Monday, October 11th 2021

Intel Core i5-12400 Could be the Next Price-Performance King, Beats Ryzen 5 5600X in Leaked Benchmarks

Intel's upcoming Core i5-12400 "Alder Lake-S" processor could be an interesting piece of silicon. Apparently, not all 12th Gen Core i5 desktop chips have the same core-configuration. While the top Core i5-12600K is expected to have six "Golden Cove" P cores and four "Gracemont" E-cores, some of the lower variants, such as the i5-12400, will lack E cores, and be pure P core chips. In this case, the chip is 6-core/12-thread with just P cores; 1.25 MB of dedicated L2 cache per core, and 18 MB of shared L3 cache. You'll probably get all the next-gen I/O, including PCI-Express Gen 5 (PEG slot), a PCI-Express Gen 4 CPU-attached NVMe slot, and DDR5+DDR4 memory.

Given that the Core i5-11400 is a $190 part, even with a 10-15% price hike, the i5-12400 is expected to be under $220. The only drawbacks here are expected to be locked BClk multiplier, and rather low clock speeds of 4.00 GHz. A user on Chinese social media posted alleged Cinebench R20 results of the i5-12400. It scores 659 points in the single-threaded test, and 4784 points in the multi-threaded test. Wccftech tabulated this against known performance numbers of popular chips, and found that the i5-12400 might end up slightly ahead of the Ryzen 5 5600X, a currently-$300 part. The table also puts out leaked i9-12900K numbers, which indicate why AMD is rushing with "Zen 3+" with 3D Vertical Cache, instead of next-gen "Zen 4."
Sources: 热心市民描边怪 (bilibili), WCCFTech, HXL (Twitter), VideoCardz
Add your own comment

69 Comments on Intel Core i5-12400 Could be the Next Price-Performance King, Beats Ryzen 5 5600X in Leaked Benchmarks

#1
Vayra86
Wait, where did those 800 ST points go now all of a sudden?

:rolleyes::wtf::sleep:
Posted on Reply
#2
londiste
Given that the Core i5-11400 is a $190 part, even with a 10-15% price hike, the i5-12400 is expected to be under $220.
Given that i5-11400 has RCP of $182, i5-10400 had RCP of $182, i5-9400 had RCP of $182, i5-8400 had RCP or $182, i5-7400 had RCP of $182 and i5-6400 had RCP of $182 I am kind of wondering why are your expecting i5-12400 to cost $220?
Vayra86Wait, where did those 800 ST points go now all of a sudden?
Clock speeds. The story says 4GHz and I am pretty sure 12900K 800+ ST score is at 5GHz. CB scores are very-very linear and 650-ish at 4GHz tracks well with 810-ish at 5GHz.
Posted on Reply
#3
xkm1948
Vayra86Wait, where did those 800 ST points go now all of a sudden?

:rolleyes::wtf::sleep:
If you read the source, the max turbo is like only 4GHz or something. 12900K probably turbo all the way to 5GHz
Posted on Reply
#4
The Quim Reaper
Lol..this thing, as gimped as it is with p cores and 4Ghz clock speeds, gets within 250 points of my 5Ghz 9900K in multi-core and thrashes it by ~30% in single core.

9900K..Once king of the hill, now heading rapidly to the back of the bus, old friend...
Posted on Reply
#5
Turmania
I don't know why amd is bothered about zen3+? They should release zen4 next....but then AMD has a history of shooting themselves in their own foot when they are leaders.

On the i5 12400, when most users are now in almost 5 ghz territory. Going back to 4 ghz.... I'm well aware frequency is not everything but still it holds some psychological weight. I'm sure it is a great cpu though those non k versions always were.
Posted on Reply
#6
Vayra86
xkm1948If you read the source, the max turbo is like only 4GHz or something. 12900K probably turbo all the way to 5GHz
Next step: consider why these turbo to 4 Ghz instead of 5 Ghz. Its not because they have TDP headroom, that's for sure.
So your P cores in an i5 are at least 20% worse than the P cores in an i9. Encouraging. Or those E cores are doing some hefty cheating along the way.

Either way, this is a very unusual stack of CPUs that warrants careful analysis. Each part is going to be radically different apparently. OR, the product segmentation is 100% artificial, but somehow, I have some doubts there ;)
Posted on Reply
#7
MagnuTron
Same tests show a i7-11700K beating the 5600X in ST - which does NOT correlate to CS:GO performance, which weirdly enough is a good metric on its own. I am eagerly awaiting CS:GO benchmarks vs Zen3.
Posted on Reply
#8
xkm1948
I am looking forward to this new design from Intel. Interested to see how they will approach HEDT this round. I will wait for 2025 before I ditch my X99 platform for new HEDT
MadsMagnusSame tests show a i7-11700K beating the 5600X in ST - which does NOT correlate to CS:GO performance, which weirdly enough is a good metric on its own. I am eagerly awaiting CS:GO benchmarks vs Zen3.
CS:GO loves huge cache. Entire line of Zen3 has monster level of cache
Posted on Reply
#9
bug
TurmaniaI don't know why amd is bothered about zen3+? They should release zen4 next....but then AMD has a history of shooting themselves in their own foot when they are leaders.

