Black screen 780GTX on a Mac 5,1

[h9826790]

Yeah I had read that and every time I try, seems to get a kernel panic message.

Even if I leave the 780 in the lower PCI slot and put the quadro back in above it and try and reboot I get a kernel panic till I pull the card out completely and reboot.

I can’t really see what’s happening, but I know it doesn’t boot in because when I switch it too the Native driver and reboot I never reestablish my remote connection to the other Mac I am using. Then I have to pull the card and put my Quadro 4000 back in and start over.

View attachment 795001

The 780 has 1 8 pin and 1 6 pin port. I am running one mini 6 to 6 pin (stock cable) and one mini 6 to 8 pin to the card. Which from everything I’ve seen and read that is the correct set up.

Mini 6 to single 8pin isn’t not quite a “correct setup. At most, it can work without force shutdown.

The mini 6pin is rated 75W, but 8pin can pull up to 150W.

Of course, you card has a 8pon doesn’t mean it will pull full 150W automatically from the 8pin, that’s why a single mini 6pin may be good enough to power it. However, you may able to see why it is not really a “correct” config.

May I know do you have Windows, or any PC can boot this card? If no, you can even download a Linux LiveCD, put that into the cMP, hold C to boot. If the card can work in any other OS, then may be it’s really driver issue in MacOS. But if the card can’t work in other is either, then may be the card itself is faulty.

 

[thornslack]

I would think your power cables are sufficient.

However it seems it is still possible the web driver is not actually loaded and running. I recall seeing cases where the nvidia control panel reports it is running when it in fact is not. As another user noted, you should see a gsync panel as well.

Isn't there a terminal command that can be used while the quadro is installed to verify if in fact the web driver is running? My quick quick perusal didn't turn it up, perhaps an expert here has a better memory.

 

[h9826790]

I would think your power cables are sufficient.

However it seems it is still possible the web driver is not actually loaded and running. I recall seeing cases where the nvidia control panel reports it is running when it in fact is not. As another user noted, you should see a gsync panel as well.

Isn't there a terminal command that can be used while the quadro is installed to verify if in fact the web driver is running? My quick quick perusal didn't turn it up, perhaps an expert here has a better memory.

But his card has OOTB support, how to explain that KP?

 

[thornslack]

Gk 110 B?

 

[h9826790]

Gk 110 B?

AFAIK, that A B thing was fixed few years ago.

 

[Steven84]

I had the same thought about it being the card but I work at a college and had our VP of IT test the card on a PC and it worked fine.

I was thinking this would be plug and play and if I ended up really needing EFI Screen I would have it flashed. This thing is driving me crazy.

Anything else you guys can think of?

Just for info sake the Quadro 4000 is a mac edition card.

Here is the first part of the Panic report.... let me know if I need to post more of it. I don't really know how to read one of these.

