SOLVED XMP profile and bumping RAM voltage (SOLVED)

[Gin]

Hello, I would like to know if anyone knows how RAM voltage works while using XMP profile, as far as I know after I set XMP voltage frecuency and timings are set automatically according to XMP profile values. However in BIOS I can see that the RAM voltage is set to 1.5 My question is, If I bump this 1.5 voltage while using XMP profile, lets say to 1.525 and XMP profile is giving 1.65 and 1.50 to my rams, so it will means that then it will give 1.50 +0.25 and 1.65 +025 ? or I'm missing how it works? Thank you in advance.

 

[EarthDog]

You confused me.

There isnt a memory voltage offset I know of. That said, setting the voltage manually trumps xmp voltage. That is to say when xmp is 1.35V...you get the manually overrided voltage no.matter what. No offsets or adding to or splitting it (unless your board is able to do that which I doubt).

 

[Gin]

You confused me.

There isnt a memory voltage offset I know of. That said, setting the voltage manually trumps xmp voltage. That is to say when xmp is 1.35V...you get the manually overrided voltage no.matter what. No offsets or adding to or splitting it (unless your board is able to do that which I doubt).

Sorry, probably my bad english.
I have set my mems using XMP profile, but due some booting issues I'm trying to try different configurations to fix it. I was asking because even with XMP profile loaded, I can set a RAM voltage value, but you said it's just cosmetic as XMP profile will override voltage. So in other words I could set 1.7v to rams but as far as XMP profile is selected that voltage means nothing?

 

[EarthDog]

A manual setting will override XMP.

 

[BugFreak]

A manual setting will override XMP.

This. If you set XMP it will adjust all your settings to XMP. If you then set your voltage manually, it will leave the rest of the settings in place and up your voltage to your new manual setting.

 

[Gin]

Thank you both that helps a lot. I was asking because I was not able to understand some stuff related to mem voltages.
I have
2x4GB Kingston 1600mhz
2x4GB G.Skill 1600mhz
Mobo: Asrock 970M pro3
PSU: EVGA600W 80 plus 47A
VGA: MSI GTX 1050 2GB X

So I know you may ask, dude how is even possible that you are still running that mesozoic hardware in 2020. Well I know and let's say it's just a trip to the past for fun and for experimental materia.

The system works just fine when on windows, my previous "random restart" is gone as soon as I replaced my old PSU with the new EVGA one. The issue comes on power up. If I power up computer, all fans and lights will fire but I'll get no signal to monitor, then I shut down pc, power up again and it always boot fine. So I was checking xmp profile and found that kingston uses 1.650v while g.skill will use 1.500V so I believe it is not possible for this motherboard to give different voltages to different rams? can someone give light on this, so IDK how was XMP profile working then, so I just set RAM voltage manually following your advice that this will override XMP profile voltage parameters and will let everything else as it was. I went in and set ram voltage to 1.535 as it seems reasonable and now my computer can restart fine, but there is still no signal to monitor after shut down. Restart is okay, but shutdown will give first attempt no signal to monitor. Should I just keep bumping voltage just for troubleshooting issue? or just wait this weekend so I can use my old mobo which is in my signature and that never gave me any issues.

 

[trents]

The Kingston using 1.65v at the same frequency as the GSkill suggests that it was manufactured early in the DDR3 era before the technology had been refined. The problem is going to be that 1.65v is too much for the GSkill and may make it unstable (as well as shorten it's life) whereas 1.5v is not going to be enough to give stability to the Kingston at 1600 mhz. I wonder if your issues would disappear if you were to set the RAM speed to 1333 mhz wich will proably work for both the Kingston and GSkill at 1.5v.

 

[Gin]

The Kingston using 1.65v at the same frequency as the GSkill suggests that it was manufactured early in the DDR3 era before the technology had been refined. The problem is going to be that 1.65v is too much for the GSkill and may make it unstable (as well as shorten it's life) whereas 1.5v is not going to be enough to give stability to the Kingston at 1600 mhz. I wonder if your issues would disappear if you were to set the RAM speed to 1333 mhz wich will proably work for both the Kingston and GSkill at 1.5v.

Yes, IDK why my brain wasn't doing the right math! you are right!, I just set 1.540 and it's the first time the computer have boot at first attempt after shuting it down. The real mess is as far as I remember, in my old motherboard I just set XMP profile for both of the ram brands and never looked back, IDK if that may have damaged the g.skill memory but so far so good after I set 1.540 in bios. I'll get updated computer hardware by the end of this year so lets hope this can survive until that moment. I have learned many things from this forum so a big thank you for you all, and for your patient "helping poor guy with obsolete hardware" lol.

 

[trents]

1.65 is probably not so high to damage RAM in the short run. If you ran it like that for months or years, maybe it would.

Bioses will generally do a pretty good job of of negotiating frequencies, timings and voltages in mixed RAM scenarios as long as there is not much difference in any of those parameters between the sticks. In this case, however the voltage differential was apparently too much.

I would suggest running memtest86 to check the integrity of the RAM and the stability of the configuration.