AMD's Robert Hallock Confirms Lack of Manual CPU Overclocking for Ryzen 7 5800X3D

[Deleted member 24505]

I noticed he did say CPU core frequency overclocking or core voltage adjustment, so stock vcore only?

 

[neatfeatguy]

I'm not sure why some folks are a bit put off by lack of OC capabilities of this up and coming 5800X3D.

Years ago OCing was an awesome way to push extra out of a CPU. I remember taking the AMD 64 X2 3800+ up from 2.0GHz to almost 3.2GHz.
I was able to push my PII x4 940 from 3.0 to 3.71 and that certainly helped.
I enjoyed the simplified OC capabilities of the PII and the i5-4670k that I ran at 4.4.

However, with how the current Ryzen CPUs manage boosts and how overclocking tends to give, overall, a minimal increased performance over letting the system manage boosts, I couldn't care less about being able to push an overclock on these CPUs.

 

[bug]

I'm not sure why some folks are a bit put off by lack of OC capabilities of this up and coming 5800X3D.

It's not a big deal, technically.
But for years, AMD rubbed it in Intel's face that they only sell fully unlocked CPUs, enabling users to extract all they want from them. This CPU flies in the face of all that. And it's worth pointing out.

 

[Taraquin]

Literally who OC's Ryzen clock, tune the memory and IF and that's it.

Pbo and curve optimizer can easily net you 5-6%. Even if pbo is disabled, using negative CO is a big win since it can raise allcore frequency by several hundred MHz within pwr budget at same consumotion/temp.

Static OC I agree is dead on Ryzen 5k. On Ryzen 3k it could be really good sometimes if you got a golden sample 3600 and could run 4.4@1.25v vs 4GHz stock.

 

[neatfeatguy]

It's not a big deal, technically.
But for years, AMD rubbed it in Intel's face that they only sell fully unlocked CPUs, enabling users to extract all they want from them. This CPU flies in the face of all that. And it's worth pointing out.

On noes....! One CPU that goes against the grain.

I suppose I can understand if something like that bothers some.....but it's no sweat off my back.

 

[SL2]

But for years, AMD rubbed it in Intel's face that they only sell fully unlocked CPUs, enabling users to extract all they want from them. This CPU flies in the face of all that. And it's worth pointing out.

An AMD representative have told us about the lack of manual overclocking more than a month ahead of the launch, and it has been pointed out numerous times across the net.
I don't see any piece of info missing so far. Repeating this news title isn't worth anything.

 

[ThrashZone]

Hi,
Might of been a meaningful story if there were at least some tests to confirm catching the 12900k as gaming champ but without that info it's just youtube waste.

 

[SL2]

Hi,
Might of been a meaningful story if there were at least some tests to confirm catching the 12900k as gaming champ but without that info it's just youtube waste.

Well the info about the lack of manual OC is important regardless.

I guess it's too soon for details like that, but personally I don't expect the 5800X3D to be the faster one. Then there's the 12900KS.. which seems to have been news to AMD as well.

 

[Makaveli]

if this chip can do a 2000 FLK then PBO+CO with DDR4000 memory I think it will be pretty quick.

 

[ThrashZone]

Well the info about the lack of manual OC is important regardless.

I guess it's too soon for details like that, but personally I don't expect the 5800X3D to be the faster one. Then there's the 12900KS.. which seems to have been news to AMD as well.

Hi,
12900ks is just intel milking the market before amd releases this chip.

As far as this chip and being locked to turbo clocks/ voltages it may not even need to be oc'ed and IF it gets close to either 12900-ks who really will care at that point 12900ks is 800.us

Sucker bait

 

[bug]

On noes....! One CPU that goes against the grain.

I suppose I can understand if something like that bothers some.....but it's no sweat off my back.

It depends on how much you're willing to read into it.
CPU unable to sustain manual overclock like its siblings can mean that it has not undergone sufficient testing (wrt overclocking), meaning AMD somehow felt a need to rush it out. Or it can mean the 3D cache raises some problems AMD was unable/unwilling to fix.

TL;DR It makes 5800X3D make like an epeen CPU ("look at me, I'm faster than Intel"), just like Intel made epeen out of 12900k when the gave it that humongous TDP limit just to be able to claim they can beat AMD.

 

[ThrashZone]

Hi,
Cache oc'ing has been known to fry a chip.

 

[bug]

Hi,
Cache oc'ing has been known to fry a chip.

Nobody suggested the cache should be overclocked. Quite the contrary, manually overclocking this probably affects the 3D cache when it shouldn't, that's why it was axed.

 

[ThrashZone]

Nobody suggested the cache should be overclocked. Quite the contrary, manually overclocking this probably affects the 3D cache when it shouldn't, that's why it was axed.

