Dell XPS M1730 Owner's Lounge, Take 2

I thought your heating problems was with your X9000 CPU not the GPUs

 

I have a lot of heating issues ¬.¬' From what I've been told in this thread, my CPU is on the high end of normal - it's nothing to be overly concerned about.

My biggest issue seems to be graphics card. When gaming I hit 94 degrees within 5 minutes no problem. CPU doesn't get OVERLY warm, maybe 68 max when stressed. The ORTHOS stress test pushed it up to 93/94, but I would never stress it as much as ORTHOS does.

 

You are totally right BatBoy. I went into some games that I play often and they run great. I wont be benchmarking my laptop anymore since I only did it at first to check if something is defective but given that I got a high score I think everything is running well. I also got great scores in windows experience scores except for memory which was only 5.0 for an 8Gb. Maybe you can shed some light on this since I am totally clueless about it. I also noticed that whenever I right click on my deskptop it takes like 5 seconds for the little window to show up. Any suggestions? Thanks again.

 

My lowest score was also memory. The problem here is that WEI doesn't just check the amount of memory, but also the speed. It's basically expecting us to have 1033MHz DDR3 to get on the higher end of the spectrum. Since our systems can only hande DDR2 667MHz, there's no real way to boost that score. I'm not sure if putting in 800MHz sticks that downclock will work. But to be honest, WEI is a very rough indicator of anything. It was Microsofts attempt at making game/application ratings easier, but it misses out some important aspects, so don't put too much weighting on it.

Just to point out I think they've changed the system in Windows 7 - I get a 6.4 with my RAM. They changed the maximum from 5.9 to 7.9 I believe, so it still seems a bit more accurate in portraying your system is comparison to others.

 

I would say 93/94° is way way too much for a CPU even when running prime, on the contrary 94 for the GPU is pretty high too but nothing to really worry about. My current 8800MGTX (replaced 2-3 weeks ago runs at ~85 for the 1st GPU and 110 for the second one, I would say that this is concerning)

 

I going to quote a Dell tech who has worked with these cards for a while, and replaced four of mine. I told him that I getting around early 90s for GPU during intensive moments. His words, "That's perfectly fine." I said, "Are you sure?" His response, "The threshold for the new cards with the new BIOS is actually 130 degrees; you will know when your cards are overloaded, and it might not even be due to temps." I agreed with the last part, seeing as I'd sometimes get artifacts even at low temps.

In your case, Anxionkt, I'd be worried. But, the reason being that the difference between the two cards is too much! 110 and 94? They should normally be operating within a 1-4 degrees range of each other. Clearly one of the GPUs has issues. Will it blow up? Based on what the tech told me, probably not.
---------------------------------------------

Question for all gamers:
I've got a problem playing 'The Witcher'. When I play it under Windows 7, the game runs smoothly with single gpu. When I run it under vista, it stutters like mad. I realised that under Vista, the enhanced edition actually uses light effects from Dell's lighting system, which is cool. This feature is not available in Win7. So I did a test, and removed the lightfx folder from the game, just to test. And what do you know, it's running smoothly, just like it did in Win7. Why do the lights cause the game to stutter so horribly with very choppy frame-rates? Any help or insight would be appreciated.

 

thank you for giving my issue some concern. Yes, with sli disabled, there is no streaking. I just tested it now to make sure that 186.03 was consistent with 185.85.

and of course, with sli enabled, I get this when I click on any server:



Just a note that it does not happen on my laptop's display and hitting print screen shows no sign of the streaking. It only shows up on my HDTV. But to rule out my tv, it doesn't happen with older drivers (179.xx), and it happens on other TV's.

I'm definitely certain that it has something to do with the "handshake" protocols that go on when using HDMI. I think nvidia has been narfing something that they don't even realize...

 

Does anybody here play CSS and have an hdmi-dvi cable?

get one at mycablemart.com for like 5 bucks, gold plated.

