Intel Core i9-12900KS just 9% in faster Cinebench R23 Multi-Core compared to 5950X

Published by

Click here to post a comment for Intel Core i9-12900KS just 9% in faster Cinebench R23 Multi-Core compared to 5950X on our message forum
https://forums.guru3d.com/data/avatars/m/248/248994.jpg
It beats the Ryzen 5950X by 9.2% but consumes something like 30-40% more electricity to do it, right? I don't believe that's a sustainable way to make progress in 2020's anymore. I'd guess it was also one of the reasons why Apple wanted its own silicon, instead of sticking to Intel CPUs.
data/avatar/default/avatar19.webp
Kaarme:

It beats the Ryzen 5950X by 9.2% but consumes something like 30-40% more electricity to do it, right? I don't believe that's a sustainable way to make progress in 2020's anymore. I'd guess it was also one of the reasons why Apple wanted its own silicon, instead of sticking to Intel CPUs.
The "funny" thing is that 12900k (with fast memory) beats 5950x in fps per watt with up to 40-50%, but in cinebench it's the other way around. So what do you care about most? Most people here seams to love Cinebench score more than games 🙄
https://forums.guru3d.com/data/avatars/m/248/248994.jpg
nizzen:

The "funny" thing is that 12900k (with fast memory) beats 5950x in fps per watt with up to 40-50%, but in cinebench it's the other way around. So what do you care about most? Most people here seams to love Cinebench score more than games 🙄
Do gamers even need such a CPU? Cinebench would reflect better the kind heavy compute work one would put their 16-core CPU through.
data/avatar/default/avatar11.webp
Kaarme:

Do gamers even need such a CPU? Cinebench would reflect better the kind heavy compute work one would put their 16-core CPU through.
Do Cinebench players need a 16 core cpu? Wait is fun, like "low" fps in games 😉
https://forums.guru3d.com/data/avatars/m/227/227994.jpg
Kaarme:

Do gamers even need such a CPU? Cinebench would reflect better the kind heavy compute work one would put their 16-core CPU through.
You don't even need a K model for gaming, and it's a shame these barely get any review coverage. A 12600 or 12700 is all you need if you plan to keep a CPU for 6-8 years. It's why i went with a 11700, because i don't plan to upgrade until 8 years from now.
https://forums.guru3d.com/data/avatars/m/248/248627.jpg
Ppl need to realise that scores are not at all comparable anymore all these CPU's from Intel with ecores will bench completely different in workstation benches vs game benches, ecores are almost useless in gaming but help tremendously in workstation applications. Amd in a sense is just brute forcing everything on performance cores till they come up with something of their own. As someone else has mentioned here I'd also like to see more benches of non K/X CPU's as I feel like there's alot more demand from those when they're available.
https://forums.guru3d.com/data/avatars/m/246/246171.jpg
nizzen:

The "funny" thing is that 12900k (with fast memory) beats 5950x in fps per watt with up to 40-50%, but in cinebench it's the other way around. So what do you care about most? Most people here seams to love Cinebench score more than games 🙄
What you describe is a matter of Intel pushing their chips too much. CPUs are practically never the bottleneck in modern games, so, when it doesn't have to boost all cores (or really, any core), it's going to run at a much more efficient frequency. If Intel's E-cores are what's primarily playing the game, that would make an even bigger difference. I'm sure if Intel lowered their boost clocks by 8%, they would see a huge efficiency increase. Whether it would be better than a 5950X, I'm not sure, but they would still be faster in Cinebench, probably no different in games, and the efficiency (and in turn, heat) probably wouldn't be noteworthy. Bear in mind too that lower max boost clocks would mean more sustainable boost clocks. As usual, Intel cares more about peak performance above all else. That's why you get people like @TheDeeGee who realize that a 12700 is a much more sensible choice, because it's so much cheaper and so much more efficient with minimal losses (at least compared to the 12700K).
https://forums.guru3d.com/data/avatars/m/165/165018.jpg
TheDeeGee:

You don't even need a K model for gaming, and it's a shame these barely get any review coverage. A 12600 or 12700 is all you need if you plan to keep a CPU for 6-8 years. It's why i went with a 11700, because i don't plan to upgrade until 8 years from now.
on the other side of that coin I went originally with a 1600X (knowing I was going to upgrade) then a 3700X also knowing I was going to upgrade and here we are nearing the end of the line where I have to decide if I want to go with a 5800X(recently got a 3090 so I want resizable bar) or the 5800X3d and call it a day.
https://forums.guru3d.com/data/avatars/m/263/263205.jpg
Cinebench and Blender are my favorite games. Only 9% uplift means no buy from me.
data/avatar/default/avatar34.webp
NCC1701D:

Cinebench and Blender are my favorite games. Only 9% uplift means no buy from me.
You hadn't bought it anyway, even if it was 100% uplift 😉 You love 3 series Ryzen too much 😀
https://forums.guru3d.com/data/avatars/m/273/273323.jpg
Kaarme:

