I installed the update to rc8 and than I installed the Unraid rc8 nvidia driver for rc8 

I installed the update to rc8 and than I installed the Unraid rc8 nvidia driver for rc8 

You know you can just download RC8 straight from the Nvidia plugin, no need to update and restart twice.

Sent from my Mi A1 using Tapatalk

I did restart the system. But I will try and restart the system again when I get home. I updated Unraid to rc8 restarted and than installed the plugin and restarted again. But the gpu. 

Is not kicking into gear.  Hahahaha

Is not kicking into gear.  Hahahaha

From what you've posted already, everything is working other than Plex is not using the GPU. Recheck your Plex settings and make sure the media you're trying to play is actually something your GPU can transcode.

Sent from my Mi A1 using Tapatalk

18 minutes ago, CHBMB said:

You know you can just download RC8 straight from the Nvidia plugin, no need to update and restart twice.

17 minutes ago, acozad1 said:

I did restart the system. But I will try and restart the system again when I get home. I updated Unraid to rc8 restarted and than installed the plugin and restarted again.

 

The point he was making was there is no need to update Unraid separately. This "plugin" itself is actually a custom build of Unraid. Simply installing the rc8 version of this "plugin" IS installing Unraid rc8, but customized with the necessary drivers.

So the only thing I haven’t got is that my intel i7 8700k igup kicks in and is doing the transcode not the nvidia p2000. I’m testing with a 4K to 1080p in Plex. 

Gotcha about the RC8 install. No need to do both updates. 

So here are my PCI system list. It is showing the igpu and my nvidia p2000. I have looked in my bios and there is no where were i can disable it, so i am hoping there is a way in unraid to disable this to have my p2000 take over. thanks everyone.

18 hours ago, maxse said:

Hey guys,

I need some pleaseeee! So I jumped on a Dell sale for a p2000 about a month ago and just didn't have a chance to set it up with unraid although I did install it into my server.

 

Well I followed the instructions and I can't get the plugin to recognize it. "NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.   "

 

I then typed lspci -v, and it wasn't even listed, so unraid doesn't see the card? I went into the BIOS on my relatively new motherboard that I just bought, and it was already set to use the PCIE as the main graphics card and multimonitor was set to disabled (I currently have an i8400 and was using the igpu to transcode before. Even doing this, still the same result...

 

The fan on the card itself is spinning so I'm assuming that means the card is getting power? Could it be something wrong with my motherboard when both PCI slots are being used? (I'm using an IBM m1015 flashed to IT). I'm not sure what all this means. I would really appreciate some help guys. I'm running an ASRock H370M Pro4 board if that helps with the most recent BIOS

Can anyone help me with this please? I've tried multiple setting in the BIOS, and the card is still not being seen. i'm not sure if it's an unraid issue? I remember I had to modify the go file to get transcoding working with the igpu not sure if this could affect it? i would reall appreciate some help with this guys, please

1 hour ago, maxse said:

Can anyone help me with this please? I've tried multiple setting in the BIOS, and the card is still not being seen. i'm not sure if it's an unraid issue? I remember I had to modify the go file to get transcoding working with the igpu not sure if this could affect it? i would reall appreciate some help with this guys, please

We just package the drivers and modify the docker component to utilise them.  I can't offer a reason why your card isn't being seen.  Sounds more like a hardware issue to me.

2 hours ago, acozad1 said:

So here are my PCI system list. It is showing the igpu and my nvidia p2000. I have looked in my bios and there is no where were i can disable it, so i am hoping there is a way in unraid to disable this to have my p2000 take over. thanks everyone.

Pretty sure in Emby you can select which GPU to use, is there not a similar setting in Plex?

20 minutes ago, CHBMB said:

We just package the drivers and modify the docker component to utilise them.  I can't offer a reason why your card isn't being seen.  Sounds more like a hardware issue to me.

Thank you. I wasn’t sure why I kept getting the Nvidia smi error. Didn’t know if it wasn’t being seen by the plugin?

 

i just checked online for my mono and it says this: Slots: 2x PCI-Express 3.0 x16 Slots (single at x16, dual at x16/x4),

 

Im using an LSI controller in one of the x16 slots. Does this mean that the other x16 slot is only going to be at X4? I get so confused by this. Not sure if that’s the issue?

9 minutes ago, maxse said:

Thank you. I wasn’t sure why I kept getting the Nvidia smi error. Didn’t know if it wasn’t being seen by the plugin?

 

i just checked online for my mono and it says this: Slots: 2x PCI-Express 3.0 x16 Slots (single at x16, dual at x16/x4),

 

Im using an LSI controller in one of the x16 slots. Does this mean that the other x16 slot is only going to be at X4? I get so confused by this. Not sure if that’s the issue?

Yes it will only be at x4 and you'll get an nvidia-smi error because it can't find a card to communicate with.....

So in Plex there is no option to choose which gpu to pick. In plex its just that box you check and thats it to my knowledge. But I am a dumb dumb. So it seems to be working now. But my CPU seems to be still loading up when there is transcoding going on. From 4k to 1080p for my guests who only have a 1080p TV. The watch nvidia-smi shows my GPU working and its showing 3% to 6% but again the CPU is loading up like crazy. Alot of red lines in the dashboard which freaks me out.

