Intel Core i7-10700

[W1zzard]

In our Intel Core i7-10700 review, we're taking a look at one of Intel's most affordable 8-core/16-thread processors. Its low TDP of 65 W makes it power-efficient, but also limits performance. We unlocked that limit and gained up to 30% real-life performance without ever risking an unstable system.

Show full review

 

[yeeeeman]

Interesting product. The 10700f should be even better value.
So judging by the numbers, the 10700 at stock consumes a bit less power than 3700X while being within 3% of 3700X performance??? Just wow.
A 14nm process, 6 years old now gives you the same performance and efficiency as a state of the art 7nm product?
Not saying this CPU is the winner in this comparison, but technically it is remarkable how Intel can use such an old part to tie in performance and power consumption with the 3700x.

 

[Cheeseball]

This one does seem to be a nice alternative for gaming, especially if you're not looking to overclock. The only problem is the memory limitation of the budget and mid-range motherboards as DDR4-2666 DDR4-2933 will hamper its overall performance. Most users won't be buying [expensive] Z-class motherboards just to put a non-K class CPU in it. Intel should've spec'd the B and H boards up to DDR4-3200 max.

 

[ppn]

It needs at least DDR4-4266. I mean 10600K scales linearly almost with memory this having 33% more cores need to be fed properly. It really is 165 watts TDp at 4.6GHz nThreads AVX. Underclocking this CPU by 25% saves 100 watts power. Actually H410 supports 2933. 10700F unlocked to 200 watts beats 10600K for the price.

 

[Cheeseball]

Actually H410 supports 2933.

Corrected my reply above.

 

[TheinsanegamerN]

Under the OCing section, it says that raising the power limits dramatically affects multicore performance. Now, am I blind, or are there no graphs showing this improvement? I'd be curious to see the clocks it sustains at higher PL1 TDP settings.

Also, I faintly remember back in the ivy bridge/haswell days that you could raise the boost bin on non K CPUs, so a 4.0 GHz boost hypothetical 4590 non-k for single core could be raised to 4.4, and the multi core would go from 3.6 to 4.0, ece. Or maybe it was only 200 mhz. Is this possible on the 10700? It's clock rates are so high already that if this boost tuning worked then you could hypothetically raise the 10700 to run at the same all core boost as a 10700k, and could do the same with a 10900, negating the purpose of the higher priced K series much how the 3800x is kinda pointless when the 3700x gets within 50 MHz boost.

If this worked, the $50 difference between a 10900-10900k/10700-10700k could make up the difference for the more expensive Z490 motherboard for memory overclocking.

 

[W1zzard]

Under the OCing section, it says that raising the power limits dramatically affects multicore performance. Now, am I blind, or are there no graphs showing this improvement? I'd be curious to see the clocks it sustains at higher PL1 TDP settings.

"Max Turbo"

 

[QUANTUMPHYSICS]

I won't be upgrading till we get DDR5 RAM. The 10 generation CPU - even the i3, are damn powerful so I'm very excited for the Core i7.

 

[HugsNotDrugs]

With the power tuning limit adjusted the 10700 seems like a great CPU, until you look at the price.

I see the 10700 being popular with OEM systems, not retail. If a consumer wants better value by foregoing a non-K CPU they will probably end up saving even more money by going with AMD.

 

[Space Lynx]

@W1zzard

So am I correct in assuming, that like in previous Z390 boards when you could "enable boost" it would set all your cores to max out of box, in this case it would be 4.8ghz all 8 cores no downclocking. Does that still exist for Z490 non-oc'able chips? I am a bit confused because you mentioned Turbo Boost several times, but is there a static turbo boost option where it will stay 4.8 at all times? I know MSI and a couple others had this feature for Z390 non-k chips last gen. Wondering if same holds true here.

 

[Rowsol]

It's amazing how cool it runs when at stock.

 

[Joss]

With the power tuning limit adjusted the 10700 seems like a great CPU, until you look at the price.