 

Yep - I had the same problem. You'll need to do the following to fix it:

1. Open REGEDIT via the Run Command and look for this key:

HKEY_CLASSES_ROOT\Directory\Background\shellex\NvCplDesktopContext

2. Once located, right click on it and DELETE it.

3. Open up Notepad and paste in the following:

Windows Registry Editor Version 5.00
[HKEY_CLASSES_ROOT\Directory\Background\shell\Nvidia Control Panel]
[HKEY_CLASSES_ROOT\Directory\Background\shell\Nvidia Control Panel\Command]
@="c:\\windows\\system32\\nvcplui.exe"

4. Save the file to the Desktop as: NV_FIX.reg

5. Run CCleaner's registry sweep just to be sure everything is tidy.

6. Open up the .reg file you created and then OK the prompt to merge the entries to your Registry.

7. Reboot and enjoy a nice speedy context menu.

 

Works like a charm. Thanks and rep One thing though. I installed CCleaner and whenever I scan the registry for issues something comes up and I fix them then I scan again and more things come up on and on and on. Any clues why that is? Thansk again BatBoy.

 

WOW!

Thanks! Solved that issue for me too! =)

 

You had 9800 GTX's replaced 4 times???

 

So, last Friday I was able to secure Vista 64bit and MediaDirect 3.5 thru Dell XPS chat without any issue. It arrived Monday, after installing and setting up, everything is performing quite nicely.
Only tested CODWaW so far for gaming. After about 2hrs of MP here are my temps and some video
I am running my X9000 @ 3.2GHZ undervolted to 1.2625v and my 8800m GTX are OC'd to 600, 900 and 1550 using NVIDIA system tools. I still think the 186 version is running a little hot compared to 185s:
I used FRAPS to record which killed my fps, I am getting 65-90fps everything max'd
http://www.youtube.com/watch?v=zFfoJZNjwS8

 

I'm not worried at all since i'm getting it replaced but 110 after 20s of gaming is not normal lol. The drivers crashes/or the whole pc BSODs in a matter of seconds. I guess it's just that there is no contact between the heatsink and one of the GPUs but after 4 cards with the same problem (not overheating) and other things not working after getting replaced by a refurb I'm giving up. (my last MB crashes when the ethernet card is used intensively etc)

On my next Dell PC I'll probably try to avoid having to get parts replaced by dell as long as I can, the tech himself told me that once they start replacing parts like the GPU they come back to replace parts quite often on this computer since refurbished parts are either not repaired at all or poorly repaired. Our only luck would be to get a part from a whole refurbished computer which had a completely different problem

Not 9800GTX, 8800GTX. The 1st one was perfectly fine IMO but it was too hot for dell and they didn't want me to clean the fans (it was just the fans after 1 year of 24/7 use). The other ones had artifacts even at low temps and after replacing the cards and motherboard 3 times they realized the problem came from the wiring of the chassis.

 

Ohh man, something is seriously wrong with a 8800GTX running at 110 degrees..
They shouldn't go higher than 84...

 

Ok I've just done a clean install and want to know if anyones had a run with Nvidia's 186.03?

I was happy on the 185.66 but I'm having a headache reinstalling them with the modded inf 3.291.

So has anyone had any joy with Nvidias latest offering?

Also I'm following Batboys driver inst' order, can I leave the GPU driver for now and keep on going with the rest of the list?

 

Yes, you can skip over the GPU driver for now...

Give the 186.03 a try. I still haven not had a chance to really do a serious 2 hour long gaming run to test thermals - but so far so good. I did an initial test run with 3DMark06 and scored in the 11400 range. This was with no OCing of the GPUs and not disabling any services/startup items.

If you don't like it, you can always head over to 185.81 (what I've been using for some time now until I installed 186.03).

 

Cheers Batboy. I downloaded it in case so I'll post what I find.

 

Batboy thanks for the context menu fix - worked great for me too! Rep+

 

To Axionkt:

Your M1730 is going to be exchanged for a totally new Alienware M17X? What's the configuration of this "new" laptop please?
(I'm french and I've got a lot of problems with my XPS M1730. SLi 8800M GTX cards already exchanged 2 times and the new card is also KO in SLi (not in mono GPU mode)...

