Does which SATA port you use effect boot time?

[DeathtoGnomes]

So, are you saying there IS actually a difference??????????????????????

go into the BIOS and disable fastboot and watch the load screen. Then watch again after clearing NVRAM (Clear CMOS? NOTE: I think its spelled that way, its been a while since the old win98 PC died.). After that boot, the BIOS will "Recall" what is where and boot faster.

I expect that if you repeat this process with the SATA plugged into various slots you might be see the difference visually. NOT 100% sure its been a couple decades since I delved this deep.

 

[dorsetknob]

Sata port 0 boots to win 10
Sata port 1 boots to win 8.1
Sata port 2 Boots to win 7
Sata Port 3 Boots from USB/Optical Drive

at Bios Initialization Press F11/F12 to select Boot Drive

or 1st Boot Drive has Bootstrap loader to Select OS to load.

 

[DeathtoGnomes]

Sata port 0 boots to win 10
Sata port 1 boots to win 8.1
Sata port 2 Boots to win 7
Sata Port Boots from USB/Optical Drive

at Bios Initialization Press F11/F12 to select Boot Drive

or 1st Boot Drive has Bootstrap loader to Select OS to load.

interesting way to multi-boot. hmmph never considered this way.

 

[The Egg]

A 3rd party SATA controller can take additional time, as well as have its own BIOS/Firmware screen with a pre-set delay during boot.

In almost every case, a 3rd party controller is going to be inferior to the main chipset controller, so avoid using them unless absolutely necessary. I usually disable them in the BIOS as part of my initial setup.

 

[EarthDog]

go into the BIOS and disable fastboot and watch the load screen. Then watch again after clearing NVRAM (Clear CMOS? NOTE: I think its spelled that way, its been a while since the old win98 PC died.). After that boot, the BIOS will "Recall" what is where and boot faster.

I expect that if you repeat this process with the SATA plugged into various slots you might be see the difference visually. NOT 100% sure its been a couple decades since I delved this deep.

All you see are things initializing and what is on the port. I am not sure what you are trying to say here at all. As I understand things, It doesn't 'recall' anything. It initializes the ports and reads what is on the end. After it is done with POST/initialization, it then looks for boot order which is determined by the BIOS/user.

A 3rd party SATA controller can take additional time, as well as have its own BIOS/Firmware screen with a pre-set delay during boot.

In almost every case, a 3rd party controller is going to be inferior to the main chipset controller, so avoid using them unless absolutely necessary. I usually disable them in the BIOS as part of my initial setup.

Yes.. that is a given.

 

[dorsetknob]

interesting way to multi-boot. hmmph never considered this way.

Its how i have always done multi boot
a corrupted HD does not stop me immediately using Computer

 

[EarthDog]

Can we keep this to the subject at hand..

 

[DeathtoGnomes]

All you see are things initializing and what is on the port. I am not sure what you are trying to say here at all. It doesn't 'recall' anything. It initializes the ports and reads what is on the end. After it is done with POST/initialization, it then looks for boot order which is determined by the BIOS/user.

i may have got that part confused ...

 

[Solid State Soul ( SSS )]

A motherboard bios is reading all the sata ports during POST, however it is true that native ports are faster than the ones run by a third party controller

 

[EarthDog]

A motherboard bios is reading all the sata ports during POST, however it is true that native ports are faster than the ones run by a third party controller

Indeed, but again that wasn't the question here.

The question as outlined clearly in the first post asks if there is a difference on the same controller, just different ports.

 

[arbiter]

A motherboard bios is reading all the sata ports during POST, however it is true that native ports are faster than the ones run by a third party controller

Yea if your board has extra ports on it which usually are color coated so native build in controller are 6 of 1 color and what ever # is another. Mine has 8 total, 6 are intel and last 2 are asmedia or something. You want to use intel ones since they generally right to cpu without 3rd party chip to go through.

 

[Deleted member 67555]

CMOS is why you don't have to wait for the controller to select the proper drive or set the date.
Without the CMOS (battery) then yes it would be slower to boot because you would have to manually set your devices in boot up.
I believe the last PC without CMOS was prior to HDD's

I remember waiting for the PC to check the memory then a: drive 3.5", b: 5.25" and finally C: the OS drive... And back then HDD priority was set by the cable and a jumper on the HDD'S.. One was master and the other was slave.

