g1s new bios but does not solve cpu problem

ftp://dlsvr02.asus.com/pub/asus/nb/g1/G1SAS203.zip

Just like the title says. I do not know what this bios brings but it's new. Sadly the cpu problem is not fixed. When charging battery until it's above 90% the cpu will not go till 2.2ghz. It just stays at 1.2ghz. Will have to wait for the 204 version. (since 204 on the g2s does not have this issue) also the 204 g2s version on the g1s does not have the problem but the keyboard stops working. thank god i make a batch file for reflashing the g1s version back

Cheers

Edit:
"Changed the Memory Refresh Mode select from 7.8 usec to 3.9 usec.
Note: This is the project leader and Sale PM request.
Correctly the Japan keyboard Matrix.
Update Azalize AUDIO verb table for G1S of ALC-660D.
Correct the Direct Media button for G1S."

 

Has anyone else tried this bios update? Does it fix the grainy screen issue?

 

uuuuh hello?
Intel microsoft and asus all have their utilities which reduce the clock of your cpu on battery life. This is done on purpose. Disable the power savings option in bios. remove power4gear by asus. set windows to max performance. then you should be able to get the full clock out of it. i dont know why you would want to. it reduces the battery life to under 45 minutes even less if you push it

 

sorry misread your post.
dont have any such issues here.

 

any news on this? I didn't see the update on Asus's support site so I don't know if I want to try it.

 

That's what I was thinking. To thread starter, where did you get that file location from?

 

by simply browsing the asus ftp server.

to Geared2play.com. if you have a g1s YES you have the problem.

discharge battery until 70% for e.g. While on battery make a super pi 1M. You will have about more or less 22sec. Plug in ac adapter and make a super pi 1M again you will have more or less 42Sec. You will only have about 23sec while charging battery when it's above 90%. It has been proved by other users. You can see with cpu-z 1.40.5 the cpu speed.

Don't tell me that you do not have the problem because you have it too. I was hopping that more and more users would mail asus about this.

the change log is on
http://de.asustreiber.de/download.php?view.623

 

i do t know what superpi is but i ve tested and restested g1s with 3dmark and pcmark and the cpu was always read correctly and the scores alsways as expected. The battery was not plugged in at any time at all. If i had to guess there are 2 problems that may be affecting you
1. some kind of software that you loaded, like cpuz or what have you for monitoring
2. conflicting issues between your software and the power saving software that comes with the laptop from asus or vista it self.
If you want my help i ask you do 2 of the following things
1. use the restore dvd to restore windows to factory state and do not install any other programs.
2. Use vistas own property feature or 3dmark to read the current cpu clock and no other utility. if you find that your problem is still present i will ask my tech to look into it

 

adolfotregosa, have you tried this update yourself? If so, does it fix the graininess issue?

 

the problem does not exist WHEN the battery is out !! like i said put in battery and discharg it untill below 90%, say 83%.
http://www.xtremesystems.com/pi/super_pi_mod-1.5.zip
make a super pi 1M wile on battery. Plug in ac adapter and try to achive and same score ! use cpu-z to see the speed..

 

I have the bios. Do you want some kind of picture with digital camera ? Sorry but i never understude the "graininess issue" :S

and also the change log does not talk about the "graininess issue"

Edit- Maybe that only happens on 1650 res screens ! mines 1440

 

Like i said - it has been proved
http://forum.notebookreview.com/showthread.php?t=135161

 

I can confirm that I have tried what adolfotregosa suggested and I found similar results. Super PI when the laptop is at 85% and unplugged finished in 22 seconds (I ran it three times with similar results). Upon plugging in the laptop and immediately running Super PI to 1 million, it finished in 38 seconds (again, I ran it three times). I tried again when the laptop was at 94%, plugged in, and Super PI completed in 23 seconds (again, three times). Vista, however, reported the processor at 2.2Ghz the entire time.

However, I am not sure that the processor isn't running at 2.2Ghz, anyway. There could be something else bogging the system down when it is in a lower battery mode. The processor could be chugging away at 2.2Ghz but something else might be taking priority away from Super PI. What would be interesting to see is if someone else running Vista on a completely different model of laptop would have this same problem, or testing it with/without Power4Gear. Also, it would be interesting to see if there are any "real world" differences. Just because Super PI takes longer under this situation doesn't necessarily mean that rendering a video or gaming will be noticeably affected.

