Jump to content

DIY eGPU Macbook experiences


oripash

Recommended Posts

I have a question about my 10.8.5 Macbook Pro (Mid-2012), which has a default Intel HD 4000 graphic card. Recently, I purchased an external graphics card (NVIDIA GeForce GTX 650 2048MB GDDR5) and fixed it to my PCIe expansion chassis, which is properly hooked up via my thunderbolt connection and running. However, my laptop (hooked up to an external monitor) doesn’t recognize the eGPU. For example, under the "Graphics/Display" section in "System Properties" in "About This Mac", it does not list my eGPU.

Is my Graphics Card compatible with my Macbook Pro? If not, then do I have to tweak certain settings as a super user in terminal in order to make it work? Or do I have to upgrade to Yosemite and use the method that you have prescribed? Or do I have to purchase an entirely different graphics card?

Link to comment
Share on other sites

MVC, I get that I made a dumb mistake. I did of lot of research to make sure my toaster was compatible with my toast and not enough on whether I could plug it into the wall. So, guys, help me avoid another dumb mistake.

Is there a cable available that connects a mini 4 pin PCB-mounted power connector to the 6-pin connector required by the card? Not much point in modifying the Akitio to split the input power per Nando's schematic here, or getting a OWC Helios instead that already has this connector, if I can't actually get the power to the card.

The whole point of this build is to keep this portable, which to me means that I can close up the case and throw it in my backpack. So the Dell DA2 solution is out, since the BPlus swex wouldn't fit into the case. The R7 250X draws 95 W max per Wikipedia, and the Akitio board draws 10W as discussed in the Dell DA2 thread, so my understanding is that it should be no problem to get this to work with an external 120W power brick plus split input power and an internal cable from the 4 pin to the card. So it's just this internal cable that's the limiting factor.

I'm fine with stripping and crimping cable wires, so if I need to custom-build such a cable I would just need to know which wire to attach to which.

Thanks in advance,

M. Burrito

The OWC Helios Mercury costs more than the AKiTiO Thunder2. It might be a better fit to your requirements given it comes with a 120W AC adapter, the extra J6 4-pin header and (I believe!) a 4-pin to 6p PCIe cable as listed in the manual "+75W (supplemental power cable)". All you'd need to do then is plug it in, wire the J6 lead to your R7 250X and off you go.

AKiTiO is aware of the differences of the slightly higher specced Helios Mercury offers compared to the near identical Thunder2 as per http://forum.techinferno.com/enclosures-adapters/8848-%24299-owc-mercury-helios-16gbps-tb2-discussion-2.html#post123036 .

@MaximumBurrito: OWC Helios Mercury or modded AKiTiO with a 4-pin to 6-pin cable would fit perfectly to your needs, but there are still a couple of concerns:

1) AKiTiO has stability problems at least with Asus R9 series cards no matter how you power it. What is your card model? In your place I would first test with a normal ATX PSU + a molex-to-barrel adapter if it stays stable, and after confirmed would consider the 4-pin to 6-pin mod.

Would be great if AKiTiO or OWC could sell this 4-pin to 6-pin cable separately. I wonder how does it look like because the other end has two 12V pins, and 6-pin has three 12V pins for 75W. In theory possible to make such a cable of course. Maybe I could try some day my GTX 750Ti powered that way, without additional capacitors, and taking power from the modular ATX PSU to the DC jack alone. Another interesting power source would be J11 which is straight connected to J6 and has 4 big solder points. If your GPU 6-pin connector is positioned upwards, you need a low profile 6-pin. However, it seems that R7 250X 6-pin connector is often placed horizontally:

http://media.sapphiretech.com//images/prods/2434/11229-07_R7_250X_2GBGDDR5_DP_HDMI_DVI_PCIE_C05_635460415996339121_600_600.jpg

2) AKiTiO has a poor ventilation for GPUs. If your card has sink plates vertically, heat might become a problem when GPU is stressed normally at 60C temperature, and I would be worried if it goes close to 80C, because that will result in plastic smell from the melted cables touching the hot GPU (happened with my GTX 980). So just a word of warning. At least, put some heat resistant plastic shield between the GPU and the 4-pin cable. R7 250X apparently blows hot air to every direction and AKiTiO’s small fan might not help much. Drilling upper / side vents would be the solution, or just keep the outer shell removed when you do OpenCL computing and put it back when you travel.

