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

[Valantar]

Would it not be possible for the 3D cache to run off a separate voltage plane?

Probably not - it interfaces directly with the on-die L3 cache in a way that's supposed to be entirely transparent to the CPU. Making that work while having those run off separate voltage planes sounds ... complicated, if not impossible, given how you'd have signalling between the two cache blocks at different voltages that would then need conversion, adding latency, making the cache die slower to acces, making for inconsistent performance and unpredictable pipeline stalls.

 

[Taraquin]

I think thats the one, there is a graph somewhere where it shows consumption across all benches, and yes the 12900k is both the fastest and the most efficient compared to the 5950x. Ill find that once im on my pc, im on the phone right now.


Well the 5950x has 33% more threads yet we are still comparing then, so does it matter?

I dont know, all i remember about 10900k was people claiming its an oven toaster etc.,not realising it is as efficient as the 5800x

In most scenarios it is not as efficient as 5800X.
Energy efficiency SuperPI and CB

Since 5800X and 5950X has same power limits stock they generally use the same power when stressed max (140W-ish) and in multicore 5950X would win.

I never said 10900K was a toaster, 11900K on the other hand

 

[JustBenching]

How have you come to that conclusion? Some workloads it hits crazy power usage right? Or am I misunderstanding something. Or are you talking with capped power?

Is it the case its a good chip but just brought to market in a bad way with its shipping configuration?

Exactly, its shipped with a stupid high power limit.

I don't think my 5800x is so far off with power consumption to 12900k either though.

Im pretty sure there is a huge difference in all core workloads. For example i think I can hit 15k (thats how much 5800x gets right?) cbr23 at around 45-50w

In most scenarios it is not as efficient as 5800X.
Energy efficiency SuperPI and CB

Since 5800X and 5950X has same power limits stock they generally use the same power when stressed max (140W-ish) and in multicore 5950X would win.

I never said 10900K was a toaster, 11900K on the other hand

Yes the 11th gen was pretty atrocious.

What you are seeing there is power consumption, not efficiency.

 

[Taraquin]

Exactly, its shipped with a stupid high power limit.


Im pretty sure there is a huge difference in all core workloads. For example i think I can hit 15k (thats how much 5800x gets right?) cbr23 at around 45-50w


Yes the 11th gen was pretty atrocious.

What you are seeing there is power consumption, not efficiency.

Show me where 10900K is more efficient? Overall consumption i CB is far lower on 5800X than 10900K, yet it scores higher. It's also faster in single using lesd energy. There may be a few very Intel optimized apps where 10900K is more efficient, but generally 5800X wins in almost everything.

 

[JustBenching]

Show me where 10900K is more efficient? Overall consumption i CB is far lower on 5800X than 10900K, yet it scores higher. It's also faster in single using lesd energy. There may be a few very Intel optimized apps where 10900K is more efficient, but generally 5800X wins in almost everything.

I think its GN that runs 30 minute long blender runs and it shows both at the same efficiency

 

[ThrashZone]

Show me where 10900K is more efficient? Overall consumption i CB is far lower on 5800X than 10900K, yet it scores higher. It's also faster in single using lesd energy. There may be a few very Intel optimized apps where 10900K is more efficient, but generally 5800X wins in almost everything.

Hi,
I'd be surprised if anyone buying Intel would be doing it because Intel is efficient

Intel has always been the choice for overclocking which is the opposite of what efficiency is in other words overclocking Intel systems can effectively score higher in benchmarks

 

[Taraquin]

I think its GN that runs 30 minute long blender runs and it shows both at the same efficiency

This one with 10900K long term pwr limited:

But in general 5800X beats the crap out of 10900K in almost every scenario on efficiency. In gaming stock it performs better and uses 30W less in the TPU test.

 

[JustBenching]

This one with 10900K long term pwr limited:
View attachment 240309

But in general 5800X beats the crap out of 10900K in almost every scenario on efficiency. In gaming stock it performs better and uses 30W less in the TPU test.

Yeav thats the one I think. Do you see it beating crap? I dont.

 

[Taraquin]

Yeav thats the one I think. Do you see it beating cheap? I dont.

Beating cheap? I dont understand what you mean...

 

[JustBenching]

Beating cheap? I dont understand what you mean...

Sorry, autocorrect. Do you see it beating the crap out of the 10900k? They are pretty much identical in efficiency

 

[Valantar]

Sorry, autocorrect. Do you see it beating the crap out of the 10900k? They are pretty much identical in efficiency

That graph is of power consumption, not efficiency. In the 12900K review, the companion graph for performance is at 10:18, and shows the 5950X - at that power draw (see 24:38) - finishing the render in 9.3 minutes compared to 16.5 minutes for the 10900K and 17.5 minutes for the 11900K. So, for this workload, the 5950X is dramatically more efficient than the 10900K, and even slightly beats the 12900K (9.4 minutes, so really not by a lot), despite it running with Intel's ludicrous 241W unlimited boost "stock" settings against the stock 5950X's limit of 144W. The 12900K can be reined in dramatically and will run far more efficiently if the power limits are lowered, but it's still slower in Blender than a 5950X. But the 5950X definitely beats the crap out of the 10900K in this testing, even when both are limited by similar long-term power limits.

