Jump to content

Anyone mobile Kepler 6xxM working?


gothic860
 Share

232 posts in this topic

Recommended Posts

Thanks for the answer but you still keep saying that I have both cards. I don't, please look at the G55/75 specs, there is no idle switching to Intel.

But if I do, why do the older gen G series work? They too must be, as Micky1979 said:

 

 

Please, I'm not trying to argue with you, I just refuse to give up that easily, even if what I do is pointless lol

Yep, maybe this is an insurmountable problem

 

Micky

Link to comment
Share on other sites

Here another Alienware running with qe/ci

 

here goes the darwindumper and ioreg: http://www.sendspace.com/file/27tho7

.

 

He is using Macbook Pro 5.1 which doesn't have any Intel GPU on it, I´m not very familiar with intel GPU´s but I know it need to be initialized in some way and one is based in the model from smbios. If your are using Macbook Pro 10.1 as your model, this model has an Intel GPU and an Nvidia GPU and the first in been activated is the Intel GPU, and it is the same in your system but ohhh!!! you don´t have any active Intel GPU, and for the Optimus experience we know that an Nvidia Card can´t be activated as a second one (I have read some theories about how optimus work). For the other side if you try to use your Intel GPU like any other Optimus user you can´t because there is not active.

 

Try using any other smbios like iMac13,1 or 13,2 or MacPro or simply use Macbook Pro 5.1 like he does, I know you will lost CPU Power management but is just for testing purpose. (Use NullCPUPowerManagemnt for avoid KP)

 

If you try iMac smbios use this "Mac-FC02E91DDD3FA6A4" as your smboard in smbios.plist this.

 

My two cents

Good Luck

 

Sorry for my poor english

  • Like 1
Link to comment
Share on other sites

Thanks for the suggestion but this made no difference for me, Chameleon gets it stuck and Clover reboots...

 

 

What exactly did you do? What changes? Please be more specific.

Link to comment
Share on other sites

Sorry, here's what I did: I used an iMac13,2 smbios.plist to boot with Chameleon, then an iMac13,1 smbios then a MacPro5,1 one. They changed nothing, it was still stuck after NVGK100hal etc etc. I also tried those smbios files with Clover, which had the same behavior as before and that is rebooting instead of getting stuck. It seemed to me the smbios doesn't influence anything in this case.

 

If I were to take a guess, I think the problem lies in the DSDT. How can an Alienware R4, not 3D, with Optimus that can't be controled, boot just fine and actually use the discrete NVIDIA when using iAtkos ML2? That's just a mystery...

Link to comment
Share on other sites

.

 

He is using Macbook Pro 5.1 which doesn't have any Intel GPU on it, I´m not very familiar with intel GPU´s but I know it need to be initialized in some way and one is based in the model from smbios. If your are using Macbook Pro 10.1 as your model, this model has an Intel GPU and an Nvidia GPU and the first in been activated is the Intel GPU, and it is the same in your system but ohhh!!! you don´t have any active Intel GPU, and for the Optimus experience we know that an Nvidia Card can´t be activated as a second one (I have read some theories about how optimus work). For the other side if you try to use your Intel GPU like any other Optimus user you can´t because there is not active.

 

Try using any other smbios like iMac13,1 or 13,2 or MacPro or simply use Macbook Pro 5.1 like he does, I know you will lost CPU Power management but is just for testing purpose. (Use NullCPUPowerManagemnt for avoid KP)

 

If you try iMac smbios use this "Mac-FC02E91DDD3FA6A4" as your smboard in smbios.plist this.

 

My two cents

Good Luck

 

Sorry for my poor english

 

Thanks for the reply, I will try it here.

 

just for adding, I think I discovered some conflicts on IQRs and hardware addresses, see the files in the zip:

I put the files of a notebook with working acceleration and the mine (iddpioneer) for compare purposes

http://www.sendspace.com/file/9xgbk6

Link to comment
Share on other sites

I'm going to start with the basics for a second... built USB installer from retail app using a "mainstream" tool that I suppose isn't supposed to be mentioned here, but I'm sure it doesn't need to be. Lol.

