Intel Core i5-11600K vs. AMD Ryzen 5 5600 vs. Core i5-10600K

After that SiSoft Sandra slide:
"Intel’s 10th gen Core series was particularly weak for AES-256 encryption, but here we’re looking at a huge 57% performance uplift for the 11900K, placing it 6% ahead of the 5600X. An incredible performance uplift and this is the first benchmark we’ve seen where Intel now beats AMD."

Shouldn't that read 11600K, or have I just not had enough coffee yet?

/there's never enough coffee
Hold off on the coffee for now, you're right and I've now fixed it. Thanks for pointing that out.
 
Great article. It'll be interesting when 12th gen launches on 10nm, although the Question is will it be able to clock as high because the current 10nm chips from intel have struggled to clock anywhere near as high as 14nm, so I can't see 12th gen bringing much accept lower power consumption.
11th gen is a right off just like 7th gen was when Ryzen first launched.
 
I've got friends looking at building PCs and I've been suggesting to them for a while now that a 10400 or 10600 is where the value is. I built two Zen 2 systems for myself last gen but now that everybody is chasing the AMD dragon the prices have inflated and availability has plummeted. Meanwhile the price has been falling on 10th gen Intel, and you can pick them up anywhere.

To be honest I've considered this whole generation a write off for both CPU and GPU. Ryzen 5000 offered a reasonable value proposition at MSRP when it was released, though as expected AMD took their performance crown and ran with it to increase prices which somewhat dampened that. But they don't exist at MSRP, so price to performance is pretty shabby in real terms. I'm sure I don't need to talk about the GPU situation to anyone here.

While I myself did enjoy the schadenfreude at seeing Intel get their comeuppance after years of overpriced mediocrity, let's not pretend that AMD is some shining beacon of consumer friendliness. They will leverage their superior product to improve their bottom line, like every corporation ever. Pushing an anti-Intel hivemind is only going to allow AMD to increase their margins in the long run.
 
The AMD 5600 still makes the most sense when you factor in the cost of a cooler for the hot-running Intel part. Then there's the actual cost of burning all that power - more expensive motherboard etc.
The only difference is availability...But here in Britain at any rate, the AMD 5600's are widely available.
So, nice try Intel....But no cigar.
 
Let's push the 5600X to a 240W and see how it goes. So even the 6 core is now a power hungry toaster. Now thanks.

Although Intel is garbage at the moment I do hold out hope for Alder Lake being a great improvement. Maybe when Meteor Lake refresh rolls out, assuming Alder Lake isn't a flop, it may be a viable alternative to Zen 5+/6. Yes it's 2023 at the earliest.
 
AMD may have overcharged for the lastest and greatest, is that what you are refering to? I could be wrong but I do not remember AMD charging more than Intel in the last 22 years. I do remember the race to 1GHz though.
and AMD creating the X86/64 something Intel said consumers didn't want or need and then Intel colluded with Microsoft to delay the consumer level 64 bit operating system until Intel could catch up to AMD.
 
"The Core i5-10600K is $224 and the 10600KF model (no graphics) posing an even more attractive proposition at $200. Those 35% savings could then go into something else"

Savings? They will disappear when you need a new motherboard because Intel has a new CPU that needs one every 18 months!!
 
Is there any point to even consider K-skus at this point(unless you're just overclocking for fun)? $50+higher power consumption and temperatures for 3% performance boost seems very questionable at this price point - and you can save even more by getting a cheaper motherboard. Assuming Anandtech's findings about 10700 hold true for Rocket Lake, there isn't even difference in boosting behavior in many cases, because motherboards don't respect the PL2 max duration by default.
 
From an economic perspective, I am not sure if Intel should be happy with 11th gen.

All SKU use the same 270mm2 die and the 11400F‘s retail price is between what the i3 8300 and i3 8350k retailed for and these chips only had a die size of 126mm2. Both on the same process, give or take a few +

Without taking yields and wafer area lost due to the large die size into account, Intel has to give you more than twice the silicon for the same price on the same process. On top of that RKL has a new core and the R&D expenses plus backporting costs also need to be earned. They surely can‘t be happy about that.
 
Look at how miraculously, Intel can now offer such a CPU for that price.

They are only doing this now because AMD is kicking their arrogant and corrupt @sses.

If not, that CPU would easily cost us over US$700!

That's why I refuse to give them any money and hope that they die.
I pity people who are stupid enough to genuinely believe Intel is worse than AMD or that AMD is worse than Intel.

Neither of them give a dam about you. They just want your money...
 
"The Core i5-10600K is $224 and the 10600KF model (no graphics) posing an even more attractive proposition at $200. Those 35% savings could then go into something else"

Savings? They will disappear when you need a new motherboard because Intel has a new CPU that needs one every 18 months!!
5000 series is also on a dead end platform.
 
Let's push the 5600X to a 240W and see how it goes. So even the 6 core is now a power hungry toaster. Now thanks.

Although Intel is garbage at the moment I do hold out hope for Alder Lake being a great improvement. Maybe when Meteor Lake refresh rolls out, assuming Alder Lake isn't a flop, it may be a viable alternative to Zen 5+/6. Yes it's 2023 at the earliest.
Let's push the 5600X to a 240W and see how it goes. So even the 6 core is now a power hungry toaster. Now thanks.

