Question Ryzen 9 7950X3D versus Intel i9-14900KF ?

[WutisDat]

Hello guys, Im having a little bit of a dillema here. Im not a specialist in this field so I thought I might get some of your insight on this and I plan to use the build mostly for occasional gaming here and there.

I would like to know which one of these CPUs are better, since both cost about the same in my area.
1) Which one consumes more energy per year (whether its in "IDLE" or "GAMING" mode)
2) Is Noctua NH-D14 as cooler, gonna handle its cpu heat?
3) Which one is more future proof?
4) Which one is better and would you pick it?

Thanks a lot for your opinion

 

[Darkbreeze]

You can decide that for yourself, but honestly, whichever configuration you can get for less after factoring in the desired motherboards and memory is the one that probably makes the most sense because their performance is very similar and their power usage per year is nearly identical. What might likely make the clearest decision is the cost of a suitable, compatible board and memory, depending on what you want to go with and what's available in your region. Keep in mind both of these are substantially powerful, high power draw CPUs so you really want a board with a very good VRM configuration. You do not want to skimp on the motherboard or the power supply.

https://www.cpubenchmark.net/compare/5684vs5234/Intel-i9-14900KF-vs-AMD-Ryzen-9-7950X3D


Also, the NH-D14 is very old and outdated. I would not recommend it these days unless you know for SURE that the memory you plan to use is going to be short enough to fit under it because the D14 is not offset to the rear nor is it elevated like most heatsink models these days. The board might also play a role as not all boards have the DIMM slots sitting in the exact same position in relation to the CPU socket. Some are slightly closer or further than others. Some might also be slightly further or closer to the top of the board. If I knew what country you were in it would be easier to try and recommend a cooler that would actually be available to you. If you already have the D14, then it's probably at least worth trying to see if fitment is an issue because surely it's performance is good enough.

Neither of them is more or less future proof because there is no such thing. Nothing is proof against the future. The second you buy ANY hardware component, it IMMEDIATELY begins to become outdated and since these are both compatible with the latest chipsets for their respective platforms and are both DDR5 compatible, they are both of a relatively equal age.

Personally, if it was primarily for gaming, I'd probably go with the Intel platform as the 3D vcache doesn't seem to offer as much benefit for gaming as it did on previous generations. And that seems to be the consensus on most reviews I've looked at like this short comparison.

https://www.techradar.com/computing/cpu/14900k-vs-7950x3d

 

[CelicaGT]

If you are only gaming, neither. 7800X3D or an i7 14700K. Spend the money saved on more GPU, faster storage, or a better PSU.

 

[Darkbreeze]

If you are only gaming, neither. 7800X3D or an i7 14700K. Spend the money saved on more GPU, faster storage, or a better PSU.

I would certainly not disagree with this either. Except in very specific circumstances, the hype for top tier flagship models is usually not nearly as accurate for the average person as it is for somebody running them under lab conditions like reviewers do usually. I think either of those CPUs you named is plenty enough for any gaming configuration these days. Considering most of it is written to be usable at least on much lower end hardware, at some point you get diminishing returns when go beyond those models. Pretty much been that way for both camps for many, MANY generations. Best thing out there doesn't always mean you get anything extra out of what you're doing for it having that distinction.

I'd say that with the money saved, increasing the quality of the PSU you go with and at least getting a slightly better MODEL of whatever tier of graphics card you decide to go with, is a much wiser path.

 

[WutisDat]

If you are only gaming, neither. 7800X3D or an i7 14700K. Spend the money saved on more GPU, faster storage, or a better PSU.

Thanks for the suggestion, Im actually reconsidering the decision based on your reply. Its also 200e cheaper too but... I have also found this and Im little confused there are 2 sites comparing the suggested CPU and I got this:
https://www.cpubenchmark.net/compare/5299vs5723/AMD-Ryzen-7-7800X3D-vs-Intel-i7-14700KF
(Over here it shows that intel gets twice as good score) why is that?

https://cpu.userbenchmark.com/Compa...0KF-vs-AMD-Ryzen-7-7800X3D/m2212020vsm2081998
(Over here it's close score)

Thanks for the insight.

 

[WutisDat]

You can decide that for yourself, but honestly, whichever configuration you can get for less after factoring in the desired motherboards and memory is the one that probably makes the most sense because their performance is very similar and their power usage per year is nearly identical. What might likely make the clearest decision is the cost of a suitable, compatible board and memory, depending on what you want to go with and what's available in your region. Keep in mind both of these are substantially powerful, high power draw CPUs so you really want a board with a very good VRM configuration. You do not want to skimp on the motherboard or the power supply.

https://www.cpubenchmark.net/compare/5684vs5234/Intel-i9-14900KF-vs-AMD-Ryzen-9-7950X3D


Also, the NH-D14 is very old and outdated. I would not recommend it these days unless you know for SURE that the memory you plan to use is going to be short enough to fit under it because the D14 is not offset to the rear nor is it elevated like most heatsink models these days. The board might also play a role as not all boards have the DIMM slots sitting in the exact same position in relation to the CPU socket. Some are slightly closer or further than others. Some might also be slightly further or closer to the top of the board. If I knew what country you were in it would be easier to try and recommend a cooler that would actually be available to you. If you already have the D14, then it's probably at least worth trying to see if fitment is an issue because surely it's performance is good enough.

