[SUCCESS] Gigabyte Designare Z390 (Thunderbolt 3) + i7-9700K + AMD RX 580

[hitINdaGs]

This seems to suggest that Photo applications will experience the largest gains.

So far, everybody who has tested real life profit from the boost reported little to no difference. Or am I missing out on a recent test?

 

[CaseySJ]

@iRamon,

So should we give up on the chip??

Although it's not looking too promising, we might still try a couple of things:

• Erase the Winbond chip using flashrom.
• Leave it erased and boot back into macOS and observe the outcome.
  • Chances are that RP05 will be empty in IORegistryExplorer.
  • The Thunderbolt device tree should disappear.
  • If the device tree does indeed disappear, it means there is still some communication happening between the controller and the firmware -- and hence still some hope...
  • If so, we should program the original firmware back and see what happens to IOReg at RP05.
• There is a small chance that this might reset the Winbond chip and take you back to the original working state.

 

[iRamon]

@iRamon,

So should we give up on the chip??

Although it's not looking too promising, we might still try a couple of things:

• Erase the Winbond chip using flashrom.
• Leave it erased and boot back into macOS and observe the outcome.
  • Chances are that RP05 will be empty in IORegistryExplorer.
  • The Thunderbolt device tree should disappear.
  • If the device tree does indeed disappear, it means there is still some communication happening between the controller and the firmware -- and hence still some hope...
  • If so, we should program the original firmware back and see what happens to IOReg at RP05.
• There is a small chance that this might reset the Winbond chip and take you back to the original working state.

Just out of curiosity and for the sake of learning from experiments, I'm in.

I'll pulll a late-nighter again and disassemble my system, erase the Winbond chip, reassemble and see if RP05 gets cleared.

If that still doesn't work, then I believe we've done all there is that can be done, gaining a huge amount of true hackin-toshing experience along the way.

 

[lambusser]

Hello Everyone!

Just thought I'd report my RadeonBoost+Vega (v1.4) results.

I run three monitors. Two are Vizio TV's connected to my liquid cooled Vega 64 Frontier Edition. I use Club3D CAC-1080 adapters (with the latest firmware) and without WEG I could not get them to work. Only my native DP connected LG 5k would work. So...

• Needed to keep WEG in order for everything to work (all three monitors)
• No fan/noise issues with RadeonBoost+Vega v1.4
• Performance changes...attached

Thank you to everyone who is working on this. Great results so far and no stability/power issues!

Lam

 

[SuDoDmz]

is there a way to fix it?

the better Question here is "how?"

I've been tracking your posts until now with fascination and have to tell ya: you're doing an amazing job not giving up on your board, investigating and also thereby teaching us. I'm wearing glasses and am shortsighted, that's the main reason, why I never had problems with chips and soldering, at all. And though wearing glasses I can't help but think something's missing in the picture you took of your Winbond-chip, to be specific: the solder joints to the left(I think?!) look like they're missing some components, though we'd have to double-check with an AIC, or, even better, an actual Designare board.

 

[SuDoDmz]

@iRamon got it

you can call me a whacko, but I think I found your solution

did a quick comparison to one of @CaseySJ 's early Winbond-Chip pictures (during investigations). And if I'm not going completely blind, there most certainly IS something amiss. I already thought those joint bubbles don't look "clean", but like something's been soldered and then "ripped" off. And those tiny thingies (don't know the english terms, sorry) could be your possible answer for missing communication and SSPX'. But believe me, resoldering those is "gonna be a bitch", quote me on that. Now look closely and compare.

@CaseySJ if your test bench is not assembled in an airtight case, maybe you could shoot another photo, for analysis, if it isn't too much hassle if you have a case like mine, it would be perfectly understandable not wanting to do the ****-job of disassembling (again)
I also don't know how he would get the proper soldering equipment and parts, another issue, where we depend on your help

 

[AlexComputerWiz]

@CaseySJ, going from 10.14.5 to 10.15.4. In the upgrade guide you posted, you showed how Catalina splits the OS and User onto separate logical drives. I followed your original suggestion of putting MacOS on an NVMe and everything else on a 2.5 SSD, which I am running currently. Will I run into any issues because my User folder is already split up?

 

[Meroje01]

** Micro-Guide: True One-Key Wake from Sleep Without Changing DarkWake **
Please do not quote this micro-guide in its entirely. Post a link instead.​

Hi (new account, not too sure what happened to the other one because I had a password saved)
Just reporting that I do have one-key wake without this kext+patch, not too sure how though. Maybe it could be the USBPorts.kext that I think you were providing at some point ?

PS: also reporting a flawless TB patch with the powered by psu method on raspberry.

 

[NCMacGuy]

@iRamon got it

you can call me a whacko, but I think I found your solution

did a quick comparison to one of @CaseySJ 's early Winbond-Chip pictures (during investigations). And if I'm not going completely blind, there most certainly IS something amiss. I already thought those joint bubbles don't look "clean", but like something's been soldered and then "ripped" off. And those tiny thingies (don't know the english terms, sorry) could be your possible answer for missing communication and SSPX'. But believe me, resoldering those is "gonna be a bitch", quote me on that. Now look closely and compare.

@iRamon

Yes indeed there are some "thingies" missing on the board in question. Here's a not so great photo of my board showing the "thingies" evident.

 

[iRamon]

the better Question here is "how?"

I've been tracking your posts until now with fascination and have to tell ya: you're doing an amazing job not giving up on your board, investigating and also thereby teaching us. I'm wearing glasses and am shortsighted, that's the main reason, why I never had problems with chips and soldering, at all. And though wearing glasses I can't help but think something's missing in the picture you took of your Winbond-chip, to be specific: the solder joints to the left(I think?!) look like they're missing some components, though we'd have to double-check with an AIC, or, even better, an actual Designare board.

Thank you @SuDoDmz for your words of encouragement, and it's good to hear that the posts of my journey have been valuable for you, hopefully others as well ... I felt more like bloating this thread with just my pains to be honest!

While my Raspberry was erasing the Winbond, I checked this thread again and noticed your posts and your observation on missing components.

After checking, I can tell you, there's absolutely nothing wrong with your eyes!!
During my initial attempt to flash the Winbond, I took photo's with my camera. I just had a look, and you are spot on: on each of the red area's you marked, there were components before that are no longer there.

You solved the mystery! ... Have a look at the my detailed shot of the board before I started flashing ... the three components highlighted are no longer there!

That also seals the deal for me to order a new Designare motherboard ... there's no way I'm going to solder this, or spend money on a professional to do it for me.

Thank you for reaching out and spotting this!

The mystery of not knowing/understanding what was causing the odd behaviour was killing me (proverbially), pushing me to keep searching for an answer and dragging @CaseySJ along. Your observation concludes the quest and gives peace of mind, along with my normal hours of sleep

A big thank you to @SuDoDmz and @CaseySJ!