Updates from Asus

Hi, thought it would be good to post these in there own thread so they dont get lost in the black hole that is this forum.

These were posted today -

 

I'll tell you what shortens the life of the components. Crappy thermal paste with big air pockets in it shortens the life of the components. Furmark will let you know within three or four minutes whether you have incorrectly applied thermal paste. If you do, you fix it and probably lengthen the life of the GPU a great deal. If you have bad paste and don't know it, you'll run hot for a year or two and it will burn out two days after your warranty expires. It's a question of running hot once to see if your system is OK versus running hot... not as hot, but hot enough every time you play a game. I think running hot every time you fire up Crysis or Bad Company 2 is just as bad or worse than running Furmark a few times to test.

If you do have bad paste, once you replace it you can run Furmark all you want without ever getting as hot as you did playing games prior to repaste.

 

Thank you for the update antisniperspy. My semi-rant was not directed at you. It was directed at laptop companies that don't like customers finding and pointing out flaws with benchmarking tools. The comments boil down to "Its not us. Its them. They certified all this hardware and their drivers won't run on the hardware they certified" followed by a nice helping of "You guys shouldn't be running benchmarks anyway."

 

"excuse me while i avoid this recall"

 

I know all is good , Just posted them to keep everyone updated that they are working to fix it. As for the temps I really dont have a comment but I have been looking for the spec sheet and it will clear up a lot of things for some people.

 

This confirms what I was thinking all along.

Didnt make sense to me that there was a hardware problem.

Everything runs flawless with stock drivers and 209 stock bios.

When they get this fixed,our laptops will be an amazing work of art lol

 

Glad you did! I would never have seen these remarks otherwise. It makes sense to me. I'm still on stock Bios and vBios and drivers and have no problems to report. Gamers are notoriously impatient, so Asus will have to take that into account.

 

So looks like a video BIOS is coming? Hopefully one that does down clock and modifies the vCore accordingly.

 

Where was this originally posted?

 

I definitely agree with you Jody. I'm very pleased that an update is in the "near future", but I'm sorry- I don't buy into the idea that the machines are meant to run at 92C +, when repasting seems to almost universally lower temps by at least 5 degrees.

But if they fix the GSODs soon I'll be satisfied- repasting is pretty simple even with the disassembly, compared to trying to work out video Bios issues ourselves.

Edit: The first was posted in the "Have you ever GSOD on your..." poll/thread, in direct response to one of HellCry's comments that the flaws are due to hardware issues. The second was in "G73: An Ongoing Battle".

 

Think about who is saying Furmark is dangerous, a retailer. It is dangerous IF and only IF the product is fauty in one way or another.

If the thermal paste isn't proper, then yeah, it will show it. If the system cannot take stress, yeah, it will show it.

It is not like people are running Furmark for fun. They are doing it to determine if the product they paid a ton of money for works. PERIOD!

To say it is dangerous and reduces the life of the component, then the product wasn't properly engineered in the first place. We all bought our systems to run SOFTWARE on them. We want to make sure they are up to the task before we end up hours and hours into a game to find it crashes.

Some games can, and do, push things very close to where Furmark does. Are you saying that those games are just invalid because they max out the hardware? If so, then why are we paying a ton of money for a top-end product if it cannot perform???

I have dual GTX480s in my desktop machine. They generate a TON of heat, and I have had to engineer the cooling solution in that system so that it can dissipate that heat otherwise the cards will shut down. Now, to do that means I hae extra slot-fans on the heatsink side of the front of the card (nice for them to engineer that solution as it works great and reduces temps by 5-6c), but if i decide to triple-sli the setup, there is no way I can have a fan the way I have it, which means temps will probably get out of control. But, here is the rub, as I have a warranty on these cards, so if I decide to get another one and they burn themselves out, the manufacturer is required to provide me with a replacement under warranty.

Furmark destroying components is a joke, as it will only show a weakness as it already exists, not creating new ones. Everything has a lifespan, and sure maxing parts out will reduce thier lifespan over not stressing them, but, again, these are top-end machines, not very different from high-end sports cars, and as such it is not only designed to be run maxed out, it is fully expected, and if not, then it is not engineered properly.

