1812

No worries, that's part of the learning curve is breaking things!

No additional power needed. you should only have to add the gpu and it’s associated sound device. The quick way is make a “junk” vm with no disks. Assign the card there and make note of the addresses. Save but not start. Go back in, change to xml view, scroll down to the hostdev section towards the bottom, copy both devices (from the start of. Hostdev to the /hostdev) then go into your OS X vm xml view, scroll down to same area, paste it in, then remove the line in each hostdev section that starts with “ address type='pci'” (it will be auto populated correctly) then save and boot. you can sometimes run into issues using vnc and a gpu, so beat to remove the vnc lines.

I’ve used evga and Asus w/out issue. That would work Ootb. Butnahop around, as you may be able to get a 730 for the same price and get a little bump in performance. I currently use a 710 on one Mojave vm and a 730 on another.

So if you did it correctly, then you know your onboard isn’t supported by Mojave. Since it wasnt stubbed or isolated from unRaid, yes you lose the terminal or GUI if you booted into that mode. If you reboot it will return. Your cheapest plug and play solution is a gt710, 720, or 730 as it uses builtin macOS drivers. To fix your vnc issue, change the port on the vnc section of your xml to something higher as another device took the original address. Or, in the xml file you sent posted earlier, if it was from the vm while shutdown, then paste it over what is in there now and you’ll be back to how it was. dealing with macOS virtualization is a bit like nailing jello to a tree. Once you understand how to do it, it’s slightly less of a pain. Bit still more work than windows 10.

You can replace it for now, or if you know how to add the complete hostdev section, then do that for the VGA.

I'm a bit pinched for time but here is some info: You do not have a GPU assigned to your vm, only vnc connection. So your reported 4mb "GPU" is reporting correctly. You have assigned an audio device (00:1b.0) the xml line looks like this: find it and replace with and then delete Save the xml. Then try booting with a monitor attached. It may look glitched out until the boot screen or may be perfect. Or may show nothing. If you get an error via unRaid, report back. I didn't have time to try and lookup the model and see if OSX will run it or not, but sometimes you gotta shoot from the hip and try things. If this is the only GPU in your system, you may have to isolate it from unRaid with vfio.pcids.

I was going to look at identifying that for you in your diagnostics. But if you want to check, go to tools>system devices and look for VGA devices.

post your xml for the vm and complete system diagnostics (tools>diagnostics) What you're showing us appears to be vnc connection and not GPU output.

I use 14,2 on high Sierra and Mojave. Maybe I missed it, but are you trying to passthrough onboard graphics? and if so, have you isolated it from unRaid properly?

assuming the xml is correct, possibly corrupt Mojave installer app. consider deleting and redownloading? a pain, but worth ruling out.

don't worry about the loading image having multiple images. mine freaks out sometimes and tiles a few until the graphics driver loads the login window. What I described is how I've always done it when I"m setting up one for remoting access/vnc only.

Something the same or less of the monitor you are using to access it with.

yes, change in the tianocore bios.

about a year and a half ago I bought the linux standard edition, which shows up as "server edition." My license says server edition and I'm currently at 6tb and have gone up to 10 or so before without hitting any limit. also of note, there is a way to run dual installs (or as many as you want) on the same registered license. but really you should just give them your money.

inside nexcloud, download external storages, then:

Disable uefi and only boot legacy. Uefi is finicky on proliants. You can sometimes do it, but often you have to hit f9 (or whatever) to access the boot menu and select the drive even if uefi is enabled in the bios.

Pm me your problem precisely and I’ll walk you through it. I’m due for some community service time to make up for my other “less than shinning example of desired member of the form” times. (This is a legit offer of help from someone who had to figure it out and has it up and running.)

2 Mojave and one high Sierra on a server running 6.7rc1 with no issues.

Thanks for this. I'm wondering if this is a windows only issue as on MacOS it reports the correct lane width (at least on one machine I have) and hits at or near bare metal benchmarks for gpu. I'll try to play around with it over the next few days!

how about you share with the class?

Just means that anyone with access to the folder in nextcloud can add/modify/erase files to that share/path. If you’re fine with that then it’s no biggie. glad to help!

Edit docker settings for Nextcloud Add path/port/variable Give it a name, then how you want it presented inside the container (ex:/thiscrap) then the path on unRaid share you want available. If you’re just doing online viewing, set to read only if you like (I don’t think it will complain) save container should restart Login to nextcloud web gui Go to settings by clicking on the letter on the top right corner Scroll down to administration section, click external storages click add storage select local under configuration-location put in the container mapping you used before (/thiscrap) Give it a folder name, add users or groups, then save. And done.

External storage works when you set the paths properly. It threw me for a small loop when I first set it up, but I use it without issue.

That information is available in the clover boot menu under options>cpu tuning. this image shows 2 different ones I run on 2 different servers. I don't notice any improvement in changing bus in cinebench or in Geekbench.