 

sorry to say but it does affect
With battery:


Without battery:


With battery:


Without battery:


Edit - 3dmark 2006 - 1024x768
With battery:


Without battery:


Edit - Battery above 90%

 

I updated to the new BIOS, nothing new on the screen issue, its beginning to seem more like a hardware thing.

 

Okay, I too used 3dmark06 to test the G1s’ performance under the same type of situation that Super PI was used for earlier. When I ran 3dmark06 with the battery discharged to <60%, on battery power (i.e., not plugged in), and the Power4Gear set to “High Performance,” my results were as follows (I’ve also attached screenshots of all of this, in the order I present them in this post):

3dmark Score: 3774 3DMarks
SM 2.0 Score: 1533 Marks
SM 3.0 Score: 1375 Marks
CPU Score: 1928 Marks

I then plugged the laptop in so that it was charging (Power4Gear is still set to “High Performance” and I ran 3dmark06 again. The results this time were as follows:

3dmark Score: 3485 3DMarks
SM 2.0 Score: 1554 Marks
SM 3.0 Score: 1389 Marks
CPU Score: 1074 Marks

Through all of this, 3dmark06 listed the CPU as performing at 2195MHz. Notice that the CPU score drops from 1928 on battery to 1074 when plugged in. So it seems like both Super PI and 3dmark06 are giving the same type of result: when the charge of the laptop is below a certain point, the CPU will perform better on battery then when plugged in. However, whether this is due to a drop in processor speed or something else, I don’t know. Both Vista and 3dmark06 are listing the processor speed at right around 2.2GHz. I should note that Everest, however, reports that the CPU is running at 1200MHz. Which one is reporting correctly? Well… given the performance indicators, I would guess that Everest is more accurate. However… all of this being said, I’m not sure there is really anything any of us can do about it (except for maybe call and ask Asus about it – which I plan to do now, as I am convinced that this is not a user-error). However, I would still be interested to see what would happen if someone ran the same tests on a different model of laptop using Vista.

 

ok, you will call me crazy but i DID flashed the g"2"s bios on the g"1"s. I tested the 204 and 205 revision of g2s and the problem goes away, but also the keyboard stops working. To force the bios I had the run winflash /force. Everest is the ALLMOST the right one. Personally i only trusth cpu-z 1.40.5 since it added support for the dynamic fsb. Santa rosa cpu fsb goes from 100 to 200 like the multiplier goes from 8x fsb 100 till 11x OR 12x fsb 200 when it auto-overclocks ONE core(. The minimum mhz is 800. Everest does not fully support the santa rosa cpu yet. Cpu-z does now.

From wikipedia :
"Intel® Dynamic Acceleration Technology
The processor will support Intel Dynamic Acceleration Technology mode on select platforms. The Intel Dynamic Acceleration Technology mode feature allows one core of the processor to temporarily operate at a higher frequency point when the other core is inactive and the operating system requests increased performance. This higher frequency is called the opportunistic frequency and the maximum rated operating frequency is the guaranteed frequency."

 

I just want to note that right at 90%, Everest went to reporting the processor running at 2200MHz (see picture).

 

i can only thank you for your time. please call asus like i did. I really liked this fixed :S

 

I will call Asus this afternoon. What I am wondering is if the problem stems from how the laptop charges. Is it possible that, to charge faster, the laptop automatically puts more power into charging the battery than into allowing for maximum performance until it reaches a certain threshold? Interestingly enough, even at 97% charge, my 3dmark06 scores are still lower than when running on battery at <60% (the first picture is charging at ~97%, the second picture is on battery at <60%).

 

I have to test this over the weekend. never noticed it mainly because i only plug the battery in to test battery life anytime i do review a new product. Maybee you are right. Maybee there are some soft issues between vistas battery saving util, asus power4gear, and some others that you guys have loaded. Its a good idea to run these tests on a fresh load before any of your own software is loaded
On a side note one thing did jump out at me. g2s has a 120w adapter. g1s has 90watt. i know that a 90w adapter will work with g2 but will overheat and shut down if you are pushing the notebook. I wonder if this was done on purpose becuase the g1s 90w adapter is just barely enough and it may draw power from the battery when needed in case the adapter shuts down.? just a thought. other then the screen resolution there is no difference in power requirements between the g1s and g2s

 

I agree - that is a good idea... if I have some time this weekend, I'll restore my laptop to factory settings and run the tests fresh.

