F8Sv-C1 question

I recently purchased the F8Sv-C1. Although I've been very impressed by this notebook, there are couple of issues that I'm trying to find answers to.

First, is there an option to control the fan speed on this thing? I would love to blast the fan to maximize the cooling, especially when I stress the GPU. Going upwards of 80 degree celcius doesn't sit right with me.

LCD - I got the model that has the AUO lcd screen. Light leakage is very noticeable from the bottom of the screen. The visual sweet spot on this thing is ridiculously tiny due to its horrendous vertical viewing angle. I tried contacting ASUS regarding this a few days ago but so far I got no response.

Has anyone successfully had their crappy LCD replaced by ASUS? Is this something that I need to "DIY"? If so, is there a guide for this?

Thanks, I will appreciate any help.

 

Anyone? (tumbleweeds...)

 

1. No.

2. All 14" screens have a narrow sweet spot for viewing with narrow verticals - it's a consequence of the glass. I think what you're describing is actually the color inversion near the bottom, which can look sometimes like light leakage but isn't. In any event, Asus generally only replaces LCDs for stuck pixels.

 

... or other hardware malfunctions.

Does the GPU go upwards of 80deg on load, or during idle? If it's on load it's not a problem -- nowadays GPUs can get quite hot. If it's during idle, then it's not really that good.

In that case, make sure that the GPU downclocks properly when idle, check the clocks with Riva Tuner, it should use the "Low Power 2D" profile automatically when not under load. If it's not using it, update your video drivers from laptopvideo2go.

 

Thanks for the responses. Yes, it hovers around 80c during load, which is about 25c more than what my desktop 8800gts 512 does at load... and it's air cooled without blasting the fan at 100% (more like 70% at 1200 rpms). Yipes.

I'm currently using rivatuner but it's been acting up... gpu clocks going out of whack on its own. Might be the new driver I installed - modified 175.16 from notebookforums.

I'm surprised there is no way to control the fan on this thing. I would rather prefer an option to blast the fan, even if it was loud, than have the gpu core temp reach 80c. Sheesh that's almost boiling point. I guess I'll use one of those notebook coolers to see if it helps any.

I've read in the forums that the AUO lcd screens are especially bad for the F8Sv series. CMO(?) screens are much better and many came equipped with these. Unfortunately, I'm one of the unlucky ones.

I guess I'll be performing some surgery on this thing to replace the lcd...

 

the only fan controls i've heard of working on the asus notebooks is if they have the ati video card.

you can try atitool and the beta's to see if functionality is extended for your model. and if it works! please tell us!

 

The laptops have a much more difficult job of cooling the components than a desktop, simply because there is less space to do it and the components are cramped together, heating each other up, which makes it even more difficult. Therefore the temperature difference is to be expected.

What do you mean by that? It is normal for the GPU to clock up and down according to load, that is done for power and heat savings. If you have 3 different sets of clocks that it uses, then it's normal, that's how it should work: lowest is Standard 2D, next is Low Power 3D, highest is Performance 3D.

Probably the engineers at ASUS set the thresholds after a good amount of thinking as to the consequences (at least I hope so ): heat versus fan noise + wear.

 

While this is true, the biggest heat creating device within this notebook is the gpu while gaming. Even the cpu, 2.5ghz peryn, at load does not get up to 80c+. The highest I got the cpu was 60c during orthos stress test. If the gpu was adequately cooled by a faster fan, much of this heat issue would be resolved.

I'm no newcomer to rivatuner. For one reason or another, the gpu is downclocking while I'm in the game. This shouldn't be happening. I think it may be the driver set I'm using. I am going to uninstall it and try a different set to see if it fixes the problem.

I wouldn't give the ASUS engineers too much credit here since many of the notebooks that are equipped with some of the 8000m series nvidia cards are having critical overheating issues. I mean if the engineers were doing their jobs, they would have caught this before ASUS shipped them out, right? Or maybe they just took Nvidia word at face value. So who should we demand to be fired? Nvidia engineers or ASUS engineers for not catching the heating issues?

Anyhow, I just placed a notebook cooler underneath the notebook and my gpu temp at load is 74c. Not too bad but haven't really pushed the gpu just yet. More graphically demanding games such as Crysis will surely add about another 10c to that.

I unscrewed the fan out earlier to see if it could be replaced with a faster one. Doesn't look too promising though... The fan has 4 wires which tells me that it can be controlled. BIOS update by ASUS could fix this...