The largest majority here are worried about a stable product that will operate as expected. We are not the ones that are expecting to overclock or push things beyond where the are supposed to be at. But we do expect the system to be able to run to the edge of it's specifications and do it reliably.

 

You can click the button to the right of his name in each quote to get to the post.

 

@TheBeast

I totally agree. The heatsink is still not properly applied. Temperatures really don't have to go a lot above 90C.

Also, if he doesn't like Furmark, run a proper game and see what happens. Metro 2033 comes within 4 degrees of Furmark maximum burning stress test.

89C with repasted 5870m in Furmark maximum stress. 85 with Metro 2033.

 

So then what about us that never got a GSOD and can run any version of the driver with no issues what-so-ever. If this was a timing issue, EVERYONE would see these problems. Whats the source of this info?

 

This really is the only source and it has been said that it happens at different times to everyone on different games. Maybe the gpu's are made of differenet parts and come from different manufacturers.


Profile

 

not every application GSOD's.

Curious,have you taken the most popular GSOD programs and tested them?

 

I hit this thing with everything I got. With the 10.7a's I ran furmark all night and when I woke, it was just fine.

I can play BC2 for hours on end. Some people said the old warcraft is a sure way of GSOding. Nope that ran just fine.

I use max, maya and the adobe suite as well when im on the go. Ive never ever gotten a GSOD. The only thing I did was remove the audigy software cause it was causing steam to crash.

Thats why I think the proof is in the pudding.

 

Only proof I see is on a laptop with functional hardware.

Stock driver and 209 are stable on all configs.

MSI vbios is also stable on all configs.

That points to a problem with the stock vbios and ATI certfied drivers that are not from asus otherwise those other bios's would not be reliable.

Now you may have a GPU with diffrent parts that somehow bypasses the vbios problem,but regardless with an update all laptops should work imho.

 

No, it just means your GPU was assembled with parts that are operating within specification. Unfortunately, not all of them are. Feel blessed and enjoy your non-GSOD machine.

 

Ever thought maybe those that GSOD are the working gpu's that needs their timings changed. And yours is the defective gpu thats why your timings are already different to begin with? If people are able to report the msi bios being a fix, it already shows that the hardware works. It just needed some fine tuning. Just because yours and a few others don't have this problem, doesn't mean there isn't a problem with your system.

 

That doesnt make any sense. If im not having any issues with it, then there is no issue.

All im saying is that it make NO sense why some work and some dont with the EXACT same software under the exact same circumstances. I think there are different revisions on cards and such and thats what needs to be explored.

 

It makes perfect sense ,and has been explained ad nauseam.

 

Some cards might have the correct timing set by the factory, dunno.
Ad nauseam, +1 to you sir.

 

Jody,
I have no problems with someone pointing out flaws with the product as it certainly assists us in addressing the issue. I have no problems with users running benchmarks either, in fact I do it numerous times a day. Before I came to work at ASUS all I did everyday was point out flaws in computer products so they could be improved in the future.

However, my comment about Furmark and on the CPU side LinX is that these programs do not necessarily reflect real world scenarios with either how well an application/game runs or actual platform stability. They basically overstress the hardware to the point where we sometimes see short term failures after a certain number of runs and generally it is in the PWM subcomponents on either the GPU card or on the MB even though these components are speced well above the minimum requirements from the manufacturer.

If a test goes beyond the failure points designed into a system and creates unnecessary failures then the user has a right to know this. Let's say the redline on my $15k engine is 6800 rpm and Ford says to not to exceed it but I decide to extend the rev limiter to 7600 rpm with a software program or just ignore it altogether because somebody told me I needed to stress test the engine this way to ensure I have a stable car.

I then drive around a road course at 7600 rpm for a couple hours and the engine actually survives. I now go home thinking I have a "stable" car. Does that mean there will not be a problem in the future even though I just shortened the life expectancy of that component even though it worked perfectly? Let's say I do this a few more times and then the engine develops a problem, do I then blame Ford?

