G51j Blue screen of death problems

@5150Joker: http://forum.notebookreview.com/showthread.php?p=5575275#post5575275

it doesn't alternate between throttling CPU and GPU (if that's what you're looking for). the CPU remains throttled while furmark is rendering.

i'll check with those settings in the next few mins.

ran prime95 with those for 5 mins, then started up furmark. CPU goes down, GPU stays at 500/800

 

is the 96C idle temperature after you run furmark?

 

those are a second or two after quitting furmark (prime95 still running). idle is around 62-65c for the GPU and 57-60 CPU.
if anyone with the bsod's wants to check, does THRM register as 0c? (this is under xp, 7, and ubuntu)

 

Hi Thalanix,

Actually what I need is for both furmark and prime95 to be run at the same time along with GPU-Z. While prime95 is crunching numbers, have furmark set to full screen and watch the GPU clocks and the graph. To give you some background of what I'm talking about, take a look at this thread: http://forum.notebookreview.com/showthread.php?t=439546

 

they were. i let prime95 bake the CPU a bit while fetching some popcorn.

continuing @ the thread you posted.

 

lol i felt so bad when i heard they released the i7 version of the G51 like a month after purchasing it. this thread has pooped on other ppl's parties, while it has raised my self-esteem..

soz guys, i feel ur pain, the BSOD and stuff. the days when i tried OC'ing my laptop to beat Joker's score, i got like at least 7 BSOD in a row b4 i said F this, im done.

i really think its the i7 problem? btw, are the 3 ppl who aren't experiencing the problem have a different "upgraded" CPU (820/920)?

and to joker, i think its confirmed that the 0.95volted GPU in G series dont throttle, to find answers, you're gonna after to urge real M15x gen. 2 users (or dell notebooks since 1645 had this problem as well) instead of different notebook manufacturer owners.

 

What I find strange is that the Clevo's are using 120W PSU's, but they have no throttling issues whatsoever despite having the 920XM and the GTX 280M.

 

Well asus also using 120w on g51, and according to killwatt im still going over on the furmark+ibt, i guess there is + over the psu.

 

The max rating written on a PSU isn't the "true" max rating. Good PSU manufacturers always give a little bit more leeway for the PSU. For example, my 120W Li Shin has a real max rating of 150-160W. The M860TU, every M570xU, and the new W860CU and W870CU all use the same 120W PSU and not a single person has reported power issues at factory settings. I don't know about the Delta branded PSU's that Dell is using, but in my opinion I think they are incapable of maintaining a clean current at a high power draw. This is most likely because they don't have enough chokes or don't have good quality chokes. This also explains why they are smaller compared to other high output PSU's.

 

I no-longer experiance BSOD in any of the games that I play (COD4, BF2, DoD:S). At first when I had this issue I only went through 4 BSOD's overall in BF2 and CoD4. I don't know whether it was the video card driver change or because I'm now running every game in Extreme Turbo mode, but it has stopped the problem in its tracks.

 

The G51's 120w power supply is made by Delta. So I'm assuming it is not a power supply issue with the AW.

 

Each individual PSU is different. I wasn't implying that Delta makes mediocre PSU's, rather I am saying that the 150W PSU that the M15x is using isn't enough when pushing the CPU and GPU to it's limit. The 120W PSU's by Delta may be of good quality. The same goes for my older PSU's. I've worked with 90W Li-Shins and they couldn't supply enough power to a notebook I was working on. But a 90W PSU from EPS can churn out clean power all day long.

 

we'll all see if the problem lies in the PSU come monday...hope its not someth. that will screw the dozens of alienware M15x users (as in it wont require physically sending it back for repair/replacement; maybe a BIOS fix, PSU replacement or somth. along those lines)

 

The issue is still out in the open. Nobody knows if it is a motherboard issue, a PSU issue, or if 5150Joker just has a lemon. Until other M15x users crosscheck with him using the PSU test guidelines or when 5150Joker gets his 210W PSU on Monday, anyone's guess is just as good as mine at the moment.

 

Hi everyone,

After looking at last 20+ page, I have a general question here:
Assuming once they identified the problem isn't hardware problem; Should I follow the steps provided (in here) when I buy this laptop in near future OR the problem should already be resolved by them when they got the fix ?

(Meaning if they happen identify it's a BIOS problem and release a BIOS fix before I make a purchase, should I do the BIOS update or they will definitely have this problem fixed in the coming production and so I don't need to do the BIOS update anymore?)

Sorry if my wording is too long-winded.

 

Yes I do, they are already working on it and I did ask them to test Borderlands and COD, I think the issue could be the RAM timing in the BIOS.

