Overclock.net banner
13021 - 13040 of 19226 Posts

· Registered
Joined
·
113 Posts
I noticed new issue , if you have even the best quality raiser cable even if you put the setting bios to gen3 or what ever your cable is , you have to drop your overclocking setting i have strix 3080 with raiser cable i cant boot PBO, and have to drop memory to 3600 and fclk to 1800 other than that WEHAAAAA, crashing, or even not booting,
test it on three type of raiser cables i have
That's odd but for what it's worth, a lot of people even historically in this thread haven't been able to make Gen4 work with higher overclocking. It's kind of weird but a lot of people have to run it as Gen3 or else it implodes stability. I guess with a riser cable I could see something similar happening. I'm not sure why that's the case but there's precedent for it.

Gen4 is kind of a joke at the moment and has 0% gaming difference but yeah I don't know if it's a Gigabyte problem or AGESA problem or what.
 

· Registered
Joined
·
198 Posts
That's odd but for what it's worth, a lot of people even historically in this thread haven't been able to make Gen4 work with higher overclocking. It's kind of weird but a lot of people have to run it as Gen3 or else it implodes stability. I guess with a riser cable I could see something similar happening. I'm not sure why that's the case but there's precedent for it.

Gen4 is kind of a joke at the moment and has 0% gaming difference but yeah I don't know if it's a Gigabyte problem or AGESA problem or what.
Just setting the Pcie configuration to Gen4 is the way to enable Gen4 in GB boards right?

I just set it up and do not face any issues, does it happens only when using it on new graphic cards?
 

· Registered
Joined
·
113 Posts
Just setting the Pcie configuration to Gen4 is the way to enable Gen4 in GB boards right?

I just set it up and do not face any issues, does it happens only when using it on new graphic cards?
Yeah that enables it. It works for me and I have it running myself, but I'm also not pushing my FCLOCK above 1867 at the moment. It's possible that if we get another good BIOS where I can push it up to 1900 and above I may need to turn it down to Gen3. I know in the past in this thread a lot of people have had problems unless they turned it off or else they get WHEA/POST problems. Not sure if it only applies to new GPUs but also the new GPUs are the only ones that use it.

For example this guy but also I remember several others: (Gigabyte X570 AORUS Owners Thread)
 

· Registered
Joined
·
198 Posts
Yeah that enables it. It works for me and I have it running myself, but I'm also not pushing my FCLOCK above 1867 at the moment. It's possible that if we get another good BIOS where I can push it up to 1900 and above I may need to turn it down to Gen3. I know in the past in this thread a lot of people have had problems unless they turned it off or else they get WHEA/POST problems. Not sure if it only applies to new GPUs but also the new GPUs are the only ones that use it.

For example this guy but also I remember several others: (Gigabyte X570 AORUS Owners Thread)
No problems running @ 1967 FCLK, Maybe it all depends on the right BIOS and right CHIP.
It may also be the new cards which utilize gen4 and not the old ones. Guys with new GPU's can confirm
 

· Registered
Joined
·
645 Posts
Cpu-z Bench
Single 690.6
Multi 13313.3

5950X X570 AORUS XTREME BIOS F31o
Ballistix MAX BLM2K16G40C18U4B 4x16GB (3800 IF1900 16-16-19-16-36 1T)
PPT 300
TDC 230
EDC 230
PBO Scalar 10X
MaxCPU Boost 100
Curve Opt ALL Cores ALL Core
Curve Sign Negative
ALL Core Curve Magnitude 50
View attachment 2468594
How you decide to go with 230 edc ?
and no weha with those setting?
and your graphic is rx550 too slow that why you not getting any unstability try to test these setting on rtx 3080 or 3090 if you can have access to them and check i was doing all the initial testing with geforce gt710 graphic but when pluging rtx 3080 all the stability story changed .
 

· Old crazy guy
Joined
·
4,938 Posts
is this the same bios for rev 1.0 as well? does anyone have a link for f31k for rev1.0 as well?
Same version for all releases
 

· Registered
Joined
·
645 Posts
Just setting the Pcie configuration to Gen4 is the way to enable Gen4 in GB boards right?

I just set it up and do not face any issues, does it happens only when using it on new graphic cards?
We Where talking when you use extension cable for graphic card with out extension cable gen4 is work automatically if you have x570 mainboard and gen4 supported graphic and cpu like ryzen 2 and 3 .
 

· Registered
Joined
·
247 Posts
test also dual rank b-die i could not get lower than 53.9 stable but you are ram experet with your magic super tight timing i used for single rank was great but iwant 32 gig ram on my system. hope you get us something work better on dual rank ,wish you where local to me will let you try my dual rank before you buy one.
Hehe thanks, for 32gb you will need to tweak the RttPark-,RttWr- and RttNom Values.
If you wanna go higher than 1900FCLK, you might need to use 2T or GDM On.
You will probably loose the timings a bit to, loosen up the TRFC etc.
 
  • Rep+
Reactions: dr.Rafi

· Registered
Joined
·
121 Posts
How you decide to go with 230 edc ?
and no weha with those setting?
and your graphic is rx550 too slow that why you not getting any unstability try to test these setting on rtx 3080 or 3090 if you can have access to them and check i was doing all the initial testing with geforce gt710 graphic but when pluging rtx 3080 all the stability story changed .
EDC230 scored the best.
Neither 220 nor 240 produced better results.

It is good with no WHEA errors.

The 6900XT will arrive this weekend and I will be testing the Gen4 and SAM.
 

· Registered
Joined
·
247 Posts
EDC230 scored the best.
Neither 220 nor 240 produced better results.

It is good with no WHEA errors.

The 6900XT will arrive this weekend and I will be testing the Gen4 and SAM.
Whats ur PPT, TDC? Just curious.
 