Hi,
Can't individualize oc'ing man well I guess you're attempting to

It's all oc'ing to me, one for all and all for none in this case irony

 

[TheinsanegamerN]

Non-issue, imo. Memory and Infinity Fabric settings remain unlocked, and if curve optimizer also works, this should be as versatile as the other Ryzen processors, IMO. The automatic tweaking algorithm on these processors is optimized to near perfection, and admittedly, above my own manual overclocking skills.


When did that happen? Not to brag but... my mobile RTX 3050 can OC and it OCs hard... it will run >2.1 GHz... insane little chip

Back in 2015:

[GeForce Forums] Nvidia has officially blocked 900M overclocking

"Unfortunately GeForce notebooks were not designed to support overclocking. Overclocking is by no means a trivial feature, and depends on thoughtful design of thermal, electrical, and other considerations. By overclocking a notebook, a user risks serious damage to the system that could...

forums.anandtech.com

The backlash was harsh enough that nvidia eventually backed off.

I'm not sure why some folks are a bit put off by lack of OC capabilities of this up and coming 5800X3D.

Years ago OCing was an awesome way to push extra out of a CPU. I remember taking the AMD 64 X2 3800+ up from 2.0GHz to almost 3.2GHz.
I was able to push my PII x4 940 from 3.0 to 3.71 and that certainly helped.
I enjoyed the simplified OC capabilities of the PII and the i5-4670k that I ran at 4.4.

However, with how the current Ryzen CPUs manage boosts and how overclocking tends to give, overall, a minimal increased performance over letting the system manage boosts, I couldn't care less about being able to push an overclock on these CPUs.

Because taking away control from end users is generally seen as a bad thing. Even if the gains are minor, that doesnt mean people like platforms being more locked down. AMD has long run on the platform of having unlocked CPUs across the lineup, and now have become massive hypocrytes by locking down their own hardware. AMD fanbois will rip on intel for making locked CPUs and "having lotsof locked SKUs" yet will defend AMD when they lock one of their own chips.

It's a slippery slope. Its one CPU, or its just the low end, oh its just the ryzen 7 and lower, its not like you can do much, ece ece.

On noes....! One CPU that goes against the grain.

I suppose I can understand if something like that bothers some.....but it's no sweat off my back.

You're comparing a whole lineup to one single half-generation old, possibly limited in availability, almost niche, overprized SKU.

It's not like you don't have options. Drop the bitterness.

Like I said, it's OK when AMD does it. Intel does it, the community will rip them apart, nvidia does it, the forums light up. AMD does it "lol why so bitter bro, its not like it matters bro".

This is what AMD fanbois call "mindshare" and claim all their competitors benefit from.

 

[Valantar]

Because taking away control from end users is generally seen as a bad thing. Even if the gains are minor, that doesnt mean people like platforms being more locked down. AMD has long run on the platform of having unlocked CPUs across the lineup, and now have become massive hypocrytes by locking down their own hardware. AMD fanbois will rip on intel for making locked CPUs and "having lotsof locked SKUs" yet will defend AMD when they lock one of their own chips.

It's a slippery slope. Its one CPU, or its just the low end, oh its just the ryzen 7 and lower, its not like you can do much, ece ece.


Like I said, it's OK when AMD does it. Intel does it, the community will rip them apart, nvidia does it, the forums light up. AMD does it "lol why so bitter bro, its not like it matters bro".

This is what AMD fanbois call "mindshare" and claim all their competitors benefit from.

IMO it's more of a case of "okay, it seems reasonable to limit this if there's significant risk of damage to the hardware if left open". The same line of reasoning isn't applicable to any traditional CPU (there is risk, but it isn't significant - you need to try quite hard or be entirely devoid of knowledge to damage a conventional CPU with OCing), hence why the context makes this reasoning applicable here and not in other cases. This would be especially true if settings that are safe for every other 5000-series CPU are suddenly highly dangerous to this one. Does this constitute the start of a slippery slope? Unlikely IMO - if anything, features like PBO+ and CO are proof that if anything, AMD is working to provide better, smarter OC functionality as time goes on. It is of course entirely possible that this constitutes an about-face, and that they will indeed be cracking down on OCing from now on. But that would be quite surprising, and a significant break with (even recent) actions from their side.

 

[bug]

@Valantar So, AMD fixing whatever causes the "significant risk of damage" before launching this is not an option to you? You just go for whatever AMD says, hook, line and sinker?

 

[SL2]

Like I said, it's OK when AMD does it. Intel does it, the community will rip them apart, nvidia does it, the forums light up. AMD does it "lol why so bitter bro, its not like it matters bro".

This is what AMD fanbois call "mindshare" and claim all their competitors benefit from.

You're still failing to see the difference between a single SKU and a whole mobile GPU generation, and calling people fanboys doesn't change that.

