Swap 13700k for 7800X3D

[R0H1T]

Ok guys, there's some misleading here.

One year ago, i created this thread here , it was a huge fight to put the CPU running fresh, and i've managed to do it when i swaped my CPU cooler to Dark Rock Pro 4 and some settings (basic stuff) in BIOS.

Now, another situation, my motherboard, like i said it's this one: https://rog.asus.com/motherboards/rog-strix/rog-strix-z790-h-gaming-wifi-model/helpdesk_bios/
I was running the BIOS ROG STRIX Z790-H GAMING WIFI BIOS 2102, until some days ago, i've upgraded the BIOS to the most recent one 2402, and then the CPU started to overheat again, a lot of settings (new ones) changed in BIOS interface, i tried to apply the same settings i had before the update and now i just can't stabilize the CPU again.

And yes, the thermal paste is correctly applied, the airflow is correct and i have a better airflow case then i had before when i created that thread. I'm just wasted of this because i don't have time and patience to be around the settings and the CPU all the time just to keep it fresh.

Try to get a refund instead & then upgrade to AMD. I wouldn't go back to Intel because they're yet to admit the full extent of the issue & looks like they're just trying to reduce their liability to a minimum here! Remember after the new ucode "fix" your CPU will run slower, even if just a few percentage points

 

[Dr. Dro]

Nah! With proper case cooling his Dark Rock Pro 4 cooler is more than capable of providing enough cooling.

That's true. OP either has a mounting problem or unlocked the CPU's wattage so the cooling is overwhelmed. Or his case is a total hotbox.

 

[phanbuey]

If you do change your CPU - then don't get an ASUS board. Both AMD and Intel are struggling on those boards big time, massive OV problems. I just helped a friend with a 14700K and a ASUS with some brutal LLC auto settings that were the same deal - instant 100C.

Grab an MSI or Giga, you may have better luck. Also kryosheet if you can do it.

 

[evernessince]

To be precise - Raptor Lake processors. Issue happens on 35W T SKUs as well. The lower end CPUs that are badged as 13/14th but actually contain Alder Lake 12th gen silicon (i3, lesser i5 SKUs) are immune to the problem. But this isn't relevant to the case at hand.

Correct, entirely my bad for not specifying "and up". Thank you for specifying all the caveats.

Ok guys, there's some misleading here.

One year ago, i created this thread here , it was a huge fight to put the CPU running fresh, and i've managed to do it when i swaped my CPU cooler to Dark Rock Pro 4 and some settings (basic stuff) in BIOS.

Now, another situation, my motherboard, like i said it's this one: https://rog.asus.com/motherboards/rog-strix/rog-strix-z790-h-gaming-wifi-model/helpdesk_bios/
I was running the BIOS ROG STRIX Z790-H GAMING WIFI BIOS 2102, until some days ago, i've upgraded the BIOS to the most recent one 2402, and then the CPU started to overheat again, a lot of settings (new ones) changed in BIOS interface, i tried to apply the same settings i had before the update and now i just can't stabilize the CPU again.

And yes, the thermal paste is correctly applied, the airflow is correct and i have a better airflow case then i had before when i created that thread. I'm just wasted of this because i don't have time and patience to be around the settings and the CPU all the time just to keep it fresh.

Looking at your past thread looks like you've been through the ringer...

I've seen quite a few people report this on the Intel sub-reddit as well where they had settings dialed in and have to redo them. You can see some of them here: https://www.reddit.com/r/intel/comments/1ek2n5l
Have you tried enabling Intel failsafe for SVID? You may have to manually lower voltages to get them in line with what they were at before.

Assuming you want to go through all that song and dance again. I can understand if you don't given the history here. Plus the CPU has been in use for over a year which makes a greater degree of degradation possible.

evernessince and others who are suggesting this is related to the micro-code bug are jumping to unfounded conclusions!

I say again, the OP has NOT indicated he is experiencing any instability or crashing problems.

Therefore, those advising he RMA the CPU are giving bad, or at least premature advice!