Yeap, I just looked, you're right! It's $65 more than an R7 3700X.
In fact, the better buy is the 2700X for less than $200.
All this talk about the last drop of performance is nauseating already; how many of us game competitively at (very) high refresh rate? Sometimes I have the feeling the tech press writes exclusively for half demented teenagers.

 

[Space Lynx]

Yeap, I just looked, you're right! It's $65 more than an R7 3700X.
In fact, the better buy is the 2700X for less than $200.
All this talk about the last drop of performance is nauseating already; how many of us game competitively at (very) high refresh rate? Sometimes I have the feeling the tech press writes exclusively for half demented teenagers.

I only play single player games, but I find them more fun at higher refresh. Diablo 3 at 60 fps versus 165 fps for example, was a much more fun experience, everything just looks so smooth.

 

[Caring1]

All this talk about the last drop of performance is nauseating already; how many of us game competitively at (very) high refresh rate? Sometimes I have the feeling the tech press writes exclusively for half demented teenagers.

They're catering to the average consumer with their bigger, faster and, must have the latest tech mentality.

 

[watzupken]

Yeap, I just looked, you're right! It's $65 more than an R7 3700X.
In fact, the better buy is the 2700X for less than $200.
All this talk about the last drop of performance is nauseating already; how many of us game competitively at (very) high refresh rate? Sometimes I have the feeling the tech press writes exclusively for half demented teenagers.

You still need to factor in the additional cost of a good after market cooler as well to allow this sort of boost speed. The Ryzen 7 3700X comes with a beefy Wraith Max cooler, while the Intel chip comes with a crappy thin cooler which is not fit for purpose when it comes to achieving the advertised boost speed.

 

[Space Lynx]

You still need to factor in the additional cost of a good after market cooler as well to allow this sort of boost speed. The Ryzen 7 3700X comes with a beefy Wraith Max cooler, while the Intel chip comes with a crappy thin cooler which is not fit for purpose when it comes to achieving the advertised boost speed.

This is a good point, I have seen the 3700x cooler in person, and it is indeed quite impressive for basically being free.

 

[ppn]

AT this point intel needs to bundle this with 224 watt cooler to be able to match 3700x in Cinebench20, and return the surplus difference in the power bill accumulated for 7 years 7/7 operation.
I cant even justify measuring 250+ watts at the wall.for just the CPU load alone.
For game it can work perfectly fine with the box, since out of the box it runs at the same 4.2GHz frequency and power as 3700x, at lower voltage .
but is probably noisy even for 65 watts, always better to replace it, I can reuse my deepcool single heatpipe cooler that can cool 100 watts to 66C in under 1000 rpm silent mode.

 

[Space Lynx]

I just hope the rumors about ryzen 4000 chips being on 5nm+ and releasing this Fall/winter are true... AND that i can score a msi x570 tomahawk assuming they ever come in stock

 

[barra]

You still need to factor in the additional cost of a good after market cooler as well to allow this sort of boost speed. The Ryzen 7 3700X comes with a beefy Wraith Max cooler, while the Intel chip comes with a crappy thin cooler which is not fit for purpose when it comes to achieving the advertised boost speed.

the only thing the wraith prism is good for is selling to unsuspecting 3600 owners

 

[Space Lynx]

the only thing the wraith prism is good for is selling to unsuspecting 3600 owners

wraith prism is great for 65 watt 3700x. there is no reason to OC on amd chips at all, so the wraith prism makes a lot of sense for a lot of people. especially if your willing to set an aggressive PWM curve for it. this is a redeeming factor about AMD, they make gaming so much more accessible / cost effective than Intel/Nvidia, and this is true even back in the AGP days. it is one reason I am not giving up on them yet. if it was not for AMD and ATI I would never have been able to afford to play games like Stronghold in 2001, or Age of Empires 2, or even WoW in 2003-2004 beta. They gave me some amazing memories. I already decided I am going to give AMD one last chance. big navi and ryzen 4800x. POSSIBLY i will settle for the soon to be released 3800 XT cpu... we will see.