Edited May 4, 20197 yr by acozad1

33 minutes ago, acozad1 said:

So in Plex there is no option to choose which gpu to pick. In plex its just that box you check and thats it to my knowledge. But I am a dumb dumb. So it seems to be working now. But my CPU seems to be still loading up when there is transcoding going on. From 4k to 1080p for my guests who only have a 1080p TV. The watch nvidia-smi shows my GPU working and its showing 3% to 6% but again the CPU is loading up like crazy. Alot of red lines in the dashboard which freaks me out.

 

Have you bothered reading some of this thread or researching what plex is doing?

Plex is only using the GPU for encoding. It still needs to decode the video using the cpu. It also have to use the cpu for the audio decoding and encoding, so it will still use alot of cpu.

Really. I thought that my Nvidia P2000 would pick up a lot of the heavy-lifting. I did know that the audio would go through the CPU but that  Plex Media Server will attempt to use hardware-accelerated video codecs when encoding and decoding video. Dam. Well thanks for the help on this. I guess I may need to upgrade my CPU from i7 8900k to an AMD Threadripper or something like that.

On 5/2/2019 at 11:53 PM, pimogo said:

Hi thanks for this! Just a clarification. I'm using  a previous version of your script that specifically calls out to NOT transcode mpeg4.  In the new script, for a similar effect, would i remove all but mpeg4 or leave all but mpeg4. 

 

CODECS=("mpeg4" ) or  

CODECS=("h264" "hevc" "mpeg2video" "vc1" "vp8" "vp9")

 

Thanks!

	Available codec options are:



	h264 (default) H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10



	hevc (default) H.265 / HEVC (High Efficiency Video Coding)



	mpeg2video MPEG-2 video



	mpeg4 MPEG-4 part 2



	vc1 SMPTE VC-1



	vp8 (default) On2 VP8



	vp9 (default) Google VP9"



	[/Code]

You need not change anything for mpeg4 to be excluded. Everything with default next to it is ALLOWED to be transcoded by the GPU and *should* have good support on most Nvidia cards. Everything without default can be specified and most of them *should* work, but can have problems. (Notably, mpeg4 and VC1)

Suffice it to say that leaving the "#" before the CODECS line should be sufficient for most users.

6 hours ago, CHBMB said:

Yes it will only be at x4 and you'll get an nvidia-smi error because it can't find a card to communicate with.....

Oh boy

Edited May 5, 20197 yr by maxse

Really. I thought that my Nvidia P2000 would pick up a lot of the heavy-lifting. I did know that the audio would go through the CPU but that  Plex Media Server will attempt to use hardware-accelerated video codecs when encoding and decoding video. Dam. Well thanks for the help on this. I guess I may need to upgrade my CPU from i7 8900k to an AMD Threadripper or something like that.

But it says all that at the beginning of this thread..... in the readme.

And there are ways to get it to decode talked about in the thread.

You could have saved yourself a lot of time by reading.

Sent from my Mi A1 using Tapatalk

2 hours ago, maxse said:

Oh boy

x4 speed should still be fine, main problem is your card isn't showing up.  Tried swapping the cards around?

  solved

Edited May 8, 20197 yr by maxse

23 minutes ago, maxse said:

Yes, thank  you! I posted a thread in the general support didn't want to keep bothering people in this thread.

 

But yes, so I have 2 "X16" slots. I disconnected my LSI HBA from slot #2, plugged in the p2000 into slot #1, and nothing. Then plugged in into slot #2 and it got detected! So Im thinking okay, let me try to plug the M1015 HBA into slot #1, no dice. Actually the computer powered on and off quickly three times before ultimately powering on, but the now the HBA wasn't detected and only the p2000 was. So nothing seems to be detected in the "x16" slot 1. 

Not sure if  have a defective board or what. I was thinking that X4 was the issue and maybe one is always set to that. I think the IBM m1015 flashed to IT requires 8x so that's why it was getting detected.. But now you're saying even at 4x I should have no issues running the p2000?

 

So I started looking at yet another newboard, but seems like there's no matx with 2 x16 slots that are actually x16. I'm still frustrated because I don't get how it's not even false advertising...

But anyway, how does everyone else seem to use a graphics card while using an HBA at the same time? Do I need to at least find one that has a dedicated X16, and also a dedicated X8, and perhaps an X1 for an expander? I couldn't find a single matx intel 1151. Yikes

Looked at you motherboard and I don't see any clear reasons why one of the cards doesn't work. But, as you said, the card in slot 4 is going to be 4x and if you also have a card in slot 2, it will be 2x. That might make an issue.

you could also try to change the pcie gen in the bios for the slot.

Okay will try changing the gen to 2?

im still learning all this now. I put the x1 card on a different slot which should limit the second x16 to x2 just x4. 

 

But it seems that x4 could bottle neck the p2000 even if I could get it to work somehow? I’m shocked more people aren’t having this issue. 

 

I’m okay with just getting another board at this point. But I don’t even know which one to get! What can I use to be able to use the HBA and the p2000 without bottle necks??

Has anyone else been getting their card disappearing and reappearing?  It works perfectly for a while but then just starts connecting and disconnecting.  Only way to fix it is a server reboot.  If I use watch nvidia-smi it will show up then disappear over and over.

 

i am using rc8 and a p2000.  Doesn’t seem reproducible.  Seems like it happens randomly.  But I only had this patch for two days and it’s happened twice already.