By the way, R7 250 does not require 6-pin, it takes all its juice from the PCIe slot. 250X is more power demanding and therefore 6-pin required. There are 750Ti models that draw the power only from the PCIe slot (beware that some of them exceed the specified 75W!)

Link to comment
Share on other sites

@MaximumBurrito: OWC Helios Mercury or modded AKiTiO with a 4-pin to 6-pin cable would fit perfectly to your needs, but there are still a couple of concerns:

1) AKiTiO has stability problems at least with Asus R9 series cards no matter how you power it. What is your card model? In your place I would first test with a normal ATX PSU + a molex-to-barrel adapter if it stays stable, and after confirmed would consider the 4-pin to 6-pin mod.

Can you advise how many AKiTiO Thunder2 units you note have been tested along with how many PSUs and how many R9's to come to the conclusion that "AKiTiO has stability problems at least with Asus R9 series cards no matter how you power it".

I'm aware of your failed 3dmark11 tests but wonder if you've kept a tab of how many other people have had the same failures with their AMD R9 + AKiTiO Thunder2?

Link to comment
Share on other sites

Would be great if AKiTiO or OWC could sell this 4-pin to 6-pin cable separately. I wonder how does it look like because the other end has two 12V pins, and 6-pin has three 12V pins for 75W. In theory possible to make such a cable of course.

Some Molex -> 6-Pin cable only have two 12v rails (the outer two) and don't use the middle pin/rail/plug. Guess that isn't really necessary. The 3 ground rails are still necessary though.

  • Thumbs Up 1
Link to comment
Share on other sites

Can you advise how many AKiTiO Thunder2 units you note have been tested along with how many PSUs and how many R9's to come to the conclusion that "AKiTiO has stability problems at least with Asus R9 series cards no matter how you power it".

I'm aware of your failed 3dmark11 tests but wonder if you've kept a tab of how many other people have had the same failures with their AMD R9 + AKiTiO Thunder2?

There was a thread for AKiTiO stability issues that is currently merged to AKiTiO’s main thread. There does not exist a single stable AKiTiO + TB2 Mac AMD R9 series implementation as far as I know - if there is one, I would be interested in that information. My tries have been with a new Asus R9 270X and used Asus R9 280X cards, they all crashed on Win8.1 and after some time on OS X as well in low power conditions only. However, I do have stable AMD cards with AKiTiO too, so the issue is series or vendor specific. I have tested with 3 different ATX PSUs (one over 1000W) and one Flex PSU. Tried without risers and with quality risers, normal power configurations and soldering. No much difference. NA211TB was stable with those cards, but AKiTiO was not.

Some other AMD instability references:

http://forum.techinferno.com/implementation-guides-apple/7453-2014-15-macbook-pro-iris-gtx760%4016gbps-tb2-akitio-thunder2-win8-1-%5Bcall%5D-2.html#post102475

http://forum.techinferno.com/implementation-guides-apple/7453-2014-15-macbook-pro-iris-gtx760%4016gbps-tb2-akitio-thunder2-win8-1-%5Bcall%5D-3.html#post103160

http://forum.techinferno.com/implementation-guides-apple/7453-2014-15-macbook-pro-iris-gtx760%4016gbps-tb2-akitio-thunder2-win8-1-%5Bcall%5D-3.html#post103250

http://forum.techinferno.com/diy-e-gpu-projects/7910-diy-egpu-macbook-experiences-25.html#post118820

http://forum.techinferno.com/diy-e-gpu-projects/7910-diy-egpu-macbook-experiences-24.html#post118687

http://forum.techinferno.com/diy-e-gpu-projects/7910-diy-egpu-macbook-experiences-25.html#post118813

http://forum.techinferno.com/implementation-guides-pc/7388-15-lenovo-w540-r9_290x-gtx780ti%4010gbps-4gbps-sonnet-ee-se2-pe4l-2-1b-%5Bgothic860%5D.html "Akitio was unstable even with an powered riser"

http://forum.techinferno.com/implementation-guides-apple/7604-2013-13-macbook-air-gtx670%4010gbps-tb1-akitio-thunder2-win8-1-%5Bithildin%5D-2.html#post120419

Link to comment
Share on other sites

Doesn't OWC also add a could capacitors.

