Do I Lose Perf Going from CL16 to CL18 RAM? Intel 10850K

E

edo101

Limp Gawd
Joined
Jul 16, 2018
Messages
510
As an Amazon Associate, HardForum may earn from qualifying purchases.
It's a little slower but with ram not running out is more important than -2CL. DDR4-3600 CL16 is really common though so you'll just have to decide if the price on this kit is worth a small speed hit sometimes compared to what it would cost to get ram that's as fast as what you have for faster. That's about $3 for the same speed, since you can get 2x16GB G.Skill DDR4-3600 CL16 from Newegg for $63. No RGB on that though. Then there are also higher clocked 2x32GB kits for around that price.
 
zandor said:
It's a little slower but with ram not running out is more important than -2CL. DDR4-3600 CL16 is really common though so you'll just have to decide if the price on this kit is worth a small speed hit sometimes compared to what it would cost to get ram that's as fast as what you have for faster. That's about $3 for the same speed, since you can get 2x16GB G.Skill DDR4-3600 CL16 from Newegg for $63. No RGB on that though. Then there are also higher clocked 2x32GB kits for around that price.
Click to expand...
If I got that, I'd be mixing and matching DIMMs though. WOuld have 4 dimms but 2 are from different make
 
Intel systems aren't super sensitive to memory latency increases. The difference may show up in some benchmarks but its not the type of performance difference you'd notice actually using the system.
 
pendragon1 said:
^^^that. you would have to go looking for it.
Click to expand...
Dan_D said:
Intel systems aren't super sensitive to memory latency increases. The difference may show up in some benchmarks but its not the type of performance difference you'd notice actually using the system.
Click to expand...
Ok so i won't notice fps drops in my games then? Ad if there are, it would be a couple of frames?
 
Upgrade... if anything, you'll see more of a overall system performance increase going to 32gb over the 16gb kit, if even it's a little higher in latency. I've never had any issues with that brand either.
 
Odigo said:
Upgrade... if anything, you'll see more of a overall system performance increase going to 32gb over the 16gb kit, if even it's a little higher in latency. I've never had any issues with that brand either.
Click to expand...
pendragon1 said:
doubt it. if that.
Click to expand...
Dan_D said:
Intel systems aren't super sensitive to memory latency increases. The difference may show up in some benchmarks but its not the type of performance difference you'd notice actually using the system.
Click to expand...
Why is this Cosair: https://www.newegg.com/corsair-32gb-ddr4-3600/p/N82E16820236607?Item=N82E16820236607

More expensive than the rest: https://www.newegg.com/corsair-32gb-ddr4-3600/p/N82E16820236607?Item=N82E16820236607

Going off newegg BF sale:
https://www.newegg.com/Components/EventSaleStore/ID-45?N=100006650 600561665 100007611
 
zandor said:
It's a little slower but with ram not running out is more important than -2CL. DDR4-3600 CL16 is really common though so you'll just have to decide if the price on this kit is worth a small speed hit sometimes compared to what it would cost to get ram that's as fast as what you have for faster. That's about $3 for the same speed, since you can get 2x16GB G.Skill DDR4-3600 CL16 from Newegg for $63. No RGB on that though. Then there are also higher clocked 2x32GB kits for around that price.
Click to expand...
Why is this Cosair: https://www.newegg.com/corsair-32gb-ddr4-3600/p/N82E16820236607?Item=N82E16820236607

More expensive than the rest: https://www.newegg.com/corsair-32gb-ddr4-3600/p/N82E16820236607?Item=N82E16820236607

Going off newegg BF sale:
https://www.newegg.com/Components/EventSaleStore/ID-45?N=100006650 600561665 100007611
 
yeah the top two are the same thing. if you are comparing them to the others on the third link, they cost more before of theyre corsair, have rgb bling and are 3600 c18. use the promo code to bring them down to $63.
 
edo101 said:
Hi this is my current RAM: https://www.newegg.com/ballistix-16gb-288-pin-ddr4-sdram/p/N82E16820164205?Item=N82E16820164205

However it is 16GB and I find i often have to close my Chrome to play games to be safe. I use a lot of tabs. I am on a Z490 tachi with my 10850k at 5ghz. I was thinking of grabbing this kit:

https://a.co/d/06lY1HF

For 32GB but it's at CL18 not CL16. Does this matter for games and video editing?

ALso is that brand reliable?
Click to expand...
If you have a bunch of stuff open in the background it won't make a real difference. You have to keep in mind that any benefits shown from 2 cas lower or tuned memory timings in benchmarks are being done in an isolated scenario without other things running in the background.

edo101 said:
If I got that, I'd be mixing and matching DIMMs though. WOuld have 4 dimms but 2 are from different make
Click to expand...
You don't want to mix memory kits, you'll have to run them at JEDEC stock speeds (probably 2133) if you do which WILL kill your performance.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
D-EJ915 said:
If you have a bunch of stuff open in the background it won't make a real difference. You have to keep in mind that any benefits shown from 2 cas lower or tuned memory timings in benchmarks are being done in an isolated scenario without other things running in the background.


You don't want to mix memory kits, you'll have to run them at JEDEC stock speeds (probably 2133) if you do which WILL kill your performance.
Click to expand...
I actually just mixed and matched the Gskill I got with my Crucial and all that got slowed down were the timings for the Crucial. The Gskill actually sligjtly improved. I am now at 48GB @3600mhz at 18-22-22-38
 
edo101 said:
I actually just mixed and matched the Gskill I got with my Crucial and all that got slowed down were the timings for the Crucial. The Gskill actually sligjtly improved. I am now at 48GB @3600mhz at 18-22-22-38
Click to expand...

That's actually what happens. You are going to have to run at the speed of the slower memory assuming your memory controller can handle all 4 sticks, etc.
 
kirbyrj said:
That's actually what happens. You are going to have to run at the speed of the slower memory assuming your memory controller can handle all 4 sticks, etc.
Click to expand...
And just to trriple check, I am not losing much going from 16-18-18-38 to 18-22-22-38 ?
 
From loose to looser is not going to make that much of a difference. They might make it up in the subs.
 
edo101 said:
And just to trriple check, I am not losing much going from 16-18-18-38 to 18-22-22-38 ?
Click to expand...
edo101 said:
Loud and clear. I don't really know what these timings mean but I went and changed two of them and this is what I got now:
View attachment 695755

Does this look right to you? I take it, it matches the clocks fo the gskill? And how much perf do you think I am losing?
pendragon1

Also how do I test it?
Click to expand...
pendragon1 said:
however you want
the main timings you add +4 to each ie: 18+4=22...
leave the sub timings auto.
clear?
Click to expand...

dude... add 4 = 20-22-22-42
youd probably have to look for it.
one thread for the same issue would help too....
 
pendragon1 said:
dude... add 4 = 20-22-22-42
youd probably have to look for it.
one thread for the same issue would help too....
Click to expand...
This one was different. This one was about buying the new ram. The other thread was about mixing ram.

Anyways, How can I test to ensure stability? I'm paranoid lol
 
edo101 said:
This one was different. This one was about buying the new ram. The other thread was about mixing ram.

Anyways, How can I test to ensure stability? I'm paranoid lol
Click to expand...
oh.
memtest, occt, just use it...
 
I've been using testmem5 recently with the ABSOLUT @anta777 config. I found it on Reddit and it seems to work pretty well. Catches a few things that memtest doesn't. Put it this way, if it passes that then I've never had a problem with the memory causing issues on my computer.
 
You must log in or register to reply here.
Back
Top