Installed 10.8.3 update to that same USB installer, so I could avoid having to update during one of many install/mistake/re-install phases.

 

Computer is a Toshiba Qosmio X875-Q7390 (info here http://us.toshiba.com/computers/laptops/qosmio/X870/X875-Q7390/)

 

What is strange though, is I managed to get my nVidia GTX670M recognized in "About this MAC" without even trying. I ran the myHack (all options one at a time) no added kexts or tweaks yet, and all of the sudden my Graphics card shows up, with a system profile of "MacBook Pro 8,1" I will post a picture in a second. What I'm wondering, is it enough for OS X to be able to identify my card? What can I do now to ensure it will be fully supported with QE/CI assuming (and it seems like it isn't yet) that there is more to be done. Everything on my laptop is working fine OOB aside from Wi-Fi (intel wifi card isn't supported) Graphics (QE/CI), though I'm wondering if this theoretical quirk can count as progress, and sound (I use the Voodoo driver for that, so I'm not worried about it too much) everything else seems okay.

 

I did have to alter my installer though, removed the nVidia Resman kext from S/L/E and I make sure to do that on the system if and when I update.

 

I'm wondering if something in the myHack pack injected the correct information for my card into the drivers, and if the resman kext were in place before doing that again, if that would therefore also alter that kext to make my card fully supported.

 

I don't know. Thought I'd run it by you guys and see if I'm just dreaming, or if there is potential. Here's the screen shot as proof.

 

http://bit.ly/10OXtZj

Link to comment
Share on other sites

I've got Lenovo W530 with Kepler Quadro K2000M on board and also have stuck or reboot after DSMOS has arrive message in log. Can't boot with Kepler at all even in save mode. And don't know what to check next

Link to comment
Share on other sites

I've got Lenovo W530 with Kepler Quadro K2000M on board and also have stuck or reboot after DSMOS has arrive message in log. Can't boot with Kepler at all even in save mode. And don't know what to check next

 

Josh, if you're getting a black screen... the easiest way around it is to open up your drive (or if you haven't gotten install to work yet then open the usb drive) from the drive root go to the folders system/ library/ extensions. Delete nVda resman.kext and your black screen problem should go away.

 

I'm trying to figure out why my GTX670M became recognized out of the blue (and to isolate what I did to make that happen) my hope is that the answer to that will bring us some ideas or possibilities towards getting QE/CI on some of these discrete mobile cards

Link to comment
Share on other sites

I'm going to start with the basics for a second... built USB installer from retail app using a "mainstream" tool that I suppose isn't supposed to be mentioned here, but I'm sure it doesn't need to be. Lol.

Installed 10.8.3 update to that same USB installer, so I could avoid having to update during one of many install/mistake/re-install phases.

 

Computer is a Toshiba Qosmio X875-Q7390 (info here http://us.toshiba.co...870/X875-Q7390/)

 

What is strange though, is I managed to get my nVidia GTX670M recognized in "About this MAC" without even trying. I ran the myHack (all options one at a time) no added kexts or tweaks yet, and all of the sudden my Graphics card shows up, with a system profile of "MacBook Pro 8,1" I will post a picture in a second. What I'm wondering, is it enough for OS X to be able to identify my card? What can I do now to ensure it will be fully supported with QE/CI assuming (and it seems like it isn't yet) that there is more to be done. Everything on my laptop is working fine OOB aside from Wi-Fi (intel wifi card isn't supported) Graphics (QE/CI), though I'm wondering if this theoretical quirk can count as progress, and sound (I use the Voodoo driver for that, so I'm not worried about it too much) everything else seems okay.

 

I did have to alter my installer though, removed the nVidia Resman kext from S/L/E and I make sure to do that on the system if and when I update.

 

I'm wondering if something in the myHack pack injected the correct information for my card into the drivers, and if the resman kext were in place before doing that again, if that would therefore also alter that kext to make my card fully supported.

 

I don't know. Thought I'd run it by you guys and see if I'm just dreaming, or if there is potential. Here's the screen shot as proof.