 

it has do be a acpi bug. It's not powergear, it's not vista, it's not xp , god, the problem exists in linux too !! i have done tests with the g2s bios with a usb keyboard and the ac-adapter worked really well. I really sucks. I can't play or do something faster when charging the battery. I never saw a laptop with this beavior. it's just plain stupid.

Just think about it. You are playing something, have to use battery for some reason for 20min, then have to wait for the laptop to have full power again. it just does not enter my head that this is a feature ! if this exists because of the 90W adapter i'll buy a 120W , i don't care. i just wan't full power when i need it, not when the laptop wants !!!

g2s has a bigger screen, so it has a more powerfull light. It has a tv tunner. The screen on a laptop is a power ungry thing

EDIT look at this !!!

Battery has 4.8A/hour and operates at 14.8 v

P = u * i

14.8 * 4.8 = 71.04 W / hour

The battery lasts more or less 1h when doing something heavy

So, if the adapter is 90W - 71.04W that the laptop uses when in full load then you have 18.96 Watts for changing the battery.

ANOTHER EDIT
http://www.asus.com/products.aspx?l1=5&l2=74&l3=463&l4=0&model=1675&modelmenu=2

G2S - ac adapter

Output: 19V DC, 4.74A, 90W
Input: 100— 240V AC, 50/60Hz universal

 

Well, what does the new BIOS do then?

 

did you see the first post ?????
"Changed the Memory Refresh Mode select from 7.8 usec to 3.9 usec.
Note: This is the project leader and Sale PM request.
Correctly the Japan keyboard Matrix.
Update Azalize AUDIO verb table for G1S of ALC-660D.
Correct the Direct Media button for G1S."

 

Thanks, from where I'm at, ASUSTreiber.de doesn't respond. No need to get jumpy there.

 

sorry, didn't mean anything
try

de.asustreiber.de

 

Alright, I completely restored my G1S to factory settings. The only modification I made to the factory settings was that I installed 3dmark06, Everest, and Super PI to the "Data" partition. I used the three programs to test my laptop's performance at 1) 100% charged, plugged in, 2) <90% charged, on battery, and 3) <90% charged, plugged in and charging. All tests were done with Power4Gear running in "High Performance" mode.

These tests gave me the same information that I received from doing the same tests on my personal install of Vista x64. That is, the CPU performs BEST when running on battery. It performs almost as well (but not quite) when plugged in and charged at or above 90%. And finally, it performs worst when plugged in and charging, but below 90%. Obviously, this is not a good thing (especially as, even at 100% charged, if the laptop is plugged in, it still doesn't perform quite as well as on battery).

The pictures are in the order initial100%/battery/charging<90% (the first six are in this post, the final three will be in the next). I know you were expecting these results adolfotregosa, but I just wanted to fully rule it out. I have already emailed Asus about this and plan to give them a call as well.

 

And here are the other three pictures.

 

solargaze27

Thank you so much for helping out !! finnaly i managed to call someone atention !!

My father also agrees that this is not a feature but a bug. He is an electronics expert and says that the ac adapter is enough because the laptop "stops/pauses" charging the battery when the hardware asks for power.

 

Yes, I also believe that it must be a bug. It just would not make sense that the laptop should perform better on battery than when plugged in. Thank you for bringing this to our attention. Hopefully BIOS 204 (if it is, in fact the BIOS) will fix it.

Also, Eddie, I know you are a major Asus distributer (or whatever the title is - pardon me if I have it wrong). If this is really a bug with the G1S as we believe it is, do you know of anyway other than calling/emailing Asus' main technical support team to report bugs to them? Thank you!!

 

Like i said. The g2s bios (204 and 205 ) does not have this problem when installed on the g1s ! but i lost the keyboard.

 

hm....i always thought plugged in works better? by ur plugged in is it with or without a battery?

 

for me, the two options. pluged in with 100% battery or without battery. The BIOS acpi has bugs

 

Well, all we can do now is wait for the next update.

 

everest was a bad idea. It has been the cause of many problems in this forum in the past. you should have left it out of your testing. you should have relied on 3dmark scores only. but its your laptop and your testing. installing a utility that has been known to cause an array of strange issues on other laptop is just not part of the "scientific method"

 

Easy Eddie.... .