I only brought this up as I see quotes like "I ran Furmark or LinX for eight hours straight without a problem so my system is stable" and later seeing comments like "Battlefield 2 Bad Company will not load quickly or has stuttering in heavy action sequences" or "Cinema 4D r11.5 will not complete a rendering, please assist me" from the same users. That is why saying a system ran Furmark or LinX fine is not going to help us assist users who have a problem in their application or game. I have several drivers here that will pass Furmark just fine but there are problems in the newer games and vice versa. Would you rather hear as a user that the platform will run Furmark for a certain number of hours or that you can actually play Mafia 2 or render 3D images in Cinema 4D for several hours without a problem?

That said we do run Furmark and other stress test programs to push the hardware and see where it could develop problems over its life cycle. We will make changes if necessary to improve the quality of a unit. Unfortunately I came into this situation later than I should have and for that I apologize to the users who are having problems with their particular configurations.

I will be browsing this forum on a consistent basis and starting next week I will have a Technical Support Specialist assigned to this forum. Our goal is to provide technical assistance where applicable or fast track the user to the appropriate CSR if a return is required or to a Service specialist if the problem is outside of our scope. Our other primary responsibility to our customers here is to create a direct communication path to R&D to provide solutions to problems or incorporate customer driven ideas into future products.

So what does that mean at this point besides some nice words?
1. We are working closely with AMD to solve the random GSOD problem with the generic drivers. At this point, we have identified a cause and are working on a solution. Whether that solution is a new vBIOS, driver set, or combination thereof is something we are still working on.

I know to several users here that this comment will sound silly, but the problem has been vexing to us as we have systems running BIOS 0209 and the 10.7a release without a single problem and then the next system will generate a random GSOD. I played the STEAM version of several games for hours on end the past few days on my G73Jh-A3 and an early MP sample without a single problem, yet a beta tester had two GSODs last night during an online session with me.

However, we could not replicate those GSODs today in the same game and level but I experienced a GSOD on the MP sample after changing the audio driver package. The interesting tidbit on this one is that the STEAM version of the game failed but the CD version did not.

We do understand for the users having this problem that it is a very frustrating experience and we are doing our best to fix it quickly. I will provide any updates as I have them. In addition, we will start working closely with the user community in regards to beta BIOS releases (testing) or perhaps driver patches to solve these issues.

2. Another flash point is the TIM application. I will have an update shortly on this subject.

In closing, we will significantly improve our presence here along with technical support and assistance where applicable. In addition, your comments will be heard at a direct level in our R&D and engineering groups. We might not act as fast as you want us to at times but I promise you that we will act, even if it means saying "Yes, we screwed that up but it will not happen again."

Sincerely,
Gary

 

omg official fixes. you heard it here folks

 

Gary,I have a question.

Does Asus have units ready to go that are stable like the one you just mentioned,or all units privy to some kind of GSOD.

Should I RMA my laptop now for one of those stable units,or wait for the fixes to come out?

If i RMA my unit,am I just spinning the dice in hope of a good unit is Basically the jist of it

 

My June manufactured G73 has not GSOD'd on me a single time. I have different video memory than others, and this may be contributing to my machine's stability.

 

Good question I would like to know this also.

 

this thread should be stickied as the official asus update thread

 

Mine is from April and I haven't had it gsod on me either. It's good to see someone from asus coming in here and giving us updates as to what they are doing to address these issues.

 

Mine is from May and it does GSOD on new drivers ex 10.7a & 10.7

 

@Gary Key:

Thank you very much for the input to the forum. I can appreciate the difficulty of tracking down a GSOD that is intermittent. There's lots of good technical trial and error type data in these threads that will be good information for your work.

My TIM was incorrectly applied but there are other owners who have zero problems with this. The users with good factory TIM still run a little warmer, but that's just a difference in the actual TIM product. I personally went from thermal shutdowns in three minutes to 88C max as long as I care to run.

