Intel Core i5-11600K

[RandallFlagg]

Best thing about Intel CPUs is you dont need a dGPU to make them work nor only limit your selection to APUs from AMD.

If youre building a computer right now and you dont already have a GPU then there is zero value in any AMD CPU currently.

This is actually crucial. Unless something changes (and reportedly it won't this year) then to differentiate the 5600X and 11600K or for that matter vs a 10700K in gaming you'd need to buy a $1500ish GPU making a complete system north of $2000.


11600K is going to have a big advantage on most desktops, at least until GPU prices come down.

 

[EzioAs]

The FarCry 5 data looks weird. It's almost the same results for 1080p/1440p. Does the game have an FPS cap or something?

 

[lightning70]

307W Full Load POWER

 

[Deleted member 206429]

This is actually crucial. Unless something changes (and reportedly it won't this year) then to differentiate the 5600X and 11600K or for that matter vs a 10700K in gaming you'd need to buy a $1500ish GPU making a complete system north of $2000.


11600K is going to have a big advantage on most desktops, at least until GPU prices come down.

Yes its really a dire situation if youre GPU-less.

To take it one step further, it puts people in a position to either buy overpriced GPUs thus continuing inflation, waste their own personal time trying to hunt down a GPU at near/at MSRP, or proliferate more E-waste buying an overpriced basic GPU to bide time until the GPUs desired come down in price.

 

[TheOne]

I figured this would be the case, which is why I jumped the gun and grabbed the 5600X when it became available for MSRP a couple of weeks ago.

 

[RedelZaVedno]

Yes its really a dire situation if youre GPU-less.

To take it one step further, it puts people in a position to either buy overpriced GPUs thus continuing inflation, waste their own personal time trying to hunt down a GPU at near/at MSRP, or proliferate more E-waste buying an overpriced basic GPU to bide time until the GPUs desired come down in price.

It's not so bad if you manage your expectations. You can find 4 gigs RX 580 on 2nd hand markets for around 170 bucks atm. It's good enough for 1080p/60fps gaming and even 1440p is older titles. It's not ideal, but it's the only decent option until mining craze ends.

 

[Deleted member 206429]

It's not so bad if you manage your expectations. You can find 4 gigs RX 580 on 2nd hand markets for around 170 bucks atm. It's good enough for 1080p/60fps gaming and even 1440p is older titles. It's not ideal, but it's the only decent option until mining craze ends.

Yes thats exactly what I said, E-waste. Its not mining either that is driving GPU demand.

 

[napata]

Sorry, but decent in terms of what?

It can't beat the 10600K in gaming.

While consuming 45% (!!!) more power:

This series is complete rubbish.

In every other review the 11600k beats the 10600k by 10%. It's not hard to find out why though: it was tested with gear 2 instead of gear 1 which gimps performance by 10%. It's like putting the IF on Zen3 at 1:2.

I get that it's not realistic to test everything else with DDR4-3600/3733 but I think the 11600k should've been tested with gear 1 DDR4-3600/3733 as it would probably outperform gear 2 DDR4-3800 by a significant amount. I've even seen benchmarks where gear 1 had an 18% performance increase from gear 2. The IMC on RL seems pretty weak.

 

[RandallFlagg]

Interestingly, looking at multiple review sites, 11600K seems to soundly defeat Zen 3 on Turing (2000 series) GPUs.

From PCGamer which is using a 2080 Ti :

 

[W1zzard]

I get that it's not realistic to test everything else with DDR4-3600/3733 but I think the 11600k should've been tested with gear 1 DDR4-3600/3733 as it would probably outperform gear 2 DDR4-3800 by a significant amount.

I tried, it does not Boot at that freq unless gear 2. Gear 1 is only designed for up to 3200, everything else is unguaranteed oc

Cl14 3209 gear 1 vs gear 2 data is in the 11900k review, it's not a huge difference

 

[Cowboystrekk]

"Looking at our power consumption results we have a total energy of 18.9 kiloJoules to complete one Cinebench run for the 11600K, against 9.6 kJ for the Ryzen 5 5600X—almost half! AMD's Ryzen 9 5950X is even only a third of that, with 6.4 kJ."

