Review Intel Core i9 10850K Review (TechPowerUp)

[Zucker2k]

With a retail price of $450, the Core i9-10850K is $50 cheaper than the Core i9-10900K. That's a 10% reduction in cost for barely a few percent in performance. Definitely consider the i9-10850K if you're in the market for an i9-10900K. What's probably more important than $50 for many people is that availability of the i9-10850K seems to be much better than for the i9-10900K. Especially in the States, as Intel CPU supply is much worse than in Europe, where you can just go out and buy an i9-10900K without having to wait for the stars to align. That's why I'm also not surprised to see terrible pricing in the U.S., no doubt from greedy merchants. Looking at our performance numbers and the pricing, I would definitely prefer the i9-10850K over the Ryzen 9 3900XT, and possibly even the Ryzen 9 3900X, for gaming and general productivity. Professionals working with rendering and simulation apps, or other similarly demanding apps, should definitely consider AMD for their rigs, as the higher thread count can make a difference. We reviewed the Core i9-10900 only recently and liked it very much. However, it's kind of obsoleted by the i9-10850K because of the small price difference. $10 more gets you an unlocked multiplier, much higher base clock, and higher power limit—just the boxed cooler is missing. Definitely worth considering. On the other hand, if the i9-10900 drops to $400, it would compete with the Ryzen 7 3800XT in an interesting reversal of 10-core Intel vs. 8-core AMD at the same price.

Intel Core i9-10850K Review - Just as Good as the i9-10900K

The Intel Core i9-10850K is the company's latest Comet Lake processor. It's just 100 MHz slower than the Core i9-10900K flagship, but much more affordable, and with better availability. In our i9-10850K review, we're taking a close look at both gaming and application performance to determine...

www.techpowerup.com

 

[VirtualLarry]

Interesting, a slightly down-binned i9-10900K, in the form of an i9-10850K.

I was just musing, "wouldn't you like to own an Intel rig, sometime"? Then I was like, "naaah".

 

[Zucker2k]

Interesting, a slightly down-binned i9-10900K, in the form of an i9-10850K.

I was just musing, "wouldn't you like to own an Intel rig, sometime"? Then I was like, "naaah".

I think the 'performance per dollar' chart is an interesting one, though, I always insist performance is not always commensurate to price.

 

[Makaveli]

I find there AES numbers are low on their ryzen builds for this review.



And from my rig with just PBO on and a Ryzen dram fast profile applied at 3200 spd. Has me matching the Core i9-10900 at 16700 MB/s

 

[DrMrLordX]

My 3900X is doing over 15 GB/s in AES. Overclocked to 4.4 GHz, it's doing over 20 GB/s.

 

[UsandThem]

They have the 9900K priced at $530, while it's been widely available for under $400 now ($370 currently at Newegg).

That places the 9900K near the bottom of the price/performance chart, when it is not that far behind any of the 10th gen flagship CPUs.

They even price the 9900KS at $600

 

[Zucker2k]

Tom's review:

Intel Core i9-10850K CPU Review: Cheaper, but Nearly Identical Gaming to 10900K

Comet Lake gets a price drop disguised as a new chip

www.tomshardware.com

 

[DAPUNISHER]

They have the 9900K priced at $530, while it's been widely available for under $400 now ($370 currently at Newegg).

That places the 9900K near the bottom of the price/performance chart, when it is not that far behind any of the 10th gen flagship CPUs.

They even price the 9900KS at $600

Yup, theoretical bang for buck charts are one of the best ways to obfuscate reality, and confuse less savvy buyers. 9900K is probably the best gaming value out there, particularly since you get a AAA game with it. Both articles lead off with the costs less! just as good! derp!, but no one is selling it for that. Hey Intel, you want your DIY buyers back? Keep slashing prices. If I can sell off enough stuff to get the wife to not go all SpaceX on me, I will build a 9900K for my son. He has a 165Hz monitor, so he could leverage it far better than me.

 

[A///]

9900K is still a good buy even if you don't game and want Intel. The 10900K is just too much to handle even with a big rad. I wonder how far they'll push their 8c RKLS parts.

 

[Markfw]

9900K is still a good buy even if you don't game and want Intel. The 10900K is just too much to handle even with a big rad. I wonder how far they'll push their 8c RKLS parts.