 

http://bit.ly/10OXtZj

Well I don't mean to burst you bubble haha, but the 670m is Fermi so other people have got it working :P But thanks for sharing that with us, glad you got it working!

Link to comment
Share on other sites

@TeoTheCool I've been googling around for around two weeks, trying to figure out how to get that "card" working... Still no QE/CI, trying everything here. If you happen to know any threads with good advice on that, I'd appreciate it. I'm coming up pretty empty handed (maybe not looking in the right places) anyhow, I realize my card is fermi, but have been having the same experience with mine, as you guys seem to be with the keplers. It's wasn't even detecting it until this morning.

 

Sorry if I've inadvertently hijacked, or diverted this thread... Wasn't my intention. Good luck with your card.

Link to comment
Share on other sites

Anyone tried the 10.8.4 beta nvidia kexts yet? Version number is 310.20.... Was hoping my Kepler K3000m would work, but I still get a gray screen on boot.

 

Just the nvidia kexts can be found in this thread:

http://www.insanelymac.com/forum/topic/250849-guide-mammoth-method-v200-hp-dvx-106x-fast-easy-comprehensive-installation-new-mammoth-boot-cd-for-i5-i3/page__st__1420

Link to comment
Share on other sites

I also tried the beta for my 660m GTX without positive results. I would say that no need to hope something from the incomong updates, the drivers should work with our hardware but the architecture is different, so we need to edit the binaries/DSDT to get it work.

Link to comment
Share on other sites

In your case if you get gray screen after apple with load spinner you can try to force load vbios to card. In Radeon world loading vbios helps to boot into guy without this crappy grey screen.

 

I've also remember one osx beta update with updated nvidia drivers after installing it I've got same gray screen and OSX cant boot into desktop...

 

Anyone tried the 10.8.4 beta nvidia kexts yet? Version number is 310.20.... Was hoping my Kepler K3000m would work, but I still get a gray screen on boot.

 

Just the nvidia kexts can be found in this thread:

http://www.insanelym.../page__st__1420

 

Tested 10.8.4 on Thinkpad W530 with K2000M and still reboot after loading

 

I want to try DSDT injection and try to load Vbios from original macbook cards hope this may help or not...

 

In my case Kepler is important cause I can't use any external monitors on my Intel HD4000

 

And strange that in previous generation of Thinkpads W520 Quadro 1000/2000 cards loads fine in time when osx start support desktop fermi... but no portable macs with fermi cards was released (2011 gen used AMD Radeons...)

 

So situation is strange with current gen of macs and pc analogs

 

 

And there is small investigation about Keplers and Current macs with it.

 

Macbook Pro 9.1/10.1 support keepers from OSX 10.7.5

 

GT 640M - 10de:0fd8 GK107

GT 650M - 10de:0fd5 GK107

 

iMac 13.1/13.2 support OSX from 10.8.2

 

GT 640M - 10de:0fd8 GK107

GT 650M - 10de:0fd5 GK107

GTX 660M - 10de:0fe0 GK107

GTX 675MX - 10de:11a2 GK104

GTX 680MX - 10de:11a3 GK104

 

Desktop keplers stats fine with GraphicsEnabler=No.

 

Mobile keplers hands after DSMOS has arrived in log and only removing of NVDAResman.kext may help to boot into GUI.

 

But in IOREG we got card detected with all connectors and configs only CI/QE is missing...

 

I've also managed that macbooks has GMUX device (don't know about all iMac models) that mux Intel and nvidia gpus and possible we should mimmic our cards to desktop ones to bypass MUXer redirection I think

 

There is some pc and mac kepler bioses and ioregs in attachment.

 

Also I've got macbook pro 15" retina with GT 650M and can help with any logs and dumps.

 

If someone have a suggestions I may help with it on real macbook retina.

kepler-bioses.zip

kepler-ioregs.zip

  • Like 1
Link to comment
Share on other sites

Don't know is it good or not but I've got some progress with Quadro K2000m on my Lenovo W530

 

OS still doesn't boot into GUI on laptop screen, but I can login and look desktop via VNC.

 

Anyway desktop is not respond on mouse gestures and show BBOD on mouse cursor.

 

post-120553-0-93457600-1366384567_thumb.png

Link to comment
Share on other sites

  • 2 weeks later...

GTX 670m is FERMI card so it should work with current drivers or so... to fix freezes you should add your card to AGPM into your macmodel definition

 

Hello, people. I have a GTX 670m GPU working QE / CI but I freezes especially when they change states. any solution? Thank you!

Link to comment
Share on other sites

GTX 670m is FERMI card so it should work with current drivers or so... to fix freezes you should add your card to AGPM into your macmodel definition

 

Hello, Josh! I am using a MacBookPro LegacyAGPM and SMBIOS 9.1 and still have freezes. If I run OpenGL Extension Viewer 3x white screen and freeze. This should be changes States. Can you help me?

 

Attached is my darwindumper and LegacyAGPM - http://www.sendspace.com/file/pvcxeh

Link to comment
Share on other sites

@joshhh - I have tested some of the vga bioses that you´ve uploaded but chameleon says: the bios file is too big to load, and then I extracted mine and the size is 64k and yours are about 90k, there´s a way to force chameleon to load it in another addres of the memory ? Or a way to downzise it ? Because I think this is the way to make our Mobile Kepler working on a hackintosh !

 

Thanks in advance.

Link to comment
Share on other sites

  • 2 weeks later...

Hello guys!

Lets try to figure out how to start mobile Kepler chips in PC laptops. Since mobile Fermi cards works well mobile Keplers should work well too because it works in latest MacBooks.

But in case of PC most of tries stops after DSMOS has arrived or simple reboot after that. Laptop models with GK107 and GK104 affected. Example Quadro K1000m/K2000m, K3000m, GT650 in Lenovo/HP/Dell/MSI laptops with ability to set nvidia GPU discreet or without any Optimus tech {censored} at all with only discreet GPU enabled (Like in my current Lenovo T510).

I've got Lenovo W530 with mobile Kepler Quadro K2000m and also got macbook pro 15" retina with GT 650M and can help with any logs and dumps.

My Quadro K2000m also doesn't start and hang after DSMOS has arrived message.

My investigation lead me through some "working" configs.

1. Remove NVDAResman.kext let OSX to boot into GUI but lack of QE/CI. IOReg contains configured GPU section with all settings and all available outputs.

2. Don't remove NVDAResman.kext and boot into save mode stuck before entering into GUI but I can connect and enter to desktop via VNC or Remote Desktop. IOReg contains configured GPU section with all settings and all available outputs.

3. Remove NVDAStartup.kext let OSX boot into GUI but lack of QE/CI. No gfx section in IOReg.

In each case I've made DarwinDumps from Lenovo W530 and also made dumps from my MacBook Retina.

Also I've extracted some Kepler vga bioses from latest MacBooks and iMacs and some PC laptops including my Lenovo W530.

All this case reminds me situation with first gen of Intel HD Graphics in Macbook Pro 2010 models. In HD Graphics case it was fixed with small patch of frame buffer.

But in case of Keplers I really don't know where to dig - nvidia drivers, card configs, vga bios difference or something else.

Some files below.

DarwinDump from RetinaBook

http://yadi.sk/d/uZCUstTp4t8nE

DarwinDump from Lenovo W530

http://yadi.sk/d/_kshhugl4t8r0

Kepler bioses extracted from bios files

http://yadi.sk/d/JUYiHng64t8mA

Kepler bioses extracted from working machines RetinaBook and Lenovo W530.

http://yadi.sk/d/RzgipiYO4t8do

 

@joshhh - I have tested some of the vga bioses that you´ve uploaded but chameleon says: the bios file is too big to load, and then I extracted mine and the size is 64k and yours are about 90k, there´s a way to force chameleon to load it in another addres of the memory ? Or a way to downzise it ? Because I think this is the way to make our Mobile Kepler working on a hackintosh !

 

Thanks in advance.

Link to comment
Share on other sites

 Share

×
×
  • Create New...