Trouble disabling Thunderbolt

[bsh]

Hello everyone,
sorry if this is not the correct topic for this, could be graphics, or peripeherals? Feel free to move it.

I have recently bought a few upgrades to my pc, an Asrock Supercarrier mobo, a 7700k, and some RAM. After replacing my old stuff with this, I went to the bios straight and set it up to my needs. This includes disabling everything I won't ever use: onboard sound, COM port, two of the LAN chips, etc. After saving the settings and restarting, the monitor didn't come out of power saving ever, so no display. But Windows booted up from the ssd (seen on the hdd activity led), but still didn't turn the display on. On press of the power button, windows shut down normally, so I assumed everything works, except there's no display. I had to press the clear cmos button to make it display anything again. But any time I set up the bios, the display was gone. The newest BIOS didn't help either.
Took it back to the shop, including my graphics card (Asus GTX970 mini) and sound card, and we did some testing: different power supply, different cpu cooler, zero peripherals except for keyboard. Everything worked fine, until I showed them what settings I change in the bios, then it was no display again. Tried other pcie slot for the graphics, without sound card, with another graphics card: nothing.
Then we went through all the bios changes one by one, and at the end, it turned out, I can disable and change everything, except for the Thunderbolt. Once it is set to "disabled" (or "fully disabled" - don't know what's the difference between the two?), there's no more display output.

Now I don't really know much about Thunderbolt, but is this how it supposed to work? Or is it a bios bug? Isn't it just optional (hence the option to turn it off in the bios)? Block diagrams show that the grpahics card is still wired directly to the CPU's pcie lanes, not going through the Thunderolt chip, so it should still work even if TB is disabled, am I wrong?

 

[RLGL]

Don't know much about T'bolt. It does interact with the display port. Why the graphics card is affected? I checked my configuration, Asus Z170, T'bolt is turned off. I have a functioning R9-470 card. To me it appears the the BIOS file needs a re-write.

 

[bsh]

Thanks RLGL, that's what I'd expect too. Hopefully it's a bios bug and will be fixed if possible. I reported it to Asrock, we'll see what they say.

 

[lopri]

Does the BIOS have dual monitor option? If you have dual monitor on, disabling Thunderbolt alone might not solve the issue because under such a circumstance Thunderbolt is simply serving as a display signal carrier.

Try connecting the cable to one of the connectors that output signals from the onboard GPU. And also try different connectors on your video card, because video cards have a default output connector. And also make sure the monitor's input source matches with the type of display signal that you are connecting your PC with. Most monitors should pick up a signal automatically when there is no competing signal, but sometimes they do not work correctly.

Once you get a video on your monitor, make sure to disable onboard GPU / dual-GPU options in the BIOS.

Good luck and report back.

 

[bsh]

IGP multi monitor was disabled and Init display first was set to PCIe card, according to the manual this combination of settings disables the igp. TBH I didn't test igp output much, because it's rather irrelevant to me, I won't ever use it. I didn't check if the board outputs anything on the onboard DP connector when TB is disabled, perhaps it is trying to use the igp in that case, despite it being disabled, but that would definitely be a bios bug then. I'm waiting for asrock supports findings, they did reply and said their R&D is looking at this.
I think I'll put this back in my pc on the weekend and (finally) start using it, with TB enabled (for now).

 

[bsh]

Update: just got a response from Asrock:

"We would like to update our curent status of this issue to you.
After we disable "Intel Thunderbolt function", we also can see the same symptom you reported.

We checked with our R&D team, and it's related to the Intel original refernce code, and may have Intel help to check it."

Kudos for Asrock for actually caring about customer feedback. Hopefully they'll be able to provide a fix in a BIOS update.