I think the 9900k is still to hot, but it is a good gamer and a decent value. The 10900k is DEFINITELY way too hot

 

[TheELF]

The 10900K is just too much to handle even with a big rad.

This has been grossly blown out of proportions because reviewers only measure maximum peak power draw instead of average power draw over the duration of the benchmark.
It's as if some review today would still only show max FPS...would you put any amount of trust into anything they would say after that?

From the same article used in the OP.
The 10850k overclocked to 5.1Ghz is actually using less energy at 5.1 to complete a task than the stock 3900x or xt needs to get to just 4.6 or 4.7 in single threaded, and most probably the AMD parts don't even hit those clocks.

, and although the power draw in multithreaded is higher than ryzen,if you keep the intel parts at stock the energy they need to complete a task is not out of place,it's in the middle field.
And less energy than needed for single thread.

 

[DrMrLordX]

This has been grossly blown out of proportions because reviewers only measure maximum peak power draw instead of average power draw over the duration of the benchmark.

If you don't spec cooling for a 250W+ heat load, the chip will never reach max all-core boost of 4.9 GHz for the "default" tau of 56s. Assuming your motherboard is playing by the rules, which it might not be.

If you don't push your chip that hard on all cores . . . remind me why you didn't just get a 9900k again?

 

[ondma]

If you don't spec cooling for a 250W+ heat load, the chip will never reach max all-core boost of 4.9 GHz for the "default" tau of 56s. Assuming your motherboard is playing by the rules, which it might not be.

If you don't push your chip that hard on all cores . . . remind me why you didn't just get a 9900k again?

Remind me why it matters to you if someone *does* want a 10850k. One is free to buy what they like or feel fits their needs, are they not? Might not make sense to you, but someone else might feel differently.

 

[DrMrLordX]

Remind me why it matters to you if someone *does* want a 10850k.

Oh wow, you went there? After all these years of nearly everyone remarking on why anyone would want to buy such-and-such product, you pull a question like that out of your hat?

Where were you when people were being eviscerated on a daily basis for even thinking about buying Piledriver? Not here, not sticking up for those people with their tiny edge cases and/or questionable purchasing decisions.

 

[DAPUNISHER]

Oh wow, you went there? After all these years of nearly everyone remarking on why anyone would want to buy such-and-such product, you pull a question like that out of your hat?

Where were you when people were being eviscerated on a daily basis for even thinking about buying Piledriver? Not here, not sticking up for those people with their tiny edge cases and/or questionable purchasing decisions.

Both my FX 8 series cost me less than what a i3 was selling for at the time. But reading the forums, I'd committed crimes against the gods.

Intel needs to keep reducing prices, and understand that nothing else they do is going to bring a lot of us back. Point being: I had a FX space heater because the bang for buck was great, and it is the only way I am buying a 9 or 10 series space heater.

 

[TheELF]

If you don't spec cooling for a 250W+ heat load, the chip will never reach max all-core boost of 4.9 GHz for the "default" tau of 56s. Assuming your motherboard is playing by the rules, which it might not be.

That very much depends on the software,if you run something simple like a game you will be able to run at the default 4.9 indefinitely with a simple cooler.
The more complex the software yes, temps will rise or performance will drop or cooling will have to be better.
Gamers nexus with his standard blender test of monkey heads showed that the 10900k only draws 130watt so that is pretty much doable indefinitely with a 125W cooler you will just get slightly higher temps.
Cinebench 20 uses up 200watt so there yes you would need a better cooler or be in danger of getting your temps too high.

Intel Core i9-10900K CPU Review: Gaming, Overclocking, & Benchmarks vs. AMD Ryzen | GamersNexus

stub Intel’s continuing to bring the heat, literally and figuratively, and is now leveling its new 10C/20T part squarely at AMD. We have a separate in-depth review coming up on the 10600K, which has interesting implications for the R5 3600 and the realm of gaming, but for launch, we need to...

www.gamersnexus.net

If you don't push your chip that hard on all cores . . . remind me why you didn't just get a 9900k again?

My point is that power draw isn't outlandish if you use the CPU in a normal fashion.
If you want to push it then yes, power draw is at 300+ and it is also irrelevant.

 

[DrMrLordX]

My point is that power draw isn't outlandish if you use the CPU in a normal fashion.
If you want to push it then yes, power draw is at 300+ and it is also irrelevant.