Neither of them is more or less future proof because there is no such thing. Nothing is proof against the future. The second you buy ANY hardware component, it IMMEDIATELY begins to become outdated and since these are both compatible with the latest chipsets for their respective platforms and are both DDR5 compatible, they are both of a relatively equal age.

Personally, if it was primarily for gaming, I'd probably go with the Intel platform as the 3D vcache doesn't seem to offer as much benefit for gaming as it did on previous generations. And that seems to be the consensus on most reviews I've looked at like this short comparison.

https://www.techradar.com/computing/cpu/14900k-vs-7950x3d

Hey thanks for help!

As for the "future proof" I kinda thought about something that can run games at good settings like 2Kres/100fps for next 5years and slowly getting behind of another few years, untill I decide to make an upgrade, preferably at that time just upgrading the GPU and NOT the whole system along with the CPU, that kinda thing Im looking for I guess.

Yeah the NH-D14 is outdated kind of as I have used that one in my old setup, I was considering to use same case(Cooler Master HAF X) I have now, along with the cooler. (Noctua NH-D14) and perhaps even PSU (Corsair Enthusiast Series TX850) all of these are kind of old, but still works so I thought I might use these till their end. Do you think they're still strong enough to support new system?

Or what RAMs, MOBO would you recommend if I actually use this stuff perhaps if I even try to go instead for the CPU Ryzen 7 7800X3D as the person below suggested?

Thanks again!

 

[CelicaGT]

Thanks for the suggestion, Im actually reconsidering the decision based on your reply. Its also 200e cheaper too but... I have also found this and Im little confused there are 2 sites comparing the suggested CPU and I got this:
https://www.cpubenchmark.net/compare/5299vs5723/AMD-Ryzen-7-7800X3D-vs-Intel-i7-14700KF
(Over here it shows that intel gets twice as good score) why is that?

https://cpu.userbenchmark.com/Compa...0KF-vs-AMD-Ryzen-7-7800X3D/m2212020vsm2081998
(Over here it's close score)

Thanks for the insight.

That's likely an all core benchmark, not something to weigh too heavily when deciding on a CPU for gaming. Synthetic benchmarks like that are typically not great for comparing real use case scenarios. For that, search out comparisons in the very games you'll play. Those can be found here at Tom's, or other places. YouTube can be a good source for this, however it's hard to find a channel that does good, empirical testing. I often recommend Gamer's Nexus and will do so again. Steve can be a little dry, and sometimes over the top with snark but their testing is often peerless. Most of their charts they feature will list results for every mainstream processor from the last 2 or 3 generations across many games from several genres. Another decent channel is Hardware Unboxed. I don't watch them as often as GN but they do have quality content, and good testing methodology.

 

[Darkbreeze]

Yeah, synthetics are generally not realistic, however, they CAN be REPRESENTATIVE when compared to other CPU models SAME synthetics, but just not necessarily for a specific use case. Some processes and applications simply not only do better on a CPU from one camp than they do on another, even weirder, sometimes they might even do better on a CPU from the SAME camp, that's OLDER. It's just weird how some of the results on almost everybody's benchmarks plays out differently. But, in this case, I don't really see any games out there, now or in the near future, that are going to see performance significantly improved over the two CPUs he mentioned. Certainly not enough to justify the cost. And in reality, in MANY cases, the extra heat if you don't have a custom loop, for these top shelf CPUs, is actually WORSE than the performance of those one step down with pretty standard yet good cooling. So there's that extra cost too if you really want to go that way and take a look at full investment.

 

[TheHerald]

Thanks for the suggestion, Im actually reconsidering the decision based on your reply. Its also 200e cheaper too but... I have also found this and Im little confused there are 2 sites comparing the suggested CPU and I got this:
https://www.cpubenchmark.net/compare/5299vs5723/AMD-Ryzen-7-7800X3D-vs-Intel-i7-14700KF
(Over here it shows that intel gets twice as good score) why is that?

https://cpu.userbenchmark.com/Compa...0KF-vs-AMD-Ryzen-7-7800X3D/m2212020vsm2081998
(Over here it's close score)

Thanks for the insight.

The 14700k is indeed twice as fast than the 7800x 3d in multithreaded applications. If you are just gaming it won't matter that much to you but it definitely makes for a snappier system. You can have some heavy workloads running on the background while you are playing your games with no hiccups on the 14700k

 

[Darkbreeze]

The 14700k is indeed twice as fast than the 7800x 3d in multithreaded applications. If you are just gaming it won't matter that much to you but it definitely makes for a snappier system. You can have some heavy workloads running on the background while you are playing your games with no hiccups on the 14700k

Show me.