If so, adding the cable to Akitio version isn't best option.

Good point. It’s the official recommendation, just my thought that I could try power splitting without them.

Link to comment
Share on other sites

Hi I have a sonnet iii with a gtx 780ti, paired with

macbook pro 2011 / AMD Radeon HD 6750M / Intel HD Graphics 3000 Graphics.

Even though I've followed all the instructions I could find: Install CUDA Driver, Unblocked NVIDIA Web Driver, Amended Kexts, my computer doesnt want to recognize the graphics card. Neither on OSX or windows.

Is it maybe because of my AMD Radeon? I only found the setup working reported by people without AMD Radeon. Is there a way to fix this? Would be a shame if i bought all this setup for nothing.

Many thanks

Julien

Link to comment
Share on other sites

Right, I have to give you all an update with my eGPU setup. I have a Late 2013 MacBook Pro with the 750m, running a stock clocked EVGA TITAN X with the Akitio Thunder2 Chassis. I have been having massive trouble booting up on OSX 10.10.3 with the latest driver (346.01.02f02) when it caused a kernel panic on boot. It works without issue on Windows 8.1.

I have thought that performing SMC resets would sort this problem out, as it fixed the problem on the first few occasions. However, I still got kernel panics during boot up. I've zapped NVRAM (and did the sudo commands again), reinstalled the driver, checked that I've edited the kexts correctly (sometimes it booted up fine so I must have done it right) and checked and repaired my permissions. I've also tried Izzard's solution earlier to see if it miraculously fixed the problem. I still had kernel panics (and random system freezes when opening Chrome).

The only way that I've "fixed" the problem so far is by going back to an earlier version (i.e. 10.10) with an earlier driver (343.01.01f01). It recognised the Titan as an "NVIDIA Graphics Device", but it worked nonetheless. It scored as high as it's supposed to on Unigine Valley as well as being recognised with the same name on CUDA-Z.

I was thinking about posting a guide, however, since I cannot get this to work properly on the latest version of OS X with the latest drivers, I might have to emphasise that.

I don't know whether the next NVIDIA driver and/or 10.10.4 would fix this issue, or is there anything else that may be tried?

Link to comment
Share on other sites

Hi Nando, I checked with OWC customer service and it seems no such cable is provided with the Helios, despite the fact that it's mentioned in that spec table in the manual. They said it only comes with the power brick and a thunderbolt cable, like the Akitio.

I'm trying to get through to an OWC engineer to see if they have any appropriate cables available, but right now it looks like I'd have to wire it together myself even if I bought the Helios. This wouldn't be too big a deal, but I don't exactly know which wire to connect to which. Are the mini 4-pin leads the same as a larger 4-pin molex connector? If so, it might be possible to replace a molex-to-6-pin cable's molex connector with the smaller mini 4-pin.

Hi goalque, going to a lower-power card (one that can draw enough power only through the PCIe connector and doesn't need an additional cable) seems like a good option. The problem is that I do want to get a card that supports OpenCL version 2.0, since it will make my code development easier.

Per Wikipedia, the only lower-wattage AMD cards that support OpenCL 2.0 are the R7 250 you mentioned, the R7 240, and the FirePro W5100. But when I looked through the Air's .kext files to see if either of these cards' PCI numbers were there, I didn't see them.

I'll doublecheck the .kexts to make sure I didn't miss a card, but those seem to be the only options based what I saw earlier.

