Jump to content

Vega Frontier Edition on High Sierra


jnolla
 Share

721 posts in this topic

Recommended Posts

Hi,

 

I've been struggling with this particular issue since upgrading to Mojave on my Hackintosh build. It has VEGA FE 16 GB.

 

I have two 4K external display configuration. At home, with two 27 inch 4K UHD BL2711U Displays. the displays are connected directly to GPU Display ports.

 

What happens is the GPU memory graph in iStat Menus shows a steady increase in utilization (right now sitting at 92%). When I open videos on Youtube or Vimeo or any streaming service in full screen mode at some point the system starts responding very slowly, and if not caught soon enough, it will become slowly unresponsive and it just can't free itself from it.

 

If caught early enough, I'm able to do one of two things (aside from force power-down) to free up video memory and restore the system to normal.

 

1. Disconnect any one of the external displays in either of the two locations.

 

2. Change the screen resolution on any one of the external displays in either of the two locations.

 

I have tried to erase my machine without being able to resolve the issue. I don't know enough about where preferences files are stored in the system relating to displays or GPU or whatever, so I don't know how to delete those. Pointers welcomed!

 

 

Aside from testing further using a non-Safari browser to run videos, and deleting any preference files anyone here might recommend, I could try with only a single display connected. It's little intrusive to my workflow, but admittedly I haven't tried this yet and it seems appropriate to do so.

 

Do these issues resonate with anyone? Any ideas?

 

Thanks in advance!

Edited by MMido
Link to comment
Share on other sites

Just got my new computer setup yesterday. i7800K, Z390, Sapphire Vega64. Oddly enough my SSD from my Macbook Pro unmodified, except for Clover, is running better than the fresh install on my NVME, lol. 

 

Something odd about my Luxmark result. 29412, but it says the GPU is running at 945MHZ. Also, trying to get the GPU fan to not be at 100% all the time. 

 

I modified the AMD1000 kext as Okrasit pointed out on page 6. I modified the data line to match the settings below.

Core: 1600MHz/1150mV, Memory: 1045MHz/1075mV, SOC: 1199MHz, Temp target: 70, Idle fan: 700rpm

 

Screen Shot 2018-12-12 at 1.52.04 AM.png

Link to comment
Share on other sites

Hi @CMMChris ,

I have the same Vega with a mobo Asus Z370 Prime A II  ... the only thing is that my GA-VEGA 64 Gaming OC 8GB doesn't go out of the sleep correctly. One monitor is displaying something but it's barely seeable because it wakes up in a very dark dimmed mode and the main screen stays black... darkwake=0 at the moment.

 

And I'm using the VegaTab_64_best.kext made by @meaganmargaret... which gives me nice performance and keeps my Vega very silent. But the wakeup problem is annoying me, also because it's quite random: last evening system worked perfectly in a cycle of SIX sleep/wakeup times. This morning at the 1st sleep It makes the black screen again... I really don't know where might be the culprit!
 

Link to comment
Share on other sites

Just saying... a silent Vega 64, especially OC, isn't anything good. While your GPU temp might seem fine, the VRMs tend to get too hot. I recommend at least 2600 RPM under full load in macOS. Voltage should be limited to 1075. HBM clock not more than 1020.

 

Regarding your sleep issues: Likely not caused by the Vega but your clover config. Feel free to upload your EFI and I will have a look.

Link to comment
Share on other sites

2 hours ago, CMMChris said:

Just saying... a silent Vega 64, especially OC, isn't anything good. While your GPU temp might seem fine, the VRMs tend to get too hot. I recommend at least 2600 RPM under full load in macOS. Voltage should be limited to 1075. HBM clock not more than 1020.

 

Not able to do it myself yet! Would you be so kind to pass me your kext? 

2 hours ago, CMMChris said:

 

Regarding your sleep issues: Likely not caused by the Vega but your clover config. Feel free to upload your EFI and I will have a look.

Sure! Here it is:

 

EFI

 

Thanks in advance! 

Link to comment
Share on other sites

@hackaro I don't use a kext. Please read through the past couple of pages of this topic, I posted an Excel sheet you can use to create your PowerPlayTable. Then inject it via Clover to the ACPI path of your GPU.

Regarding your clover config: Please tick "AppleRTC" and "KernelPM" in "Kernel and Kext Patches", save, reboot and see if sleep works.

