Acer Aspire 8930G Owner's Lounge

Discussion in 'Acer' started by hoangtruong33, Dec 31, 2008.

  1. johnbrown105

    johnbrown105 Newbie

    Reputations:
    0
    Messages:
    8
    Likes Received:
    0
    Trophy Points:
    5
    Hello All,

    I have an 8930-6386 (2.53 GHz, 1680x945 screen). It can charge the original battery, but not two replacements that I have tried. Has anyone successfully replaced the original battery (14.8V, 4800mAH, 71Wh, AS07B42)?
     
  2. jgenemans

    jgenemans Notebook Enthusiast

    Reputations:
    0
    Messages:
    17
    Likes Received:
    0
    Trophy Points:
    5
    JohnBrown105,

    I have the 8930G-7665 2.0 GHz quad core Q9000 with 1920x1080 screen and have no problems with any of the batteries listed below:

    Generic, AS07B72, 14.8V 4400 mAH, 65Wh &
    Acer, AS07B72, 14.8V 4800 mAH,
    Acer, AS07B32, 14.8V 4800 mAH

    All of these are 8-cell batteries, the lower voltage 6-cell won't work, the key is voltage (after form factor) at 14.x volts. There are a few more models that will work, but many more that won't but will fit.

    Other upgrades I've done include the 750GB Seagate 7200RPM hybrid drive (#1 recommended upgrade - SSD speed, HD storage) + 1TB 5200RPM + 6 GB RAM.

    I have not messed with video upgrade (I'm using the 9700M with updated video drivers to utlize the CUDA processors), but contemplating CPU replacement as 2 of the 4 temperature sensors have malfunctioned causing heavy computing issues where I have the throttle down the CPU to prevent overheat shutdown. I am also considering the BIOS upgrade that was posted in

    brebs,

    Thanks for posting this version of the BIOS. What other versions of the BIOS do you have? I am assuming this version will run virtual machine. Will it also work on the Quad Core Q9000?

    Has anyone tried the version that brebs posted? I usually lurk and wait to hear from other people's experience, but like the prospect of undervolting my CPU for the above stated reason.


     
  3. brebs

    brebs Newbie

    Reputations:
    0
    Messages:
    5
    Likes Received:
    0
    Trophy Points:
    5
    Just what can be googled. What do you want?

    Yes. I run qemu, in Linux.

    I don't see why not. The DSDT contains entries for 4 CPUs:

    Code:
        Scope (_PR)
        {
            Processor (CPU0, 0x01, 0x00000410, 0x06) {}
            Processor (CPU1, 0x02, 0x00000410, 0x06) {}
            Processor (CPU2, 0x03, 0x00000410, 0x06) {}
            Processor (CPU3, 0x04, 0x00000410, 0x06) {}
        }
    I have ;) But just me, AFAIK, so far.

    My T9400 CPU (dual-core) normally also gets throttled, when both cores are fully utilized for a few minutes.

    What seems to be working nicely so far (I'm still working on it), is:

    Code:
    # Disable SLFM (changing 9 to 8)
    wrmsr 0xee 0x8eb91000
    
    # Preparation - ensure that EIST is enabled (the 5)
    wrmsr 0x1a0 0x364850489
    
    # Reset to a low value
    wrmsr -a 0x199 0x619
    
    # Undervolt at a reasonable non-optimal speed, to stay under the thermal throttling limit
    # 8 is the speed (9 is usual, a is IDA (turbo). 18 is the voltage. These are in hexadecimal.
    wrmsr -a 0x199 0x818
    I use a USB laptop cooler which sits under the laptop - it had 2 fans, but all the heat in the 8930g is in the left-hand side (video card & CPU), so I just disconnected the cooler's right-hand-side fan, to reduce fan noise.

    A great trick to increase airflow substantially is to hack away the 8930g case's plastic fan grate (I used a pair of scissors). Yes, this makes a hole in the case's underside, but it really works :cool:

    I am long-overdue for attempting to re-apply the thermal paste on both the CPU and video card.

    The thermal throttling seems to be triggered by TM2, but I haven't been able to influence it. This becomes more evident when the thermal interrupts are enabled, with "wrmsr -a 0x19b 0xffff1f". The thermal readout seems to get stuck ("rdmsr 0x19c -f 22:16") :(
     
  4. jgenemans

    jgenemans Notebook Enthusiast

    Reputations:
    0
    Messages:
    17
    Likes Received:
    0
    Trophy Points:
    5
    brebs,

    Thanks for the quick reply. I predominately use my machine for HD video processing, and have balanced the GPU and CPU in Sony Vegas and get HD frame-rates in near real time, as long as I can get about 75% utilization on each, it rocks, however, if the CPU climbs above it I have to use Throttlestop to enforce more aggressive throttling (the opposite of its intended function of overclocking) based on thermal profiling to prevent the CPU from shutting itself down.

    btw, do you know of a program that actually works where I can manually force the fan to maximum to slow the heat-build up so that it can cope?

    I am using win7-64 home premium.

    If I am in a rush to process a long video, I will completely remove the underside cover, and place the machine on a A/C register with fan turned to 'on' (not auto -- then no throttling is needed) to complete the job 25-33% faster.
     