 

**IF** it's a software/BIOS issue, and it's fixed with a BIOS release, it's not a big issue to flash your laptop to the latest BIOS if it doesn't come with it already. You'll be able to tell from the BIOS version number whether it's the latest or not. Guides everywhere.


I would suspect wattage ratings of PSUs are as dodgy as they can be on desktops. Yumcha PSUs often claim "600W" because they have huge 3.3V and 5V rails which are barely used for anything, but are woeful when it comes to the 12V rail/s which are used to power the most hungry components, CPU & GPU, and they can fail to output 250W on 12V rail.

I wonder if the wattage of PSUs is input or output (i.e. does it take into account efficiency of conversion or not). I.e. "100W 19V 5.xxA" but converted at 75% means 75W available at computer end of plug as opposed to 100W (which at 75% would mean ~133W going into the adapter)?

How hot do these adapters get when under high power draw (e.g. gaming)? I wonder if some kind of powerbrick "thermal throttle" is cutting in when they get really hot, like 80C+

The maths of 45W CPU + 65W GPU + screen + rest of laptop + externals < 120W powerbrick doesn't seem to add up.....

 

The rated wattage is for output.

 

I'm not experiencing the problem and my specs are identical to most others here: i7-720QM, 4GB DDR3-1067, GTX 260M, etc.

I wonder how many people actually do have this problem. There seems to be about 3-5 people here who have reported it (one no longer experiences it) and 3 people who can't make it happen on their deck. Awful small sampling sizes to determine if its a 'real' problem or not, either way.

I wish more would report here, yes or no on their decks.

 

I'm one of those not experiencing the issue, however I haven't had the time lately to extensively game due to work, but after I nail off my current project I be more than happy to do extensive testing (in 1-2 weeks time) with the multitude of games that I've been neglecting over the past month

 

I have a Asus G60J-JX112V, i7-720QM, 4GB DDR3-1067, GTX 260M .

And I only had bsod on Dragon age origins after 30 m of game play , I also have been playing cod 6 but i never played it more than 30 m no BSOd here

 

Yeah sure they are. I bet they're playing Borderlands for 30-40 minutes to see if it'll crash just because the you asked them too...

But hey - maybe they are. My experience with RMAing laptops has certainly been vastly different from yours if you are so optimistic about their quality of service. Me, I've never had a good RMA experience, on either personal machines or business ones unless they were completely dead.

As far as the BSODs. It's clearly a driver/BIOS issue. Stop 124 errors are almost always caused by BIOS/mobo or video card issues and are triggered by HAL. Borderlands runs just fine on my laptop with the DOX drivers. Stock drivers it'd blue screen every 30-40 minutes. Of course with the DOX drivers it flickers now and then. The people who downgraded to XP and aren't having the issues anymore are probably just not triggering in XP HAL to BSOD - hell I don't even think there is a stop124 for XP...

Beta drivers from NVIDIA are fine too. This is all of course with the 204 bios update.

It's still not perfect, as I do get crashes once in a while, like every other day... But it's far from being a show stopper at this point.

If it was ram timings it would affect all sorts of things. Not just a handful of specific games. Lets see:
Dragons Age - No problems.
Dirt 2 - No problems.
Prototype - No problems.
Red Faction 2 - No problems.
E:Total War - No problems.
Warhammer - No problems.
Left 4 Dead 2 - No problems.
Encryption/Decryption/Decompression etc - No problems.

All the above ram intensive tasks run just fine. If you mess with memory timings on any PC you know that with incorrect or just overly aggressive timings you'll be unstable in everything you do. Not just COD4 and Borderlands. Nor would downgrading to XP fix it.

In short it's a driver/bios issue.

 

Hey thalanix and/or anyone else that might have the equipment. I was wondering if you possibly had a more powerful AC adapter to test the 51j with. Reading about the Dell issues and how similar they are (throttling of the CPU in this case, GPU in theirs) made me wonder if the cure could be the same.

I still lean to thinking that the cooling on the 51j just isn't designed for pushing both cpu and gpu to their stock limits (let alone overclocking), but I'd really be interested in seeing how the deck does with a more powerful AC adapter. I can't imagine the heat levels of both CPU and GPU non-throttled on this deck, but who knows... Maybe the one fan and mediocre heat pipe layout on the 51j is better than I believe. Clearly, Asus knew there was either a cooling or power issue with the design or they wouldn't have stock underclocked the GPU. But even stock underclocked, both can't be pushed to their limits at the same time. So, is power draw or cooling the culprit in this case?

Edit: I guess this is not a part of the BSOD issue, so maybe its own thread... But then again, who knows yet?

 

Actually I originally thought that the BSODs were related to heat or OCing. But even underclocked and sitting on a laptop cooler WITH the bottom of the laptop removed so that the air was blowing right at the cooler intake, it'd still crash in Borderlands - even with the GPU temperature below 80C.