Unless you guys have some ideas on how to wire up my own cable or another lower-power OpenCL 2.0 card that could be used by the Mac, neither option looks good right now. If neither work, then it looks like I'm stuck pulling the Akitio apart and using an external power supply :/

  • Thumbs Up 1
Link to comment
Share on other sites

Hi Nando, I checked with OWC customer service and it seems no such cable is provided with the Helios, despite the fact that it's mentioned in that spec table in the manual. They said it only comes with the power brick and a thunderbolt cable, like the Akitio.

I'm trying to get through to an OWC engineer to see if they have any appropriate cables available, but right now it looks like I'd have to wire it together myself even if I bought the Helios. This wouldn't be too big a deal, but I don't exactly know which wire to connect to which. Are the mini 4-pin leads the same as a larger 4-pin molex connector? If so, it might be possible to replace a molex-to-6-pin cable's molex connector with the smaller mini 4-pin.

Hi goalque, going to a lower-power card (one that can draw enough power only through the PCIe connector and doesn't need an additional cable) seems like a good option. The problem is that I do want to get a card that supports OpenCL version 2.0, since it will make my code development easier.

Per Wikipedia, the only lower-wattage AMD cards that support OpenCL 2.0 are the R7 250 you mentioned, the R7 240, and the FirePro W5100. But when I looked through the Air's .kext files to see if either of these cards' PCI numbers were there, I didn't see them.

I'll doublecheck the .kexts to make sure I didn't miss a card, but those seem to be the only options based what I saw earlier.

Unless you guys have some ideas on how to wire up my own cable or another lower-power OpenCL 2.0 card that could be used by the Mac, neither option looks good right now. If neither work, then it looks like I'm stuck pulling the Akitio apart and using an external power supply :/

The larger 4-pin molex pins are removable, but the pins are two big. The closest connector that you can use is a floppy drive connector:

http://forum.techinferno.com/enclosures-adapters/7205-us%24189-akitio-thunder2-pcie-box-16gbps-tb2-95.html#post121721

But as AKiTiO does not have two 4-pin sockets ready (only the J4), it won’t be easy. I would utilise J4/J11 or both of them for 75W output to a 6-pin, and keep the TB card back powered. There are two free 2-pin sockets (FAN & LED), they have the same pin size as the J4 socket, and they can be detached. In theory you can join them for 2 X 12V and 2 X GND at J6 rear side (I have done the 12V part already here). You can’t do this without soldering. And this is my hypothesis, please do not modify anything if you have no experience. We have seen some fried boards.

Device ids can be added in the kexts if it’s not included already, but unfortunately R7 250 is codenamed “Oland” and there is no accelerator block so I guess you won’t get OpenCL support in OS X.

Link to comment
Share on other sites

Right, I have to give you all an update with my eGPU setup. I have a Late 2013 MacBook Pro with the 750m, running a stock clocked EVGA TITAN X with the Akitio Thunder2 Chassis. I have been having massive trouble booting up on OSX 10.10.3 with the latest driver (346.01.02f02) when it caused a kernel panic on boot. It works without issue on Windows 8.1.

[…]

The only way that I've "fixed" the problem so far is by going back to an earlier version (i.e. 10.10) with an earlier driver (343.01.01f01). It recognised the Titan as an "NVIDIA Graphics Device", but it worked nonetheless. It scored as high as it's supposed to on Unigine Valley as well as being recognised with the same name on CUDA-Z.

[…]

I don't know whether the next NVIDIA driver and/or 10.10.4 would fix this issue, or is there anything else that may be tried?

Hi Mark,

Glad you got it working. This is interesting to me because it confirms my latest theory: that 10.10.3 (14D136) and/or 346.01.02f02 simply does not work, at least not for me with my GTX 970 and 2011 Mac mini.

I have tried "everything", like you, even totally from scratch and in different orders multiple times. So, for now I just have to give up and - like you - hope that the next revision of the OS / Web drivers fixes it.

In the meantime, I've ordered an MSI GTX 970 to try instead of my Asus one.. .but I don't have high hopes.