*** Panic Report ***
panic(cpu 0 caller 0xffffff7f88adf692): "bool IOAccelDisplayMachine2::display_mode_did_change(uint32_t): vendor driver returns false"@/BuildRoot/Library/Caches/com.apple.xbs/Sources/IOAcceleratorFamily_kexts/IOAcceleratorFamily-378.26/Kext2/IOAccelDisplayMachine.cpp:490
Backtrace (CPU 0), Frame : Return Address
0xffffffa749c13770 : 0xffffff800786c1c6
0xffffffa749c137c0 : 0xffffff8007995274
0xffffffa749c13800 : 0xffffff8007987544
0xffffffa749c13870 : 0xffffff800781e1e0
0xffffffa749c13890 : 0xffffff800786bc3c
0xffffffa749c139c0 : 0xffffff800786b9fc
0xffffffa749c13a20 : 0xffffff7f88adf692
0xffffffa749c13a50 : 0xffffff7f88b0c4a1
0xffffffa749c13a90 : 0xffffff7f886b4a76
0xffffffa749c13b40 : 0xffffff7f886a19bc
0xffffffa749c13b80 : 0xffffff7f886abf09
0xffffffa749c13ba0 : 0xffffff7f886af91f
0xffffffa749c13c20 : 0xffffff7f886a717a
0xffffffa749c13c70 : 0xffffff7f8870e8ab
0xffffffa749c13cc0 : 0xffffff8007e6cc2f
0xffffffa749c13d10 : 0xffffff8007ec95a7
0xffffffa749c13d70 : 0xffffff80079426a2
0xffffffa749c13dc0 : 0xffffff80078716e0
0xffffffa749c13e10 : 0xffffff800784ea3d
0xffffffa749c13e60 : 0xffffff800786154b
0xffffffa749c13ef0 : 0xffffff800797171d
0xffffffa749c13fa0 : 0xffffff800781e9e6
Kernel Extensions in backtrace:
com.apple.iokit.IOGraphicsFamily(519.20)[0E507098-E375-39B7-9AE2-AFF2C7A5D961]@0xffffff7f88698000->0xffffff7f886e0fff
dependency: com.apple.iokit.IOPCIFamily(2.9)[AEA4C7F5-CCB0-338B-B471-CF28A9792522]@0xffffff7f88094000
com.apple.nvidia.driver.NVDAResman(10.3.2)[6DE4BC26-EEC1-379B-AD41-532FB0A529CD]@0xffffff7f8870a000->0xffffff7f889e3fff
dependency: com.apple.iokit.IOPCIFamily(2.9)[AEA4C7F5-CCB0-338B-B471-CF28A9792522]@0xffffff7f88094000
dependency: com.apple.iokit.IONDRVSupport(519.20)[64AB0898-0DF1-32DF-850A-8840C3478439]@0xffffff7f886ee000
dependency: com.apple.iokit.IOGraphicsFamily(519.20)[0E507098-E375-39B7-9AE2-AFF2C7A5D961]@0xffffff7f88698000
dependency: com.apple.AppleGraphicsDeviceControl(3.20.13)[55E796DB-EFBA-34DB-8148-E834A4383AD2]@0xffffff7f886fe000
com.apple.iokit.IOAcceleratorFamily2(378.26)[AE312652-F3AF-3BFD-AEFF-FE4CCCC33BA9]@0xffffff7f88ad4000->0xffffff7f88b6cfff
dependency: com.apple.driver.AppleMobileFileIntegrity(1.0.5)[4FAE8181-B0B4-3A30-B0F3-77F48D2486E5]@0xffffff7f88aa4000
dependency: com.apple.iokit.IOSurface(211.15)[37BCC686-AF99-3381-AC4F-E54A347E0F40]@0xffffff7f88ab9000
dependency: com.apple.iokit.IOPCIFamily(2.9)[AEA4C7F5-CCB0-338B-B471-CF28A9792522]@0xffffff7f88094000
dependency: com.apple.iokit.IOGraphicsFamily(519.20)[0E507098-E375-39B7-9AE2-AFF2C7A5D961]@0xffffff7f88698000

 

[JamieWestphall]

@Steven84, did you ever get it to work? I can't get mine working still.

 

[MIKX]

I ' feel ' that its a power problem.

I previously used a "DUAL SATA to GPU 8 pin " cable -
https://www.aliexpress.com/item/5pc...Power-Supply-Cable-20cm-SATA/32832007302.html

So, if you can live with out two internel SATA HDD ports .. ( use the DVD drive cage for 2 HDDs )

Setup

1. From cMP motherboard one 6 pin to 6 pin cable for the GPU 6 pin socket

2. The above DUAL SATA to 8 PIN from the two ( now empty ? ) internal SATA sockets to the GPU 8 pin socket.

Can't hurt to try this.

 

[skizzo]

I ' feel ' that its a power problem.

I previously used a "DUAL SATA to GPU 8 pin " cable -
https://www.aliexpress.com/item/5pc...Power-Supply-Cable-20cm-SATA/32832007302.html

So, if you can live with out two internel SATA HDD ports .. ( use the DVD drive cage for 2 HDDs )

View attachment 824506

Setup

1. From cMP motherboard one 6 pin to 6 pin cable for the GPU 6 pin socket

2. The above DUAL SATA to 8 PIN from the two ( now empty ? ) internal SATA sockets to the GPU 8 pin socket.

Can't hurt to try this.

I am under the impression a single SATA port can provide ~50W, so two of them makes ~100W. With 8pin potentially pulling up to 150W I am not sure this is a real improvement. If anything, sounds more like a lateral adjustment if you will.

I would recommend using dual SATA to a 6pin PCIe connector. To be safe, I would recommend at least 3 SATA ports going to a single 8pin if the 8pin is needed to be powered via SATA ports.

However, if the poster has two SATA ports open for power, then they should use a dual SATA to 6pin cable for the 6pin of the GPU. Then use the two mini 6pins on the logic board to power the 8pin on the GPU, use a dual mini 6pin to single 8pin cable for that. This would be the best, safest, and recommended power setup

 

[MIKX]

The onboard Apple mini 6 pin should be perfectly OK for the GPU 6 pin but I do agree about TRIPLE SATA to 8 pin to the GPU 8 pin socket - to be safe.

( Internal spinner HDDs are a thing of the past since SSD & bootable NVMe M.2 )

I use an EVGA PowerLink to my Sapphire Dual-X 7970 without any issues. ( had to cut off a little bit of the plastic fan shroud though.)

Single motherboard 6 pin + dual SATA to 8 pin. X-Plane 11 at hi-rez settings runs fine