300 MHz does not equate to destroying anything.
-
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.
Associate
I suspect this will still be the leading high-end gaming chip (short of using chillers and suchlike on the high core count variants). I would agree with Steve from GN that there's no sign of AMD thinking they can get near 5GHz all cores and yet Intel will chip a 5Ghz all-core gaming chip on the 9900KS, which will probably win on single-threaded performance too, in terms of IPC.
It's a strong launch from AMD and they are closer but it's not yet a return of the Thunderbird or Clawhammer days. Maybe they'll win on price but I'm still not convinced they'll top the fps tables, benchmarks, etc.
It's a strong launch from AMD and they are closer but it's not yet a return of the Thunderbird or Clawhammer days. Maybe they'll win on price but I'm still not convinced they'll top the fps tables, benchmarks, etc.
Soldato
- Joined
- 7 Dec 2015
- Posts
- 3,043
This is not entirely true though, with occasional exceptions. When I run demonbuddy for a dead game, 9600K gives me 30 fps while 2700X gives me 23fps. Roughly Zen+ @ 4.3GHz matches Coffee Lake @ 3.3GHz, just because that bot is written for Intel and runs choppy and laggy on Ryzen (bad optimization).
If Zen 2 has 15% more IPC than Zen+ and 3800X can clock @ 4.2GHz for all cores it will match the framerates of 9900K underclocked @ 3.7GHz. A stock 9900K is still 27% faster than a 3800X with all cores clocked at 4.2GHz. You may think 27% is negligible, but it makes a huge difference between being able to stay alive in the game and keeps dying in the game, resulting in over 30% difference of exp/h.
This is not surprising, just like some games are optimized for AMD graphics cards and don't run that well on nVidia graphics cards.
And guess what, Cinebench is not the only benchmark. There are other measurements such like Linpack, for which 1800X is only as good as 7700K.
Soldato
- Joined
- 28 Sep 2018
- Posts
- 2,515
use cases matter. I do majority of my gaming in sim racing with a high end VR unit; Pimax 5k+ Both VR and sims love clockspeed. For that I picked the 9900k when it came out and I knew it wouldn't suffocate a 2080ti. I don't plan to upgrade my cpu/ram/mobo for the next 5 years and I'm confident that a 8c/16t 5ghz all core 9900k will have a good shelf life when I switch to a 3080ti>4080ti.
With that said, if I was buying today, I'd certainly hold the line until reviews are out for new Ryzen chips but calling the 9900 series a joke demonstrates limited perspective.
I do think the price gouging due to supply constraints also contributed to the reputation of the 9900k especially but I pre-ordered in the US and got in on the launch price so I don't have that burden. To put in perspective, my 9900k was cheaper than the 8086k pricing in the UK at that time.
Last edited:
Don't think so much troll but what Robert says below:
There is a lot of stuff I use where for various reasons Intel is still the better choice often due to being hand optimised for Intel CPUs (compiler target, machine code/assembler tricks, etc.) some years back and unlikely to ever get an update with more optimal tuning for other CPUs.
Caporegime
The Core i9 9900KS reminds me of the P4 Extreme Edition,which was a jazzed up P4,drinking power to eek out a win. This is exactly what Intel did when the Athlon 64 FX launched,as it did win some benchmarks at the time.
Soldato
Yes that's turned out to be incorrect, but doesn't take away from the point this was blatant damage control
i wonder what the price difference will be lmao
Looks kind of like a pre-overclocked 9900k for a little extra cost. the 9900k runs stupidly hot so this chip will be on fire
9900ks is just like enabling multicore enhancement in the bios of the 9900k lol but then again, what if the 9900ks turbo's to 5.5
Looks kind of like a pre-overclocked 9900k for a little extra cost. the 9900k runs stupidly hot so this chip will be on fire
9900ks is just like enabling multicore enhancement in the bios of the 9900k lol but then again, what if the 9900ks turbo's to 5.5
Soldato
Also without the chiller needed to keep it boosting to 5Ghz.
Caporegime
Are the benchmarks pre, or post mitigation patches? 
Soldato
I get that it is a great chip and does really well in gaming, but "just" £100 difference is a huge amount at 25% more cost or greater bearing in mind at stock speeds (assuming they were for all) then the 3700X was within 1% single thread and so that is a massive £170 difference for what might possibly be only a small difference.
I would assume on the single thread the CPU for both still boosted to their 4.4Ghz & 5.0Ghz respectfully. So if the 3700X is only dropping 1% whilst being 0.6Ghz lower then getting the all cores to say 4.2Ghz could possibly only net 5% or so performance different.
I get that means that 9900K would still have the performance crown but does the extra £170 which is significant really seem reasonable? That is your RAM or Mobo or new M.2 drive basically paid for.
Soldato
9000K destroy the 3800X, nope. AMD are benching the 9900K at stock speeds so 4.7Ghz, and are claiming parity or better in gaming, do you think an extra 300Mhz will destroy the 3800X, how about no.
Associate
I was kind of hoping that they will at least give us rumored 10 core.
It’s exactly why the 9900ks is coming to keep the gaming crown...
5ghz all core beats the 3800x
But we don’t know what the 3800x clocks to yet...
Associate
so far all this, intel beat ryzen beat intel is pure speculation. Have to wait for independent gaming/performance reviews. But i'll say that Ryzen will take multi tasking crown while Intel will continue to dominate gaming until more than 8 cores for gaming are used.
Soldato
I stated about stock for single thread performance which is 5Ghz (I9 9900K) vs 4.4Ghz (3700X) so where a game is single thread performance (i.e the weakness that AMD had) then they are showing it within 1%.
So if we are able to get the 3700X to close to the 4.4Ghz then it would remain within that 1%. If you look at the fact that every additional 100Mhz appears to give a 2% gain for AMD then you could extrapolate that at 4.0Ghz all core for the 3700X then you would be looking at around a 9% performance advantage for Intel and at 4.2Ghz around 5% advantage.
It would also suggest if AMD was able to release a CPU with an all core of 4.45Ghz then you could have the same performance as Intel 5Ghz all core for gaming.
I know this only simple maths speculation but to write off something that is clearly a lot less in price but within ballpark similar performance seems strange. Further to that, we are assuming that the new 9900KS is going to be £500.
I feel it may be £500 to keep AMD competitive but this is Intel and at same time could imagine them previously aiming this at the £550-600 mark as it is a highly binned chip.
Associate
I wouldn't read too much into what AMD demonstrates on stage.