The micro-code problems cause instability after degradation reaches a certain point. Not having crashing absolutely does not mean that degradation hasn't occurred. This is the primary issue with Intel not starting a recall, there will be hundreds of thousands of people with Intel CPUs that have suffered silent degradation as a result of this problem and will only realize after their warranty has expired or at a later date. Even in the best case scenario, you have shaven years off the life of your CPU.

Again, no reports of instability issues and the heat issue appears to only have resurfaced after this BIOS update. That does not suggest a micro-code issue.

The lack of a symptom does not preclude having a disease that's known to be a asmptomatic until a certain point. Particularly important given that the doctor (Intel) has already stated that our OP's patient (his CPU) is infected.

Except moving to a Zen6 processor would also require a totally different motherboard and possibly a new OS license too. A full refund on the i7 would hardly cover that.

Assuming he doesn't have a retail license, yes he'll have to spend $9 to get a new windows key. $9 is not worth nothing.

 

[Bobaganoosh]

As GN pointed out, people with affected CPUs should RMA out of an abundance of caution.

Do not just RMA things "out of an abundance of caution". That's now how that works. You can't just submit an RMA and say "a Youtuber said I should even though there's nothing wrong with it". If they're having stability issues, sure, RMA, but otherwise this sounds like there's either an issues with the mount/cooler or there's an issue with the BIOS settings. Being they suddenly had overheating issues after the BIOS update, I'd guess that's where the problem is.

OP, please indicate what settings you've changed in your BIOS and what you have for CPU settings (voltage, power, frequencies, etc. especially anything that you've modified from default). It would also be helpful to show a shot of HWiNFO64 or something showing your temperatures, power levels, and voltages. Then we can help you figure out what the actual problem is.

If you're just doing this for personal reasons and don't want to solve this problem, Newegg has a CPU trade-in program that will let you trade in the CPU (for some price) towards buying a new one. Word of warning though, some of be quiet!'s coolers tend to not work as well on AMD CPUs due to the way their plates are made and they don't tend to have any off-set mounts.

 

[R0H1T]

Except moving to a Zen6 processor would also require a totally different motherboard and possibly a new OS license too. A full refund on the i7 would hardly cover that.

Better than suffering on an unstable system, assuming it really is due to the CPU.

 

[anonuser57]

The cost of the new CPU and Mobo combined with the fact that you will not be able to get decent resale prices on your existing hardware make it not worth it. Plus your current CPU should not be reaching such high temps while gaming with your existing cooler. It is more likely you have a bad mount or inadequate airflow.

 

[docnorth]

100 C is a setup problem not a CPU problem.

The cooler you have is arguably better than the old single fan NH-U14S used here.

View attachment 357735


Can you point this part out?

TPU tests with a single fan air cooler.

View attachment 357737

Mostly agree, except that the old NH-U14S is not weaker, but (at least) on par vs DRP-4. It's difficult to find reviews for the old Noctua, but the NH-D15S can be used as comparison, it has almost the same performance rating.

 

[3DVCash]

If you do change your CPU - then don't get an ASUS board. Both AMD and Intel are struggling on those boards big time, massive OV problems.

Still? AFAIK that was sorted over a year ago on their AM5 boards. At least I haven't had any issues.

And echoing the general sentiment here, something is def wrong with your setup, OP. Whether it's the mount, a BIOS setting, your case, or a combo of all 3, there's no reason an i7 should be running that hot while gaming. I'd get to the bottom of your issue there first before deciding to jump ship.

 

[Zazigalka]

I think I saw someone, iirc one of staff members, mention that underclocking his e-cores alone resulted in some truly insane temperature drop. try that.

 

[Bobaganoosh]

I just looked at your old thread (not the whole thing, admittedly), but your voltage was quite high. If you're not overclocking, you should be able to lock the voltage (with voltage override) at somewhere between 1.3-1.4V and run everything you need. It was hitting 1.542V or something at only 5.4GHz. That's too much. You may also be able to leave the voltage on adaptive and just put in a -100mV offset or something. It's a little different how each motherboard handles that and I haven't used yours so I'm not sure the easiest path there, but I would start by trying to get that voltage lower.

Also, one of your posts seemed to indicate that you put the CPU in another motherboard, then returned that motehrboard for other problems, went back to the first motherboard (without changing any settings) and suddenly the temps were much better. This would indicate that either your thermal paste was bad (paste went bad in the tube, which can happen over time, especially if it separates), application was bad, or the mounting of the heatsink on the motherboard/CPU wasn't great. These can all happen, even to the most experienced builders, so no offense is intended here, but it's worth checking. For example, if you very carefully remove the screws and pull the heatsink straight off, take a picture of the CPU (with the paste spread as it is without touching it) and share that with us. It may prove helpful. Just an example from some of my trial and error runs previously (dealing with an issue I had a while back):

 

[ShrimpBrime]

Hello!

I have a Intel i7 13700k + Asus ROG STRIX Z790-H GAMING + Corsair Vengeance RGB DDR5 6000MHz 32GB, and after those problems with Intel degrading CPU's and now in the summer wasted of seeing that motherboard red light warning that my CPU is at 100º when i'm playing specific games are getting me wasted of having this combination and i refuse to buy an AIO just for this crappy hot CPU, when my Dark Rock Pro 4 is a beast of an air cooler.

My real question here is, what i do in my PC is 90% gaming, basically that, sometimes my Girlfriend uses my PC for some heavy renders at Archicad but it's very rare to happen, is it dumb to swap my current setup for a Ryzen 7 7800X3D and pair it with a motherboard like ATX Asus ROG Strix B650E-F Gaming WiFi, since it's new in features and already AM5 for future upgrades?

Thanks in advance for all opinions you can provide.

P.S. If this is the wrong section, please move it (System Builder's Advice??)

It's normal. Intel clocks to it's thermal max.

The Asus board. Look for a setting called "Asus performance enhancement", should be near the top of the OC page. Disable this, it's enabled by default.

If you want to run cooler without a sacrifice to performance, lower the CPU LLC to lvl 2 and test. At level 1, you'll loose performance a bit, but it'll run much much cooler. May not have a big impact on gaming, you'll have to test it.

Aside from that, XMP enabled will increase the Vccsa voltage and is on die. This also increases cpu thermals. So you could test lowering system agent voltage and see if that helps. You shouldn't need more than 1.2v for just running XMP, I'd see if you could manually set this to 1.15 or maybe even 1.10v and see if that helps.

Otherwise good luck!!

 

[Outback Bronze]

Your CPU looks solid mate, you just gotta do some undervolting.

Apparently not all CPU's are affected. Intel has stated this.

"Intel Core 13th and 14th Generation desktop processors with 65W or higher base power—including K, KF, KS and 65W non-K variants—could be affected by the elevated voltages issue. However, this does not mean that all processors listed are (or will be) impacted by the elevated voltages issue."

Here is a good read about it and lower down the page how to undervolt. He uses a 12900KS and runs a -0.1000v.

Edit: forgot to add the link, doh lol: Intel’s 'Raptor Lake' Desktop CPU Bug: What to Know, What to Do Now (pcmag.com)

"I regularly run the Core i9-12900KS in my system with the offset at its lowest allowed value of -0.100v (-100 millivolts) to help reduce heat and power consumption"

Good luck with your decisions.

 

[evernessince]

Do not just RMA things "out of an abundance of caution". That's now how that works. You can't just submit an RMA and say "a Youtuber said I should even though there's nothing wrong with it".

You are taking a snippet out of context. Intel have admitted there is a defect that causes permanent damage to 13th and 14th gen CPUs (with some limitations) and his SKU is one of the top effected. These are not remotely normal circumstances.

"nothing wrong with it"? You have zero ability to say that given it's impossible to tell if and how much his CPU has degraded. This is the problem with the lack of information from Intel.

Telling OP to keep his CPU is irresponsible.

 

[izy]

Hello!

I have a Intel i7 13700k + Asus ROG STRIX Z790-H GAMING + Corsair Vengeance RGB DDR5 6000MHz 32GB, and after those problems with Intel degrading CPU's and now in the summer wasted of seeing that motherboard red light warning that my CPU is at 100º when i'm playing specific games are getting me wasted of having this combination and i refuse to buy an AIO just for this crappy hot CPU, when my Dark Rock Pro 4 is a beast of an air cooler.

My real question here is, what i do in my PC is 90% gaming, basically that, sometimes my Girlfriend uses my PC for some heavy renders at Archicad but it's very rare to happen, is it dumb to swap my current setup for a Ryzen 7 7800X3D and pair it with a motherboard like ATX Asus ROG Strix B650E-F Gaming WiFi, since it's new in features and already AM5 for future upgrades?

Thanks in advance for all opinions you can provide.

P.S. If this is the wrong section, please move it (System Builder's Advice??)

I would say if you really want to swap your CPU you should wait for the 9000 series from AMD , get an 9700X / 9800X3D when they will be released or if you want to spend more wait for 9950X3D (if you want to render stuff fast too), i dont see any reason for you to swap your current CPU for an 7800x3d.

 

[Bobaganoosh]

You are taking a snippet out of context. Intel have admitted there is a defect that causes permanent damage to 13th and 14th gen CPUs (with some limitations) and his SKU is one of the top effected. These are not remotely normal circumstances.

"nothing wrong with it"? You have zero ability to say that given it's impossible to tell if and how much his CPU has degraded. This is the problem with the lack of information from Intel.

Telling OP to keep his CPU is irresponsible.

You have to prove there's an issue to be able to RMA things. They didn't recall it, as you said. So this is a "it's fine until you prove it isn't" situation and the responsibility is on the user to solve the issue. You just jumped right into "it's faulty, RMA!" (paraphrasing) with no evidence that it's defective. Notice I said "IF there's nothing wrong with it", because I can't tell from the information provided if there is or is not a degradation issue and I don't think you can either. You even admitted that you're just suggesting they return it because Steve said so. Intel also said that not everyone will be effected. Being that we all understand there's no recall, and not enough evidence has been provided, there's more work to do before they can RMA.

If you read my posts, you would see that I suggested if there is an issue with stability and crashing, they should RMA the CPU. If there's just an overheating issue, which seems to be the case, and the evidence provided indicates that it's a setup issue, which it does, than I tried to offer suggestions on how to fix said setup issues. That would be the appropriate and responsible next step. What you're suggesting is that they RMA the CPU with no evidence of failure. If Intel even accepts that, they'll likely send a new CPU that is the same model. Then the OP will have the same model CPU that will likely do the same thing if the operating conditions don't change (because it looks like their motherboard is giving too much voltage and it's possible there's a bad paste situation on top of that). Also, Intel hasn't fixed anything yet, so any replacement would be just as likely to have the same issues. So RMAing a CPU that isn't known to have a failure only opens OP up to possible future failures that they aren't having right now. Does that sound responsible?

 

[warriorunited]

I can't answer to everyone, just too many tips and very useful information.

My BIOS settings were basically XMP I activated, with Case/CPU FANS running at Turbo Speed instead of "Normal", since i use be quiet! silent wings fans, they are super quiet anyways, and all the rest were a stock setting and ASUS multicore enchancement enabled.

So, i've followed the suggest to disable that, and i'm using XMP II instead of I, the difference seems to be that XMP II uses RAM total default settings, not Asus Optimized.

Running Cinebench after 2mins, the results bellow (it reached 100º after a few seconds i took the screenshot).

If you want to run cooler without a sacrifice to performance, lower the CPU LLC to lvl 2 and test. At level 1, you'll loose performance a bit, but it'll run much much cooler. May not have a big impact on gaming, you'll have to test it.

Where can i find this?

 

[Bobaganoosh]

I can't answer to everyone, just too many tips and very useful information.