- - - Updated - - -

I think you are making a bigger mess of things instead of fixing here.

First up, ATI is 1002, not 300. Also, sometimes these things need to be Endian flipped, so sometimes 10DE is DE10, 1002 is 0210, etc.

But your setup isn't very different form other past builds so trust the past mods and stick to them.

I'm going to guess that your NVStartup is modded wrong.

Or that you permissions have issues.

Even the TB2 Macs that don't allow display output still show the card correctly.

You need to be sure that you still have Nvidia drivers selected, the output you show is what would be there on OSX default drivers.

Well I'm not making a bigger mess because I'm careful to make sure I can roll back after my experimenting (which I did when I realised it wasn't a true solution). We have to dabble to learn, you know! :P

NVStartup is definitely modded correctly, permissions are definitely correctly; this latest update from Mark has convinced me this version of the OS / drivers just doesn't work for this card. The nearest similar setup I've seen is a 2012 mini (not 2011) and with a Sonnet case instead of an AKiTiO.

Tomorrow I'm going to try with a different brand of GTX 970 (an MSI) and I'll report back. When I finally get it working I'll make a guide to help my fellow 2011 mini owners.

Link to comment
Share on other sites

Well, to help out I busted out a GTX970 fro Zotac and fired it up with my 2012 rMBP

I started in clamshell mode, the internal screen isn't even showing up in proflier

So whatever issue you are having isn't with the 970 device id and/or the drivers

I have a few other brands of 970 here but not Asus.

I do have an MSI Gaming Edtion as well as PNY and EVGA and MSI reference.post-31171-14494999964017_thumb.png

post-31171-14494999963594_thumb.png

Link to comment
Share on other sites

Well, to help out I busted out a GTX970 fro Zotac and fired it up with my 2012 rMBP

[...]

So whatever issue you are having isn't with the 970 device id and/or the drivers

Thanks, MVC. So it's either the fact it's a 2011 mini, or it (hopefully) just a gotcha with the Asus STRIX.

MSI Gaming Edition arrives tomorrow... **fingers crossed**.

One other thing I'm wondering about: I'm unable to use Physx in Windows because the Nvidia drivers detect the Radeon in my Mac mini (regardless of whether I disable it in Device Manager and/or uninstall the driver), and therefore disables PhysX, forcing it to run on the CPU. If nothing else, this murders my 3D Mark score because even though the "graphics" benchmark is very good, the combined test is crushed by running Physx on the 2.7 GHz dual-core i7 instead of the GTX 970. This makes me sad.

Do you know any solution to this? Should I consider getting nando's Setup software and experimenting? Thanks again!

Link to comment
Share on other sites

Well, to help out I busted out a GTX970 fro Zotac and fired it up with my 2012 rMBP

I started in clamshell mode, the internal screen isn't even showing up in proflier

So whatever issue you are having isn't with the 970 device id and/or the drivers

I have a few other brands of 970 here but not Asus.

I do have an MSI Gaming Edtion as well as PNY and EVGA and MSI reference.[ATTACH=CONFIG]14801[/ATTACH][ATTACH=CONFIG]14802[/ATTACH]

My Titan+2013 rMBP booted up to the same build with no issues around 30-40% of the time. After a clean install I only started to get kernel panics on OS X after previously booting up on Windows. Have you tried booting it up several times? Also I noticed that your MBP is one with TB1 as well, could that be another factor?

Link to comment
Share on other sites

My Titan+2013 rMBP booted up to the same build with no issues around 30-40% of the time. After a clean install I only started to get kernel panics on OS X after previously booting up on Windows. Have you tried booting it up several times? Also I noticed that your MBP is one with TB1 as well, could that be another factor?

It absolutely is, and that is why I'm so surprised that a 2011 Mini is having trouble.

For the most part, TB1 Macs are relatively trouble free compared to TB2. The fact that your 2013 rMBP was able to EVER show display output is what has surprised me. I imagine that you have seen my thread about TB2 Macs and their issues. (in stickies)

The 2013/14 MBPs with 750 have been [partially exempt. Have you tried the gfxtsatus toggle that one guy discovered?