Intel has cheaper stuff in their stable that'll get you about the same performance. That's why the product is a bit iffy unless you really do just want a cheaper 10900k. Fortunately I think the 10850k will not require quite as much cooling as a 10900k. I hope.

 

[A///]

If you're buying a 10900K or 10850K, you're part of the enthusiast crowd. Your average prosumer isn't placing down $500 for a processor and not try to extract more power at it. If we're going to look at SuperPi data, why are we looking at the, what I assume are 1M trial times? Short boosts favor both Intel and AMD. Set it to 32M and let it rip! We know Intel boosting is leagues better than AMD's algorithm. Let's see how Intel does on a sustained boost in terms of frequency, power draw and heat output (dumping).

We know there's little OC headroom on Ryzen due to the processors being pushed to the max at the factory or the silicons inability to clock to a higher factory frequency.

Intel has cheaper stuff in their stable that'll get you about the same performance. That's why the product is a bit iffy unless you really do just want a cheaper 10900k. Fortunately I think the 10850k will not require quite as much cooling as a 10900k. I hope.

I agree, but if you're not using the processor to its full potential, save your money and get a 10700K or even a 10600K. There's no point in getting a 10850K or 10900K if you don't plan on using 50% of it let alone its full capacity over the course of you owning it. It's wasted potential.

 

[Shmee]

If I were to get an intel chip now, it would likely be a 10850k or a 10700kf. They are great gaming chips with more reasonable price.

 

[coercitiv]

We know Intel boosting is leagues better than AMD's algorithm.

How do we know that?

 

[ehume]

. . . their 8c RKLS parts.

?

 

[A///]

?

Rocketlake S is supposed to top out at 8 cores according to a rumor that surfaced in June. Was rumored to utilize Willow Cove cores backported. Though the part itself may not even come out. It's been pegged at end of year to 2Q21.

 

[TheELF]

If we're going to look at SuperPi data, why are we looking at the, what I assume are 1M trial times? Short boosts favor both Intel and AMD. Set it to 32M and let it rip! We know Intel boosting is leagues better than AMD's algorithm. Let's see how Intel does on a sustained boost in terms of frequency, power draw and heat output (dumping).

You said it yourself, Short boosts favor both Intel and AMD, and that's because for short boosts they can use the highest amount of power to boost as high as possible.
The same goes for the cinebench results by the way, it finishes fast enough to have most if not all of it be run during highest boost.
If you want to show how much energy they use you want to show some reasonably high boost scenarios.

We already know that the 10850k will stick to 125W for anything that takes longer than 56secs and will stay there until the workload is finished,so anything that takes longer than a few minutes will average out at 125w.

 

[A///]

You said it yourself, Short boosts favor both Intel and AMD, and that's because for short boosts they can use the highest amount of power to boost as high as possible.
The same goes for the cinebench results by the way, it finishes fast enough to have most if not all of it be run during highest boost.
If you want to show how much energy they use you want to show some reasonably high boost scenarios.

We already know that the 10850k will stick to 125W for anything that takes longer than 56secs and will stay there until the workload is finished,so anything that takes longer than a few minutes will average out at 125w.

Yes, the P1 state. PL2 is higher at double power for the time scope. In the 3300X thread, someone mentioned they'd bought their wife an XT processor and the boost was longer and harder for the XT compared to reports of the original chip. I'm guessing AMD has improved somewhere. I'd like to see AMD have higher all core boosts in general. I'm not overly concerned about single core boost for single threaded tasks and would rather see significant IPC improvement in single threaded tasks over multiple generations.

 

[TheELF]

I'm not overly concerned about single core boost for single threaded tasks and would rather see significant IPC improvement in single threaded tasks over multiple generations.

IPC can only be measured by running software that uses that IPC, if you hit the limit of what cinebench or whatever other software can use in a single cycle then no matter how much more IPC the companies stick into that core it will not give you any more improvement, and normal software uses much less IPC than what benchmarking software uses.
It's the whole multithreaded argument all over again where everybody thinks that devs are just lazy and that's the only reason that software isn't optimized because they just can't understand that some software just can't use any more cores or IPC or ram bandwidth or what have you.

Clock boost on the other hand can be used by anything as long as some other component like the GPU or disk drive etc isn't bottlenecking you any and all software will benefit from clocks.