The other problem that almost everyone has is missing keystrokes. The laptop will randomly drop letters as you type. Disabling the touchpad in BIOS completely eliminates the problem. I have tried every fix that everyone else has tried. The latest touchpad driver from Synaptic greatly reduces the problem, but it's still there. I don't know why the touchpad driver or hardware would affect my ability to type but it does. Disabling it with the function keys makes no difference unlike disabling it in BIOS. I read in Asus forums on the official web site that this happened with previous Asus laptops that have the same keyboard and touchpad. I think its a driver conflict of some kind related to the keyboard filter driver and the synaptics driver.

I can appreciate what you are saying about Furmark. Others in the forums have made the same point. I was hesitant to run it at all. I didn't know my thermal paste was bad until I ran that benchmark. Honestly the purpose of running it was to prove that my laptop was perfect except for the missing keystroke thing. It turns out that it wasn't. However, with a little dab of IC Diamond, all is right with the world again.

The GSOD is a really big deal for a lot of users and they will greatly appreciate the demise of this issue. You may be heralded as a hero in these threads should you succeed in your mission.

 

Gary, I would like to say thank you for taking attention to these issues. It is obvious that you are actually working on these items, and some of us do understand that you, as the provider, can't do "shot in the dark" solution resolution, thus it takes time for issues to be figured out and proper solutions to be put into place.

If I may ask, could you please add one other item to your list, and that is the keyboard "missing characters typed" issue. I am a professional who does the majority of my authoring (both code and correspondence) on my laptop and this is a HUGE issue in allowing the unit to be truely "functional".

To the rest, if I may add a sidenote, my G73Jh-A3 is a late May/early June manufacture and the 5870m card in mine has Samsung K4G10325FE-HC05 memory chips on it.

I have not had a single GSOD outside of forced shutdown because of reaching max GPU temp due to faulty thermal paste. The thermal issue has been resolved and I will be testing the 10.7a Catalyst drivers and will report if there are any issues. Who knows, maybe I received a "working" 5870m.

 

Thanks Gary!! Just hearing that someone at Asus recognizes the problem and is willing to work on it is very comforting. I'm excited already!

 

also something is seriously up with the touchpad, it will respond erratically with or without the synaptics drivers its starting to get really annoying

 

I have had that problem occasionally. I found that turning off pinch zoom helps a great deal. At least with the synaptics drivers, there are so many cool settings to mess with that I can get mine where it is acting screwy pretty easily.

I removed it. Added it back. Then I turned of pinch zoom so it would stop resizing text in my web browser all the time. It's working really well now except that I'm still missing keystrokes occasionally.

 

No, we have not discovered any differences in hardware produced in March to those in early July when it comes to the GSOD issue. As such and after working with AMD, we do not think this is a hardware issue, rather one of firmware, software, and driver combinations although we are doing our absolute best to rule out any hardware problems first. I say this as hardware is easy to fix.

I would wait until we have a decision on what to do next, afterall the unit is under warranty.

 

mine doesn't work at all... any driver
even at 300/300 gpu timings it gsods...
700/1000 gsod
705/1100 gsod

9.12 gsod, 10.1 gsod etc etc 10.7a gsod

scarcraft 2 = 114 degress and shut off. 3dmark06 = instant gsod in 1 min. FFIV benchmark = overheat/gsod whichever first.
Singularity= gsod/overheat. Photoshp CS5= GSOD GSOD GSOD... awww just drew the best hand ever GSOD aaarrrgghh!

i need it for work dont know what i am going to do, if I RMA and wait for a fix then i have no work computer for 3-4 weeks and maybe get back a non working half fixed laptop like others have.

I am just s oupset, I have always bought asus motherboards, laptop, etc etc but this just is so terrible

--> just want to add that it did this since i bought it in april. I put in a intel x25m g2 SSD and reinstalled windows (to load photoshop faster)... i blamed myself the first month thinking i did the SSD or windows wrong, i wish i new about these problems so i could have given the laptop back for refund.

 

Good to see an Asus rep in here, my big thing is do I RMA it and cross my fingers or just ride it out till a software update comes about. I just did a full dissasembly to get to the GPU to apply some AS5 and have already seen my temps drop even though I have quite a few heat cycles left to go. I would GSOD on games until I did the Vbios flash and a 715/1025 OC and now I can play any game with a random line hiccup across the screen. I guess we will see what type of Software comes down the line before we all go RMAing our $1500 laptops just to have the same problems happen. Plus I don't know about you guys but even having my laptop gone for 2 weeks would be HUGE!