Thanks very much for your answer!
Best regards from France.

 

Now , either I`m missing something or fail to understand how an X2 card, basically one board, can have different temps. All XPS1730 users I know have the same temp on both cards... weird. I always assumed since that since both cards are on the same board, and the fans run together , they couldn`t have any differences. Huh. Could I have a printscreen of HWMonitor running on your system?

I`ve seen mine go as high as 86 when they were clogged with dust, and right after cleanup they went to a max of 77-78C in Red Alert 3, the one that gave me the highest temps so far...
Anything above 90C and I`d be in panic...

 

Well if you think about it logically, it all depends on the game/process. For example, if SLI is disabled - GPU2 won't be used. It'll stay at 'idle' temperatures, whereas GPU1 will shoot up. My system almost always has different temperatures - Right now they're 77/71. Although they're on the same board, not both are under the same workload. Generally speaking, if something isn't optimised for SLI, GPU1 will take all the load.

As for your second comment, yeah, although some people are saying my 94 degrees is 'hot but alright', I just don't feel comfortable with that kind of temperature. With Red Alert 3, how high are your settings? And AA on or off?

 

Here you go:

Think of it as a dual core CPU, both cores will have different temps.

EDIT: mmm I've just looked at the pic, if you can't read the temps tell me and I'll upload a bigger one.

To Eambo, did you say you hit 94° with your CPU? The X9000 is listed as 105° max, however, I've never seen CPUs hit such high temps except during Extreme Overclocking, if I were you I would really look into this.

To capitainealbator, in fact I'm getting a refund on my XPSM1730 bill (purchased in feb 2008) + warranty upgrade ~2600€ with taxes, with this I can purchase whatever I want from dell. I've been given the advice from a sales rep to keep the warranty and transfer it to the new PC (3 years for 160€ purchased 1 month ago) what would you guys do? I was thinking of getting the cash back and buy a warranty upgrade whenever I needed it but reading posts on this forum it seems the warranty was a good deal and it would probably be more expensive to buy it later.
Best regards from France too

 

Hey Axion. I use EVEREST to monitor my temps most of the time, since it shows on my display. While running ORTHOS, I got 94C on my CPU on EVEREST. However HWMonitor reported highest temp as 86.Not sure which.

I can't really understand why both my CPU and GPUs would be overheating when I've had all the cooling components replaced. I'm definitely not suffocating the vents - at the minute the system is on a propped up grill - Allowing air through the bottom AND propping up the rear. CPU is running at 65C, 3.0GHz while playing a moderate game. Not overly intensive, but not idle either.

GPU is concerning me more than CPU. Note that the CPU temp was also while I was overclocked to 3.0GHz - dunno if it would make that much of a difference. If we're going by Everst though then yeah, I hit 94 with CPU. Never seen it over 60s usually though - even when playing pretty intensive games. It was only when I ran ORTHOS.

 

I don't have an x9000 myself so I can't tell you for sure, it just seemed really high to me for a CPU. It could just be bad thermal paste application or bad contact between the heatsink and the CPU as I have at the moment with one of my GPU core hitting 110° (replaced a few weeks ago so almost brand new). You should wait for advice from people running an X9000 OCed, it's just that when reading your temps I was quite worried even if I used to spend a lot of time with overclocking and extreme cooling. Orthos stresses the CPU more than any other task but IMO it's not a reason for it to reach these kind of temps in normal condition (room temp, airflow etc), but as I said, people with X9000 will be a much better source of info

 

Here's the temps I got with HWMonitor when I stressed my CPU:



I'm having an issue at the minute actually. I took these two screenshots at the exact same time - HWMonitor and Everest. I dunno which to believe to be honest, they're both showing me two very different temperatures. Anyone got any ideas? And to clear things up, this is the color codes for the everest temps:

Blue = GPUs
Brownish = CPU/CPU Diode 1
Purple = CPU Diode 2


I dunno which to believe =-/ I really like Everest because it shows the temperatures on my LCD display, AND it shows two different temperatures for the graphics diodes. Whereas HWMonitor doesn't display, only shows one temperature - but it seems more accurate in my opinion. Opinions please

 

@ Axionkt: I PM you ...in french!