On the i5 12400, when most users are now in almost 5 ghz territory. Going back to 4 ghz.... I'm well aware frequency is not everything but still it holds some psychological weight. I'm sure it is a great cpu though those non k versions always were.
This is the locked part, probably confined to 65W. There will certainly be an unlocked counterpart, probably going all the way to 95W. If it's not ridiculously marked-up, it will be an interesting part as well.
Posted on Reply
#11
RedBear
TurmaniaI don't know why amd is bothered about zen3+? They should release zen4 next....but then AMD has a history of shooting themselves in their own foot when they are leaders.
Maybe TSMC won't have enough 5nm capacity until Apple will transition to 3nm?

Interesting numbers coming from that benchmark anyway, competition can only be good for the consumers.
Posted on Reply
#12
Asni
Well, i would fix the title: "Intel Core i5-12400 Could be the Next Price-Performance King in any non latency sensitive application".
It will be hard to realize a 2022 cpu doesn't overcome a 6700k from 2015 in 99% of the games.
Posted on Reply
#13
bug
krukGaming performance of ADL might not be what you might expect from these leaks:


Wait for proper benchmarks ...
Well, it's only reasonable performance in Cinebench is not the same as performance in any different scenario. Zen itself overtook Intel in Cinebench easily, but it took its third iteration to also close the gap in gaming.

Anyway, speculation might have set some expectations half a year ago. Today, we just have to wait for 3 more weeks and read the reviews. Speculating is pretty pointless at this point.
Posted on Reply
#15
AnarchoPrimitiv
TurmaniaI don't know why amd is bothered about zen3+? They should release zen4 next....but then AMD has a history of shooting themselves in their own foot when they are leaders.
People seem to forget that Intel literally has an R&D budget 6.5x greater than AMD and an annual revenue 8x greater than AMD (FYI, Nvidia has an R&D budget 2x greater than AMD, and that's for GPUs alone, not x86 and GPUs), so AMD is practically on a shoestring budget compared to Intel. This makes it all that more impressive what Lisa Su has been able to do with what she has... Can anyone name any other example, from any other industry, where a small entity has not only competed, but bested their competitor who has R&D and annual revenue that many more times greater than Intel has over AMD?

That being said, I think it would have been in all our interests with respect to preserving competition in the long term, if AMD had another 3-5 years of dominance in order to increase market penetration (especially in the most lucrative x86 markets, mobile, though I'm sure Intel used their influence to ensure AMD didn't have access to the top tier models for laptops until the last year or so, and enterprise where I'm sure Intel sold chips at cost to hinder AMD's access), annual revenue and to build up their financial reserves. "Enthusiasts" are so fickle (and many probably even never tried AMD despite being the better product due to their.... Loyalties) that I'm sure just two consecutive generations of AMD not outperforming Intel by 15%+ could easily rollback any gains... And if AMD is just on par or even slightly behind, it'd be even worse. I don't think people realize how crucial the last few years and next 3-5 years are for ensuring that AMD remains a viable competitor in the x86 market... If things go bad, I could easily see AMD retreating from consumer x86, focusing on semi-custom, trimming the fat and calling it a day all in the interest of ensuring profitability.

All I'm saying is, I think it's a lot easier than people imagine for us to slide right back into the early to mid-2010s of complete innovative stagnation, 4% gen-over-gen performance increases, and unjustifiably high prices for mid to top tier CPUs... As Intel has proven in the past that they'll immediately slam the brakes on progress unless a proverbial gun is pressed against their head.
Posted on Reply
#16
bug
AnarchoPrimitivPeople seem to forget that Intel literally has an R&D budget 6.5x greater than AMD and an annual revenue 8x greater than AMD (FYI, Nvidia has an R&D budget 2x greater than AMD, and that's for GPUs alone, not x86 and GPUs), so AMD is practically on a shoestring budget compared to Intel. This makes it all that more impressive what Lisa Su has been able to do with what she has... Can anyone name any other example, from any other industry, where a small entity has not only competed, but bested their competitor who has R&D and annual revenue that many more times greater than Intel has over AMD?
Happens all the time. But usually the little guy is way, way smaller than AMD is, compared to Intel, and gets bought out. For example: WhatsApp and Facebook. And sometimes the little guy doesn't sell out and then you're looking at Google vs Lycos/Altavista/others.
Posted on Reply
#17
Makaveli
Looks good but i'm tired of seeing leaks of synthetic benchmarks and need to see a full review. I'm expecting ADL to do much better than RKL against Year old Zen 3 but I need to see more.
Posted on Reply
#18
Turmania
AnarchoPrimitivPeople seem to forget that Intel literally has an R&D budget 6.5x greater than AMD and an annual revenue 8x greater than AMD (FYI, Nvidia has an R&D budget 2x greater than AMD, and that's for GPUs alone, not x86 and GPUs), so AMD is practically on a shoestring budget compared to Intel. This makes it all that more impressive what Lisa Su has been able to do with what she has... Can anyone name any other example, from any other industry, where a small entity has not only competed, but bested their competitor who has R&D and annual revenue that many more times greater than Intel has over AMD?

