Contribute
Register

How to build your own iMac Pro [Successful Build/Extended Guide]

Status
Not open for further replies.
Deleted
 
Last edited:
Wow you really are an ass to people that try to contribute .....

Good luck .... maybe it will end up like the sleep - where u still have no idea

Thanks man for all your insults! Appreciated!

And all this only because I tried to proceed with the HWmonitor approach in parallel..

You leave me absolutely speechless ... Sleep well!
 
Last edited:
@interferenc ,

I really tried my best in compiling Rehabman's source with Xcode9.1. Was on it the entire day without breaks and did not find time for anything else. In consequence, I even received insults from others for my efforts. No way for me to derive a FakeSMC.kext from Rehabman's source distribution, that does not prevent my system from boot. There is still one compilation error and plenty of warnings..

As however Rehabman's compiled kext distribution from 171017 perfectly works on my system, I propose that you directly leave him your X299 modifications of his source and ask him to compile and distribute the respective kexts and HWMonitor.app, if you feel uncomfortable to do it yourself.

No further intents, efforts and contributions from my side.

As you have seen from the previous posts, there are anyway also appreciated efforts by other people along this thread to come up with an alternative HWMonitorSMC approach, which I fully support and observe with much interest, and which indeed also already looks quite promising but certainly still requires several future improvements and further implementations.

In any case, I am really tired in receiving permanent insults and effronteries for my work and efforts. But this really has nothing to do with you and should not be your personal problem or issue.

Cheers,

KGP
 
Last edited:
@interferenc ,

Success!!! Finally!!

With the tremendous support of @RehabMan, I was finally able to successfully compile his source distribution even with Xcode 9.1, as I was finally able to successfully compile your add_skylake_x_support fork in addition. Now my system also successfully boots with the FakeSMC.kexts resulting from both source compilations.

Now the bad new:

Your CPUSensors.kext causes KP on boot:

IMG_9295.JPG


and also your other Sensor kexts cause plenty of error messages during boot.

Thus, at present the available HWmonitor information is neither complete nor very satisfactory and also erroneous concerning the GPU PWM (Aorus GeForce GTX 1080 Ti WaterForce WB 11GB extreme edition), to say it politely ;)

Screen Shot 2017-11-28 at 00.58.23.png


But I am more positive now that we also might succeed in the end with your approach after some more effort.. :thumbup:
 
Last edited:
Thanks man for all your insults! Appreciated!

And all this only because I tried to proceed with the HWmonitor approach in parallel..

You leave me absolutely speechless ... Sleep well!
I was thinking, it was nightmare but no, this guy is crazy (in that bad way) "ccosmo"

@kgp don't even waste your time for people like this one... who drink too much %% or... I don't know...
You shouldn't forget about thousands anonymous people who appreciate your work and say nothing :p
sometimes it's better than one crazy-hamster like ccosmo ... :p

Btw: General idea of HW sensor etc was great.. I was thinking about that 2-3 days ago and tried to find something which will work ... We can say "we don't need that every-day, but... Yes... sometimes it could be great to have it working somewhere :p

PS: I'm not a pro magician for ultra-compilations of nasa-scripts from x, y, z... but I hope I will be helpful/I would like ... for later tests or something like that, etc... :) :p
 
Last edited:
  • Like
Reactions: kgp
And now fantastic news for @ALL Syklake-X/X299 Users

Look at this :headbang:

Screen Shot 2017-11-28 at 01.38.20.png


I was finally also able to successfully compile the original source distribution of @interferenc https://github.com/interferenc/HWSensors ;):lol:

So much about a cowboy setting on the wrong horse :wave: What a disgusting, offensive and stupid comment by the way! :silent:. I simply know whom to trust on and whom better not to listen ;)...