You make a really good point.

 

seriously tho
anyone who has been on this forum for years like i have knows that all these utilities like everest are known to cause an array of strange issues. Does anyone think it is a coincedence that every poster here that is talking about the same issue just so happens to have everest installed? Probably means nothing but i think that installing everest negates all the research and testing you guys have done. When i say base install i mean BASE. nothing loaded but the base os from the restore disk. I can safely say that loading 3dmark and using it for checking performance is ok. Leave everest out of this if you are trying to go about this using the "scientific process". i dont know why i like that process so much btw. i did get straight A in all my science clasess all my life xept chemisty.

 

Well, a reason for that could be that Everest at the moment is a pretty old version and that could create problems when you benchmark new hardware on it..

 

Ye thats what i mean. I am not pointing fingers here. I am just sdaying that if the o.p. called me and told me this the first thing i would have him do is reinstall and avoid all these utilities especially everest. It is possible however unlikely that it is causing this issue. You can not rule it out until you remove it completely

 

please dont mention "but when i flahed the g2s bios on my g1s the problem went away". I have to assume that none of you know how to troubleshoot your laptop and how to rule out "causality". Yes that is a good point but inadmisable until you guys test your theory of flawed bios on a 100% base o.s.. I suspect that
1. either something in the bios is causing the computer to enter powersaving
or
2. Something you installed is trgering power scheme at 90%.
Either way i dont know. To rule out the possibility of user error is your first priority and you already know how to do that but none of you have done that. Once you have ruled out user error then you can point fingers @ asus. Another thing i find VERY odd is that adolfo is the first person who brought this to light even though g1s has been the most popular asus for 2 months now.

 

I believe the first initial encounters with this problem did not use everest in their testing; only super pi. And there is absolutely no reason for super-pi to screw up your bios/cpu/battery life since it's a tiny, simple, tried and tested program that that only calculates how long it takes your cpu to count Pi to a certain decimal.

Eddie, since it seems you have several resources at your disposal, I suggest you conduct your own research regarding this problem instead of chastising those who have encountered it (flawed as their methods may be); especially being an authorized asus dealer, this problem is definitely worth taking a very close look at -- for your customers sake.

I'm very interested in how this whole thing pans out, since Ive got a g1s due to arrive next week.

 

Eddie is right guys. You make a great point Eddie and you know what you are talking about.

Jack. He is saying that just having everest installed could be the problem.

 

I was not chastising. I was meerely pointing out the obvious. everest should be removed! It may not be the cause but it is POSSIBLE! i ve seen it cause stanger things in the past on unrelated models.

 

I know what he's saying with everest. All I'm saying is that I think this problem was first detected with super pi, and nothing else. I don't know if everest was originally installed when it was detected, and it's not for me to conclude.

LiveDesign, Did you have everest installed when you first tested for the problem using super pi?

Eddie, I understand everest sucks...but you have resources and experience at your disposal and you could easily just do the testing yourself and give us some feedback as well.

 

I could and i will when the time and opportunity presents it self. I was hoping the people here who are already using the g1s could solve atleast part of the task by following very simple instructions. I guess not so i will have to do it in my time off.

 

ok i was going to publish scores using my own tests however it is pointless. i found that the o.p.'s problems are the same in every g1s sold in the states. Pcmark tells the same story unfortunately. As stated by the o.p. the problem is in bios however i think your inquirys to asus may be pointless for the following reasons
1. g2s with the same power requirements has a 120w adapter. it does have a higher resolution and bigger lcd. how much more power does it need? 5w? 10w?
2. i feel the ac brick that comes with the g1s is slightly underpowered, it should have been the same 120w brick as the g1s. I think the qc in taiwan figured that out. Instead of changing the packing contents and whatever else had to be done they simply coded the bios to underclock the cpu until battery is almsot fully charged to prevent the brick from overheating which is exactly the same problem i had while using a 90w brick on a g2s that i was testing
Conclusion? i dont think this is an issue that can be fixed world wide. I may be wrong but i dont typically say these things without some shread of thought and evidence