Btw, I would be more than happy to replace a worn out fan due to high rpm than RMA or junk the whole notebook(warranty expires) due to a burnt out graphics card. But, that's just me.

 

Temps: Yep, in most ASUS notebooks there is just one fan cooling both GPU and CPU. Having two would certainly help. Not sure what are the reasons for having one, though (having more space for components, most likely?). What's worse, sometimes (always?) the fan is driven only by CPU temp so a cool-running CPU can mean an overheating GPU (G1S is a good example).

Clocks: Yeah, if it's downclocking during a game that's not good. You should indeed try new drivers, I think. Laptopvideo2go is the way to go if you have nVidia.


As to the engineers... Assuming that when you say "nvidia fault" you are talking about the recently publicized substrate faults in the 8400 and 8600 series of nVidia GPUs:

While I don't give ASUS engineers too much credit (at least not with the recent machines), I don't blame them for the nVidia trouble. If the ASUS notebooks are affected (which is not entirely clear at this point, although it's likely), I don't think we can expect a notebook manufacturer to run full GPU tests on batches of GPUs that come from nVidia after supposedly being tested and error-free (I presume that's how it happens, I can't be sure). Even more so, they cannot detect faults such as this one, which is a faulty substrate and the effects of which only show after extensive testing simulating maybe one year of normal GPU use. Such an extensive GPU test should be done at nVidia. They are responsible for this fault, not the notebook manufacturers.

On the other hand (and this is a common misconception) the widely publicized nVidia fault is NOT causing the GPUs to overheat; it makes them resistant to a smaller number of heating up / cooling down cycles; which means they will fail after a shorter period of use than would be normal. So, any overheating on ASUS notebooks is not nVidia's fault (at least not due to this particular malfunction); it is due to ASUS designers/engineers.

Personally, I think that some recent models from ASUS (and by recent I mean last 2 years) were driven by an over-enthusiastic search for performance components, at the expense of cooling efficiency and sometimes even build quality. The G1 is an example of the former. I'm not sure we should blame the engineers for this, or the marketing department

About the fan vs. GPU replacement, I agree with you, but I don't think we represent the typical notebook customer. For that typical customer, I'm afraid a broken fan means the same thing as a broken GPU: a notebook that doesn't work.

BTW, GPU burnout does not invalidate warranty unless it's due to user error (excessive overclocking etc.) but even that is difficult to prove... they'd probably just replace it.

Edit: I was assuming above that you were talking about the recent publicized faults in the 8400 and 8600 series of nVidia GPUs. Was that correct?

 

Indeed. But, both companies are definitely at fault for introducing hardware without thorough testing.

And yes, I was referring to the recent fiasco over 8400m and 8600m series cards.

As far as my F8S N-C1 goes...

Uninstalled 175.16 modded drivers and replaced it with modded 174.31 drivers from notebookforums. For good measure, I uninstalled rivatuner as well and installed EVGA precision, which is a dumbed down version of rivatuner more or less. The biggest difference between the two is that EVGA precision does not have automated 2D, low 3D, and perfomance 3D clocking. This is preferable in my situation since I could not stop rivatuner from downclocking while in game. Very annoying.

Now I just use hotkeys to overclock or underclock. Couple of button presses and I have my desired gpu clocking. So far, I've been able to successfully overclock the nvidia 9500m to 600 core/500 mem. A healthy 100mhz overclock for both core and memory.

I also undervolted my CPU using the guide posted here in Notebookreview.com. I was able to get 2.5ghz @ .975 volts. A reduction of .300 volts! Reduced my cpu temps about -6c during load when I did this. I'm sure it improved the battery life just a tad, too.

Aside from the heat and light leaking lcd, I'm very impressed by this notebook. Also, vista is not as bad as I thought it would be.

 

Nice undervolting results. Just make sure you do thorough stability tests, that's a big undervolt that you have there!

As to the GPU clocks, something is not clear to me. You're saying that RivaTuner is controlling the GPU profiles -- I though that wasn't the case, but the driver controlled the GPU profiles. What RivaTuner can do is change the clocks for the three profiles, on the other hand.

If EVGA precision can help me have explicit control over my GPU clocks in recent notebooks, I want it!! I'll look it up a bit and see what it can do.

 

Yep, I stress tested the peryn for 2 hours at .975v using orthos. Rock solid. I wish the VID at 6x multiplier could go lower than .950v though.