Edit: You can remove the following ACPI renames, they are useless since the devices are not present in your DSDT:
- XHC1
- OSI
- HEC1 (use HECI to IMEI instead)
- IDER
- CAVS

Edited by CMMChris
Link to comment
Share on other sites

17 minutes ago, CMMChris said:

@hackaro I don't use a kext. Please read through the past couple of pages of this topic, I posted an Excel sheet you can use to create your PowerPlayTable. Then inject it via Clover to the ACPI path of your GPU.

Regarding your clover config: Please tick "AppleRTC" and "KernelPM" in "Kernel and Kext Patches", save, reboot and see if sleep works.

KernelPM is unnecessary on Asus motherboards as MSRx02 is free. PM is native tough. I will immediately try the other one you said and report back.

 

As for the Excel sheet, well I'm a noob on that... And basically I don't understand where I have to find those values to put in your worksheet... 

 

PS: I'm  making also the other modifications as you edited ur post... 

Edited by hackaro
Link to comment
Share on other sites

Try this

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

I am playing Rise of the Tomb Raider with this and have no overheating or crashes.

 

Bildschirmfoto 2018-12-12 um 21.33.27.jpg

  • Thanks 1
Link to comment
Share on other sites

1 hour ago, CMMChris said:

@hackaro I don't use a kext. Please read through the past couple of pages of this topic, I posted an Excel sheet you can use to create your PowerPlayTable. Then inject it via Clover to the ACPI path of your GPU.

Regarding your clover config: Please tick "AppleRTC" and "KernelPM" in "Kernel and Kext Patches", save, reboot and see if sleep works.

Edit: You can remove the following ACPI renames, they are useless since the devices are not present in your DSDT:
- XHC1
- OSI
- HEC1 (use HECI to IMEI instead)
- IDER
- CAVS

 

These modifications didn't change a bit the behaviour! Randomly my VEGA does not wakes up properly... now I will study your Excel and try to generate a KEXT with that.

 

I will also look into the BIOS of my VEGA because I read somewhere that its VBIOS was updated... I need a Windows installation though... 

Link to comment
Share on other sites

1 hour ago, CMMChris said:

Try this


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

I am playing Rise of the Tomb Raider with this and have no overheating or crashes.

 

 

OK! Tell me if I'm right: 

that string is HEX, I convert it to BASE64 with CC HEX Converter and I put it on my config.plist in Devices section like this: 

 




	<?xml version="1.0" encoding="UTF-8"?>

	<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">

	<plist version="1.0">

	<dict>

	    <key>Devices</key>

	    <dict>

	        <key>Properties</key>

	        <dict>

	            <key>PciRoot(0x0)/Pci(0x1,0x0)/Pci(0x0,0x0)/Pci(0x0,0x0)/Pci(0x0,0x0)</key>

	            <dict>

	                <key>PP_PhmSoftPowerPlayTable</key>

	                <data>

	                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

	                </data>

	            </dict>

	        </dict>

	    </dict>

	</dict>

	</plist>



	 



	[/CODE]


and I remove the kext!

Edited by hackaro
Link to comment
Share on other sites

updates: downgrading Windows 10 to Windows 8 I was able to successfully update my  GA-VEGA 64 Gaming OC 8GB correctly to F2 BIOS version. 

 

Now my config looks like: 