It seems like the justification for this is the 14 nm lithography of the Intel chips vs. the 7 nm lithography of the AMD chips. I read many times before that supposedly the density of Intel's process is almost the same as TSMC's 7 nm process. Is this true and if so, how does the similar density begin to explain the energy consumption difference between the chips?

Intels 14nm is slightly less dense than a theoretical Samsung 10nm, and Intel 10nm is very close to TSMC 7nm. Compared with GloFo 14nm and Samsung 14nm Intel 14nm is superior.

 

[ODOGG26]

Honestly its really not hard to find a 5600x at msrp these days. While it seemed to have gone up atm, it gets restocked often and goes right back to msrp at places like newegg, amd and other various sites. Plus microcenter and other local stores always have them at msrp. So yea its really not hard to find mrsp 5600x. I think Intel did good on the price but im not really impressed with the power consumption of the card compared to its predecessor.

 

[btk2k2]

$350 https://www.newegg.com/p/pl?d=5600x&N=100007671&isdeptsrh=1
$270 https://www.newegg.com/intel-core-i...=11600k-_-19-118-235-_-Product&quicklink=true
$230 https://www.newegg.com/intel-core-i...=10600k-_-19-118-124-_-Product&quicklink=true

+ Cooler + motherboard costs.

I think when comparing CPU costs you need to consider platform and cooling costs as well.

Interestingly, looking at multiple review sites, 11600K seems to soundly defeat Zen 3 on Turing (2000 series) GPUs.

From PCGamer which is using a 2080 Ti :
View attachment 194542

View attachment 194544

For some reason at a certain threshold of CPU/GPU loading comet lake and seemingly rocketlake perform better with turing. If you swap to ampere this oddity goes away.

Wizzard did a great writeup that explains this in detail.

 

[Why_Me]

128€ - Intel Core i5 10400F 6x 2.90GHz So.1200 BOX - Sockel 1200 | Mindfactory.de
188€ - Intel Core i5 10600KF 6x 4.10GHz So.1200 BOX - Sockel 1200 | Mindfactory.de
258€ - Intel Core i7 10700F 8x 2.90GHz So.1200 TRAY - Sockel 1200 | Mindfactory.de
389€ - Intel Core i9 10850K 10x 3.60GHz So.1200 WOF - Sockel 1200 | Mindfactory.de

These are the only CPUs I'd consider calling a decent (not impressive) value atm. 275€ price tag for 6 core CPU in 2021 that can't beat competition is hardly a value by any merit.

The i5 11400F / i7 11700F paired with a B560 board should be a hit with gamers.

https://www.bhphotovideo.com/c/product/1621133-REG/intel_bx8070811400f_core_i5_11400f_2_6_ghz.html/
Intel Core i5-11400F $167.95 New Item - Coming Soon

https://www.newegg.com/intel-core-i7-11700-core-i7-11th-gen/p/N82E16819118237
Intel Core i7-11700F $365.49

https://www.newegg.com/p/N82E16813119384
ASUS PRIME B560-PLUS LGA $119.99

 

[evernessince]

Interesting. Tom's used a 3090, while TPU used a 3080. So really just goes to show these next gen CPUs don't help gaming much without throwing in a near $3000 GPU, which is going rate for a 3090 right now.

Anyway I have no intention of buying a GPU over $500 vs my 2060KO which sells for that used these days, which means I won't be buying for quite some time.

Was really more interested in the application performance, and the 11600K is pretty good there.

Tom's also forgot to include AMD system specs for some reason.

Interestingly, looking at multiple review sites, 11600K seems to soundly defeat Zen 3 on Turing (2000 series) GPUs.

From PCGamer which is using a 2080 Ti :
View attachment 194542

View attachment 194544

Intel Rocket Lake (14nm) Review: Core i9-11900K, Core i7-11700K, and Core i5-11600K

www.anandtech.com

Anand tested with a 2080 Ti and the results are nowhere near what those charts are showing. PCGamer didn't list AMD system specs either so it's hard to compare.