 

actually now that i think about it. years ago an amd model which supported a lower watt mobile and dtr cpu (z80k) had a similar issue but different solution. z80k had a 90w adapter. it worked just great with a mobile amd as it did with a more demanding dtr. for a few months i could not figure out why in the world asdus did never aprove the dtr version of the same cpu even though it worked very well. I figured it out some months later when the rma for the 90w adapter sold with any dtr cpu started coming in while the same adapter sold with the mobile worked juswt fine. as it turned out the 90w when used with the dtr overheated and died while the mobile had no such issues. We replaced the 90w adapter for 120w for about 3 dozen customers over the course of the next year at our own expense as they were all out of warranty with asus. though this may be a very different problem, the cause remains the same, the adapter for the g1s is underpowered. the solution is much easier. code the bios so that while the battery is charging to 90% the cpu is only drawing half of what its potential. once it reaches 90% the the current to the battery slows down a bit and there is maybee 10 min left till full charge, let it go full throttle because they just gained a few watts back and the adapter will not overheat in sucvh a short time (90%-100%). i am not a major in this, (i cant even wire a simple jack into the wall, i cant even tell you why there are 3 wires into your house hold jack) but i can tell you i am fairly sure this was done on purpose to the g1s

 

Hi !! By the way you write it's simple to know that you know what your saying, but you have to believe on others people experience too.
This is only maths when it comes to electronics. Let's forget the g1s.

My father has a laptop with a 19v - 3.42A ac adapter and with a 4A battery.
This mean that the laptop while operating independant of charging or not the battery IT will never consume more than 3.42A, the hole computer. This means that the battery if in good condition have to last MORE than 1h, since it has 4A. And this is true. the laptop when using all of it power lasts for an 1.5 hours. Why ?? Because the laptop will never use more than 3.42A / hour.

The G1S and G2S are sold with a 90Watt adapter or the home page of asus is lying. Again, lets do math.

The ac adapter has 4.74A , so the laptop cannot consume more than 4.74A or it would be a faulty design ( and i do not believe this). The battery as 4.8A. So with full load the battery will last 1h. It's simple math.

So you ask HOW did i discover this. By playing a game, test drive unlimited with the laptop like it came from asus. No everest, no nothing. just power4gear that came pre-instaled.
In one day i have 20fps, then i had 30, then 20, then 30! and this was bugging me until I has again on the 20fps when my father unpluged the ac adapter and the 30fps came back !! It was when i realised that I had ALLWAYS 30fps when i was on battery and "ocasionally" when on ac adapter, because it was charging the battery

This was what happen. with the laptop just like it came out of the box. no drivers messing, no nothing.

Please stop saying that the 90W is not enough because the g2s bios does not have this problem on the g1s i'm getting tired of this, i just liked the problem solved or at least a oficial something from asus. I know that you have power to help us g1s users. Please, please do

thank you so much

 

Speaking of scientific methods, the scientific method asks us to replicate results under varying conditions as we attempt to generalize from a specific finding to a general population. I used Everest (the very newest version) just to check the CPU clock speed because what Vista was reporting was not jiving with how the CPU was performing (the CPU was performing like that prior to installing Everest). I used Everest because adolfotregosa had already used CPU-Z to discover the same thing. Two different utilities reported the same results. That is, I duplicated something using a slightly different method and found confirming results. It isn't proof, but it is evidence. So far, nothing has been falsified.

I had not heard/read that the latest edition of Everest was horribly flawed. I have only been a member of this forum for around 2 months, and I have not encountered that in what I have read. I apologize for not knowing the history of the flawed program and I won't use it anymore for any monitoring. Thank you for bringing this to my attention! I will agree that programs can cause very strange issues to happen on a computer that would not be expected. I work in IT, and we often will tell our customers that computers can have a mind of their own (and then we laugh reassuringly so that the customer knows that we are joking and does not think we are going to perform strange Voodoo rights to fix their computer).

One way or another, all adolfotregosa - and then later, I - have been trying to do is bring something to the attention of G1S users that we think is fairly important. The G1S is supposed to be an enthusiast laptop. Enthusiasts demand a lot from their laptops. Had I known this CPU issue prior to purchasing the G1S, I would have either looked for a different laptop, or waited until a solution was presented. When you spend $2000 for something, you want to get what the specs suggest you should be getting.

Thank you to both adolfotregosa and Eddie for investing the time and energy into looking into this issue... even if we didn't all always agree