<key>PciRoot(0x0)/Pci(0x1,0x0)/Pci(0x0,0x0)/Pci(0x0,0x0)/Pci(0x0,0x0)</key>
			<dict>
				<key>PP_DisableDIDT</key>
				<integer>1</integer>
				<key>PP_DisablePowerContainment</key>
				<integer>1</integer>
				<key>PP_FuzzyFanControl</key>
				<integer>0</integer>
				<key>PP_PhmSoftPowerPlayTable</key>
				<data>
				tgIIAQBcAOEGAADuKwAAGwBIAAAAgKkDAPBJAgCvAAgA
				AAAAAAAAAAAAAAAAAAIBXABPAkYClACeAb4AKAF6AIwA
				vAEAAAAAcgIAAJAAqAJtAUMBlwHwSQIAcQICAgAAAAAA
				AAgAAAAAAAAABQAHAAMABQAAAAAAAAABCCADhAO2A+gD
				GgRMBH4EsAQBAX4EAQGEAwAIYOoAAABAGQEAAYA4AQAC
				3EoBAAOQXwEABAB3AQAFkJEBAAZssAEABwEI0EwBAAAA
				gAAAAAAAAByDAQABAAAAAAAAAABwpwEAAgAAAAAAAAAA
				iLwBAAMAAAAAAAAAAMDUAQAEAAAAAAAAAABEIwIABQAA
				AAABAAAAAFgCAAYAAAAAAQAAALh8AgAHAAAAAAEAAAAA
				BWDqAAAAQBkBAACAOAEAANxKAQAAkF8BAAAACChuAAAA
				LMkAAAH4CwEAAoA4AQADkF8BAAT0kQEABdCwAQAGwNQB
				AAcACGw5AAAAJF4AAAH8hQAAAqy8AAADNNAAAARobgEA
				BQiXAQAG7KMBAAcAAWg8AQAAAQQ8QQAAAAAAUMMAAAAA
				AIA4AQACAACwrQEABAAAAQgAmIUAAEC1AABg6gAAUMMA
				AAGAuwAAYOoAAJQLAQBQwwAAAgDhAACUCwEAQBkBAFDD
				AAADeP8AAEAZAQCIJgEAUMMAAARAGQEAgDgBAIA4AQBQ
				wwAABYA4AQDcSgEA3EoBAFDDAAAGAHcBAAB3AQCQXwEA
				UMMAAAeQkQEAkJEBAAB3AQBQwwAAARgAAAAAAAAAC0Mb
				ZAAoCjIACgBUA5ABkAGQAZABkAGQAZABAAAAAAACAR0H
				3ADcANwALAEAAFkAaQBKAEoAXwBzAHMAZABAAJCSl2CW
				AJBVAAAAAAAAAAAAAAAAAAAAAAACAtQwAAACEGDqAAAC
				EA==
				</data>
			</dict>

is this correct or do I have to add something?

 

Preliminary tests are telling that black screen is still there anyway... :wallbash:

Link to comment
Share on other sites

Your sleep wakeup issues don't have anything to do with the PowerPlayTable. The issue is something else. What are your BIOS GFX settings? Do you have IGPU dual monitor or something enabled?

Regarding the hex data: No need to convert it. Just use Clover Configurator to set-up your device properties. It will convert the data automatically.

  • Thanks 1
Link to comment
Share on other sites

7 minutes ago, CMMChris said:

Your sleep wakeup issues don't have anything to do with the PowerPlayTable. The issue is something else. What are your BIOS GFX settings? Do you have IGPU dual monitor or something enabled?

Regarding the hex data: No need to convert it. Just use Clover Configurator to set-up your device properties. It will convert the data automatically.

yes, I have iGPU enabled and multi monitor enabled, which in Asus's language means that it will use both the iGPU and the PEG0 infact in MacOS I have this:

 

System Res -> Graphics

 

but this is standard for a 18,3 sysdef and I've always done this way since my RX580 successfully.

Link to comment
Share on other sites

I have read almost the whole thread, very appreciative of the work you have all done here.

 

I am curious about one thing. I play games mostly and I am still crashing (computer restart) when taxing the card with High settings and with DisableFanControl=1. I don't care about noise, honestly it can be a jet engine. I just want speed and stability.

 

My kext looks like this:

 

<key>aty_properties</key>
                        <dict>
                                <key>PP_ACDCGpioDisabled</key>
                                <integer>0</integer>
                                <key>PP_DisableACG</key>
                                <integer>1</integer>
                                <key>PP_DisableAVFS</key>
                                <integer>0</integer>
                                <key>PP_DisableClockStretcher</key>
                                <integer>1</integer>
                                <key>PP_DisableDIDT</key>
                                <integer>1</integer>
                                <key>PP_DisablePCCLimitControl</key>
                                <integer>0</integer>
                                <key>PP_DisablePPTuning</key>
                                <integer>1</integer>
                                <key>PP_DisableULV</key>
                                <integer>1</integer>
                                <key>PP_Falcon_QuickTransition_Enable</key>
                                <integer>1</integer>
                                <key>PP_PhmUseDummyBackEnd</key>
                                <integer>0</integer>
                                <key>PP_ToolsLogSpaceSize</key>
                                <integer>524288</integer>
                                <key>PP_DisableFanControl</key>
                                <integer>1</integer>
                        </dict>

 

Is PP_DisablePowerContainment=1 needed if I am not using a power table? I updated kext cache using the manual method. 