TechSpot tested with a 3090 and found the 5600X faster on average: https://www.techspot.com/review/2220-intel-core-i5-11600k/

It's really a matter of which games are tested.

The i5 11400F / i7 11700F paired with a B560 board should be a hit with gamers.

https://www.bhphotovideo.com/c/product/1621133-REG/intel_bx8070811400f_core_i5_11400f_2_6_ghz.html/
Intel Core i5-11400F $167.95 New Item - Coming Soon

https://www.newegg.com/intel-core-i7-11700-core-i7-11th-gen/p/N82E16819118237
Intel Core i7-11700F $365.49

https://www.newegg.com/p/N82E16813119384
ASUS PRIME B560-PLUS LGA $119.99

The 10400F is currently the best budget gaming CPU so if the 11400F is on par with that it should be considered. The 11700F is a waste of money though same as the 10700K. You are only going to see gains over the 10400F if you have a very high end GPU at 1080p and lower. The additional cost of the CPU itself, the motherboard, and a high end CPU cooler to cool that thing make it not worth the very limited potential benefits. The only reason to spend more is if you want the best regardless of cost and in that case you should be buying the best, not a CPU that barely beats out budget models for twice the total platform cost.

 

[Why_Me]

Tom's also forgot to include AMD system specs for some reason.

Intel Rocket Lake (14nm) Review: Core i9-11900K, Core i7-11700K, and Core i5-11600K

www.anandtech.com

Anand tested with a 2080 Ti and the results are nowhere near what those charts are showing. PCGamer didn't list AMD system specs either so it's hard to compare.

TechSpot tested with a 3090 and found the 5600X faster on average: https://www.techspot.com/review/2220-intel-core-i5-11600k/

It's really a matter of which games are tested.



The 10400F is currently the best budget gaming CPU so if the 11400F is on par with that it should be considered. The 11700F is a waste of money though same as the 10700K. You are only going to see gains over the 10400F if you have a very high end GPU at 1080p and lower. The additional cost of the CPU itself, the motherboard, and a high end CPU cooler to cool that thing make it not worth the very limited potential benefits. The only reason to spend more is if you want the best regardless of cost and in that case you should be buying the best, not a CPU that barely beats out budget models for twice the total platform cost.

Intel Core i7-10700 vs Core i7-10700K Review: Is 65W Comet Lake an Option?

www.anandtech.com

 

[r9]

10700k is $249 at Microcenter would pick that over 10600k any day of the week.

 

[RandallFlagg]

It's really a matter of which games are tested.

That's actually 90% of the correct answer. I remember when TPU got beat up because gen 10 beat Zen 3 on aggregate in its game review. I had looked up a german site that review like 20-30 games, and about 1/3 of them a 10900K would beat 5900X. It just so happened that a lopsided number of those games were used in TPUs reviews - which they've used for years.

Then there are factors like RAM speed, CL, and even XMP setting and even motherboards. One has only a few clicks to see that each item can affect performance a few percent on a cpu, and it compounds.

The 10400F is currently the best budget gaming CPU so if the 11400F is on par with that it should be considered. The 11700F is a waste of money though same as the 10700K. You are only going to see gains over the 10400F if you have a very high end GPU at 1080p and lower. The additional cost of the CPU itself, the motherboard, and a high end CPU cooler to cool that thing make it not worth the very limited potential benefits. The only reason to spend more is if you want the best regardless of cost and in that case you should be buying the best, not a CPU that barely beats out budget models for twice the total platform cost.

I am seeing 11400 11500 show up on geekbench and PCMark now. Geekbench scores are impressive, PCMark not so much. They seem to have excellent single and low thread count scores, comparable to a 5600X. Multi-core, they fall behind (PCMark).

I think the 11400 and 11500 effectively obsolete the 10400 / 3600 / 3600X for any new build though. 11400 is selling for $189 at microcenter and shows up with a 30% higher single thread performance vs 10400 on geekbench.

 

[HenrySomeone]

10700k is $249 at Microcenter would pick that over 10600k any day of the week.