The system boot with all resulting FakeSMC and Sensor kexts works absolutely flawless :thumbup:

There are certainly things yet to be improved: My CPU has 18 cores and not 8 :thumbup: My AOI pump is on the CPU Fan header (4736 rpm), and my 20 Fans are on Mainboard Fan 1 (1470 rpm) and 2 (not displayed). Other few things might be missing (e.g. NVME information etc. ) and my water blocked GPU certainly does not have 1470% PWM as well as other sensor data values might not be totally correct at present (fan speed, core temps and frequencies, etc.).. ;)

But the first glance is already amazing and a big step towards the right direction :thumbup::thumbup: And wouldn't it be boring to remain without future work and efforts :lol:

Please find attached below the compiled FakeSMC and HWSensor kexts as well as the respective HWMonitor.app for test purposes!

Gorgeous work, @interferenc :thumbup::thumbup::thumbup: And gorgeous lessons learned by @RehabMan :thumbup::thumbup::thumbup:

I am extremely grateful to both of you :!::!:

A pleasant night and sound sleep to everybody :wave:

KGP
 

Attachments

  • HWSensors-Release.zip
    3.4 MB · Views: 150
Last edited:
And now fantastic news for @ALL Syklake-X/X299 Users

Look at this :headbang:

View attachment 295409

I was finally also able to successfully compile the original source distribution of @interferenc https://github.com/interferenc/HWSensors ;):lol:

So much about a cowboy setting on the wrong horse :wave: What a disgusting, offensive and stupid comment by the way! :silent:. I simply know whom to trust on and whom better not to listen ;)...

The system boot with all resulting FakeSMC and Sensor kexts works absolutely flawless :thumbup:

There are certainly things yet to be improved: My CPU has 18 cores and not 8 :thumbup: My AOI pump is on the CPU Fan header (4736 rpm), and my 20 Fans are on Mainboard Fan 1 (1470 rpm) and 2 (not displayed). Other few things might be missing (e.g. NVME information etc. ) and my water blocked GPU certainly does not have 1470% PWM as well as other sensor data values might not be totally correct at present (fan speed, core temps and frequencies, etc.).. ;)

But the first glance is already amazing and a big step towards the right direction :thumbup::thumbup: And wouldn't it be boring to remain without future work and efforts :lol:

Please find attached below the compiled FakeSMC and HWSensor kexts as well as the respective HWMonitor.app for test purposes!

Gorgeous work, @interferenc :thumbup::thumbup::thumbup: And gorgeous lessons learned by @RehabMan :thumbup::thumbup::thumbup:

I am extremely grateful to both of you :!::!:

A pleasant night and sound sleep to everybody :wave:

KGP

Beautiful :thumbup::thumbup::thumbup:
Confirmed, its running perfect on my 7900X,

* Incase if someone needs help how to install that, drop all the kext files on "Other" folder of your EFI partition,
drop HWMonitor.app to Application,

Tada, :mrgreen::mrgreen::mrgreen:
Thanks @kgp, great work man! :clap::clap::clap:
 
And now fantastic news for @ALL Syklake-X/X299 Users

Look at this :headbang:

View attachment 295409

I was finally also able to successfully compile the original source distribution of @interferenc https://github.com/interferenc/HWSensors ;):lol:

So much about a cowboy setting on the wrong horse :wave: What a disgusting, offensive and stupid comment by the way! :silent:. I simply know whom to trust on and whom better not to listen ;)...

The system boot with all resulting FakeSMC and Sensor kexts works absolutely flawless :thumbup:

There are certainly things yet to be improved: My CPU has 18 cores and not 8 :thumbup: My AOI pump is on the CPU Fan header (4736 rpm), and my 20 Fans are on Mainboard Fan 1 (1470 rpm) and 2 (not displayed). Other few things might be missing (e.g. NVME information etc. ) and my water blocked GPU certainly does not have 1470% PWM as well as other sensor data values might not be totally correct at present (fan speed, core temps and frequencies, etc.).. ;)

But the first glance is already amazing and a big step towards the right direction :thumbup::thumbup: And wouldn't it be boring to remain without future work and efforts :lol:

Please find attached below the compiled FakeSMC and HWSensor kexts as well as the respective HWMonitor.app for test purposes!

