• 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.

6700k 4.5Ghz vs 5820K 4.5ghz

lumik

lumik

lumik

Associate
Joined
4 Mar 2016
Posts
62
Both cpus runs at 4.5ghz with 3000Mhz DDR4 CL14.58520K uses 3600Mhz uncore.

enjoy
Fallout 4 1920x1080 all max
6700k
scene1-74.2fps
http://abload.de/img/fallout4_2016_03_02_2e8acq.png
scene2-40.7fps
http://abload.de/img/fallout4_2016_03_02_2iqxrm.png
5820k
scene1-65.1fps
http://postimg.org/image/qeklbpfnl/full/
scene2-35.1fps
http://postimg.org/image/f08q3rfwp/full/

War thunder CPu benchmark 1920x1080 all max
6700K
2x6k6u.jpg

5820k
1wfkc3.jpg


GTA5 benchmark 1280x720 all max no AA only pass4 results
6700K
Min/max/avg
Pass 4, 53.946125, 209.006317, 103.409103
5820k
Pass 4, 42.305767, 210.377731, 105.616760

Crysis3 1024x768 all max SMAA
6700k 106fps
http://abload.de/img/crysis3_2015_09_23_199oome.png
5820k 130fps
http://s10.postimg.org/5jjuhjbvt/crysis3_2016_03_04_11_50_29_360.jpg

EDIT:6700K uses GTX970 1500/8000
5820K uses 980TI 1329/7000
 
Last edited:
look at crysis 3 ? that doesnt use all the cores ;)

you need a varied and more amount of games.

bf4 shreads the i7 6700k for eg.
 
Crysis3 is only game up today that uses 12 cores.But its very old already(2013game)
yeah its only game where 5820k is faster because game uses 12cores.
 
Why the weird resolutions? Test at 4k or 1440p if you're comparing 6700k's vs 5820k's. If you're paying for an i7 'k' CPU for gaming in 2016 you should be running at least a 970 or equivalent at bare minimum and preferably a 980ti or equivalent...

Who plays gtav at 1280x 720 or Crysis at 1024x768 in 2016!

I'll take proper reviews at a minimum res of 1080p showing 1-2 fps difference between the 6700k and 5820k in games where the 6700k wins out with a 200mhz overclock advantage

http://www.kitguru.net/components/cpu/luke-hill/intel-core-i7-6700k-i5-6600k-skylake-cpu-review/8/

Of course where an app can make use of the 12 threads of the 5820k it spanks the faster per core 6700k by a far more appreciable margin...

http://www.kitguru.net/components/cpu/luke-hill/intel-core-i7-6700k-i5-6600k-skylake-cpu-review/5/
 
This Test was made in CPu bottleneck scenes.You see fallout4 and war thunder are CPU bottleneck even at 1920x1080 with max settings.

GTA5 and crysis have lower resolution, but keep max details to eliminate GPU bottleneck to show what cpu is faster.

Btw all reviews on net sucks except few polish and hardware.fr(and digitalfoundry)..They all test in wrong games or wrong setting or in wrong scenes.They are GPU bottleneck.

6700k is faster in todays games.
Crysis3 is bonus game.Its very old already from 2013 but it shows how much 5820k will be faster if game uses 12cores.It have 50% more cores, but its 20% faster.

Btw crysis3 uses all cores to simulate physx on that long GRASS in that only one level.Rest game isnt that demanding.
 
Last edited:
~>Dg<~ said:
look at crysis 3 ? that doesnt use all the cores ;)

you need a varied and more amount of games.

bf4 shreads the i7 6700k for eg.
Click to expand...

Why didn't you do some proper benchmarking and post the results of your before and after, not being funny but things like "BF4 30 fps better minimums" I just can't believe unless I see some evidence, because for all the videos and results I've seen there's barely anything between well overclocked CPU's going as far back as the 2500k.