Because you have, like I said, plenty of options. Alder Lake offers great performance, and slower and faster models are critically acclaimed. Don't like Intel? Well AMD is about to launch more CPU's, even though they're not really new anymore. AM5 is maybe 6 months away if you want something newer, and right after that there's Raptor Lake.

Comparing all that to a situation when Nvidia in reality had very little competition in mobile gaming is just very misleading and biased.

You have options, don't be bitter.

https://pcpartpicker.com/products/cpu/#F=96,98

 

[Valantar]

@Valantar So, AMD fixing whatever causes the "significant risk of damage" before launching this is not an option to you? You just go for whatever AMD says, hook, line and sinker?

Let's see ... the cache die can't handle more than 1.35V, and it seems its voltage is tied to vCore. Other Ryzens routinely run vCore above 1.4V. How, exactly, do you propose this "fix this before launch"? By significantly restructuring the Vermeer die to separate out a cache voltage rail? Or by magically overcoming what is likely a limitation of the high density TSMC 7nm cache libraries used for the cache die? Because that's likely what such a fix would require. So no, I don't see that as a likely option, no.

As for whether this constitutes "going for whatever AMD says", that ... I'll leave that for you to judge. I'll side on the side of "let's not allow idiots to break their chips too easily" on this one. You're welcome to disagree.

 

[ThrashZone]

Hi,
OEM's lock chips all the time not much is made of it just another job for @unclewebb to deal with throttlestop

 

[SL2]

It's a slippery slope. Its one CPU, or its just the low end, oh its just the ryzen 7 and lower, its not like you can do much, ece ece.

It seems like you don't understand why this happened at all. It's just AMD being mean and greedy?
You really haven't seen the video, or you just don't believe what they say. Your reasoning right there is a slippery slope, and all you do is adding FUD.

AMD had three options.

1 - Launch the 5800X3D like they say the will, no OC.
2 - Not launching at all.
3 - Allowing OC even though AMD knows that the CPU will break.

Not a hard choice really.

 

[Turmania]

I personally would not touch a cpu that has voltage limitations. It does not install confidence in me that it will be long lasting product even under normal conditions. As well it does not install confidence in me that when it starts to heat up it can shut down the system during a session. Why risk it?

 

[Valantar]

I personally would not touch a cpu that has voltage limitations. It does not install confidence in me that it will be long lasting product even under normal conditions. As well it does not install confidence in me that when it starts to heat up it can shut down the system during a session. Why risk it?

Why would it shut down if it's being kept below tJmax? These limits are a widely documented part of the spec; keep within them and you should be fine, and if not then that qualifies you for a warranty replacement of the CPU, as it's not performing as per specifications.

 

[Makaveli]

I personally would not touch a cpu that has voltage limitations. It does not install confidence in me that it will be long lasting product even under normal conditions. As well it does not install confidence in me that when it starts to heat up it can shut down the system during a session. Why risk it?

Why would a cpu not last long with the manufacturer setting a voltage limit and someone keeping it at stock.

I'm not seeing the logic in this post.

 

[TheinsanegamerN]

You're still failing to see the difference between a single SKU and a whole mobile GPU generation, and calling people fanboys doesn't change that.

Because you have, like I said, plenty of options. Alder Lake offers great performance, and slower and faster models are critically acclaimed. Don't like Intel? Well AMD is about to launch more CPU's, even though they're not really new anymore. AM5 is maybe 6 months away if you want something newer, and right after that there's Raptor Lake.

Comparing all that to a situation when Nvidia in reality had very little competition in mobile gaming is just very misleading and biased.

You have options, don't be bitter.

https://pcpartpicker.com/products/cpu/#F=96,98

It's hypocritical of AMD, ater using intel's locked CPUs as a talking point, to then lock their own CPUs.

Options from other companies existing is a red herring argument.

It seems like you don't understand why this happened at all.

Higher voltage can damage the cache. Not hard to understand.

It's just AMD being mean and greedy?

It's not like AMD jumped the price on their CPUs by 30-50 percent with the 5000 series, or refused to support the 400 series chipsets until public backlash forced their hand, or like they did the same thing with the 300 series.

Newsflash, AMD is a corporation. All corporations are greedy, and need to be held to task when that greed spirals out of control.

You really haven't seen the video, or you just don't believe what they say. Your reasoning right there is a slippery slope, and all you do is adding FUD.

AMD had three options.

1 - Launch the 5800X3D like they say the will, no OC.
2 - Not launching at all.
3 - Allowing OC even though AMD knows that the CPU will break.

Not a hard choice really.

"muh FUD" - every fanboi ever.

AMD could have, you know, allowed OC and locked the voltage so the cache doesnt get hurt. Just an idea.

Stop accusing me of FUD when all of your arguments rely on baseless handwaving of any points the opposition makes.