  5. brebs

    brebs Newbie

    Reputations:
    0
    Messages:
    5
    Likes Received:
    0
    Trophy Points:
    5
    You should definitely be undervolting the CPU, for starters.

    I have not found a way to control the fan (I only use Linux) - I've tried changing CPU MSRs in countless combinations, and also the Nvidia settings. However, I don't believe it would help, because a 100% fan is still overwhelmed by a 100% CPU and 100% video card.

    So, currently I throttle the video card (Nvidia 9700M GT), using this script (in Linux):

    Code:
    #!/bin/bash
    
    temperature() {
    	# https://wiki.archlinux.org/index.php/NVIDIA
    	# Also:  nvidia-smi -a | grep Gpu  and  nvclock -T
    	# nvidia-smi -q -d TEMPERATURE | grep Gpu | cut -d: -f2 | cut -d' ' -f2
    	/usr/bin/nvidia-settings -q gpucoretemp -t
    }
    
    
    throttle() {
    	echo "In throttle"
    	s=20
    	while true ; do
    		t=`temperature`
    
    		# To see what speeds are supported:
    		# nvidia-settings -q GPUPerfModes -t
    
    		if [[ $t -ge 100 ]] ; then
    			# Lowest possible
    			pspeed=169
    			mspeed=100
    			s=5
    		elif [[ $t -ge 96 ]] ; then
    			pspeed=275
    			mspeed=300
    			s=10
    		elif [[ $t -ge 93 ]] ; then
    			pspeed=400
    			mspeed=300
    			s=10
    		else
    			# No throttle
    			resetSpeeds
    			s=15
    			if [[ $t -lt 77 ]] ; then
    				# Can go back to normal relaxed mode
    				applySpeeds
    				return
    			fi
    		fi
    		echo "t=$t pspeed=$pspeed mspeed=$mspeed"
    		applySpeeds
    		sleep $s
    	done
    }
    
    
    resetSpeeds() {
    	pspeed=575
    	mspeed=800
    }
    
    
    applySpeeds() {
    	if [[ $pspeed -ne $appliedPSpeed ]] ; then
    		/usr/bin/nvidia-settings -a GPU3DClockFreqs=${pspeed},${mspeed} # > /dev/null
    		appliedPSpeed=$pspeed
    	fi
    }
    
    
    # Start
    renice -n 1 -p $$ > /dev/null
    
    if [[ -z $DISPLAY ]] ; then
    	# For nvidia-settings
    	export DISPLAY=":0.0"
    fi
    
    # Enable clock changing
    nvidia-settings -a GPUOverclockingState=1 > /dev/null
    resetSpeeds
    appliedPSpeed=-100
    applySpeeds
    
    while true ; do
    	t=`temperature`
    
    	if [[ $t -ge 80 ]] ; then
    		throttle
    	fi
    	sleep 40
    done
     
  6. pablocruz

    pablocruz Newbie

    Reputations:
    0
    Messages:
    6
    Likes Received:
    0
    Trophy Points:
    5
    I bought this one back in August with no problems!! Not sure if this is what you are looking for!!
    8Cell 14 8V Replacement Battery for Acer Aspire 8930 8930G 5520 5715Z AS07B72 | eBay
     
  7. pier89

    pier89 Newbie

    Reputations:
    0
    Messages:
    3
    Likes Received:
    0
    Trophy Points:
    5
    Hi everyone,

    I have a quite strange problem with my 8930G (Intel Core 2 P8400, 4Gb RAM, 320Gb HDD, Geforce 9600M GT): it suddenly turns off (as if the power cable was unplugged) and restart, while loading the OS. I've tried to remove RAM modules (using only one at a time), and to remove the hdd and start from a live distro on pendrive, but the problem is not related to these components. Instead, after the motherboard warms up, the problem does not appear anymore (indeed, I also tried to let the pc cool down, then warm it up with the hairdryer). I really do not have any idea about which component might be faulty.

    Also, it is not an OS fault, since I tried booting both Windows and Linux.
     
  8. brebs

    brebs Newbie

    Reputations:
    0
    Messages:
    5
    Likes Received:
    0
    Trophy Points:
    5
    Does the machine stay up if you just stay within a BIOS menu?

    I recently had to change the PSU, because its output was not steady. That made the motherboard produce a quiet clicking sound about every 2 seconds, IIRC.
     
  9. pier89

    pier89 Newbie

    Reputations:
    0
    Messages:
    3
    Likes Received:
    0
    Trophy Points:
    5
    I have not tried it, but if I leave it hanging on GRUB boot menu, it stays up (and the fan is always on). Even if I run memtest86 from GRUB list, it completes the test without powering off. I had problems with power supply some time ago, but the notebook powered down WITHOUT restarting, so this time I think it could be something else. Also, I do not hear any clicking sound from the notebook.
     
  10. pier89

    pier89 Newbie

    Reputations:
    0
    Messages:
    3
    Likes Received:
    0
    Trophy Points:
    5
    I was wondering if there's some way to do manual hardware probing (that is, manually enable devices over PCI and USB bus) in order to identify the device that causes hardware failure.
     
Loading...

Share This Page