That is a fantastic price! Honestly the cheapest I've ever seen a cpu so close to the very top in performance hierarchy, especially counting for inflation. Just a shame it's exactly the other way around with gpus (although that's probably the main cause of this discount in the first place - people just aren't building gaming rigs, especially not high-end-but-just-short-of-the-very-top ones which is exactly where this cpu slots). Still though, if you just need a lot of processing power (with minimal graphical one), it's one of the best time to buy (between 10400, this one and 10850k).

 

[Solid State Soul ( SSS )]

The most impressive value would be the i5 11400F, especially compared to the 300$ 5600X

 

[r9]

The most impressive value would be the i5 11400F, especially compared to the 300$ 5600X

No doubt! Can't beat it for $130.

 

[RandallFlagg]

Sorry, but decent in terms of what?

It can't beat the 10600K in gaming.

While consuming 45% (!!!) more power:

This series is complete rubbish.

TPU used gear 2 (desynchronized the memory bus on RL) in an effort to get RL up to the Ryzen favoring DDR4-3800. They had switched to that after they got mobbed when Zen 3 failed to beat Comet Lake on their standard DDR4-3200 CL14 platform. Not even all Zen 3 will run DDR4-3800, but that's their new standard.

By switching away from a very common and consistently useable memory setup for all platforms (DDR4-3200 CL14) to the one that the AMD crowd favored, their benchmarks now cripple the Intel platform and frankly aren't going to properly reflect RL. RL actually would have performed better in most of TPUs benchmarks at DDR4-3600 gear 1.

Computerbase.de talks about this. They state that even the 11900K won't do over DDR4-3733 gear 1 reliably. By going to gear 2 to make up the difference, one would need to run DDR4-4400 or higher.

Sites using DDR4-3600 gear 1 show Rocket Lake generally beating Zen 3 in games.

Gear 1 vs Gear 2 makes about a 3-5% difference, which on your chart up there is enough to vaunt the 11600K above the 5600X, like it's doing on other sites.

I still prefer the site because they have productivity benchmarks like compiler times, virtualization, SQL, java and so on that rarely appear elsewhere.

However, if you are looking for gaming benchmarks for real world gaming scenarios where people don't intentionally hobble their computer in games by desynching the memory bus, you probably should look elsewhere.

 

[Turmania]

If you are a brand new system builder these days may god help you. That is all I can say these times.

 

[Deleted member 205776]

So, takeaways about this gen, nothing except the i5-11400F is worth buying. Well in my eyes nothing is worth buying since the socket is practically dead, unless you like changing boards when Alder Lake is out.

 

[W1zzard]

TPU used gear 2 (desynchronized the memory bus on RL) in an effort to get RL up to the Ryzen favoring DDR4-3800. They had switched to that after they got mobbed when Zen 3 failed to beat Comet Lake on their standard DDR4-3200 CL14 platform. Not even all Zen 3 will run DDR4-3800, but that's their new standard.

By switching away from a very common and consistently useable memory setup for all platforms (DDR4-3200 CL14) to the one that the AMD crowd favored, their benchmarks now cripple the Intel platform and frankly aren't going to properly reflect RL. RL actually would have performed better in most of TPUs benchmarks at DDR4-3600 gear 1.

Computerbase.de talks about this. They state that even the 11900K won't do over DDR4-3733 gear 1 reliably. By going to gear 2 to make up the difference, one would need to run DDR4-4400 or higher.

Sites using DDR4-3600 gear 1 show Rocket Lake generally beating Zen 3 in games.

Gear 1 vs Gear 2 makes about a 3-5% difference, which on your chart up there is enough to vaunt the 11600K above the 5600X, like it's doing on other sites.

I still prefer the site because they have productivity benchmarks like compiler times, virtualization, SQL, java and so on that rarely appear elsewhere.

However, if you are looking for gaming benchmarks for real world gaming scenarios where people don't intentionally hobble their computer in games by desynching the memory bus, you probably should look elsewhere.

Excellent analysis of the situation. The differences are small though, check the 11900K review, it has Gear 1 vs Gear 2 data.

> even the 11900K won't do over DDR4-3733 gear 1 reliably
I'm not even getting 3600 Gear 1 on my 11900K, I tried. That's why I have 3200CL14 Gear 1 vs Gear 2