The only thing that has helped me was updating the BIOS and switching drivers for the video card.

 

Yeah, thats puzzling. Not only that heat wasn't the issue, but clearly power-draw either as just updating drivers wouldn't make much of a difference. That is, of course, assuming that newer drivers aren't underclocking or throttling with a different profile. But I haven't seen anything like that in my testing either. Both the stock and beta Nvidia drivers seem to have the same power profile... ie: they look the same under idle and load in GPU-Z.

So, again, maybe the throttling isn't the BSOD issue... but its not very clear yet based on such small samples. However, throttling in itself is an issue if its affecting performance under normal loads (games, etc). My freezing under MW1 might be related.

 

Hmm, would the G71's 150w AC adapter work for the G51J?

 

If you spliced the DC plug from the M15x PSU to the G71 PSU, then yes.

 

You might want to read your post and the previous one a few more times soviet.

 

Whoops, reading too fast again. I need to stop opening so many Dell/AW tabs.

But still, it will work, haha.

 

No, it still wouldn't, lol. Unless you *really* jammed that m15x plug in there hard and got lucky!

But I get what you were going for.

 

As long as the adapters are the same, I don't see why not. Do you have a G71 brick?

 

Lol, yeah. The M15x plug is too fat for the G51 DC jack. Besides, we don't want any aliens probing our notebooks, haha.

 

it might not be a power issue. the gpu is underclocked 10% and undervolted 2%, which would probably be 15 or so less than what it's actually rated for. on notebookcheck, both the 260 and 280 are 75w. if draw scales linearly with cores, then it would be 85% of 87.5% of 75w. that would bring it a little under 56w. add in the screen and it should be extremely close to 120w (probably a little over).

running p95 and furmark still turns it into a fireball when throttled, so there's a better chance it's bad cooling.

no watt meter or extra psu to check. anything outputting the same voltage will work.

 

I hadn't worked out the wattage, so thanks. Yeah, thats pretty damn close or over spec and I have no idea who or how good the psu maker is. Regardless, heat is already too high here so I imagine you're right. Still... I'd like to see it done for verification.

 

if there's a .85v in the vbios, it might be stable. too bad intel integrated p-states in the CPU, otherwise that could be undervolted as well.

perhaps if asus added voltage control to the bios?

 

what is the voltage on the GPU? (must've missed it somewhere in the thread) the G51VXs are at .95v

 

And fan RPM and temperature control. That'd go a long way to helping avoid some of this way-too-high heat. I'd love to spin up my fan much earlier on the heat scale than Asus has chosen.

 

definitely agree with you there...if I could get my fans on max at ~70C I'd never break 75c under full load.

 

70c might be a little early, 75 seems more like it. would probably just slow the heating rather than dropping temps.

atm, the card is the same as the g51vx, which is at 0.95. lower clocks are usually at lower voltages from what i've seen, and some lucky people should be able to lower their 3d clock voltage while holding stable.

 

Nope, but I can grab my friend's G71's adaptor sometime. I'll post back the results whenever I get the chance (won't be for the next week though, too much work)

 

That'd be great, Hydeo... whenever you can (sooner, better ). Just make sure it indeed runs at the same voltage: 19v on my brick.

 

Hi,

I read all previos 30 pages and have the same BSOD issue
So we are more than 3.

I have upgraded RAM to DDR3 1333Mhz and Intel SSD HDD

__________________
Asus G51J

 

Yup...I'm waiting for a fix.

 

New drivers came out TODAY. I have not had a chance to try them yet being at work and all...but I suggest anyone that has issues with BSODS grabs them right now...

http://www.nvidia.com/object/notebook_winvista_win7_x64_195.62_whql.html

 

Hey same here: I have 1333 ram and Intel 160 GB SSD =] , and of course also BSODs

 

I've just downloaded them, and after class, I will test it extensively...AKA Call of Duty for hours on end.

 

I bought this laptop, and I am suffering from the exact same problem as the gentlemens here. From the event log:

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000124 (0x0000000000000004, 0xfffffa800414e8d8, 0x0000000000000000, 0x0000000000000000). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 120409-18735-01.

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Played Oblivion for about 4 hours and got this problem. However I played TF2 for the same amount of time and didn't have this problem yesterday.

 

Have you updated your BIOS to 204 and are you using the latest NVIDIA driver?

 

Had a try with the new drivers today... still no luck (for me)... Other people have had better luck?? I have since returned my g51j but I had to pay a re-stock charge because ASUS wouldnt accept there was anything wrong with it... i only play MW2, and it would BSOD every 20 minutes or so... oh well...

 

i had bsod's with 195.55 and 195.62. every 10 minutes with the latter.