It beats the Ryzen 5950X by 9.2% but consumes something like 30-40% more electricity to do it, right? I don't believe that's a sustainable way to make progress in 2020's anymore. I'd guess it was also one of the reasons why Apple wanted its own silicon, instead of sticking to Intel CPUs.
It's honestly incredible what Apple has achieved with the M1 chips. The performance considering the amount of heat generated and power pulled is astonishing.
https://forums.guru3d.com/data/avatars/m/197/197287.jpg
nizzen:

The "funny" thing is that 12900k (with fast memory) beats 5950x in fps per watt with up to 40-50%, but in cinebench it's the other way around. So what do you care about most? Most people here seams to love Cinebench score more than games 🙄
While i get what you're saying, you're crippling a cpu to claim efficiency. It'd be like claiming a semi truck is amazing at gas mileage, but only when it has no trailer, rather then seeing how a semi truck performs with a maximum loaded trailer, its intended use.
https://forums.guru3d.com/data/avatars/m/246/246171.jpg
BlindBison:

It's honestly incredible what Apple has achieved with the M1 chips. The performance considering the amount of heat generated and power pulled is astonishing.
Much of what they achieved is in software. Apple has spent a lot of time optimizing their platform. We often complain about how locked down their platform is, but this is why for so many years, iPhones could directly compete (or sometimes outperform) flagship Androids despite having significantly inferior specs. x86 just isn't that efficient of an architecture, but it stuck around for so long because of compatibility.
https://forums.guru3d.com/data/avatars/m/291/291873.jpg
Pitty. Intel could do better in the ARM space. They definitely could do better in x86 space. 9% is unfortunate because of the required amperage.
https://forums.guru3d.com/data/avatars/m/222/222700.jpg
It is pretty amazing what Intel has done with these new chips. Granted the power consumption is out of control and I believe it deserves to be criticized more by the press, but that Thread Detector that they built and the architecture is really cool! Hopefully they can get the power under control!
data/avatar/default/avatar02.webp
Coupe:

It is pretty amazing what Intel has done with these new chips. Granted the power consumption is out of control and I believe it deserves to be criticized more by the press, but that Thread Detector that they built and the architecture is really cool! Hopefully they can get the power under control!
Fps per watt is VERY good too with fast memory in games. Noone care about Cinebench powerdraw any more. We have Threadripper for that game 😛
data/avatar/default/avatar02.webp
schmidtbag:

What you describe is a matter of Intel pushing their chips too much. CPUs are practically never the bottleneck in modern games, so, when it doesn't have to boost all cores (or really, any core), it's going to run at a much more efficient frequency. If Intel's E-cores are what's primarily playing the game, that would make an even bigger difference. I'm sure if Intel lowered their boost clocks by 8%, they would see a huge efficiency increase. Whether it would be better than a 5950X, I'm not sure, but they would still be faster in Cinebench, probably no different in games, and the efficiency (and in turn, heat) probably wouldn't be noteworthy. Bear in mind too that lower max boost clocks would mean more sustainable boost clocks. As usual, Intel cares more about peak performance above all else. That's why you get people like @TheDeeGee who realize that a 12700 is a much more sensible choice, because it's so much cheaper and so much more efficient with minimal losses (at least compared to the 12700K).
Almost every new game is cpu bound, if you have good enough gpu. You are talking about slow gpu that are gpubound in every game. That's the difference. Not everyone playing games in 4k 60hz with v-sync on 😉 Maybe everyone with Ryzen 1, 2 and 3 series 😛
data/avatar/default/avatar06.webp
Aura89:

While i get what you're saying, you're crippling a cpu to claim efficiency. It'd be like claiming a semi truck is amazing at gas mileage, but only when it has no trailer, rather then seeing how a semi truck performs with a maximum loaded trailer, its intended use.
What you are saying is everyone MUST play cinebench because of many cores? Like Ryzen 1700x and 1800x was very good for rendering per core and very efficent, but it was VERY slow in games. So everything depends on what you are using the cpu for. Is it efficent in gaming or rendering. Ryzen cpu's is very good, there is no doubt. Too bad very few ACTUALLY do rendering with the cpu's.... They just playing cinebench....
https://forums.guru3d.com/data/avatars/m/266/266726.jpg
nizzen:

The "funny" thing is that 12900k (with fast memory) beats 5950x in fps per watt with up to 40-50%, but in cinebench it's the other way around. So what do you care about most? Most people here seams to love Cinebench score more than games 🙄
nizzen:

Fps per watt is VERY good too with fast memory in games. Noone care about Cinebench powerdraw any more. We have Threadripper for that game 😛
intel's new power gating tech+thread directing tech is definitely impressive(more so for avx-512 ironically), however I feel that Its not quite as impressive when you consider that alderlake launched a year later, with a brand-new socket, with brand-new memory tech and is on a monolithic die, all of which saves them a considerable amount of power from the start. In this case imo its more impressive that amd's product is able to compete as well as it does despite being 'old' normally when you see new product, its "new product is better than old" in a more dramatic fashion. as for playing cinebench, everyone always wants to see what products/things/people do when they are pushed to their limits, rather than when they are lightly loaded, even if its a more relevant workload, a quirk of human psychology I think.