Gorgeous work, @interferenc :thumbup::thumbup::thumbup: And gorgeous lessons learned by @RehabMan :thumbup::thumbup::thumbup:

I am extremely grateful to both of you :!::!:

A pleasant night and sound sleep to everybody :wave:

KGP

I can confirm with my 7900X too :headbang:
I think I have everything without GPU :p
This is really beautiful! :D This is what I needed :p

Thanks to @RehabMan @interferenc and of course Big THanks to @kgp Great job!!! :clap::thumbup::thumbup::thumbup:
 
Thank you, it was really nothing. I hope you guys understand why distributing binaries is a bad idea. I could have had any kind of malware in it, and you would never know. On the other hand, you can read source codes (or at least check the changes I made, to see how little they were), and compile it yourself, and can be pretty sure there is nothing fishy in it.

Also, since we are on the bleeding edge with unsupported cpus, learning the tricks of self compiling stuff is not a wasted effort, you are going to need it again :)

I will try to check why it tops out at 10 cpus, but I would much prefer to have @RehabMan to find out :)

Well there are certainly more issues we have to talk about and to improve things might require the collaboration of several people involved in the development of HWSensing.

Let me know whenever you find time to discuss the details.

However, at first place the automatic detection and implementation of all cores shall be the primer goal. Will be tricky to squeeze all the information into the tiny window. We might also have to think in a redesign if the entire gui.
 
Last edited:
Serious Warning for @ALL !!

@interferenc,

the apparently improper implementation of the onboard fan headers of the ASUS Prime X299 Deluxe within the HWMonitor.app can cause severe damage on your system!

Apparently the CPU-OPT and Chassis 2 Fan Header are not properly implemented and not recognised, thus in consequence all PWR Fans connected two these two headers stop working as soon I launch HWMonitor. Only PWR Fans connected to the Chassis 1 Fan header continue working with HWMonitor!

ASUS-Prime-X299-Deluxe-Fan-Headers.png


This is indeed a sever issue which asap requires a fix!

Note that there are even more fan headers on this board, which might not be properly implemented! And one might also want to use the ASUS Fan/Sensor Extension Card delivered with the board, which implements 4 additional fan headers and 3 sensor headers!

I am currently not able to use HWMonitor, as I am running into high danger to potentially overheat and damage my system! Out of 20 System Fans, 12 Fans fail with HWMonitor. Only the 8 PWR fans connected to PWR Fan Hub 1 that further connects to the onboard Chassis 1 Fan Header work with HWMonitor and are recognised as Mainboard Fan 1. All 7 PWR Fans connected to PWR Fan Hub 2 that further connects to the onboard Chassis 2 Fan Header and all 5 PWR Fans connected to PWR Fan Hub 3 that further connects to the onboard CPU_OPT Fan Header are not recognised by HWMonitor and stop working as soon I launch the HWMonitor.app. Also note that the pump on the CPU_Fan Header is at leased recognised by HWMonitor, although the latter onboard header is also improperly implemented and shows up as Mainboard Fan 2! Finally not even the Chassis 1 Fan Header seems properly implemented so far - connected fans in fact run in CPU Idle for CPU Temps between 16 and 30 deg C at 60% of 1500 rpm max. (hardwired by myself in BIOS) and for CPU Temps between 30 and 60 deg C at 80% of 1500 rpm max. (also hardwired by myself in BIOS), thus the displayed 1470 rpm cannot be realistic and true and would rather correspond to 100% fan speed hardwired in the BIOS for CPU Temps >60 deg C, which is definitely not the case under CPU IDLE conditions and cannot be confirmed by visual nor acoustical inspection of the actual IDLE fan speeds or the displayed CPU core temperatures.
 
Last edited:
Status
Not open for further replies.
Back
Top