• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intel Core i9-12900K allegedly scores 30K points in Cinebench R23

Minstadave said:
I get that on mobile that small efficient cores are useful but the small cores seem largely superfluous on desktop don't they? I'm not sure what they'd add over just using the silicon for full cores. It might help eek out some more performance per watt but Intel has shown they don't really care about efficiency much on the high end.
Click to expand...
4 small cores fit in the space of one large core yet provide much more performance.

Intel obviously thinks this is the way forward as its the small core counts than are being increased every generation from now on.

Raptor lake will have 8+16 and meteor lake 8+24 etc.

Rumours are that Zen 5 maybe implementing a big little design also.
 
Last edited:
Rumour has it these chips will still use 250W and peak 400W so not to sure how much efficiency the little cores will bring. Also will be interesting to see how these perform in Windows 10 as i suspect teh leaked results are using Windows 11 and its new scheduler
 
cliffy said:
Rumour has it these chips will still use 250W and peak 400W so not to sure how much efficiency the little cores will bring. Also will be interesting to see how these perform in Windows 10 as i suspect teh leaked results are using Windows 11 and its new scheduler
Click to expand...

That's my point I guess. Tacking on a butt load of Atom-esque cores to pad out the multithreaded performance seems a messy solution to Intel's efficiency problem.
 
Grim5 said:
Intel's results speak for themselves.

the little cores are there to improve performance without blowing out the power budget.

Yes you're asking why don't they use 16 big cores and the answer is that no one wants a 500w CPU like the old 10980xe. So 8 little cores gives them basically 9900k power consumption while having extra multithreaded performance.

Intel has figured out what you haven't, that the performance per watt and per million transistors is higher on the small cores than big cores, so they only had a couple options : make 10 big cores for 250w or make 8 big and 8 little for the same 250w and it turns out 8 little skylake cores is better than 2 big golden cove cores
Click to expand...

Intel is making some pretty Cinebench numbers I don't disagree but it's not leagues ahead of a 5950X. I wonder how this implementation translates to real world use cases, I guess it would be useful to bin off all the low level processes to these smaller cores and if you're rendering/transcoding then this solution may work nicely but currently if you need multithreaded performance you'd just go 5950X or Threadripper, I'm not sure Intel's solution is superior or just a bandaid to cover for their terrible efficiency. Be interesting to see how this evolves over time.
 
Last edited:
Minstadave said:
Intel is making some pretty Cinebench numbers I don't disagree but it's not leagues ahead of a 5950X. I wonder how this implementation translates to real world use cases, I guess it would be useful to bin off all the low level processes to these smaller cores and if you're rendering/transcoding then this solution may work nicely but currently if you need multithreaded performance you'd just go 5950X or Threadripper, I'm not sure Intel's solution is superior or just a bandaid to cover for their terrible efficiency. Be interesting to see how this evolves over time.
Click to expand...
It will be lower down the stack where it's most interesting, higher than a 5800X multithreaded performance for 5600X prices and 5900X multithreaded performance for 5800X prices. Both areas in which Zen 3 failed to deliver tangible gains over zen 2 at their respective price points.
 
Minstadave said:
Intel is making some pretty Cinebench numbers I don't disagree but it's not leagues ahead of a 5950X. I wonder how this implementation translates to real world use cases, I guess it would be useful to bin off all the low level processes to these smaller cores and if you're rendering/transcoding then this solution may work nicely but currently if you need multithreaded performance you'd just go 5950X or Threadripper, I'm not sure Intel's solution is superior or just a bandaid to cover for their terrible efficiency. Be interesting to see how this evolves over time.
Click to expand...

It will be Interesting to see. What Intel need are chips with higher efficiency big core clusters that scale past 8 cores. That’s not what is on offer with Alderlake.

The key will be scheduling tasks to the best cores for the job, but it’s also tricky part to leveraging performance.
 
Dave2150 said:
Source: https://videocardz.com/newz/intel-core-i9-12900k-allegedly-scores-30k-points-in-cinebench-r23

Intel back on top, again with an inferior 10nm (Intel "7") process. Will be fun once Intel's fab's eventually match or exceed TSMC, which won't be that long away most likely.
Click to expand...

Hi Dave wasnt rocket lake meant to blow zen3 ? you went quiet for awhile you actually bought one should never have bought one one of the worst releases :cry: I dont understand brand loyalty to that level
 
Joxeon said:
4 small cores fit in the space of one large core yet provide much more performance.
Click to expand...
Bazillion tiny cores won't make one thread execute any faster.
While some things multithread and can scale to really high core counts, some things don't benefit any no matter how many threads you make.
Basically anything in which next operation depends on previous operation won't do anything with higher number of smaller cores.

And scheduling is going to be nightmare for realizing full performance:
You don't want threads going to heavy weight core eating unnecessary amount of power/thermal budget when not needed.
But if such serial execution of commands needing thread/workload goes to lite core even momentarily that's going to hit performance.
(moving workload between cores has delay)
Also it's really hard to make smaller core without cutting out some instructions/their hardware execution unit.
And emulating such instructions would be really slow compared to hardware, while moving thread back to full core again adds its penalty.

Basically arbitrarily multihreading stuff without any inter thread dependancies, like Cinebench, is the best case.
(also for hiding horrible memory latency)
 
Joxeon said:
What we are getting this time is 10 core CPUs for £250 which when compared to CPUs like the 5600X that gave us just 6 cores but for the price of 8 then it's very compelling if the early benchmarks prove right.
Click to expand...

The prices make it un-compelling for me. Premium priced DDR5 and Mobo required at the same time as the CPU stings too much right now.
 
Joxeon said:
They have 14 core 6+8 versions going into laptops so they can't be that power hungry.

I'm guessing the small cores will focus on light background tasks and heavy workloads while the big cores will be more useful for gaming.
Click to expand...

Are they really?
That is rather interesting.
Is it the same cores as the desktop chip?
 
These chips are more aimed at laptops than desktops imo. Been big little set up with mobile SOCs for years so it is almost surprising it has taken laptops so long to catch up.
 
You must log in or register to reply here.
Back
Top Bottom