Although Intel is garbage at the moment I do hold out hope for Alder Lake being a great improvement. Maybe when Meteor Lake refresh rolls out, assuming Alder Lake isn't a flop, it may be a viable alternative to Zen 5+/6. Yes it's 2023 at the earliest.
you cannot push the 5600x to 240w since it has so much frequency headroom until it hits a wall
 
I've got friends looking at building PCs and I've been suggesting to them for a while now that a 10400 or 10600 is where the value is. I built two Zen 2 systems for myself last gen but now that everybody is chasing the AMD dragon the prices have inflated and availability has plummeted. Meanwhile the price has been falling on 10th gen Intel, and you can pick them up anywhere.

To be honest I've considered this whole generation a write off for both CPU and GPU. Ryzen 5000 offered a reasonable value proposition at MSRP when it was released, though as expected AMD took their performance crown and ran with it to increase prices which somewhat dampened that. But they don't exist at MSRP, so price to performance is pretty shabby in real terms. I'm sure I don't need to talk about the GPU situation to anyone here.

While I myself did enjoy the schadenfreude at seeing Intel get their comeuppance after years of overpriced mediocrity, let's not pretend that AMD is some shining beacon of consumer friendliness. They will leverage their superior product to improve their bottom line, like every corporation ever. Pushing an anti-Intel hivemind is only going to allow AMD to increase their margins in the long run.
the hard truth is that AMD would have done exactly the same with prices if Intel would've been in deep ****.
 
Intel is still technological leader in security holes.
AMD is slowly trying to catch up, but Intel has so many security holes, they should rename their CPUs from "<something> Lake" to "<something> Cheese".

12600 should be "Tilziter Cheese"
13600 maybe "Emmentaler Cheese"

And so on. They should always point out features they are the best in.
 
Last edited:
From an economic perspective, I am not sure if Intel should be happy with 11th gen.

All SKU use the same 270mm2 die and the 11400F‘s retail price is between what the i3 8300 and i3 8350k retailed for and these chips only had a die size of 126mm2. Both on the same process, give or take a few +

Without taking yields and wafer area lost due to the large die size into account, Intel has to give you more than twice the silicon for the same price on the same process. On top of that RKL has a new core and the R&D expenses plus backporting costs also need to be earned. They surely can‘t be happy about that.
This is precisely why the consumer desktop Rocket Lake range tops out at 8 cores. The likes of the 11900 is 34% bigger than the 10 core 10900K, in terms of area (it's not as high, though), and 53% larger than the 9900K. It's worth noting that this single die is generating 19 SKUs - far more than the likes of the die used for the 10900K - so Intel are clearly expecting to cover costs via this route, rather than having multiple die formats.
 
From the article: "Time to buy Intel?" and "this new hybrid 14nm design"
giphy.gif
 
This is precisely why the consumer desktop Rocket Lake range tops out at 8 cores. The likes of the 11900 is 34% bigger than the 10 core 10900K, in terms of area (it's not as high, though), and 53% larger than the 9900K. It's worth noting that this single die is generating 19 SKUs - far more than the likes of the die used for the 10900K - so Intel are clearly expecting to cover costs via this route, rather than having multiple die formats.
True, they probably don‘t have any dies to throw away.

That said, weren‘t previous smaller dies that were not up to task also passed down the SKU ladder, I.e. an i3 becoming a Pentium or Celeron.

As far as I remember, Intel even used harvested 10C dies for some of their 6C Comet Lake SKU in spite of having I think four different die types from 4C-10C.
 
While I myself did enjoy the schadenfreude at seeing Intel get their comeuppance after years of overpriced mediocrity, let's not pretend that AMD is some shining beacon of consumer friendliness. They will leverage their superior product to improve their bottom line, like every corporation ever. Pushing an anti-Intel hivemind is only going to allow AMD to increase their margins in the long run.
Oh yes I'm waiting for AMD to strike a deal with Dell to lock out Intel processors from that market.... to strong-arm retailers by offering them deep discounts if they refuse to sell Intel... to create an AMD C++ Compiler (ACC) that will detect GenuineIntel tag and force Intel processors to run on generic unoptimized x64 path...

oh wait wrong era. 🙄

AMD has a long way to go to become a hypercapitalistic monster like Intel and I doubt they will ever reach that low, so I'm not worrying on that front. If and when they do, count on me being the first to blame Intel 100% for being uncompetitive and allowing AMD to ride roughshod all over them.

AMD will take 0% of the blame in this hypothetical scenario.

Just as Intel took 0% of the blame before the Ryzen era.
 
Last edited:
That said, weren‘t previous smaller dies that were not up to task also passed down the SKU ladder, I.e. an i3 becoming a Pentium or Celeron.

As far as I remember, Intel even used harvested 10C dies for some of their 6C Comet Lake SKU in spite of having I think four different die types from 4C-10C.
They did - when you have a surfeit of barely functional dies to use up, then they will always get used in different family SKUs. We see this sometimes with GPUs, where larger chips get used in models that would typically have the medium/small-sized die.

The 11th gen Core desktop range is essentially just two dies: Comet Lake Refresh for i3, and Rocket Lake for i5, i7, and i9.
 
Back