Keyboard Lag Debugging

Time for a new problem, I'm sick of looking at Hairy Donuts.

I did some research on this problem, and I found this information on the ATI Hotkey Poller:

"Atiedxx.exe and atiesrxx.exe are the AMD External Events Client Module and AMD External Events Service Module. Both processes are executed when the Windows Service AMD External Events Utility is started. This service was known as the ATI Hotkey Poller in Windows XP and Windows Vista."

Now, I don't know about you, but I sure don't use any hotkeys with CCC, so I disabled both processes, and typed up a nice long document and noticed the only mistakes in the document were my own typos. I noticed a very noticeable drop in dropped characters, so I wanted to share the love.

For a more permanent fix and more info on what ATI installs with CCC, see this link.

Have fun experimenting and let me know if this helps.

 

I for one appreciate you flashing the brains out of your laptop and running Hairy Donut Tests for us. I notice keyboard lag right away. I will disable and test tonight. I'll report back.

 

Interesting find Chastity.

 

One thing I have noticed since day one is that, while I occasionally miss keys, if I lightly press individual keys on the corners they will consistently not register. It's very hard to distinguish between which of these keystrokes are missed because of hardware and those where I didn't press the key hard enough.

 

Wow, nobody has any feedback on this yet?!

 

I disabled the external events thing. I never noticed keyboard lag before and I don't notice any now.
I type at over 200 WPM on a normal day.

 

I still notie the keyboard missed keystroke issue and typing 30-45 wpm. I have to type at 10-15 wpm or it to not miss keystrokes. Honestly, I think this is more important to fix than the GSOD issue since the GSOD issue is patched by old drivers

 

Uninstall the synaptics drivers and you'll be rid of the lag for the most part while still having a working (albeit very unresponsive and "dull") touchpad while this is worked out.

More important then the lag is the fact that it registers all keystrokes without the synaptics drivers.

 

then the mouse speed is awful

 

Well I'm ordering a keyboard under warranty so lets see if that fixes this problem

 

If you have the same issue most of us have, it's a software problem and not solvable by hardware as far as I know. ASUS does know about it and are working on it though .

 

For me, the results are the same as Kaelang's. Never really had any issue with lag or dropped keystrokes. One thing to note is that I did not have Atiedxx.exe running, only atiesrxx.exe and atieclxx.exe, no idea what the second one is. Disabling atiesrxx.exe ended up with no difference for the keyboard.

 

Stopping the Atiedxx.exe and stopping the ATI event service did nothing to help with it keeping keystrokes... ie, I still am losing characters...

 

Chastity, I am still missing keystrokes but it is definitely less. When I first downgraded from 211 to 209 (the touchpad was enable by default) last night to do the vBIOS that you posted, the missed strokes were horrible. Now they are just slightly annoying. I will leave the touchpad enabled for the work day tomorrow. I type a lot at work so I will know more by the end of the day.

It's a good "thinking outside the box" attempt even if it doesn't work, but I think it is helping.

UPDATE: I did several tests with the touchpad soft disabled, AMD service on/off, touchpad on, etc. I'm pretty sure that it's not helping. I'm still missing keystrokes at a certain rate. It's not horrendous, but its annoying. I can't leave it like that tomorrow or all my IMs and emails with co-workers will look like I spell like a second grader.

 

Uninstalling the synaptics driver does not solve the problem for me. Removing the ribbon cable and gently scraping the black stuff off the contacts doesn't solve it. The latest synaptics driver doesn't solve it. And finally disabling the ATI hotkey service doesn't seem to solve it either. However this latter suggestion does seem to reduce it greatly if I disable the touchpad with the function key while typing posts here. It's still there, but it's at a low enough frequency that it will make you think you are making typos when that is not the case.

 

I can also confirm im still missing keystrokes with this. Thanks for trying Chastity but i think well need either a new keyboard or new synaptics driver...

 

I don't think that changing the keyboard will fix the lag. Asus has long been known to have keyboard lag problems ever since the M50 series. Have found a permanent fix to that. Hopefully this will also work with the newer notebooks.

http://forum.notebookreview.com/4038483-post97.html

 

I can barely type 50 WPM....

Ya someone try it out.. but we'll loose functionality of synaptics driver.. Asus needs a driver update soon... Keyboard lights fixed so this next as far as keyboard goes .. we're halfway there..

 

FYI. I have installed the latest synaptic drivers from the synaptic website (just installed on top of Asus) they work well for me, no lagg, no issues and I get the two-finger stuff.

 

I don't recommend turning this service OFF. If my memory is correct, in the past I shut this service down and the fans of my x1900xtx would not turn up when needed.
This could be totally unrelated but just sharing a very old pas experience I had with this service.

 

The fans are controlled through BIOS. I'm not finding that the service in question helps or hurts the keyboard problem so I turned the service back on. I'm always up for some experimentation to get rid of the missed keystroke problem, though! I started looking around for any other possible solutions last night after this didn't work and I came up with nothing.

 

Uninstalled Synaptics and keyboard is fine now. The drivers don't do much other than have that gestures crap which I don't really use. I can wait till a proper working ASUS Synaptics driver is out. Keyboard is so nice without the lag.

 

The drivers give you the "awesome" ability to turn the touchpad off in windows (which cant be done in any way without them), so you dont get your mouse jumping all over the place when your palm is resting on it preventing you from typing, so this solution is really awful, at least for me... I rather just disable the touchpad in bios entirely and am at least able to type normally but thats also a pain in the rear when i actually want to use the touchpad again...

 

I just have touchfreeze with it so it's fine... Mouse moves no problem but it doesn't click .

 

That still isnt really a solution, but if youre satisfied with it good for ya .
Im waiting for Asus to release an updated ATK or Synaptics package. Until then touchpad stays disabled in Bios.

 

I'm pretty sure all the problems of the keyboard and touchpad came with the BIOS update. I received my unit with 206 from factory. Upon received, I installed the touchpad driver 15.0.0.9 (non asus) to enjoy the two finger scroll feature (not available in asus package.... do not why!?!?!). everything was fine until update to BIO 209 when the problem appears, but, since I used to use mouse, the touchpad was not very noticeable. Now I'm using only the touchpad but I can't go back since I'm in vBIOS v93 with BIOS 211.

I just tested many different touchpad driver packages (from asus and others), with two finger and without them, but the problem persist.

 

Can you keyboard laggers try this:

http://dlcdnet.asus.com/pub/ASUS/nb/Drivers/KB_filter/KBFilter_WIN7_64_1003.zip

 

The big issue here is that disabling the touchpad in BIOS is the ONLY 100% fix for everyone...

Some systems work fine on stock drivers, others are fine if you uninstall Synaptics, yet others (like mine) have issues even with synaptics uninstalled...

Only disabling in BIOS is the fix that works... which isn't a fix because I bought a laptop to use the touchpad, not a mouse.

I have a sneaking suspicion that this problem will be ignored on the G73 like it has been on countless other models... Too much liability if they figure out they have a hardware issue. Ignore the problem and it will go away, right?

 

I will try it now
. Thanks!

 

what's new in this package, it's the same version that came in the dvd drivers ana already posted at asus website.

 

I'm curious to see who is using it, and who isn't, and what results. I know I wasn't since I'm on a fresh install.

 

I'm using since the very first time in my fresh install, 'coz is a controller installed by the drivers dvd install

 

Is the only reasonable solution to this keyboard lagging problem disabling the mousepad via the BIOS? I'd rather not do that because /sometimes/ I want to use it just not very often and the keyboard lag is pretty much irritating as hell with the touchpad on.

Thanks.

 

As it stands right now, the best recommendations about fixing input lag is to first try uninstalling the Synaptics drivers and give it a go. This has resulted in less-than-stellar performance on the trackpad (and its less-than-stellar already). The only known 100% fix it to disable the trackpad in the BIOS and stick with an actual mouse.

 

I'm getting pretty desperate with the touchpad/key lag issue. Today I formatted my hard drive and started over. I tested for key lag at different points while loading it back up to see if there was any software incompatibility that caused it.

Over the months that I've had the laptop, I've loaded so many different keyboard filters and touchpad drivers that I really needed format to get to a ground state.

The short version is that Asus synaptics drivers cause key lag on my laptop no matter what I try. I am using the standard MS driver with Touchfreeze right now and I miss zero keys. However, the touchpad can be pretty erratic without the driver. The one annoyance that touchfreeze doesn't solve is the scrolling area of the pad. It doesn't "throw away" the scroll input so while I type, the text box is constantly scrolling up and down.

I'm going to download the source code for touch freeze and see if I can add two features... the ability to ignore touchpad scrolling and the ability to turn off the mouse with some sort of key combination.

In effect, I'm going to try to roll my own solutions because I have lost faith in Asus/Synaptics solving this problem. I know that Gary is/was trying, but I'm thinking they don't even know what is causing it. It seemed that Gary had to really dig to find a machine that would reliably show the problem. That is very odd to me since my survey showed at least half the respondents miss keystrokes.

I'm tired of having to dig my mouse out of my bag every time I want to surf the internet for fifteen minutes. Then I've got to smooth out the sheet ont the bed if that's where I happen to be or the optical sensor will go goofy if it passes over a wrinkle. Or I'm on the couch mousing on the arm of the sofa and every time I reach for my Diet Coke my $100 mouse slides off the arm and hits the tile floor. This is a crappy problem to have for six months with a pretty expensive laptop. I can understand a laptop having problems with a peripheral mouse and third party drivers but when it's the built-in pointing device with factory drivers it is ridiculous.

Sorry for the rant. I'm sure most people won't even read it, but I feel better now.

 

You seem to be a candidate for a Darkfield mouse. I was using one yesterday on a cotton hoodie that I threw on top in the passenger seat. Never missed a track.

 

I agree.. this problem has gone too long.. have u tried hp synaptics drivers? They seem to work for me although i nver had lagg..

 

I am using the Logitech Performance MX Darkfield as I type this. I love that mouse. However, I would like to be able to leave it in my bag sometimes when I am not going to be programming/gaming/designing circuits. If I'm just going to check my favorite G73 forum and the news, I would just boot up on battery and use the touchpad for half an hour... if it worked.

I have not tried any HP drivers. At least, not on purpose. I have downloaded just about everything that everybody says works and nothing has solved it. Chastity used a 15 point something. I found that on Asus but it was for a different laptop. I tried it and it still missed keys. I tried the ones on Synaptics and I just love those! I like the two finger scrolling, but I miss keystrokes. The stock ones and updated stock ones don't help either.

 

What i have tried:
- Disabling touchpad in BIOS: Works, but obviously not a solution.
- Uninstalling Synaptics drivers: Works, but again not a solution.
- Uninstalling ATK package: Works, not a solution either.
- Installing every different ATK seperate driver or older/newer package, Keyboard filter, every Synaptics driver i could possibly find (Stock, HP, Dell, Clevo...) : Does not work.
- Formatting and doing a fresh factory install or doing a Windows 7 professional one i have: Does not work.

 

This problem is obviously a "margin of error" problem just like the GSOD error was. Ie, some people have it, some people don't, some worse than others.

Hopefully Gary and his group of problem solvers will take this as a serious problem and give it as much attention as they did the GSOD issue.

If it is a hardware problem, great, provide a fix and work out a proper streamlined RMA experience as some of us will refuse to use the "normal" RMA channels where equipment is constantly damaged and abused. Nothing worse than sending your machine in for a problem only to get it back with damage or worse, more problems.

If not, then kick the software engineers in the pants and tell them to start doing thier homework and build in for a margin of error (which always HAS to be done on "real-time" eventing like input) and get a fix ASAP. (I have been a systems engineer in real-time manufacturing, warehousing/distribution and IT systems for almost 20 years, so I have some actual experience in this area, as the problems seen here are when people go on the "theoretical" specs of how to accomplish something more than the "applied" aspect of how it actually works)

 

Disabling the touchpad in the bios doesn't seem to fix my keyboard issues. the spacebar doesn't always strike no matter what options I configure. I've debated disassembling it and trying to lubricate some of the components to see if that makes it work better, but I don't want to ruin it either...

It feels to me that the BIOS just needs a tweak to increase the Keyboard voltage.

 

I don't have keyboard lag. My G73 was a refurbished model, so there's something that used to be wrong with it, but since I've had it nothing has ever gone wrong. I'll probably/hopefully never know what used to be wrong with it.

I'm just wondering though, if this was a software issue, wouldn't an external keyboard be affected by lag too? Can someone with keyboard lag test that?

On a side note, when I first install ati drivers, I always turn off hotkeys in the Catalyst window and in windows services.

 

I doubt that an external keyboard would be involved in any way with this issue. The internal keyboard uses a PS/2 interface and any external unit would be USB. External keyboards do not miss keystrokes to my knowledge. It is only the internal keyboard and only when the touchpad is enabled and using a Synaptics driver. Interestingly, removing the ATK software will also alleviate the problem. Also of interest, I've done some research and the ATK package is not just found on Asus laptops. Other brands of laptops also use this software to operate media keys and such. The ATK executables are signed by Asus. Either Asus makes this package and it goes out with keyboards they make for other laptop manufacturers or Asus has recompiled the package after modifying it for the keys on this particular model.

 

I don't really care who 'made' the problem, but simply put there is a problem. I hate having my touchpad disabled in the BIOS as I do now cause if I'm just sitting down to check something real fast I don't want to have to locate my mouse and go through the process of turning it on just for that 2 minutes Im browsing the internet. I love the keyboard without the lag tho.

 

None of us care who made it. I'm trying to come up with theories and disable/enable services, modify process run time priority, tweak settings, or whatever we can come up with as a workaround. It's just a generic discussion to try and promote ideas. I just found a C# example on how to hook keyboard and mouse input like Touchfreeze does so we can inspect what is coming into the laptop. Maybe we can determine which app is throwing away keystrokes.

I downloaded the source code for TouchFreeze, but it's written in C++. I do manage code written in C++ but I don't work with it all the time. It will take me until the cows come home to unravel that app and bend it to my will. I'm a lot quicker in C# so I'm thinking of throwing together an app that counts keystrokes and then counts input characters on the screen. Perhaps knowing if there is a discrepancy in the numbers will help. If we hook keyboard input globally and count every keystroke, then compare that with what gets passed to a text box, we can determine whether the missed keys are even making it into the windows operating system. If they are not "hookable" globally, there is nothing we can do. If they aren't getting to windows then there could be some sort of electrical problem where having both PS/2 items powered up on the bus causes a under-voltage or under-current situation. If they are getting into windows but not making it into a tex box, then it is definitely a driver problem.

 

That's close to impossible.

If you are telling the truth my friend, then you must have absolutely amazing fingers.

 

That sounds like a lot of work... Let us know what you can come up with .

 

You do understand that C# is a higher level abstraction of C++, thus there is more latency inherent in the number of libraries required. C++ is damn near assembly (and considering you CAN put assembly code inline with C++ code and it will compile native) you have less latency within the code for every command executed.

Here is the key... It is not "voltage" that is the issue. It is a polling event by the ATK package. Synaptics is run on MANY MANY systems all without problems. Just because you can disable the trackpad in BIOS and have the problem go away does not mean it is that component. Two components sit on that input channel, the keyboard and the trackpad. BOTH have to share the input (exactly the same as it was with PS/2 keyboards/mice, they both shared the input), thus if the ATK poller is dropping messages or mangling them, you have problems on both inputs, and many of us DO have problems on both the keyboard (missing characters) and trackpad (not responding to inputs properly).

It is more than likely going to take a BIOS revision and a ATK package update to fix this. Synaptics is not the issue, just anotehr symptom of the same problem.

 

Umm... yeah. I'm pretty familiar with the differences between C++ and a .Net language. This code sample demonstrates using C# to hook keyboard and mouse input via the Windows API. If one hooks the native API call, it is identical to the method used in programs like TouchFreeze.

The ATK package is not exclusive to Asus laptops. It runs on MANY MANY laptop models made by other manufacturers as well. Since we don't have the ATK source code, none of us has a way to determine whether or how often it polls the keyboard.

The ATK package has no problem reacting to keyboard input in the absence of the Synaptics driver. It's only when both are installed that the problem arises. To further complicate the matter, apparently there are some laptops where the two work perfectly well together and others where they do not.

I suspect that the ATK package is the culprit myself. That is the purpose of the exercise. No one said their was a "voltage problem" with the laptop. I said that if the keystrokes are not making it into windows then the problem is between the operating system and the hardware. This could be for any number of reasons including BIOS, PS/2 bus, trackpad design flaws, keyboard design flaws, etc.

 

I wrote a little test program, but it looks like the windows hook catches the input from the keyboard only after it has passed through the drivers. Therefore it is useless for determining where the keystrokes are lost.