 

If the card is not able to control its own temps with firmware it seems strange to me that it requires a power table just to stop crashing.

 

I am on Mojave 10.14.2 and Sapphire Vega 64 Reference.

 

Kind regards.

Edited by evforeverr
Link to comment
Share on other sites

On 11/1/2018 at 8:41 PM, surfinchina said:

 

And just for the record, it's a Vega FE and works OOB on HS and Mojave - excepting I need the fixup kext to allow for 2 monitors on DP.

 

uhm... @surfinchina what fixup kext are you talking about? I have exactly 2 DP monitors and they don't wakes up correctly after sleep... can you provide me the Kext? Could it interfere with the PP_PhmSoftPowerPlayTable I inserted under Devices section of Clover? 

Link to comment
Share on other sites

3 hours ago, hackaro said:

 

uhm... @surfinchina what fixup kext are you talking about? I have exactly 2 DP monitors and they don't wakes up correctly after sleep... can you provide me the Kext? Could it interfere with the PP_PhmSoftPowerPlayTable I inserted under Devices section of Clover? 

This one. It's pretty usual for the vega. I don't know if it interferes with the powerplay table because I don't use any of that stuff.

VegaGraphicsFixup.kext.zip

Link to comment
Share on other sites

4 hours ago, CMMChris said:

VegaGraphicsFixup is part of WEG. And no, wakeup issues are anything but usual.

 

At this point I'm quite clueless about the wakeup issue... what else can I try? I can still send back this Vega to Amazon and try another one... but which one... 

Link to comment
Share on other sites

sometimes i got random issue with temperature of vega 56, after system restart the temperature of the card without load start up from 30° to 62° - fans 1,100 rpm, activity about 0-3%. If i run luxmark - result is normal, after load the temperature does not go down; after system restart the card is working normally. on pp_table when i set Target Temp  more then 48° the problem start up at random more often, then Target Temp 48° - problem happens rarely. I will try test other values of Target Temp.

Link to comment
Share on other sites

On 12/12/2018 at 2:33 PM, CMMChris said:

Try this


B6020801005C00E1060000902C00001B004800000080A90300F0490200320008000000000000000000000000000002015C004F02460294009E01BE0028017A008C00BC0100000000720200009000A8026D0143019701F049020071020202000000000000080000000000000005000700030005000000000000000108330433043304330433043304330433040101330401018403000860EA00000040190100018038010002DC4A010003905F010004007701000590910100066CB00100070108D04C01000000800000000000008083010001000000000000000070A7010002000000000000000088BC010003000000000000000088D5010004000000000000000030370200050000000001000000B87C0200060000000001000000B87C0200070000000001000000000560EA00000040190100008038010000DC4A010000905F0100000008286E0000002CC9000001F80B0100028038010003905F010004F491010005D0B0010006C0D401000700086C39000000245E000001FC85000002ACBC00000334D0000004686E0100050897010006ECA30100070001683C01000001043C41000000000050C3000000000080380100020000A08601000500000108009885000040B5000060EA000050C300000180BB000060EA0000940B010050C300000200E10000940B01004019010050C300000378FF0000401901008826010050C300000440190100803801008038010050C300000580380100DC4A0100DC4A010050C30000060077010000770100905F010050C300000790910100909101000077010050C300000118000000000000000BE4126400AC0D3700230054039001900190019001900190019001013200370002012307F700F700F700510100005B0069004A004A005F007300730064004000000097609600904B00000000000000000000000000000000000202D4300000021060EA00000210

I am playing Rise of the Tomb Raider with this and have no overheating or crashes.

 

Bildschirmfoto 2018-12-12 um 21.33.27.jpg

 

OK, I have started using your spreadsheet. It's fantastic, thanks for the hard work.

 

These settings are also good, I get over 30K with Lux ball. Why is the P0 Core, P1 Core etc. running at 1075 v though? Did you need this for stability, wouldn't it save some temps to lower the voltage on those P states?

Link to comment
Share on other sites

 Share

×
×
  • Create New...