 

Mine does the same thing.
Right now in HWmonitor my GPUs both show 65C but in rivatuner gpu0 is 59C and gpu1 is 65C both idle. Just wish I could see the temps of my X9000 CPU

 

Well I just got my x9000 yesterday and plugged it in. I also notice that everest usually reports temps of roughly 5C higher than hwmonitor. In any case, with the cpu overclocked to 3.0ghz (anything higher makes my fan spin at max and it's quite loud), after 20 mins of orthos my temps were 75/85C. Not too bad I'd say considering that there's overclocking going on.

On the flip side, it seems my gpu (after putting AS5 on it) is running much cooler according to everest while playing world of warcraft. It'd used to be in the mid upper 60's, and now it's in the mid upper 40's O_O. Well, off to attempt some undervolting.

 

What temps are you getting on the x9000 while playing WoW?

Also very nice temps on the graphics cards. I wish mine ran at 40C =-P

 

What is your CPU voltages at 3.0ghz? My BIOS sets mine at 1.475v

 

My cpu temps were roughly in the 60's while WoWing. As for the temps I posted, that was all on stock voltages. I'm hoping for a 10C drop after undervolting.

Side note: Yes, I went from a 2.1ghz to an overclocked 3.0ghz but in all honesty I can't really see much performance gain in WoW where I figured it was cpu heavy. I did gain about 7 fps in Crysis though. Maybe since the fps fluctuates like crazy in WoW it's hard to notice a smooth steady gain.

EDIT: Btw, is my cpu being undervolted? Look at the voltage for my highest multiplier....it's......kind of very nice and ridiculously low, to the point where I think it's not actually undervolted. But based on the way the temps look, it sure is considering stock temps is 75C/85C. Also, aren't x9000's supposed to have 17x multipliers? Or is that because people are going up to 3.4ghz?

 

Update on my HW Monitor Pro beta testing.

Franck is doing a marvellous job. Tested two builds now, 1st to detect the fans and their RPM, the second is preliminary fan control.

1st went flawlessly. Getting accurate readings on all 3 fans.

2nd went very well. Only one issue and that's with GPU fan 2 (fan 2/3 in the screenshot). Both CPU and Fan 1 (GPU 1) the speeds are setting perfectly and maintaining their speeds. Fan 2/3 (GPU 2) is setting, but lasts only a second. I think this is due to the 2 fan settings overriding each other, but I'll let Franck work that out.

At the moment we have manual settings for 0, 50, 100. Hoping Franck will be able to get at least an extra one in there (75%?) as 100 is VERY fast.

Beta 1 showing the fans (System is GPU 1, Fan 1 is GPU 2)


Beta 2 showing prelim fan control (CPU + GPU 1 setting @ 50%).


Awesome work by Franck.

 

Looks cool. Although I can't imagine anyone leaving fans at 100% since they're LOOOUUUDDD.

 

Wow, looking forward to being able to control the fans speed after all this time. Kudos to you guys for working so hard on this project

 

Hey congrats man, and yeah, running @ 3.2-3.4 turns on the jet engines lol

 

Wow very awesome. Looking forward to it. Now all we need is LCD support and seperate GPU temperatures and HWMonitor is <3 =-D

 

Just download everest and it'll show up on the lcd for all your temps. You can even position them where you want them to display. Although everest temps are like 3 degrees higher than hwmonitor.

EDIT: And for whoever is curious my final voltages are:
6x-12x @ 1.0000V
13x @ 1.0250V
14x @ 1.0750V
15x @ 1.1250V

I tested them about 30 minutes each (except anything under 12 because I figured if 12 can handle the lowest voltage, the even lower multipliers should also).

