Overclock.net banner

3770k vs 6700k In Fallout 4, Arma 3, Rainbow Six Siege. With core\cache\memory scaling ddr3-1600c11-2133c9 ddr4-2133c15-3000c12-4000c17

16K views 36 replies 14 participants last post by  white owl 
#1 ·
Test System
3770k -- 6700k
Asus Maximus V Gene Z77-- Asus Maximus VIII Gene Z170
Samsung DDR3-1600c11 2x4GB 1.35v dual sided -- G.Skill DDR4-3200c14 2x8GB single sided
EVGA GTX1070 SC (OC=+75 core +750 mem +112 powertune use on all tests)
64g Samsung 830, 1tb Sandisk ultra II, 3tb WD green
Seasonic XP 1050 < tad overkill for the current system which has topped out at 306w from the wall
Silverstone TJ08-EW
Swiftech Apogee drive II
200mm Phobya rad + Silverstone Air Penetrator AP181
EK VGA Supreme HF
Win10


Average fps in fallout 4 are not perfectly accurate im afraid due to it being a short run with low fps at the start and high fps at the end if i was a second out starting\stopping the test it could throw the results a bit but the minimum seem very consistent


For arma 3 i used a stratis time demo average fps are nice and consistent but the minimum were very random so i left them out


R6 your typical gpu bottlnecked game most ram reviews use here you can see the 3770k at stock with the gtx1070 oc is faster than the 6700k oc with stock gtx1070
R6 had the gpu pegged at 100% load while arma stayed well under being cpu bottlnecked


lightroom also sees over 20% improvement from higher ram speeds when exporting but only ~10% when rendering previews

time taken to export 150 24mp .nef to jpeg
2133c15 3:20
3200c14 2:50
3866c16 2:39

time taken to build 150 24mp 1:1 previews
2133c15 6:03
3200c14 5:33
3866c16 5:23

XTU, AIDA64, Timings

DDR3
3770k@4.5GHz 2133c9 1.5v


DDR4
6700k@stock SPD-2133c15 Single channel

6700k@stock SPD-2133c15

6700k@4.7GHz core 4.2GHz cache SPD-2133c15

6700k@4.7GHz core 4.7GHz cache SPD-2133c15

6700k@stock 3866c16 t1 1.4v

6700k@4.7GHz core 4.7GHz cache XMP-3200c14

6700k@4.7GHz core 4.7GHz cache 3000c12 t1 1.42v (unstable)

6700k@4.7GHz core 4.7GHz cache 3866c16 t1 1.4v

6700k@4.7GHz core 4.7GHz cache 4000c17 t2 1.4v
 
See less See more
14
#4 ·
I thought all this time my 4.5ghz 2500k would last me another 2 years till I started doing more and more research into sky lake and the 6700k...now I really believe that the upgrade was worth it. Incredible performance and I'm Killin it in even single core performance over a 2500k @ 4.5ghz...
 
#5 ·
saved a bit of other data with afterburner when runing these tests not sure if any of it is of interest
for example this is a run through fallout 4 on the 3770k
 
#6 ·
I just finished a new Mini-ITX form factor build in one of the smallest, coolest looking cases I've ever had... the aluminum, Lian-Li made Ncase M1 v5.

Components:
Asus Z170i Pro Gaming MB
i7-6700k
Noctua C14 HSF
nVidia 1080-FE
Corsair Vengeance LPX DDR4-3200 (2x8GB)... problem...
1TB SSD
Corsair SF600 PS
2 Vardar 120mm fans exhausting out on side panel
M-Audio AV40 (2.0, powered) speakers
Seenheiser PC360 headset

I can only get one stick of the above memory to run, so I'm running single-channel now, unfortunately. I also have 2 sticks of Corsair -3000Hz I will try. I'm thinking I have to raise the voltage just a bit. If that's not it, then that memory slot could be bad or that memory stick. It's amazing how fast this rig is in Battlefield 1, even with single-channel memory.

It was a difficult build for me... extremely tight quarters for everything to fit.

Any thoughts on the memory situation?
 
#7 ·
so according to Test , best Spot is DDR4 3000 with Cl 12 but unstable.So DDR4 3400 CL 14?
 
#8 ·
Thanks for taking the time to do this.
I was /am on the fence upgrading my 4770k which is stuck at 4.2 stable to a 7700k.
My MSI M Power Max Z87 board is starting to get a little flaky, but it is still usable.
Most reviews were only showing a ~2-10 FPS gain. I had a feeling with high clocks and some fast DDR4 the gain could be much more.
Zen, Vega, 1080Ti are so close though I may as well wait a few more months. I got money burning a hole in my pocket right now though!
 