· Registered
Joined
·
16 Posts
Hi all,
maybe you can help me troubleshoot my setup?

I have really hard crashes when using [email protected] with cpu&gpu.
Hard crash meaning instant loss of power and black screen. When i start the PC again it needs 30s with orange RGB lights to auto-enter bios and auto-reset to "stable" settings. BIOS is the current 30o. It happened before 30o too, but i dont know if it happened with the 21 version.

I suspected PBO and RAM overclock first and dialed everything back to safe, but it still happens. It seems to crash even on the "optimized default" settings. All drivers are up to date. My power consumption measured from the wall outlet is 460w when folding full throttle and i have a 550w PSU. Is there anything I can do to narrow down the search?
I can also exclude cooling, CPU and GPU are watercooled over 3x360mm radiator space.

thanks!
 

· Old crazy guy
Joined
·
4,938 Posts
Hi all,
maybe you can help me troubleshoot my setup?

I have really hard crashes when using [email protected] with cpu&gpu.
Hard crash meaning instant loss of power and black screen. When i start the PC again it needs 30s with orange RGB lights to auto-enter bios and auto-reset to "stable" settings. BIOS is the current 30o. It happened before 30o too, but i dont know if it happened with the 21 version.

I suspected PBO and RAM overclock first and dialed everything back to safe, but it still happens. It seems to crash even on the "optimized default" settings. All drivers are up to date. My power consumption measured from the wall outlet is 460w when folding full throttle and i have a 550w PSU. Is there anything I can do to narrow down the search?
I can also exclude cooling, CPU and GPU are watercooled over 3x360mm radiator space.

thanks!
I suspect the PSU can't handle the power spikes.
You can't see the spikes when measuring the AC power consumption.
Probably there's a voltage drop and the board resets.
 

· Registered
Joined
·
16 Posts
I suspect the PSU can't handle the power spikes.
You can't see the spikes when measuring the AC power consumption.
Probably there's a voltage drop and the board resets.
it sure feels like a PSU crash, silent "clack" and off everything goes. PSU is a beQuiet! Straight Power 11, which is not a low-end one.
Can higher total wattage PSUs automatically handle those spikes better? If yes, what wattage should i go for, i am ready to buy a PSU just for the sake of finding the culprit. Should 750W be sufficient?
 

· Old crazy guy
Joined
·
4,938 Posts
it sure feels like a PSU crash, silent "clack" and off everything goes. PSU is a beQuiet! Straight Power 11, which is not a low-end one.
Can higher total wattage PSUs automatically handle those spikes better? If yes, what wattage should i go for, i am ready to buy a PSU just for the sake of finding the culprit. Should 750W be sufficient?
Yes, if it's a good one 750W should be enough.
Quite surprising since it's a beQuiet... but we don't know which kind of spike this is.
A good 550W should be able to peak at 650W but maybe the constant load it's reducing this margin.
My advice is to look for something on Amazon to test and if it's not working send it back.
 

· Registered
Joined
·
356 Posts
it sure feels like a PSU crash, silent "clack" and off everything goes. PSU is a beQuiet! Straight Power 11, which is not a low-end one.
Can higher total wattage PSUs automatically handle those spikes better? If yes, what wattage should i go for, i am ready to buy a PSU just for the sake of finding the culprit. Should 750W be sufficient?
Sounds like the PSUs OCP trips. Look at how high the peak Amperage can go on the PSU - you probably trip it with the spikes from the 2080 Ti. If it worked before, the PSU probably degraded over time. Also: The straight power series isn't really high end. I'd go Dark Power Pro 11 750, Seasonic PX/TX-750, or a good Corsair.

See the LTT PSU reference table.

Edit: The specs of your PSU; also: it's a multi rail one, so unless you're using 12V3 and 12V4 for the GPU you're limited to 36A (432W peak) instead of 40A (480W peak), which is still cutting it close with the 2080Ti:
2468631


As multi GPU is pretty much dead, I'd go with a good single rail PSU such as the ones I've mentioned.

Edit 2: Have you tried using the higher rated rails for the GPU?
Edit 3: If you're on a budget, try one of the Fractal Design ION+ Platinum, they're decent as well.
Edit 4: And remember, most of your other stuff also occupies the 12V rails, such as fans and pumps, at least when connected via board header.
 

· Registered
Joined
·
13 Posts
Hey y'all! :)

Yesterday, I bought a X570 Aorus Master and equipped it with a brand new 5950X.

While the results in Cinebench R20 are quite good at stock, I am a bit disappointed in the FCLK tuning.
No matter what I try (changing around VDDG and SOC voltages, including LLC, going from F31j to k to o and back, ...), my PC will not boot with a FCLK above 1867Mhz (no matter if the RAM is synced or not, the RAM alone goes beyond 4000 MHz if I leave the FCLK at 1867MHz or below). Once I set the FCLK to 1900, my board gets stuck in a boot loop, always ending up on 07.

Does anyone else have that problem or did experience this in the past? Is this connected to the CPUs memory controller or is this connected to the Gigabyte board?
I have a MSI X570 Ace on the way to test this independently, but I was wondering if there's ways to remedy this without actually unboxing the alternative board :)

Lastly a question related to motherboard returns (in case I'd want to go for the MSI). Is there anything special I need to be aware of when returning a board (within the 14 days "right to return"-period)?
The only thing I've actively done is removing the plastics on the heatsinks, slotting in the CPU, RAM and GPU; attaching the WiFi antenna and updating the BIOS. Any potential risk of discussions when returning the board or should it be a "no brainer"?

Thank you in advance for your support, much appreciated. Should you have any questions w/r/t the performance of my hardware as is, please shoot!
 
13021 - 13040 of 19226 Posts
Top