As for precision... the only issue with it is that it only allows me to underclock the card to 400 core/ 300 mem...

And I noticed something disturbing today... the gpu was using all three performance profiles without rivatuner. Turns out this is a feature within the gpu known as powermizer...

The odd thing is, the gpu now stays at the manual profiles until I close out of a game. Only then it underclocks to the powermizer profiles. The gpu is currently running at 392/783/301 and hasn't budged since the restart.

However, I didn't notice any downclocking during gaming since I uninstalled rivatuner and 175.16 drivers.

I also downloaded the powermizer switch but didn't have a reason to use it yet. If the gpu underclocks again during gaming then I will definitely use the switch.

 

Precision is just a customized, slimmed down version of RivaTuner that eVGA had the guy who wrote it convert for them and put a new GUI onto. I'm using it on my desktop 8600gt, because I'm too lazy to mess with anything else and I like the simplicity.

If it works for your notebook gpu, it would certainly be big news for the Asus folks!

 

That's just what I said above. GPU profiles are controlled by the driver (specifically PowerMizer), not by RivaTuner.

Could be conflicts between PowerMizer and the EVGA application, which leads to the profiles not being switched intuitively.

I think in later driver versions this switch only enables you to remove the cap on the GPU frequencies when on battery. Just a note.

 

Which program did you use to underclock the CPU? I have tried NHC and it could not read the voltage values...

 

I guess he used RMClock, it's the most recently updated software (although it's still quirky with Santa Rosa and later).
See my Tips and Tricks for some details, or the undervolting guide mentioned in the quote.

 

Finally, success!!

I used the Powermizer Switch as described here:

http://forum.notebookreview.com/showthread.php?t=273276

Now no more underclocking during gaming when it is plugged it to an outlet. You can disable powermizer during battery use as well if you so choose so. I can now control my overclock and underclock with a touch of couple of keys and don't have to worry about the gpu underclocking my card when I don't want it to.

And yes, I used RMClock to undervolt my peryn T9300. It's an awesome program. Use this guide:

http://forum.notebookreview.com/showthread.php?t=235824

I also uninstalled Evga Precision and reinstalled Rivatuner mainly because I can manually underclock lower via Rivatuner. My manual underclock setting sits at 261/522/205. My manual overclock is 594/1188/504. Works great so far.

 

Nice job, that PowerMizer switch. I didn't know about it.

 

mechguy:

I figured out exactly the same issue with my G1Sn-B1, and have asked ASUS to consider releasing an updated BIOS, at least for their gaming/graphics intensive notebooks. It is so ridiculuos to notice that they just simply left out the ability of the fan to configure to gpu temps., just like the cpu temps and that they are ignoring the issue (especially with the faulty chips dilemma going around). At least it seems to me like a definate need for their gaming rigs. The fact is that ASUS tech support guys just don't care. I basically spelled out the discrepancy for them, and they say 96C is an acceptable temperature according to their support policies. They also said something about the GPU's hard-coded diode, and made an excuse like they couldn't enable the option even if they wanted to.

I have tried the PowerMizer switch, but my GPU keeps downclocking while playing Crysis. It seems my throttling threshold is 96C, and oddly the fan will boost to 100% around the same exact time, but usually the throttling wins and my game crashes. Since the fan will boost at the exact time my GPU passes 96C, it leads me to believe there is something in place, but again: It could just be the CPU thats getting heated-up by the GPU.

I have been waiting for Nibitor to fully support the video BIOS of the 9500M GS, so that I can flash some new thresholds confidently, but I'm not sure if the system BIOS is the problem, or the video BIOS. I'm sure I'll be trying Nibitor anyways, since it's my only option left. I see that changing the clocks is possible right now, but I'm hesitant to use that tool, because not all the options are populated, and I'd be risking too much by flashing default values for older systems into my chip.

I found a 9500M GS video BIOS somewhere on the net and someone has put the fan on 100% for all three profiles. Personally I'd rather just adjust the temp. threshold on the high 3D one if possible, or at least just 100% fan for the top two only.

Someone else was just changing their throttling clocks to match their performance clock and that fixed it for them, but I like to OC mine. So, if I'm running Crysis at all Very High with shaders and shadows on High (1440x900), even if it did still throttle to the high non-OC'ed frequencies, the game will still probably crash.

Such an easy fix, but yet it is so far away. Any developments?