I want to go x99 myself from a 4.6ghz 3770k, but it isn't for gaming reasons as I'm not going to kid myself it's a worthy upgrade for that alone.
 
proper benchmark i have :p

theHunter i5 3570k

1080 max

2016-02-17 18:12:26 - thehunter
Frames: 8210 - Time: 60000ms - Avg: 136.833 - Min: 109 - Max: 160

5820k

2016-02-20 01:25:06 - thehunter
Frames: 8425 - Time: 60000ms - Avg: 140.417 - Min: 116 - Max: 172

2550 x 1440 max settings

2016-02-17 15:58:37 - thehunter
Frames: 5126 - Time: 60000ms - Avg: 85.433 - Min: 68 - Max: 111

2550 1440 max 5820k

2016-02-20 01:13:16 - thehunter
Frames: 10101 - Time: 60000ms - Avg: 168.350 - Min: 81 - Max: 251


rust fantastic 1080

i5 3570k

2016-02-17 16:28:33 - Dwm
Frames: 5181 - Time: 60000ms - Avg: 86.350 - Min: 60 - Max: 98

i7 5820k

2016-02-28 16:45:14 - Dwm
Frames: 5928 - Time: 60000ms - Avg: 98.800 - Min: 88 - Max: 103

dirt rally 1080

3570k avg 95.19 fps min fps 75.87 max fps 119.51

5820k avg 95.73 min 80.53 max 123 .89


battlefield sorry i didnt save the result . as i wasnt playing it much but the fps has vastly increased. i play it at 1080.

what i would say is the real difference is the smoothness factor.it is a totally different experience.


now the thing is in some games it can be 20-30 fps different. now if you work that out at 1080 that is a huge increase.thats like a top end gpu card jump.not close.newer games will benefit even more then you get the extra multitasking increase. my main is for videoing and editing. so its 2/3 times as quick.
 
Last edited:
Does the hunter use all cores or something? At 1440p I'm surprised the score has doubled as at this resolution and above the onus is more on the GPU, doesn't seem believable at all. The 1080p results are believable.
 
I can't believe that some people are arguing about the best chip?

The 5820K is just as good at gaming as the Skylake chip and obliterates it with the extra cores at everything else.

For roughly the same price you would have to be a clown to pick the Skylake chip.
 
lumik said:
This Test was made in CPu bottleneck scenes.You see fallout4 and war thunder are CPU bottleneck even at 1920x1080 with max settings.

GTA5 and crysis have lower resolution, but keep max details to eliminate GPU bottleneck to show what cpu is faster.

Btw all reviews on net sucks except few polish and hardware.fr(and digitalfoundry)..They all test in wrong games or wrong setting or in wrong scenes.They are GPU bottleneck.

6700k is faster in todays games.
Crysis3 is bonus game.Its very old already from 2013 but it shows how much 5820k will be faster if game uses 12cores.It have 50% more cores, but its 20% faster.

Btw crysis3 uses all cores to simulate physx on that long GRASS in that only one level.Rest game isnt that demanding.
Click to expand...

But no one plays at these resolutions its like claiming a 5820k is superior to a 6700k by contriving an unnatural memory bandwidth scenario where's a 5820k with its quad channel controller beats a 6700k with its dual channel controller by a large margin ...

http://www.pcworld.com/article/2982...e-shocking-truth-about-their-performance.html


Of course no normal usage scenario would replicate this sort of scenario so its academic much like running a game at a very low resolution with a powerful overclocked cpu but with what is now a rather mid range graphics card (looks like a 980?)
 
no one plays at 1920x1080?From when?Fallout 4 runs at 1920x1080 max details.War thunder again 1920x1080 max details.In both games 6700k is faster around 10-15Fps.
GTA5 1280x720-Both have same Fps 6700k have better minimums.if its 1920x1080 both will have same avg fps and it will be GPU bottleneck at around 85Fps.6700k will still win in minimums.
Crysis3-5820 is faster in 1024x768.In 1920x1080 game will be GPU bottleneck around 105fps on both cpus.
 
Last edited:
lumik said:
no one plays at 1920x1080?From when?Fallout 4 runs at 1920x1080 max details.War thunder again 1920x1080 max details.In both games 6700k is faster around 10-15Fps.
GTA5 1280x720-Both have same Fps 6700k have better minimums.if its 1920x1080 both will have same avg fps and it will be GPU bottleneck at around 85Fps.6700k will still win in minimums.
Crysis3-5820 is faster in 1024x768.In 1920x1080 game will be GPU bottleneck around 105fps on both cpus.
Click to expand...

