Memory timing level 2, causes weird artifact like behaviour, only during boot screen.

[electronicworld]

Hello everyone, i am using memory timing level 2 on my rx 580. I like it because it gives me an additional 3%-4% fps boost.

During the boot i am getting this;

However those artifact like lines only appear during the boot screen when using the setting. I dont experience any instability in a variety of games, benchmarks or when just idling.

I have checked for memory errors, overheating and artifacts in any applications but it is just fine everywhere.

Please share your thoughts with me if it is safe to keep using this setting and why it might be happening

Thanks.

 

[Imperator]

Well, during my tests i noticed sometimes errors like you have.
What i did ?
I play with my video card for testing purposes.
I am aware of all the outcomes.
But as i said i do it on purpose.
When you choose a wrong timing for your memory, yes indeed you will receive that error.
The only way to correct it is to flash back your card with original BIOS.
Do not loose your original BIOS file.
That will be the base for further modd.
You play, you are not satisfied, you flash back.
And so on

 

[electronicworld]

Hey there, its the original bios i am using, just changed the setting through amd radeon wattman.

 

[Imperator]

OK. Then reset to default and try again to see the result.
Keep us informed

 

[electronicworld]

Yup, resetted the setting to automatic works just fine now. Does it mean it isnt stable with that setting even if this only persists on bios screen ?

 

[Imperator]

I am curious: by mistake it is ASUS ?
Cause one of my card (testing card) RX580 is ASUS and the memory chips are exposed they do not have on all of them that thermal pad or some cooler.
That makes them the worst candidate for overclocking.
If you do not have ASUS, then it is the moment to blow the card (dust is the worst enemy).
Then if you feel that you have enough skills, you should open the card and clean it.
After that you need to repaste the GPU and check also the status of the thermal pads (in case it needs, replace some of them)

 

[electronicworld]

Replaced both thermal pad and thermal paste. The case has good cooling i can tell by my cpu, and gpu never exceeds 75c celsius on the core.

 

[Imperator]

What model it is and manufacturer ?

 

[electronicworld]

Sapphire pulse rx 580 8gb

 

[Imperator]

Hmm, Sapphire have a very good cooling system.
So yes i believe you when you say you have good temps.
So there must be something about the video card.
It could have been used for mining, or aggressive overclocking so it does not resist any more to o/c.
Who knows....

 

[RJARRRPCGP]

Hmm, Sapphire have a very good cooling system.
So yes i believe you when you say you have good temps.
So there must be something about the video card.
It could have been used for mining, or aggressive overclocking so it does not resist any more to o/c.
Who knows....

Back in the ATI-AGP era, Sapphire had too small of heatsinks, IIRC.

 

[Imperator]

Ha ha , good old days.
True. Back then they indeed had so small cooling systems.
But on RX400 and RX500 series they use the right system.
I can confirm even on HD 5000, 6000 and 7000 series cause i also have.
Instead ASUS made it wrong (at least i own a RX580 where memory chips are exposed )

 

[eidairaman1]

Yup, resetted the setting to automatic works just fine now. Does it mean it isnt stable with that setting even if this only persists on bios screen ?

Post is very low resolution, dont worry about it, up the performance and play

Back in the ATI-AGP era, Sapphire had too small of heatsinks, IIRC.

All of them did because they weren't needing to be huge to run stock,even nvidia.