 

Gary, I would hope we get a video BIOS fix. I have no problems with a GSOD though.

But I just want better downclock support with modified vCore support. When I compare the Asus video BIOS and the ones from MSI or Alienware ours is a little on the simple side. Sent you a personal message about this also. A lot of people would like this support if you read the forums.


Also, I want Asus to stop voiding warranty for people that actually saved you expensive warranty claims by properly applying the paste on the GPU themselves. I'm more technically skilled than the Asus RMA people and I just saved you an RMA now because the GPU gets to warm and one in the future because the GPU went dead because of these constant high temperatures.

It's a slap in the face of everyone who saved you the RMA because they can do a better job than the Asus RMA.

 

Thanks for the update, but I believe it when I see it released.

 

I talked to an asus rep in a chat a few days ago and you do not void the warrenty by replacing the thermal paste. The only way you void the warrenty is if you damage something while doing it. If you guys want to see the email I could post that also.

 

Is that a worldwide policy?

 

Where did you hear Asus is voiding warranties?

To replace the ram in the machine you have to disassemble down to the same point to do the upgrade, so how would changing the TIM constitute a voiding of the RMA?

Plenty of video card vendors allow you to swap out thier heatsink/fans for waterblocks and more efficient cooling devices without voiding your warranty.

If they were to void my warranty because I accomplished a "fix" that was within the scope of user-ability, they had better be able to prove that my "fix" was the problem for the failure that I am RMA'ing under. There were no stickers that were removed that could constitute a "automatic voiding because of removal" or anything of that nature.

It is well known and understood that laptops may need to be partially disassembled to provide for upgrades, so opening up the laptop itself cannot be grounds for voiding.

Key here is that our laptops are our primary machines for a lot of us. They are also expensive machines with specifics in regards to finish and looks, thus sending in for RMA is a horror story waiting to happen in regards to turn-around time (most of us cannot be without our machines for a variety of reasons, including work), as well as wondering if you are getting back exactly what you sent in (getting a perfect display is paramout to many laptop consumers and the idea of getting back a faulty display because some tech wasn't paying attention when he was working on multiple machines is nightmare material), having physical damage to the unit (which we already have heard reports of), or having something else not working when you get it back that worked just fine before you sent it in.

On top of all that, in a previous career I was a computer tech certified by dozens of vendors in repairing thier systems (ie, laptops), including every major player like HP, Compaq, Dell, IBM, etc, etc, so the idea that I wouldn't be qualified to make a simple TIM replacement is laughable.

Another issue is the damn fet falling off the bottom of the laptop. I have a missing pad myself that is MIA, thus I cannot just glue it back on. Does anyone have a part # for the feet on the G73Jh?

 

My video memory was manufactured by Samsung where some users reported having Hynix chips. I would call this a hardware difference, and I'm sure there are other slight variances. My June manufactured G73 does not GSOD (with either stock or reference drivers), but it took 2 returns to get one that worked properly.

I don't believe for a second that this isn't a hardware related issue unless different machines are running different versions of the video firmware.

 

The problem is probably a bit like overclocked CPU ; one can tolerated it and another cpu identical will just crash on heavy load with the same overclock.
Its probably the same with some timming in the vbios here. Some gpu / memory can handle and some may not.

Its just my opinion/ guess tho..

 

I second that... No two chips are the same.

 

So which video firmware version do you see in Catalyst Control center?

 

Gary,

Thanks for coming on to these forums and keeping us in the loop. I know it makes me feel a bit better about my recent G73 purchase and ASUS in general.

Now I have a simple request to make. Is it possible for ASUS to begin to update their website's driver section for the G73Jh more often? There are clearly new versions of drivers available on ASUS' site for most components located elsewhere, but for some reason the G73 section still lists older ones (some are 4 versions out!). These machines sure do come with a lot of bloat so for the owners who are opting for a clean windows installation having this updated database would be extremely helpful.

Thank you and your team for your personal attention. I eagerly await a solution to the GSOD flaw!