call), they claim that your SERVICE TAG belongs to another region, (eg in my case Aust) so they cannot service you.
many many calls to some useless or unco-operative level 1 , call centre staff later, they say you can transfer your swrvice tag to the region you are in and then we can servie you, fine i say, how long does this take ....."oh, about 10 business days!! then u can take yuor unti to an authorised service centre"
(u can imagine the discussion afer this.
anyway found an excellent manager of ME Support centre at call centre and all sorted, but here is my advice, even though i had purchase the best availabel warrantly package online dell site Aust, netx day busienss replacement and 3 year, there was no option SPECIFICALLY mentionign International Warranty!! so even at the authorised service centre staff, theyc an bring up your purchase item list , biull of materials, from your service tag, but as ther is NO specific line item stating International Warranty, they wont touch it.
BE WARNED ANY DELL owners who are business travellers !! and dont live in their counry of purchase

so, in any case eventually got all aprooved and local Service Centre did drop off machien and they did replace the notherboard.

qucik tip for all, the call centre manager ( excellent guy) did ask me to do the followig test to proove that the LCD screen is ok, and the GPU was definitley at fault.
from OFF machine, press and hold he "d" key on the keayboard and momentarily oress the "power button".
thsi imidialty puts the screen into a dispolay lof solide colors and looping, eg white, black, red, green, blue etc etc dont recall the exact color sequence, but importantly this prooves that your LCD screen is OK, and the fault is elsewere, ie the GPU.

got my Mobo replace in les than 24 hours and al ok so far.
interesting ly, teh temps in I8Kfanare still similar ranging betwen 40-52 C.

question:
1) as i was not present to witnes the mobo change, how can we check and determin what version of mobo is installed in the M1330 ?

2) how can we tell what version of Fan assembly is installed ? as other mentioned they also had their fans replaced in this tread elsewere.
also bios is still at A08 as before.

my papaerwork, from servcie centre only says motherbord replacement, but no further specifics on version or part number etc
thanks

 

I8Kfangui is not accurate software.
USE http://www.cpuid.com/hwmonitor.php
also update your bios to A11/A12 version.

 

Exscuse me guys, how can I see the version of my motherboard?

We bougth this laptop 10 days ago; the gpu haven't burned down yet, but it's quite hot where I rest my palms (especially the touchpad).

The strange thing is that my laptop was one of six laptops our department has bougth. I seem the only one to be having this problem .

Thanks.

 

gdistasi, turn over yr Dell, u will see many bar code stickers, one of them will have a code Axx, eg A01 etc, this will be yr version of mobo.

Everyone one else, how can you tell the version of yr mobo , if u know it has been changed ??
eg Dell service centre had changed my mobo,(i was not there) so my sticker of A01 is not relevant now, can we check by any software ? or bios ??

thanks

 

Ugh! I couldn't take the infernal constant-on fan noise with BIOS A12, so I'm back to A09. I'm going on a year with my 1330, and aside from some hyper-bright pixels on the original screen (CMO which was replaced with a Toshiba), I haven't had a lick of trouble, even while gaming. So I figure if my GPU dies, it was going to die anyway. Running the fans full time would've just delayed the failure, and I would've had to listen to fan noise that much longer Oh, yes, I have the 8400 GS.

 

I was just kidding about sending out a new heatsink.. Of course Dell wouldn't do that.. .But saying that the thermal pad is decent enough cooling shows a lack of understanding of basic heat rejection mechanisms in electronic circuits. A thermal pad isn't designed to take away heat at the rate that is required in the setup of a graphics processor.
A thermal pad is used mostly when the undelying chip is very delicate, and could be damaged by application of a force even a little in excess of the design parameter. Basically in processors without a lid, like the older athlons, durons etc... Also, it's thermal conductivity is two orders of magnitude less than a metal interface of similar thickness

Besides, there is nothing to prove that all nvidia chips are faulty. If that were the case, the vostro 1400, inspiron 1420s should also be affected. I haven't seen such cases in any signinficant numbers..

The arguement used by the proponents of "all chips are faulty" brigade, that the "heat cycling" is what damages the chip, is flawed as well. A chip isn't designed to run continuosly at over 80°C.
Bottom Line- Inadequate cooling.

 

The mobo revision sticker is put under the first DIMM.. You'll have to remove the memory module cover, and then the first SO-DIMM.. The sticker underneath that shows the version of the current mobo..

 

So what is the solution?
I updraded bios to A12 last night and downgrade to A06 by now (with an external monitor, of course)
But the issue still presents.
I'm getting mad with this m1330
x(

 

The solution for the display issue? A mobo replacement..

 

Hi Sesshomaru,
checked and confrimed my replacement mobo is again an A01, like my original one A01.

So anyone know what this means ? potentially same issue GPU burnout appearing in similar time frame 5-6mnths?

anyone know if there is any difference with more recent mobo version A0x ?

cheers

 

I'm not in U.S or in supported country of DELL.
So what should i do by now?
How to have mobo replaced?

 

MrNhim,
are you still under warranty ?

note Dell is not in every county, but they have authorised service company agents in some countries and regions.
I assume you need to conatct the nearest auth service agent/region/closest country to you and speak to them.
this again assuming you are still under waranty.
if not under warantyyou can still buy or get the auth servcie guys to replace mobo but they will charge $$

 

Nope.. The GPU dies due to insufficient cooling provided by the stock heatsink assembly. The latest revision is A04(or 05).. They still are shackled to the same cooling assembly, so if you use the laptop like before, it'll die again..

I'd suggest getting in touch with dell intl. support, and see if they are willing to ship out a replacement board to you...