 

[barra]

wraith prism is great for 65 watt 3700x. there is no reason to OC on amd chips at all, so the wraith prism makes a lot of sense for a lot of people. especially if your willing to set an aggressive PWM curve for it. this is a redeeming factor about AMD, they make gaming so much more accessible / cost effective than Intel/Nvidia, and this is true even back in the AGP days. it is one reason I am not giving up on them yet. if it was not for AMD and ATI I would never have been able to afford to play games like Stronghold in 2001, or Age of Empires 2, or even WoW in 2003-2004 beta. They gave me some amazing memories. I already decided I am going to give AMD one last chance. big navi and ryzen 4800x. POSSIBLY i will settle for the soon to be released 3800 XT cpu... we will see.

I found the prism to be pretty average, sure it kept temps mostly below 80 degrees on my 3700x but it was far from quiet doing so. Best thing it did was cover half the cost for a better air cooler. You wont be disappointed giving AMD another go, though do expect some software inconsistencies on both CPU chipset and GPU software packages

 

[R0H1T]

Interesting product. The 10700f should be even better value.
So judging by the numbers, the 10700 at stock consumes a bit less power than 3700X while being within 3% of 3700X performance??? Just wow.
A 14nm process, 6 years old now gives you the same performance and efficiency as a state of the art 7nm product?
Not saying this CPU is the winner in this comparison, but technically it is remarkable how Intel can use such an old part to tie in performance and power consumption with the 3700x.

The 3700x can also be tuned to perform a lot better, especially with manually tuned RAM oh btw did you forget the RAM support & of course the locked multiplier for the 10700 chip? Not to mention 14nm++ isn't 6 years old, neither is skylake ~ that's patently false

Your claim is roughly pants on fire on the fact check meter

 

[ppn]

At $25 more there's 10700KF. 4,6GHz is the sweet spot already where the voltage starts to ramp up especially when all the low voltage dies went for the making of 10900K. 10600 and 10700 are the garbage chips but good enough, 4,8GHz at 1,3V is just sick. DDR is the big dissatisfaction here, DDR3 topped at 2133 and DDR4 is not reaching 4266 at reasonable price and voltages yet when what they are doing is just doubling the rate for almost free as the starting point and DDR5 is almost here killing all prospects for any socket available today. If i can run the CPU with DDR and NB all at sync 4,6GHz it would be perfect. That means waiting for DDR5 and 5nm+ from both companies.

 

[yeeeeman]

The 3700x can also be tuned to perform a lot better, especially with manually tuned RAM oh btw did you forget the RAM support & of course the locked multiplier for the 10700 chip? Not to mention 14nm++ isn't 6 years old, neither is skylake ~ that's patently false

Your claim is roughly pants on fire on the fact check meter

For sure, any unlocked CPU can be tweaked to do better. But simply comparing these two products from a technical standpoint, I think my analysis stands true.
I specifically said the 3700x is the clear buy nevertheless, but you seem to have a bit of fanboyism for AMD in you, hence you don't like anyone saying anything good about intel products. But that is ok...
As for the 3700x, lets see how the 3700xt will fair, I am quite curious how the balance of perf/consumption will be kept/disturbed.
14nm is 5-6 years old. Sure, they tweaked the knobs here and there to make it run faster, but nevertheless it uses the same transistors, same machines, same tech, same library so it is the same process. And I am reiterating. For such an old piece of tech, I think it is impressive that it can keep up with the latest and greatest from amd/tsmc. This shows that AMD has a lot more work to do on the power saving schemes and that 7nm power advatange is the main reason why zen 2 parts are so good. This also shows that technically the skylake core is more efficient when it comes to work done per unit of power used vs zen 2 core. But the 7nm usage turns the tables in favor of amd. Only just.

 

[R0H1T]

Yet you don't seem to have any bias, towards Intel, is that what you're claiming? I highlighted the specific points of your post ~ which you should know are incorrect but whatever floats your boat

And before you go on to nitpick on this reply, why don't you post evidence to backup your initial claim that everything's (nearly) the same as it was 6 years back? For instance there's no SKL in 2014, so no backtracking room there.