Core i9 10900K Review with benchmarks finds its way to the web in Asia
Click here to post a comment for Core i9 10900K Review with benchmarks finds its way to the web in Asia on our message forum
Andrew LB
Whenever intel releases a new CPU, ever notice how the leaks usually involve a render test like Corona or Cinebench? Yet those tests that always favor AMDs high core count don't end up being more than just outliers in the all encompassing reviews coming in a few days. Let them play their synthetic benchmarks. imo anyone using their PC for rendering wouldn't buy either of these chips to begin with.
metagamer
I just watched that Corona video. 20m48s vs 23m50s. 3 minute difference. If you spend hours a day rendering in Corona or similar, this Intel isn't for you. Nor is the 3900x. If you render an hour or 2 a day the difference is negligible.
jwb1
There are several AIR coolers rated for 250 (+) watts, the Noctua nh-d15 (or S), the Be Quiet Dark Rock Pro 4, Coolermaster MASTERAIR MA620M, to name a few. My 9900k never goes over 85 degrees at 5.1ghz all-core when rendering in Handbrake H.265 10bit with the MA620M.
jwb1
Intel has new tech. Rocket Lake for Z490 will see more IPC as it won't be based on Skylake anymore, still 14nm, but at least they will get off Skylake. Then probably 7nm. I bet they skip 10nm for desktops.
People who buy these high end Intel chips don't care about power. Hence why Z490 mobos and pre-orders for the 10th gen CPUs are selling out already before official release date or reviews. They like the overclock, tweak, and play games mostly. Just like AMD people don't care about losing out on the faster CPU for gaming. They like their extra cores that most don't max out, outside AIDA64 or Cinebench.
Ricardo
This argument of power consumption is so tiring. People don't realize that the more energy going through a chip, the less longevity it will have. That is the main point of looking for more efficient parts, not the electric bill or the environment.
Another good one is that less power means less need for beefy PSUs, which in turn means a good PSU will remain working for longer and will be less susceptible to failure.
This applies to both CPUs and GPUs, which are the two most power hungry components of a PC - more power efficiency means better system overall.
metagamer
Totally agree man.
Explains why you run a 3700x with a 1050ti. Perfectly balanced.
jwb1
Heat and power do eventually contribute to the eventual death of components, but in this enthusiast community, I've never seen any real world data to prove this happening normally. Most people on both sides, Intel/AMD don't keep their parts long enough to ever worry about this.
Ricardo
LOL
Hey buddy, ever heard of upgrading your PC in pieces? Not everyone is rich. Besides, my old i5 3570 bit the dust, and I'm a programmer, so I couldn't wait to buy a whole new PC.
For the CPUs themselves that may be true, but GPUs and PSUs will definitely feel the stress much quicker, and big power consumption will absolutely make a difference in 3+ years of use. Especially for the enthusiast crowd, who tend to mess with every little knob to max an overclock.
nizzen
5200mhz all core is pretty much doable with watercooling:
OCUK is selling 5200mhz 10900k bundle
https://www.overclockers.co.uk/8pack-xtreme-asus-maximus-xii-formula-overclocked-i9-10900k-5.2ghz-gaming-bundle-bu-01u-8p.html
Webhiker
So to sum it up. Benchmarks where intel is doing well are important and should be published. While benchmarks where AMD are winning doesn't matter and should just be forgotten about ?
metagamer
That's not bad. I bought my 7700k from OcUK and it was a 8pack 5ghz chip and yep, it runs 5ghz all day long no problem from day one. They were delided chips and use liquid metal as TIM.
I mean if all you do is play Cinebench and compress your hard drive all day every day, and render stuff 20 hours a day then yeah, buy AMD. Nobody is denying that, AMD do bring the fun factor in those scenarios.
Webhiker
If you do a little alternative math an calculate the all core GHz and divide it with the time it takes you get : AMD doing 38.94 and intel 34.26 amount of work pr / sec
I find that interesting. You have a CPU running 4900 GHz (10 cores) all core vs one running ~4050 GHz (12 core) all core.
So that's 49.000 vs 48.600 which just shows what we already know, that Ryzen has much better IPC.
Webhiker
I actually think it's important to show how a CPU performs in both single and multi threaded workloads. According to many comments recently you don't need more that 6 cores to game. So what is the 10900K for ?
metagamer
no it doesn't
for gaming. Because it's the fastest chip for gaming. Period.
JethroTu11
AMD has pushed Intel to their 14nm limits.
jbscotchman
Our great leader's words are the only ones that matter.
Mesab67
Will be interesting to see on Wednesday, for the 10900K, just what kind of system outlay will be required for those extra 'pro-gamer' fps @1080p, wall-socket power draw included. Not knocking Intel for chasing that particular crown (though, at all costs!), they'd benefit a lot more from R&D. Hearts and mind aren't won targeting the affluent consumer only.
With next gen consoles inbound that are going to deliver very significantly higher CPU and GPU horsepower and functionality not seen for some time (thanks to a certain company), the next couple of years are going to be very interesting indeed, for everyone. Let's all hope this new competition-driven innovation (and pricing ?/!) lasts for some time.
metagamer
Here's what the review will reveal. 10900k is the fastest gaming chip. In "productivity" it is slightly slower than the 3900x. We all know how much it'll cost and how much power it consumes.
The review will not tell you anything new.
jbscotchman
Yea pretty much.
jwb1
And that there are more in-silicon mitigations. Rocket Lake is going to be more interesting as it will have a new architecture based on 14nm.