The easiest setting to try would be an offset to adaptive voltage. You could try -50mV...if you run some single and multi-threaded tests and that's fine you may even be able to run -100mV. If you run into a crash, back that number off. Keep an eye on your VID values and temperatures to see if the setting is actually working.
Edit: (^he said, not knowing where that setting is in Asus BIOS lol). I found this guide where someone was doing a similar thing with their 13900k with the same motherboard. Quote from that link:

Now for the undervolting. The ONLY setting that I messed with for this is the GLOBAL CORE SVID VOLTAGE. It will need to be put on adaptive mode. The OFFSET MODE SIGN should be negative (-). Then I started tweaking the offset voltage. I would input a value, which I started at .08, but many recommend starting with .05, then I would start my computer, run a CINEBENCH test with the minimum test duration off, and log my scores, and temps. Like I said I started with -.08. Here were my scores:

Stock Voltage :

• CINEBENCH score of 37884
• Max VID Voltage was 1.442V
• Max Package temps were 97c
• Max package power was 313W

-.08 Voltage Offset:

• CINEBENCH score of 37879
• Max VID Voltage was 1.353V
• Max Package temps were 84c
• Max Package Power was 253.67W

 

[R0H1T]

Well that thing is frying itself, aside from changing some "stock" settings use this for monitoring temps ~

Free Download HWiNFO Sofware | Installer & Portable for Windows, DOS

Start to analyze your hardware right now! HWiNFO has available as an Installer and Portable version for Windows (32/64-bit) and Portable version for DOS.

www.hwinfo.com

Hwmonitor isn't as reliable IMO & hwinfo gives you a lot more data.

 

[warriorunited]

Now using HWiNFO:

 

[Bobaganoosh]

Now using HWiNFO:

View attachment 357755

Interesting that now your voltage and power levels seem lower. Did you change any more settings since the last run? I edited my post above with a link to a guide showing where to try an adaptive offset. That said, I would think the power levels above should be ok with that CPU cooler. I'm still wondering if there's a paste/mount issue.

 

[R0H1T]

You should disable this ~

The Asus board. Look for a setting called "Asus performance enhancement", should be near the top of the OC page. Disable this, it's enabled by default.

Voltage seems ok but chip is running at its limit. If it's not showing as APE it could be something like multi core enhancement ~

Not my board but the settings are probably in the same place.

 

[warriorunited]

The easiest setting to try would be an offset to adaptive voltage. You could try -50mV...if you run some single and multi-threaded tests and that's fine you may even be able to run -100mV. If you run into a crash, back that number off. Keep an eye on your VID values and temperatures to see if the setting is actually working.
Edit: (^he said, not knowing where that setting is in Asus BIOS lol). I found this guide where someone was doing a similar thing with their 13900k with the same motherboard. Quote from that link:

I've followed that guide, i assume where he says "0.5" in BIOS the number is 0.05000, correct? Sorry i'm trying to learn about this . The rest is already set at Adaptive Mode and Negative.

You should disable this ~

Voltage seems ok but chip is running at its limit. If it's not showing as APE it could be something like multi core enhancement ~

Not my board but the settings are probably in the same place.

I've already disabled that, the results are above

 

[Bobaganoosh]

I've followed that guide, i assume where he says "0.5" in BIOS the number is 0.05000, correct? Sorry i'm trying to learn about this . The rest is already set at Adaptive Mode and Negative.

Yes lol. he means 50mV, or 0.050V.

 

[ShrimpBrime]

I can't answer to everyone, just too many tips and very useful information.

My BIOS settings were basically XMP I activated, with Case/CPU FANS running at Turbo Speed instead of "Normal", since i use be quiet! silent wings fans, they are super quiet anyways, and all the rest were a stock setting and ASUS multicore enchancement enabled.

So, i've followed the suggest to disable that, and i'm using XMP II instead of I, the difference seems to be that XMP II uses RAM total default settings, not Asus Optimized.

Running Cinebench after 2mins, the results bellow (it reached 100º after a few seconds i took the screenshot).





Where can i find this?

Ok, so first, please use HWInfo64 instead of HWMonitor so we have accurate readings.

Edit, NVM, just saw you switched already

You can find Load Line Calibration (LLC) in the Digi VRM tab. Should be above the v-core section.