 

[Taraquin]

Sorry, autocorrect. Do you see it beating the crap out of the 10900k? They are pretty much identical in efficiency

In blender, not in almost every other app and gaming. You could say that 10900K at it's best running very powerlimited can be as efficient, but in most cases it is not. In SuperPI, CB and gaming it beats the crap put of 10900K in efficiency

 

[Space Lynx]

as someone who got a 5600x on launch day, I found overclocking ryzen to be completely stupid anyway, no gains in real world gaming, so no point for me. I am fine with this decision. ram OC'ing scratches that oc'ing itch for me nowadays anyway. though I don't expect I will bother there either anymore, XMP, and away I game, life is too short

 

[JustBenching]

That graph is of power consumption, not efficiency. In the 12900K review, the companion graph for performance is at 10:18, and shows the 5950X - at that power draw (see 24:38) - finishing the render in 9.3 minutes compared to 16.5 minutes for the 10900K and 17.5 minutes for the 11900K. So, for this workload, the 5950X is dramatically more efficient than the 10900K, and even slightly beats the 12900K (9.4 minutes, so really not by a lot), despite it running with Intel's ludicrous 241W unlimited boost "stock" settings against the stock 5950X's limit of 144W. The 12900K can be reined in dramatically and will run far more efficiently if the power limits are lowered, but it's still slower in Blender than a 5950X. But the 5950X definitely beats the crap out of the 10900K in this testing, even when both are limited by similar long-term power limits.

Of course, wasnt comparing the 10900 to the 5950x but to the 5800x.

In blender, not in almost every other app and gaming. You could say that 10900K at it's best running very powerlimited can be as efficient, but in most cases it is not. In SuperPI, CB and gaming it beats the crap put of 10900K in efficiency

What do you mean very power limited? It's as power limited as the 5800x since they consume the same

In gaming im sure they are pretty similar as well, zen 3 consume a lot during gaming, dunno why

 

[Taraquin]

Of course, wasnt comparing the 10900 to the 5950x but to the 5800x.


What do you mean very power limited? It's as power limited as the 5800x since they consume the same

If you run 10900K w/o pwr limits like many do it's very inefficient, if pwr limited it can be efficient in some apps like blender.

 

[JustBenching]

If you run 10900K w/o pwr limits like many do it's very inefficient, if pwr limited it can be efficient in some apps like blender.

Well any cpu if asked to pull 250w will be inefficient. Thats pretty normal. The question is why the heck would you do that if you are interested in power efficiency? What you are saying really makes no sense, you care about power efficiency but you are going to run the cpu way above its efficiency curve cause....??

 

[ThrashZone]

Well any cpu if asked to pull 250w will be inefficient. Thats pretty normal. The question is why the heck would you do that if you are interested in power efficiency? What you are saying really makes no sense, you care about power efficiency but you are going to run the cpu way above its efficiency curve cause....??

Hi,
Keeps the argument going.

 

[Makaveli]

Hi,
Keeps the argument going.

Discussion

 

[Taraquin]

Well any cpu if asked to pull 250w will be inefficient. Thats pretty normal. The question is why the heck would you do that if you are interested in power efficiency? What you are saying really makes no sense, you care about power efficiency but you are going to run the cpu way above its efficiency curve cause....??

Intel have 2 specs, pwr limited and unlimited toggled by one bios click. Standard spec on AMD is 142W-ish limit and raising it is a much more complucated approach via many sub menus in bios.

Your statement was that they are equally efficient, mine is that they are not except in rare cases.

 

[JustBenching]

Intel have 2 specs, pwr limited and unlimited toggled by one bios click. Standard spec on AMD is 142W-ish limit and raising it is a much more complucated approach via many sub menus in bios.

Your statement was that they are equally efficient, mine is that they are not except in rare cases.

My statement is that if you care about efficiency you wouldnt be running it at 240 or 500watts or power unlocked..

In what cases is a 125w 10900k not equally efficienct to a 125w 5800x? Can you show me some numbers?

 

[ThrashZone]

Discussion

Hi,
Not really there are some people that just like to argue on lame points.
Professional forum instigators it's best to pass by them and not take the bait.

Forgot they also make snotty remarks to.

 

[Taraquin]

My statement is that if you care about efficiency you wouldnt be running it at 240 or 500watts or power unlocked..

In what cases is a 125w 10900k not equally efficienct to a 125w 5800x? Can you show me some numbers?

Gaming. Check the link I posted earlier. In gaming 10900K running stock uses 31W more while performing worse...

In cinebench 5800X is much more efficient at 125W since it already beat the 10900K stock while using less stock.

 

[JustBenching]

Gaming. Check the link I posted earlier. In gaming 10900K running stock uses 31W more while performing worse...

Thats system power consumption. No point in looking at system power, especially in gaming, cause GPU plays the biggest role

In cinebench 5800X is much more efficient at 125W since it already beat the 10900K stock while using less stock.

And its not in plenty of other applications as shown from gnexus review (vray , chromium compile, blender). Actually, correct me if im wrong, in every test gnexus ran besides adobe, both cpus consumed the same and performed the same.

 

[ThrashZone]

Hi,
The argument comparing a 8 core against a 10 core is silly and should of been the first clue to pass.

 

[JustBenching]

Hi,
The argument comparing a 8 core against a 10 core is silly and should of been the first clue to pass.

Well then you cant compare anything really. There is no other 10 core cpu, the 5950x has 33% more threads than the 12900k so we shouldn't compare them etc. And why stop at the number of cores? Let's not compare two cpus unless they are on the same node or running the same frequency either.