By 'those resolutions' I was specifically talking about 1024x768 and 1280x720. If you gaming at these resolutions with a skylake or haswell-e i7 then you have got your load out seriously wrong 1080p is the absolute minimum res that anyone should be testing at and playing at with that sort of hardware. Moat over lockers with said chips will be at 1080p + I.e 1440p 4k etc
 
lumik said:
Both cpus runs at 4.5ghz with 3000Mhz DDR4 CL14.58520K uses 3600Mhz uncore.

enjoy
Fallout 4 1920x1080 all max
6700k
scene1-74.2fps
http://abload.de/img/fallout4_2016_03_02_2e8acq.png
scene2-40.7fps
http://abload.de/img/fallout4_2016_03_02_2iqxrm.png
5820k
scene1-65.1fps
http://postimg.org/image/qeklbpfnl/full/
scene2-35.1fps
http://postimg.org/image/f08q3rfwp/full/

War thunder CPu benchmark 1920x1080 all max
6700K
2x6k6u.jpg

5820k
1wfkc3.jpg


GTA5 benchmark 1280x720 all max no AA only pass4 results
6700K
Min/max/avg
Pass 4, 53.946125, 209.006317, 103.409103
5820k
Pass 4, 42.305767, 210.377731, 105.616760

Crysis3 1024x768 all max SMAA
6700k 106fps
http://abload.de/img/crysis3_2015_09_23_199oome.png
5820k 130fps
http://s10.postimg.org/5jjuhjbvt/crysis3_2016_03_04_11_50_29_360.jpg
Click to expand...

Yep, similar to my finding when testing a 5820K @ 4.5Ghz to my [email protected]. The 6700k wins every time, minimum FPS are 10-30% higher depending on the game. Makes a big difference in vast majority of games that only use 1-3 cores.

Game developers don't give a **** about optimizing for 6+ core systems until intel release a mainstream 6 core CPU - and even then it willl take years until the average PC has one.

Quad cores will be the sweet spot for gaming for a long time.

Of course for productivity, rending, encoding etc, X99 is best by far.
 
Caracus2k said:
By 'those resolutions' I was specifically talking about 1024x768 and 1280x720. If you gaming at these resolutions with a skylake or haswell-e i7 then you have got your load out seriously wrong 1080p is the absolute minimum res that anyone should be testing at and playing at with that sort of hardware. Moat over lockers with said chips will be at 1080p + I.e 1440p 4k etc
Click to expand...

The point is to test the cpu not the gpu. If you hit the gpu bottleneck before the cpu bottleneck then you will see very little in the way of performance differences beyond better minimums. By lowering the resolution you alleviate the gpu bottleneck and get to see what impact the cpu differences provide in an unlimited gpu power situation.

Most people are gpu limited so cpu improvements will mainly affect minimums, but if we see the big gpu performance gains we are hoping for with pascal and polaris next year, then this information becomes relevant.
 
Dave2150 said:
Yep, similar to my finding when testing a 5820K @ 4.5Ghz to my [email protected]. The 6700k wins every time, minimum FPS are 10-30% higher depending on the game. Makes a big difference in vast majority of games that only use 1-3 cores.

Game developers don't give a **** about optimizing for 6+ core systems until intel release a mainstream 6 core CPU - and even then it willl take years until the average PC has one.

Quad cores will be the sweet spot for gaming for a long time.

Of course for productivity, rending, encoding etc, X99 is best by far.
Click to expand...

Care to post some evidence/reviews etc showing this?... At least at 1080p with high details?
 
Devrij said:
The point is to test the cpu not the gpu. If you hit the gpu bottleneck before the cpu bottleneck then you will see very little in the way of performance differences beyond better minimums. By lowering the resolution you alleviate the gpu bottleneck and get to see what impact the cpu differences provide in an unlimited gpu power situation.

Most people are gpu limited so cpu improvements will mainly affect minimums, but if we see the big gpu performance gains we are hoping for with pascal and polaris next year, then this information becomes relevant.
Click to expand...

Its an artificial test as unnatural as my example of memory bandwidth is.

Post some benchmarks showing performance now at typical resolutions and details settings. Early Polaris and Pascal cards are unlikely to be that much faster than top end 28nm cards are now by the time the new node is well developed Intel will.likely being shipping volume hex core plus cpu's (with cannonlake) and mainstream top end games and apps will favour more cores over marginal clock and ipc benefits
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom