Intel X3100 users rejoice! It's finally here! New Pre-Beta Drivers

hi, i'm newbie here.
i just got some problem to update my x3100 driver, i've tried this driver http://forum.notebookreview.com/showpost.php?p=5020310&postcount=1752
but the setup.exe file won't work, it says "the driver being installed is not validated for this computer. please obtain the appropriate driver from the computer's manufacturer"
and the same thing happens when i tried the official driver from here http://downloadcenter.intel.com/fil...+32-bit+version&lang=eng&strOSs=152&submit=Go!

is there anyone can help me?
now my x3100 driver version is 7.14.10.1329 which is the default driver when i bought the laptop

 

Download the zip file not the exe, then install it manually.

 

the official one?how to install manually? copy paste or what?

 

the official from intels website.

there are two files, an exe and a zip. the zip is intended for developers but is the same as the exe. Download the zip.

go to "device manager", right click on the correct device and choose "update".
then choose to install from a specific location, then choose "don't search". then click "have disk", then browse for the driver.


Edit: another way to do it is;
run the setup (exe file) and click next, a temporary file will be created in your temp folder. copy that before it dissapears. then do the same manual install through device manager.
this method is a bit harder so I suggest just using the zip file.

 

thanks, it really works.
i really want to try the moded driver, how is it? is it better than the official driver?

 

This link has some interesting information for those perhaps contemplating enabling a faster Core Render Clock:

In the case of VBIOS it's just more difficult since the BIOS can be customized quiete a lot. Have a look at crl_1588.bsf by opening it in notepad. It's included in vbios_gm965. You have to download the .zip version of the Windows driver. It includes some VBIOS developer information. At least for DOS it provides a facility to TSR the new VBIOS. Well - I know that's not very helpful.

You will also find (inspecting the .bsf file I mentioned) that graphics core clock is customizable by the OEM:

Combo $Render_Freq_Switch, " Enable Intel® Dual-Frequency Graphics Technology:", &No_Yes_List,
Help "This feature determines whether the Intel® Dual-Frequency "
"Graphics Technology (DFGT) is enabled or disabled. Intel® "
"DFGT is a graphics power savings technology that adaptively "
"lowers the render clock frequency when the system is running "
"in battery mode. Note that this technology is only active "
"when the system is running in battery mode and the LFP is the "
"only active display device. When Intel® DFGT is enabled, the "
"Default Render Clock Frequency should also be selected."

Combo $Render_Freq_Default, " Default Render Clock Frequency:", &Render_Freq_List,
Help "This feature allows the OEM to set the default render clock "
"frequency when the Intel® Dual-Frequency Graphics Technology "
"feature is enabled.

There is a Terminate and Stay Resident application to load the VBIOS to RAM (but is NO VBIOS update itself). But it's useless in Vista/XP:

crl_1588.exe /?

Intel(r)GM965/PM965/GL960 PCI Accelerated SVGA BIOS
Build Number: 1588 PC 14.34 01/24/2008 14:38:24
DECOMPILATION OR DISASSEMBLY PROHIBITED
Copyright 1999. Intel Corp. All Rights Reserved.

64K RAM Video BIOS.

There is also a generic VBIOS file crl_1588.dat. If you'd try to patch your notebook/desktop BIOS with this one, it may very likely break your notebook/desktop, since some important configuration data is missing.​

Request for X3100 driver with the vbios files

I've looked at the latest .ZIP file version of the X3100 driver and the vbios and crl_1588.dat, crl_1588.exe and vbios_gm965 files are removed. Does anyone have an older version of the drivers with these files? The customisation of the vbios for different Core Render clocks might give an idea of how to patch vbios for faster operation.

 

It's still contained in v15.12.4 drivers. E.g. here the Vista 32-Bit Version:
http://downloadmirror.intel.com/17408/a08/winvista_15124.zip

open vbios_gm965.zip and you'll find even updated VideoBIOS 1668.

 

how does updating vbios help? any dangers?

 

First of all: You can't. The only way to update the VBIOS ist by means of installing an updated system bios which contains the vbios.

Nobody here in this thread has the necessary tools to compile a VBIOS table, assemble a VBIOS and install it.

nando4 suggested patching the VBIOS, which is different from updating the vbios. Patchting the VBIOS is a crazy idea especially if the features, which you want to patch are basically unrelated to the problem you're trying to address.

Your notebook may never boot again. I mean: You switch it on and nothing happens.

 

oh ok, thx for the info 7oby=) my vbios is 1471.0 is it old?

 

Yes it's a little bit old. But that doesn't necessarily mean anything. You won't get more performance by updating the VBIOS. But if there's a more recent system BIOS available for your notebook, I'd install it.

There was once a problem with memory corruption on certain Dell notebooks ( click). This problem was fixed with a more recent VBIOS, but I'm not aware of any other notebook brand that was affected by the same problem.

 

I think I have an idea (Somehow) to install the vBios from intel.

My idea is simple which is attended to install freeDOS (Or from livecd) then running the TSR (The file provided by nando4) after that using grub4dos and hook it to windows/linux partition and boot. So in this way we can get the latest vBios under fully loaded operation system.

The way I am thinking of it is more of:

Code:

Boot loader (Grub) => Starts freeBSD => Autoload an .Bat file =>
Execute Intel TSR vBIOS => Execute grub4dos =>
(HOOKING to OS Partation to keep it loaded into RAM
*This is done without rebooting*)
=> Booting OS => Done!

Seriously I am not sure if this is going to benefit us for much (As I know usually the driver set anther vBios into RAM) unless we recompile our own vBIOS (Search for a file called IEGD_5_1_VBIOS.zip to get the general idea)


I will see if I get the time, I will buy new usb flash to try this on. (If anyone wants to try it first , be my guest )

 

I used FreeDOS 1.0 the fdfullcd.iso image:
http://www.freedos.org/freedos/files/

it is able to read FAT formatted USB-Sticks where you can drop your files such as the crl_1668.exe TSR app and grub4dos.

Once you boot the FreeDOS, you'll find your USB-Stick on C:\. You may launch "crl_1668.exe" TSR and it will indeed install the VideoBIOS TSR. Remember "crl_1668.exe" is a generic VideoBIOS which is not configured and may be unable to detect you notebook LCD. But if you attach an external TFT you should be fine (and maybe even without).

If you now try to launch grub4dos you will be presented this message:

Code:

Failure restore ROM INT 0x10 vector. Unsupported DOS, device driver or TSR.

INT 0x10 is the VideoBIOS interrupt.

 

The TSR vbios is nothing more than a TSR, it's meant for developers to mess around with, it does not load it onto your card, and I don't believe it will perform the same either.

If you can't get a BIOS update for your system with a new vbios included, then forget it.

 

Thanks for testing it, but still I will give it a try then I will tell if its worked or not.

Probably I wont be able to to do it as fast as I want Kinda today is a busy day for me as its the end of the week here. After finishing my University => Going to buy usb stick => Back to my family (5 hours driving) => Barbecue => Having fun with my friends

@Yakumo: Be a little optimistic

 

A TSR implies it intercepts perhaps bios INT 10 calls and reroutes them. A better way then. Does anyone know how to identify where the shadow video RAM is located? Would be easier to customise grub to overwrite the shadowed video RAM vbios with the file vbios copy, then load linux/windows and see how it responds.

BIOS modders can also disassemble the bios into components, including just the vbios, replace it, then flash a new system bios. Not recommended unless there is a bootblock recovery available.

 

Actually if I install the TSR and don't hook up an external monitor to my laptop, the screen stays black. Pressing Fn + F8 doesn't help. I'm lost.

Which proves my point saying: If you patch the crl_1668.dat into your system bios without having configured the VBIOS table (crl_1668.bsf) to match your notebook configuration, you may never see a display on your notebook again. Your only help then is to flash it back.

Last time I checked the VBIOS for the IEGD drivers is different from the VBIOS for the desktop windows drivers. While you can run IEGD drivers on a desktop VBIOS, the opposite doesn't work AFAIK: Running the desktop windows drivers on a IEGD VBIOS.

But maybe the package includes a useful softkick tool that allows to load an arbitrary VBIOS to memory and redirect all requests to this one. Basically a TSR, but windows compatible. However I doubt that.

Yes, that's what it does. Is it possible to "chainroot" launching Windows from DOS? Chainloading is to load the bootsector of the partition (512 Byte of the Windows partition) to memory and set the program pointer to this address. That's what the MBR does and that's also the way I launch Vista from grub4dos. I tried some MacOSX chainloaders, but basically they crashed my system.

The fundamental problem is this: A TSR program is resident and bends the interrupt vectors to point to the "new" VideoBIOS. However the concept of the TSR only lives in the DOS world. As soon as Windows boots it doesn't respect TSRs (although it might still use the redirected interrupt calls) and as such will overwrite the TSR memory space with it's own data. There is hardly anything that can be done to "protect" this area - unless you go with Windows 95. Shadowing as nando4 suggests is in this case the more promising concept.

Sees to be BIOS implementation specific. The Mobile 965 Express chipset documentation only mentions "memory shadowing" in general in section 4.9, but doesn't suggest address spaces here.

If you ignore the big warning one page before:

Code:

DECOMPILATION OR DISASSEMBLY PROHIBITED

it's possible. I don't know whether you ever dissassembled code. It's quite painful although I admit that the 64kb of the video BIOS is only roughly 400 screen pages in assembler.

One might go that road, however I think what GMABooster tries to do is more promising: Changing the Render clock on the fly. This requires unlinking the X3100 PCI Express device, which makes it more difficult once Windows has launched. I recognize that this didin't succeed yet, but if you put Linux on the notebook and connect to it by means of SSH you might be very well disconnecting and reconnecting your X3100 PCI Express device and measure whether changing the render clock does have some effect then.

And once that works, you might invest time to get this injected into Windows.

 

Vista 32-Bit v15.13.4.1829 dated 06/29/2009
Vista 64-Bit v15.13.4.64.1829 dated 06/29/2009

As it has been with the v15.13.3 drivers as well, those are useless for X3100 (GM965) users. Only useful for

Code:

*	Intel(R) B43 Express Chipset
*	Intel(R) G41 Express Chipset
*	Intel(R) G43 Express Chipset
*	Intel(R) G45 Express Chipset
*	Intel(R) Q43 Express Chipset
*	Intel(R) Q45 Express Chipset
*	Mobile Intel(R) GL40 Express Chipset
*	Mobile Intel(R) GM45 Express Chipset
*	Mobile Intel(R) GS45 Express Chipset

No changelog due to drivers from desktop mainboard section. Update : Changelog available

 

Is any way to improve (overclock) my X3100 on notebook HP 550?

 

Dual channel ram will give you a huge boost.
Also if you overclock your FSB you will see a nice improvement.
No way to overclock the GPU yet but it might be possible.
Edit: nando4s post on page 171 of this thread may give you an idea;
http://forum.notebookreview.com/showthread.php?t=128777&page=171

 

Overclock FSB is too hard for me, i will wait for GMABooster. Or i will try dual channel ram - is it really improve? With dual channel ram have X3100 better score in games?

 

Okay guys going throguh 70 pages of this thread im tired and i just wanna know which drivers suit x3100 better, the windows 7 ones or the vista ones?

 

I can't tell you which one if you are using windows 7 , but if you are using win-vista/win2008 then the modded driver worked so good with me here you can get it from here http://forum.notebookreview.com/showthread.php?p=5020310#post5020310 (Excuse me in that post , I had written better post ,but I accidentally switched off my computer than felt lazy to rewrite it all , that's why I made it brief)


To others:
Just in case if anyone still waiting me to post my test on replacing x3100 vBIOS , I had the same problem as 7oby had ,but I feel that I got a workaround for it (Only for Linux) - I can't keep trying on it anymore (Got an F on my midterm which means I have to spend more time on studying for my final)

 

I'm sorry for the "F". Can only get better. But feel free to post the workaround - we're all ear!

 

I am going to talk to my professor to see what he could do for me.
Anyway, the workaround is simply replacing grub4dos with loadlin (Or if someone know any other boot loader) and then by using vmlinuz to make linux booting from DOS (after applying our new vBIOS). For this I noticed loadlin is managed to start and show outputs but It manage to crash my system and making it reboot (This without using intel vBIOS!). Ok now after applying intel vBios, FreeBSD crash with "Illegal Instruction occurred" with some googling I noticed this http://www.mail-archive.com/[email protected]/msg05917.html

So from here I can conclude that there might be a problem in FreeDOS in supporting drivers and we need to find an alternative to FreeDOS and/or the loadlin problem is something related to a problem in my computer. For me I think the problem is more likely from FreeDOS.

I hope someone can keep trying to get this working as I will have to stop this project for some time.

From what I know he can't figure it out without a real x3100 hardware and also he might not be able to clock it more than 500 MHZ in theory.
After all , I am wishing him good luck in his project.

 

Good effort, but the DOS based vbios may not be beneficial. Why? The vbios loaded via a TSR redirects int 10 bios calls to the new vbios. Linux loaded using grub4dos will load it's own resources bypassing the TSR since that needs DOS-like programming, if it does do any INT 10 bios calls at all even.

I'm hoping GMABooster guy can figure out how to increase core render clock. I guess if yours is already at 500Mhz (see here to check) then really only way to get more graphics performance is with better drivers, dual-channel RAM setup, overclocking the FSB, or using a ViDock (preferrably $12US DIY version).

Though its questionably what performance difference GMABooster would make considering this tells us for a GMA950 EEE netbook going from 166Mhz to 400Mhz Core Render clock speed ( GMABooster):

It increases Windows 7 WEI by one point 0.1 for Desktop/Aero and three points 0.3 for 3D graphics.​

 

Can someone help me out with left 4 dead, I mean it run almost smooth but not exactly nice to play.

Is there some software I can use to lower the graphics quality?

 

Well, Its might be me. I just don't trust something unless I try it

http://www.3dfxzone.it/dir/tools/3d_analyze/index.php (Disable lighting + try few others)

 

thanks FrX, +rep just what I was looking for.

Dissabling lighting did help a little, but still not perfect yet.
The best way to play right now is if I kill my team.

 

I still don't get the performence I want. Everything is fine unless I'm looking at all of my team or a lot of enemies. When they are on screen I get ~6 fps.
With lighting disabled it is ~7fps.
When I overclock my FSB from 200mhz to 245mhz it is ~9fps.

Otherwise when I look at anything else it is ~35fps and sometimes ~40fps.

The only way to play it right now is with my team mates killed off so they don't come on screen but that makes the game not as much fun.

BTW FrX, I installed your driver and I tried the game without rebooting but there was no difference. Also after a restart there was no difference.

I messed around with all of the settings in 3d analyze but no other settings helped.

I guess for now I have to play with the low frames. It is sort of playable.

 

Win7 32-Bit v15.15.3.1855 dated 08/03/2009
Win7 64-Bit v15.15.3.64.1855 dated 08/03/2009

As it has been with the v15.15.0, v15.15.1 and v15.15.2 drivers as well, those are useless for X3100 (GM965) users. Only useful for

Code:

*	Intel(R) B43 Express Chipset
*	Intel(R) G41 Express Chipset
*	Intel(R) G43 Express Chipset
*	Intel(R) G45 Express Chipset
*	Intel(R) Q43 Express Chipset
*	Intel(R) Q45 Express Chipset
*	Mobile Intel(R) GL40 Express Chipset
*	Mobile Intel(R) GM45 Express Chipset
*	Mobile Intel(R) GS45 Express Chipset
*	Mobile Intel(R) GS40 Express Chipset

No changelog due to drivers from desktop mainboard section.

 

This is the main theme on Intel Graphics

I don´t want to overclock my CPU, I want to OC my graphics chip...

 

Anyone running Win7 RTM? If so, was there any improvement in X3100 performance. Eg: 'winsat dwm' versus Win7RC1?

 

No improvement, same as in RC for me..

 

Guys i just upgraded to Windows 7 and i downloaded the 15.12.75.1825 drivers and while playing Halo in 800x600 i get Black Bars on the left and right, is there a workaround to this or is there new drivers that fix this up?

Thanks in Advance.

 

Use the Vista drivers. No black bars and 'winsat dwm' gives better numbers.

 

Thanks, ill download the vista drivers, i hope they install okay since its Windows 7.

 

Vista drivers should work just fine on Windows 7. Windows 7 is basically the same kernal as Vista, so drivers that work in Vista work in 7.

 

Yeah i just installed the drivers and everything went fine, the black bars are gone .

 

yeah i have a clevo m72r system the small 12.1" with the 965 / x3100. just installed windows 7 and it updated the drivers and the screen went baserk. all odd colors and artifacts all over the screen. after a reboot it seemed okay but during the startup animation right when the swirls end before it goes to the welcome screen i get all dots all over the place. they look exactly like stuck pixels would look like but id say prob more than a dozen mostly at the very top edge of the screen but a few all over.

also my fan is running non stop and the temps went up a lot.

i thought maybe it was the newest one from windows update so i reinstalled windows this time trying the regular 1.1's that come with 7. same thing happens at startup but no where else. i went to the last vista driver and all is perfect.

my wei score is now 3.5 and 3.2 with the vista drivers.

the newest 7 ones were 3.9 and 3.1

i think they need to work on these there is for sure something wrong with them. ill stick with the vista ones myself for now..

 

Windows XP 14.39 and Windows Vista 15.14 drivers-GLSL 1.2 support and OpenGL 2.1.

 

where Windows XP 14.39? here I have always http://downloadcenter.intel.com/fil...+XP+Professional&lang=eng&strOSs=44&submit=Go! 14.37! please post the new drivers link

 

i wonder when intel is going to release a new driver for vista. It's been 5 months already.

 

and yeah, I only have the vista 15.12 drivers....is there a 15.14 out there somewhere or is 15.12.4.1666 the same thing. Numbering schemes always throw me for a loop

 

The latest out there is 15.13.4 for Vista. And no, 14.39 and 15.14 isn't out there yet. It's future.

 

v14.37.x : XP
v14.38.x : XP + Series 4 Chipset only
v14.39.x : XP Future

v15.12.[0-4] : Vista
v15.12.75.x : Win7
v15.13.x : Vista + Series 4 Chipsets only
v15.14.x : Vista Future
v15.15.x : Win7 + Series 4 Chipsets only

--

v15.13.4 doesn't help mooler he has X3100.

Yeah, based on this:
http://software.intel.com/en-us/art...n-intel-4-series-express-chipsets-and-beyond/

There's v14.38.x available all up to v14.38.7.5090 dated 08/03/2009, but that's Series 4 only. Situation may be similar with v14.39.x once it's available.

 

Oops, you are right. And to think I also said X3100 development will be phased out(I mean the new feature drivers, hotfix drivers will still come out).

Yea, 14.38 for XP and 15.13 for Vista and from then on are all for 4 series chipsets aka GMA 4500.

Sorry to disappoint. Maybe you can mod it or whatever but official support is over.

(I wish they do unified drivers the way Nvidia does. But maybe they thought the code would get too bloated. Anyways, X3000/X3100/X3500/X4500/4500MHD all use the same basic architecture so it should be relatively easy.)

 

Hello, I've been looking over this thread sporadically for some time. I have a lenovo tablet w/ the x3100 and 4 gigs of ram and 1.9 dual core on windows 7. I'm looking for the driver that will give me the most performance in the game unreal tournament 2004. I know the cards perfectly capable of handling it, even on the highest settings thanks to countless youtube videos and personal anecdotes. However, the driver I'm using with windows 7 will only get <30 fps regardless of my quality settings (gets the same no matter if it's on ultra high or ultra low) Perhaps one of those vista drivers would be better suited for this?

 

Well, First I got to say welcome (I know ,but I had to say it)
as you were reading this thread, I am more than sure you know that you have to try few drivers (modified or not) and with small benchmark you can determine (I posted before about PreformanceTest benchmark which is fast with benchmarking (only select 3d part) in less than 5 mins you would be able to determine which driver is good for you) also you might have the strange glitch of boosting your fps if you installed the driver without testing.



For those of x3100 users who are still thinking if either they should move to win7 or not please read forward...
I had the pleasure of using win-2008 server - stranded for long time which in simple few words, it is the same as vista-sp1 and using its kernel. Moving from win2008 to win7 I had to do some benchmark regarding its 3d GFX (especially intel drivers). I did the benchmark on Win7-Ultimate (RTM) with using 3dmark0 , 03 and PassMark PerformanceTest on x32 and x64 version of win7. The result were actually no gain of performance between x32 and x64 in 3D with one more thing that really bothered me. My performance was less than win2008! Moving from win2008 to win7 decreased my performance (I tried the benchmark with aero and without) I also tweaked win7 services all with no hope. The only thing that I noticed was the updated win7 drivers from Intel helped to increase the FPS (5~9) more. Nothing else can be mentioned. Modifying latest 4-series drivers to work with win7 causes can help (but not much), there is a bug have been mentioned before which is the strange color bug on modified drivers on win7 (using win7 driver not vista) I didn't investigate more about it , but I noticed the driver get back to normal after 5 mins (strange colors stopped) the 3d card couldn't work with 3Dmark but was fine with PerformanceTest. If I remember right , I think I didn't delete screen-shots of 3dmark benchmark but as both are smiler its a waste of time.
I didn't get so well with win7 and x3100 ,so I didn't test in any real game (I was more of disappointed of intel in their latest driver in linux which has dropped alot in performance which I can excuse Intel in this one, they do it for good , but win7 with x3100 is more of dead to me)

 

I'm not surprised you get better performance in Server 2008 than Windows 7. Server 2008 comes with most things turned off by default allowing the user to turn on ONLY the things he/she wants/needs. Windows Vista (and 7) has most everything on by default, which uses up resources. Server 2008 is super snappy for me, but I don't use it for gaming.