#9 ·
Quote:
Originally Posted by Xuper View Post

so according to Test , best Spot is DDR4 3000 with Cl 12 but unstable.So DDR4 3400 CL 14?
whats the best spot to you?
for me its whatever gives the highest stable speed in the end and after extended testing it was 3866 16-16-16 2t 1.425v
the sweat spot is likely to change from one ram kit to the next as some may handle higher frequencies better than lower latency and vise versa
 
#11 ·
#13 ·
#15 ·
Thank you very informal. Proves that higher clocked RAM now has as much of a significance as more powerful GPUs & CPU's now:

All 6700k @ 4.7Ghz overclock:

Fallout 4:
4000 Mhz - 40.2 min 68.0 avg
3200 Mhz - 37.5 min 64.2 avg
2133 Mhz - 30.8 min 51.1 avg

Arma 3:
4000 Mhz - 81.8 avg
3200 Mhz - 76.4 avg
2133 Mhz - 64.6 avg

This is also a benefit with the Z270 platform over the X99 platform. On Z270 you can get DDR4 4266mhz easy XMP profile paired with a 5.2 Ghz 7600/7700k. When I had an X99 platform I couldn't really go above DDR4 2400 Mhz XMP profile on a 5960x @ 4.7Ghz without getting into some serious manual custom memory timings. Many people were able to get up to around 3000Mhz but that's about it.

Just further proves my experience that higher clocked CPUs & Memory will always trump more cores @ slower CPU & memory clocks in 99% of all gaming scenarios. I am batting for the Z270 team right now versus X99 but I think I'll have to go the new LGA 2066/x299 platform once it's released later this year. It's been proven that once you push the 4k realm (triple 4k or the new 144mhz 4k monitors), the extra PCIE lanes and the extra CPU cores really does make a big FPS difference to push that much bandwidth.
 
#16 ·
since i was a bit bored today and have never seen it tested i decided to check what impact ram speed has on lightroom
6700k@4.7ghz with the ram at spd\xmp\max stable oc

time taken to export 150 .nef to jpeg
2133c15 3:20
3200c14 2:50
3866c16 2:39

time taken to build 150 1:1 previews
2133c15 6:03
3200c14 5:33
3866c16 5:23
 
#17 ·
Looking at this what is good speed to get for DDR4 without having to spend a lot more money?
 
#18 ·
anything around 3000mhz
i spent a little extra for a 3200c14 kit that overclocks well
you also wanted a good mb to go much past 3200mhz with skylake but it seems easier to hit higher speeds with karby
 
#19 ·
Quote:
Originally Posted by Dasa View Post

anything around 3000mhz
i spent a little extra for a 3200c14 kit that overclocks well
you also wanted a good mb to go much past 3200mhz with skylake but it seems easier to hit higher speeds with karby
Looking at the prices in Canada and it looks like I can get 32GB 3200 CL16 ~ $250. Not sure if I should stay with 16GB and save money. Could just get cheap 2400 kit for $85 for 16GB.
 
#20 ·
wether or not its worth the extra cash depends a bit on the balance with the rest of the system and its uses

G.SKILL Ripjaws V Series 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel Z170 Platform Desktop Memory Model F4-3200C16D-16GVGB $130
https://www.newegg.ca/Product/Product.aspx?Item=N82E16820232181
 
#21 ·
Quote:
Originally Posted by Dasa View Post

wether or not its worth the extra cash depends a bit on the balance with the rest of the system and its uses

G.SKILL Ripjaws V Series 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel Z170 Platform Desktop Memory Model F4-3200C16D-16GVGB $130
https://www.newegg.ca/Product/Product.aspx?Item=N82E16820232181
Yeah, those are the ones I had in mind. Probably best to wait since I will be going Zen and XMP probably does not work well there.
 
#25 ·
looking better with those tweaked sub timings
see if you can get REF Cycle Time below 400 lower the better but you will hit a point where it becomes unstable fast and could corrupt your os so its best to test with a linux boot and GSAT before going back to your main os
you can also try increase Dram Refresh interval to ~25,000 if you like it just helps the mbps in aida64 a bit but can create instability during standby loads if set to high

the others are obviously not to be ignored going by how much you have gained already

tertiary timings i ended up leaving at auto as i found minimal gain from tinkering and some minor instability but if you find something that makes a worthwhile difference in there let me know
smile.gif


there is some good info here if you spend enough time digging
http://www.overclock.net/t/1569364/official-ddr4-z170-z270-and-x99-24-7-memory-stability-thread
 
#26 ·


Played arma 3 for 4 hours +. My CPU voltage automatically set to 1.5v and 4.8Ghz multiplier. Needless to say The multiplier added +2x to clock speed. Couldn't figure out why my temps were uncontrollable with 20c jumps and max temp of 100c. Max voltage was 1.45v luckily. Had a rx480 @ 1500MHz playing 4k ultra no AA @45 to 60fps.Luckily 6700k didn't fry. I couldn't get my ram to boot with 4400Mhz OC profile.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top