 

[TheHerald]

Show me.

Show you what?

 

[35below0]

A benchmark that shows 14700K is twice as fast in MT applications.

 

[Darkbreeze]

A benchmark that shows 14700K is twice as fast in MT applications.

Exactly right. Because I'm calling BS on that.

 

[TheHerald]

A benchmark that shows 14700K is twice as fast in MT applications.

Exactly right. Because I'm calling BS on that.

What? You are disputing the fact that it's twice as fast in MT applications? Let's start with corona

https://benchmark.chaos.com/corona/v10/scores?index=1&order=desc&by=rays&my-scores-only=false

Just search for the CPUs in question

 

[35below0]

AMD Ryzen 7 7800X3D 8-Core Processor [32Gb RAM] - 6 555 120rays/s
12th Gen Intel(R) Core(TM) i7-12700K - 7 318 182rays/s
Intel(R) Core(TM) i7-14700K [64Gb RAM] - 11 111 635rays/s


Also:
AMD Ryzen 9 7950X 16-Core Processor - 15 370 436rays/s
Intel(R) Core(TM) i9-14900K - 14 916 246rays/s

These are top non-Threadripper/Xeon CPUs. The top i9 and Ryzen 9

 

[TheHerald]

AMD Ryzen 7 7800X3D 8-Core Processor - 6 555 120rays/s
12th Gen Intel(R) Core(TM) i7-12700K - 7 318 182rays/s

Also:
AMD Ryzen 9 7950X 16-Core Processor - 15 370 436rays/s
Intel(R) Core(TM) i9-14900K - 14 916 246rays/s

These are top non-Threadripper/Xeon CPUs. The top i9 and Ryzen 9

What? The 14700k (I ignored the top score) scores over 11m. The 7800x 3d is barely above 6-6.5.

What does the 12700k have to do with this?

 

[35below0]

What? The 14700k (I ignored the top score) scores over 11m. The 7800x 3d is barely above 6-6.5.

What does the 12700k have to do with this?

sorry, i got confused and picked the wrong CPU by mistake.
My bad!

 

[TheHerald]

sorry, i got confused and picked the wrong CPU by mistake.
My bad!

No problem. The fastest 14700k scores 12.7m.

Still your comparison to the 12700k shows the problem, the x3d loses to a 2 gen old i7.

 

[geofelt]

Passmark shows the
R7-7800X3D to have 24 threads and a MT rating of 34392. The single thread rating is 3758
https://www.cpubenchmark.net/cpu.php?cpu=AMD+Ryzen+7+7800X3D&id=5299

I7-14700K with 28 threads and a MT rating of 53657.
Not exactly 2x, but darn good.
St is 4479:
https://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-14700K&id=5719

 

[Darkbreeze]

Not exactly 2x, but darn good.

Not even close to twice as fast. Twice as fast would be 68784, not 53657. If you're going to say a thing, it needs to BE what you say it is. Otherwise, say it is 56% faster, which is what it is, not 100% faster, which it is not. It is literally almost exactly HALF as much faster as was said by TheHerald. Details matter.

 

[35below0]

Technicaly, The Herald was correct. And provided the benchmark.

It's fair to say 53657 is almost twice the score of 34392. It's not 68784, and no one said it was. It's certainly close enough.

 

[Phaaze88]

It's fair to say 53657 is almost twice the score of 34392. It's not 68784, and no one said it was. It's certainly close enough.

Are you and geofelt really calling 53657 'close enough' to twice that of 34392, when there's still 44% left? That's huge for these synthetic benchmarks.

 

[John Chesterfield]

It's certainly close enough.

It really isn't.

 

[35below0]

Are you and geofelt really calling 53657 'close enough' to twice that of 34392, when there's still 44% left? That's huge for these synthetic benchmarks.

I've nothing against the CPU. Objectively, as the benchmark result shows, the 14700K blows away the 7950x3d in that Corona test. But...

the claim was TWICE as fast which, equally objectively it isn't. I thought the claim was that the 14700K was almost twice as fast. That's why i said it's close enough.
I could have sworn i read "almost". this is not my day

It's certainly faster in MT, even considerably faster but NOT twice as fast.

 

[Phaaze88]

I've nothing against the CPU. Objectively, as the benchmark result shows, the 14700K blows away the 7950x3d in that Corona test. But...

the claim was TWICE as fast which, equally objectively it isn't. I thought the claim was that the 14700K was almost twice as fast. That's why i said it's close enough.
I could have sworn i read "almost". this is not my day

It's certainly faster in MT, even considerably faster but NOT twice as fast.

I saw/read those and also agree with Herald. It's when geofelt came in with the Passmark comparisons, which your post(#21) is agreeing with geofelt's claim of close enough. That's what I was concerned about.