 

[EarthDog]

I remember waiting for the PC to check the memory then a: drive 3.5", b: 5.25" and finally C: the OS drive... And back then HDD priority was set by the cable and a jumper on the HDD'S.. One was master and the other was slave.

Ahh, the days of IDE platters......

*looks off into distance reflecting...............





*snaps out of it, remembers this is about SATA ports and boot speeds.





Ok, well, looks like my thinking was about right. If anyone has anything else to add about the subject in the title/1st post, by all means!

 

[The Egg]

Indeed, but again that wasn't the question here.
The question as outlined clearly in the first post asks if there is a difference on the same controller, just different ports.

I would say no, or at least, not by any meaningful amount. Motherboards will often have 1 or 2 native chipset ports which share resources with other devices. I imagine it runs some sort of detection routine on these ports, which likely adds an imperceptible amount of time. I don't know that a drive connected to one of these ports would necessarily take longer to initialize though.

It would take some scientific testing to know for sure. And whatever you find might not necessarily apply to other boards.

 

[EarthDog]

I would say no, or at least, not by any meaningful amount. Motherboards will often have 1 or 2 native chipset ports which share resources with other devices. I imagine it runs some sort of detection routine on these ports, which likely adds an imperceptible amount of time. I don't know that a drive connected to one of these ports would necessarily take longer to initialize though.

It would take some scientific testing to know for sure. And whatever you find might not necessarily apply to other boards.

If SATA ports are disabled due to M.2 being used, nothing is plugged into them, but I bet even if there was that wouldn't matter.



In the end, I think the obvious that so long as you are using the native ports to the board boot times will not be different if booting from SATA0 or any other active SATA port (again on the native ports).

 

[R-T-B]

but isnt the boot order determined by what the user has set in the bios?

yes.

 

[The Egg]

If SATA ports are disabled due to M.2 being used, nothing is plugged into them, but I bet even if there was that wouldn't matter.
In the end, I think the obvious that so long as you are using the native ports to the board boot times will not be different if booting from SATA0 or any other active SATA port (again on the native ports).

Pretty much. Although.....some chipsets (like the P67 from my Sandy Bridge rig) had two SATA 6.0Gbit ports, and then four at SATA 3.0Gbit, but all part of the native chipset. In theory this could result in minor differences (besides bandwidth, maybe one set is initialized first), but again, nothing that I would expect to be meaningful to human perception.

Also, all current and future chipsets should be SATA 6.0Gbit for all ports, so that scenario has lost most relevance.

 

[EarthDog]

The latest reply...

On every POST my BIOS automatically sets the boot order with finding the boot drive on any SATA port.

If I move my SATA boot drive from SATA 0 to SATA4, it doesn't change the boot order in my BIOS... anyone else? I'm not sure I even understand what that response means!!!

It moves down the line (boot 1, 2, etc) regardless of physical sata port connection, mirite?

Edit: ok.. funny. The context behind the original statement seems to reference the first boot. So, you swap from port x to port y, and on first boot it takes longer. Then of course each subsequent boot takes the same time as any other port. Communication... good times.

 

[Aquinus]

If I move my SATA boot drive from SATA 0 to SATA4, it doesn't change the boot order in my BIOS... anyone else? I'm not sure I even understand what that response means!!!

It moves down the line (boot 1, 2, etc) regardless of physical sata port connection, mirite?

I think that has a lot to do with how the machine is setup. In my case, all but the 500GB drive is in a SATA RAID array. Moving disks around is practically transparent behind the away as the controller figures it out based on the RAID information stored on disk. I guess it's possible that some motherboards remember the drive info so if the disk ever disappears, the BIOS can tell. If that's the case, the boot order might stay the same. If the computer was 14 years old or something, I wouldn't expect that though.

 

[EarthDog]

I think that has a lot to do with how the machine is setup. In my case, all but the 500GB drive is in a SATA RAID array. Moving disks around is practically transparent behind the away as the controller figures it out based on the RAID information stored on disk. I guess it's possible that some motherboards remember the drive info so if the disk ever disappears, the BIOS can tell. If that's the case, the boot order might stay the same. If the computer was 14 years old or something, I wouldn't expect that though.

See edit.