Core i7 9700K Results Spotted in Geekbench Database
Click here to post a comment for Core i7 9700K Results Spotted in Geekbench Database on our message forum
Robbo9999
I'm quite surprised to see such a clear win for the 9700K over the 8700K in this benchmark for the multi-threaded portion, because the 8700K has 12 threads vs only 8 for the 9700K, I thought that in multicore tests they would be closer to each other with maybe a very slight win for the 8700K. Maybe Geekbench doesn't use 12 threads effectively, like maybe there's a ceiling for the number of threads it can use, don't know?
warlord
Nope, 9700K is just much more efficient. Gaming wise it seems it will be also a superior choice. Less and more powerful threads. That is what we really want. After single threaded performance the most important factor is the distributed power between cores/threads.
Pimpiklem
All these years Intel saying paste is fine we dont need solder.
Lets see if they apologise and admit they were mistaken and man up.
Lets see if they deserve any respect and admit when they are wrong.
Lets just see.
Noisiv
i7 8700k is what... 4.2 or 4.3 GHz on all cores
i7 9700k is 4.6 GHz @8-cores
Kaarme
9600K seems pointless compared to the other two. But I guess they need it for the full lineup.
Fox2232
Geekbench3 did not support SMT. Likely nothing changed with Geekbench4.
I think you missed the point in toothpaste. They made it with Ivy. Compare its thermals with Sandy...
Sandy @4,5GHz ~60-62°C under full load and all cores (with good air cooling). Maximum OC around 4,8-5,0GHz before thermals/power consumption was an issue.
That's for chips sold as 3,3~3,4GHz guaranteed once all cores are under load.
Ivy changed that. Not so easy, not so cool. Intel dropped toothpaste and went metal solder at time when toothpaste had too little thermal headroom. They would have trouble delivering those stock clocks with toothpaste.
And in this situation, they are risk-free from repeating another Sandy which lasted 5 years before stock CPUs started to outperform them.
Now you have chip with little OC headroom. And next chip can keep same clock, just improve IPC by 5% and it wins by that margin. Or they bump clock by 5%.
Barry J
I have my i7 8700k @ 4.9ghz all cores on air @ 1.35v for everyday use
reix2x
3.5v on core? your heatspreader most be like a car on size jaja ( i suppose it is 1.5v)
GREGIX
probably 1.35v
my sons 8700k runs 4.7Ghz 1.255v all day long so it is probably that.
MK80
My 8700K runs 4.9Ghz 1.26 super stable...
Darksword
Those single core scores are making me jealous. 🙂
My 7820X @ 4.7Ghz only gets 5786/32375
fabtech
Yes this is indeed a nice single score, but not that great in Multicore. But I really do not like GeekBench, especially when you see some scores from mobile phones that can be as good as an i7 8xxx. I don't really rely on this Benchmark test, I prefer to see real benchmark like Cinebench, Blender, X264, X265, After Effects... they are more real life usage.
My best scores on GeekBench on single core is just 5244, and the Multithread score is just 39210. On Cinebench I get 182cb / 2316cb, Blender Ryzen Logo : 17 sec.
Venix
well multiply your score with 1.08 and you will see the diference is really between 8~9% .....so really will you even notice comparing to what you have ? :P
OrdinaryOregano
"We don't need solder"
Yeah no that's not what they said. They published a paper that said Solder has a chance of cracks and that means it can break within the warranty period and... a company shipping millions of processors has to account for the number of processors it will have to replace, the defect %
Now whether it was actually something that ended up mattering is a different matter but don't misrepresent the argument with populist sentiments.
JOHN30011887
Im defo switching my 8700k for the 9700k
Aslong as it performs same or better than what iv got then im happy, cause the 8700k gets stupid hot in some games and i dont ever want water cooling
oxidized
I LOL'd at those who kept going with the story "8/8 is a downgrade over 6/12" lmao, this is probably pretty accurate, but maybe let's wait for the final results.
Luc
"These aren't the test you are looking for... move along"
Irenicus
1.35 wtf? You must have lost the silicon lottery pretty badly.
My 8600k runs at 4.8 @1.265. Even considering that the 8700k has more cores, from what I've seen 1.35 is quite high. Especially on air.
tunejunky
i have some mixed emotions about this:
i'll buy an I-9 after i said i wouldn't (chipset) just because this is the best example of the process node.
on the other hand, i'm still pissed off over the marketing.
pity the poor I-5 six cores vs 12 - 18 (depending on the final prices).
this weird issue with hyper-threading should have been a non-starter, especially in the current market where a hyper-threading six core is de rigeur at the price point.
if Intel's marketing had gotten it right, the I-5 would be a monster and the enthusiasts would get OC'ing for the I-7 with higher core speeds.
as it is AMD shrugs.
and fewer people buy enthusiast Intel.
Kaarme
As far as hyperthreading is concerned, Intel is a full-blown dissociative identity disorder patient. In these top CPUs Intel treats it like a premium feature, yet for the longest of time it also was a way for Intel to be extremely cheap and stick to dual core processors in the entry level. Even mobile i5s were dual core for many years, relying on HT to grant two extra threads. So, the same thing is both a money saver and a luxury item for Intel. It's impossible to understand what they are thinking.