The 3@ TB2 Macs I have will never, ever allow display output in OS X(exception being an eEFI rom), though after a few black screens I gave up. Maybe they would have if I kept trying. But they are much more difficult then the 2012 rMBP which practically begs you to use eGPU. (except in Windows, another story)

Anyhow, here is a dropbox link to my 14D136 eGPU modded files. I fully take all responsibility for posting them. I have added a few extra instances of the "fix" though I don't think that has anything to do with anything.

As usual, run a "repair permissions" after installing.

EDIT: Will realist when perfect

  • Thumbs Up 1
Link to comment
Share on other sites

Thanks MVC for posting your kexts. I tried them, and still had the same issues ("NVIDIA Chip Model" in System Information, and kernel panic if I try to boot OS X with the eGPU connected).

By the way, using your kexts I got this in the log: Driver "AppleHDAController" needs "IOPCITunnelCompatible" key in plist

Link to comment
Share on other sites

Hmm, well here is what I did.

I was very concerned about somehow linking to busted kexts so I posted them from the rMBP that I tested the GTX970 in. I literally dragged them from S/L/E a folder on desktop, compressed it and put it into dropbox.

I'll plug the rMBP back in with a Titan-X before I head out for drinks tonight and see if that appears in the log.

I just downloaded the link and I can see it in the plist.

Which brings up another possibility. Are ou certain that the kexts are going into the right place?

I understand Mark having issues as he is on one of the machines known for them. I just haven't seen many reports of earlier TB1 Macs having issues like this.

"Nvidia Chip Model" is what a driver that doesn't recognize a card will say.If you open the CUDA control panel it will show which driver is loaded.

EDIT: Yep, it's in log, yet the Titan-X booted and I'm running 2 displays from rMBP.

Very odd.

EDIT: Fixed it and added the pause key as well

Link to comment
Share on other sites

Glad I’m not losing my sanity just yet! ;)

And this is practically a virgin install of 10.10.3 as of a few days ago, from a freshly downloaded install image (never upgraded from an earlier version).

I will try again before I change from Asus STRIX to MSI Gaming. If either of those things cracks it, it’ll be interesting!

Link to comment
Share on other sites

OK, I've now replaced the Asus STRIX GTX 970 with an MSI GTX 970 GAMING 4G.

Again: works really great in Windows, but still kernel-panicks Yosemite and still shows up as "NVIDIA Chip Model" if I hot-plug.

I give up! :34_002:

Link to comment
Share on other sites

Hi all,

I stumbled across these forums a few months ago while trying to find an eGPU solution. I've been reading a few threads, and I decided to take the plunge and try it out. I debated building the Akitio eGPU, but I finally decided on the OSS Cube 2 (http://www.onestopsystems.com/sites/default/files/pdf/186-cube_cube2_cube3_merged.pdf). I also had an old GeForce GTX 260 laying around, so I'm using the GTX 260 for now as my test case. On my Macbook Air, I have Yosemite 10.10.3 and Widnows 8.1 installed via Boot Camp.

The issue that I'm having is that I cannot boot into Mac OS X or Windows 8.1 while the eGPU is plugged in via thunderbolt. On Mac OS X, I get a kernel panic. On Windows 8.1, I briefly see the Windows 8 logo on boot up and then it goes to a black screen. I've tried powering on the eGPU with and without a monitor attached, and I get the same result.

Mac OS X Yosemite 10.10.3:

I installed all the latest updates. I modified NVDAStartup.kext and IONDRVSupport.kext. I modified the nvidia web drivers and was able to install them (WebDriver-346.01.02f01). However, I cannot boot into OS X with the eGPU plugged in. If I plug in the eGPU after boot up, OS X sees the Cube 2 but doesn't seem to detect the GPU. NVIDIA complains that the web driver is not compatible with my current OS and doesn't seem to be able to detect the GPU either. dmsg gives me the following messages:

[FONT=Menlo]NVDAStartup: Official (forced)[/FONT]
[FONT=Menlo]NVDANV50HAL loaded and registered[/FONT]
[FONT=Menlo][AGPM Controller] unknownPlatform[/FONT]
[FONT=Menlo]X86PlatformPlugin::cgRegisterCallbackHandler - gpu internal PLimit notification already registered[/FONT]

I'm not sure if that is normal or not. I saw some posts on other hakintosh forums regarding these messages, and they suggested modifying NVDANV50HAL.kext, but I don't have this kext. I have NVDANV50HalTesla.kext and NVDANV50HalTeslaWeb.kext. Maybe this is something that has changed in Yosemite?

Windows 8.1:

I applied all the latest updates and installed the latest Intel HD 5000 driver (saw this as a suggestion in a post somewhere). But I can't boot with the eGPU plugged in, and Windows does not detect the eGPU after plugging it in after boot up. I also tried booting into safe mode while the eGPU was plugged in via thunderbolt, and I get stuck at the same black screen. I'm not sure how to debug this further and what else to try.

I appreciate any suggestions. Could the issue be the GTX 260 card? Are only specific cards supported in an eGPU setup? Thanks for any input. Hopefully I can get this working. I'll be sure to post my results if I do.

Link to comment
Share on other sites

750M model is not the best candidate for Maxwell Nvidia eGPU use on OS X. One of my contacts reported problems as well.

GTX 970 should not be a problem:

maxserve blog: (MacOSX 10.10.2 edition) How to recognize GPGPU via Thunderbolt external PCIe Box at MacOSX

I recognize that some parts of the setup procedure is taken from my NA211TB implementation. Note that "CUDA is required" is incorrect information if you just need OpenGL.

Tips for troubleshooting:

1) Apple has recently released quite a many "security updates". Confirm that your OS version matches 14D136 when you install the modified web driver 346.01.02f02.

2) Verbose boot mode and system console log always tell what is really happening. Compare the log to previous installs if you are sure that kexts are correctly modified, you have Web driver set as default, and you have done system cache clearance after setting kext-dev-mode=1.

3) I don't fix permissions. Use nano as an text editor and terminal sudo commands. Copy-pasting kexts may mess the permissions causing the kernel panic.

I am examining whether AppleIntelFramebufferAzul.kext includes Mac hardware specific information. When I compared Mid 2014 15” MBP Iris Pro and Late 2013 13” MBP Iris, the kexts were identical in Yosemite 10.10.3. Every byte. But only the 13” MBP gives Nvidia eGPU screen output, and that makes me think that Apple may knowingly restrict the new TB2 Macs for Nvidia eGPU use in OS X. I thought that the integrated graphics could be the factor, but 2014 Mac mini Intel Iris 5100 does not have Nvidia eGPU screen output, even though the Late 2013 13" MBP has the same integrated graphics. I can turn screen output on and off with 2014 rMBP (Iris Pro) in Mavericks by simply replacing a single kext, running “kextcache -system-caches", and reboot. The key is there in the binary data.

All the 2014 Mac minis are shipped with OS X 10.10. My Mid 2014 15” rMBP had originally 10.9.4 and therefore Mavericks installation from the Internet recovery is still possible.

Would be interesting if the latest Mavericks security updates enable Nvidia screen output on 2014 Mac mini as it does with my 2014 15" rMBP (Iris Pro only). But it won't be easy:

https://support.apple.com/en-us/HT201686

The only option is to try kext replacement and permission fix via Terminal commands. If successful or not, post the result in this thread:

http://forum.techinferno.com/implementation-guides-apple/8619-tb2-macbooks-allow-monitors-used-nvidia-egpus-osx.html

Interestingly, AMDs give eGPU screen output in new 2014 Mac mini too. They use different frame buffers. I can use up to three AMD cards externally whereas with regular PC Nvidia cards, I have been able to detect up to 2. 2011/2012 Mac minis are the exception, allowing up to 3 Nvidia eGPUs.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.