 

I've been using everest, however yeah the increased temperature readings is beginning to bug me. Also occasionally the driver that displays on the screen will 'reset' itself, and I have to set up the screen all over again, which is beginning to get annoying.

As for your voltages, I'll have to give them a try. I've noticed however I can't seem to find my 15x multiplier? Where about did you set this?

 

Scroll down, I also missed it too. If not, it's because I overclocked to 3.0ghz. I *think* that's how some people even have 17x multipliers.

 

I'm on 3.0GHz, not finding my 15x still =-( There's no scroll - and the dropdowns are only giving me up to 14x. This is on Performance on Demand. However if I put it on no management, it goes up to 15x. Not sure what's going on heh - not a big difference to be honest but would be nice to have. Thanks for your voltages, mine were quite a bit higher - so I'll try it at yours and hopefully drop a few degrees =-)

 

If i go below even 1.300v @ 3.0ghz my laptop locks up or bsods within minutes.

Eambo try setting everything back to defaults in the profile page first then scroll down and see if 15x is there.

 

Result..... regarding extending my warranty,

I got a phone call today from Dell as I have been enquiring about extending the warranty when it runs out in Jan. 2010

I managed to haggle them down to £120 for an extra 2 years.

The first phone call I got from Dell quoted me £150 for 1 year.

Pretty happy with that.

Goes to show how much they quote to the unsuspecting customer.


I hope nothing goes wrong with it during the 2 year period and they send me an M17....:wink:

 

I think I'm gonna bring it back down to 2.8 for now until I get the temperatures playing nice on 2.8. As far as voltage is concerned, I'm running at the voltages that zergs recommended - no issues as of yet.

 

Oh, you can haggle down the warranty price? What kind of things did you say? Or did you simply call someone else?

@Mickbt26
Yea, my super low voltage makes me wonder what's going on, which is why I had to post the screenshot to prove that I'm not going crazy, or hopefully I'm not. Maybe the AS5 application was too much or too little? Probably a ton of factors.

@Eambo
Whoa whoa, I didn't recommend the voltages, I was only posting what worked for me. I'd highly recommend you run your tests too. Or people will start blue screening and saying I photoshopped my pics! =p

 

I am new to this thread however I don't know where else to turn. I recently received a refurbished XPS M1730 as a system exchange from Dell for my older E1705 which Dell failed to fix for me on three separate occasions. The specs are as follows: Intel Core 2 Extreme 2.8 GHz, 8GB of ram, 64-bit Vista Home Premium (Service Pack 1), Nvidia 8700M GT in SLI.

I was thrilled with the fact they had sent me an M1730 when I opened the box but when I turned on the machine I got an all green background behind the dell and windows vista configuration screens. When I finally got into Vista my regular desktop had small pulsing green artifacts and lines overlayed onto my desktop background, icons, and toolbar. I restarted the machine and the dell startup screen which is normally a black background with "Dell" written over it is instead the same bright green as the artifacts only it is a solid background. Upon entering Vista the artifacts remained.

I've done some searching but no one seems to have had this problem with their M1730 from the moment they first started it up. It doesn't help either that this is my first time having Vista on a machine as I am unfamiliar with it. I really hate the idea of having to send it back to Dell, has anyone else experienced this issue?

 

If the screen is weird on the DELL logo at post then it sounds like the graphics cards have had it. I dont think there is much you can do other than sending it back im afraid.

 

Sounds like a failing video card or possibly a LCD issue - call DELL ASAP and have them fix it. They can send someone to you.

BTW, you should be thrilled they upgraded you to those specs from a E1705!

 

Hope they'll upgrade you to the 8800s or the 9800s even!

 

Thanks for the speedy replies. I figured as much since my Nvidia 7900gs failed multiple times in my e1705 so I knew the symptoms. I was just hoping it was some little quick fix that I didn't know about. I am very thrilled they upgraded me, I just hope that I can get a system that doesn't have any display issues for once!