That being said, I think it would have been in all our interests with respect to preserving competition in the long term, if AMD had another 3-5 years of dominance in order to increase market penetration (especially in the most lucrative x86 markets, mobile, though I'm sure Intel used their influence to ensure AMD didn't have access to the top tier models for laptops until the last year or so, and enterprise where I'm sure Intel sold chips at cost to hinder AMD's access), annual revenue and to build up their financial reserves. "Enthusiasts" are so fickle (and many probably even never tried AMD despite being the better product due to their.... Loyalties) that I'm sure just two consecutive generations of AMD not outperforming Intel by 15%+ could easily rollback any gains... And if AMD is just on par or even slightly behind, it'd be even worse. I don't think people realize how crucial the last few years and next 3-5 years are for ensuring that AMD remains a viable competitor in the x86 market... If things go bad, I could easily see AMD retreating from consumer x86, focusing on semi-custom, trimming the fat and calling it a day all in the interest of ensuring profitability.

All I'm saying is, I think it's a lot easier than people imagine for us to slide right back into the early to mid-2010s of complete innovative stagnation, 4% gen-over-gen performance increases, and unjustifiably high prices for mid to top tier CPUs... As Intel has proven in the past that they'll immediately slam the brakes on progress unless a proverbial gun is pressed against their head.
Heard this all before, why are you obsessed with budgets? It is not like as if AMD is short of cash. And holding performance edge once in a quarter of a decade is nothing to brag about. The real accomplishment would be to to not to reach top but to stay there, don't you think the same? A promoted team from england won Premiership title, the year they were fabourites to go down, budget is not everything.
Posted on Reply
#19
freeagent
I might have to upgrade after Alder Lake.. wait for the first year bumps to get shaken out. Hopefully I return to all core overclocking again, this boosty stuff is no fun.
Posted on Reply
#21
defaultluser
an we assume that this will be just as castrated as the Rocketry Lake 11400 (8 eus disables?)

Only Intel can find a way to make their new Core i5 processors even slower at iGPU than it's old HD 630!
Posted on Reply
#22
Vayra86
defaultluseran we assume that this will be just as castrated as the Rocketry Lake 11400 (8 eus disables?)

Only Intel can find a way to make their new Core i5 processors even slower at iGPU than it's old HD 630!
Hey they need to sell those spanking new dGPUs too!
Posted on Reply
#23
GreiverBlade
could they stop with the obvious "oh wow, the next gen from XXX beat the previous gen from XXX" ...


[personal opinion space]
also a locked 12400 beating a 5600X on price perf value? well if the gap is in double digit and noticeable in daily usage ... yeah sure ...
5600 is ~230$ for me atm 5600X is ~285$ given a choice even a 5600G would be just as good as that 12400 if IGP is needed at ~275$
(obviously given that i have an AM4 board, if i still had my 6600K and would need, indeed, a total platform overhaul, i might have thought differently, but then AM5 and next Ryzen is also on the way ... )

although i reckon the score is quite impressive against my 3600 (stock, got free but value paid by the generous donator was 185 chf ) MC 3412/SC 478 (again i wonder if it would translate with a huge gape in my daily use)

nonetheless i am glad Intel is not sitting on their thumbs ... but i kinda find their current attitude childish (not referencing leak, but rather their PR )
Posted on Reply
#24
klf
niceee,,, soo semms early leak from some website in arabic language where was proably overclocked 12900 cpu and show nearly 1000 point in single core,,,
i think it was enginering sample but "golden chips" will able give for everyday work in windows with some aio 360 cooler about 800-900 points,, of course will not possible run all cores on 100%... but on that multicore software work i have 2990WX... )) old and unsupoorted from amd...
soo i see nearly all programs in my win run on max 1-2 cores,,,, more cores need only if you do x265.. AI go best on nvidia
thast why i go to intel....and some fast pcie-5 nvme
Posted on Reply
#25
londiste
GreiverBladealso a locked 12400 beating a 5600X on price perf value? well if the gap is in double digit and noticeable in daily usage ... yeah sure ...
5600 is ~230$ for me atm 5600X is ~285$ given a choice even a 5600G would be just as good as that 12400 if IGP is needed at ~275$
Tried out of curiosity, in CPU-Z my 5600G gets 570 and 4485, so those supposedly 12400 scores are 19% faster in ST and 11% faster in MT.
5600G vs 5600X is an interesting dynamic. In my experience the frequencies end up pretty much the same but where cache matters, 5600X is measurably faster. That includes games.

Edit: My 5600G Cinebench R20 results are 21-22% lower.
Posted on Reply
Add your own comment
Dec 26th, 2024 04:07 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts