F5F Stay Refreshed Power Users Overclocking Need help OC'ing RAM

Need help OC'ing RAM

Need help OC'ing RAM

Pages (2): Previous 1 2
V
Vadux
Junior Member
8
11-08-2016, 03:49 PM
#11
cough all you want anort... that fallout bench is slanted by the fact they used effectively lower latencies as the ram got faster. if they had of use 1600 cas 8 against 2400 cas 12 there would be no difference in the fps... but they didn't... as the ram gets faster they effectively reduce the absolute latency when they drop the cas to 11. and that's where the fps increase came from.
V
Vadux
11-08-2016, 03:49 PM #11

cough all you want anort... that fallout bench is slanted by the fact they used effectively lower latencies as the ram got faster. if they had of use 1600 cas 8 against 2400 cas 12 there would be no difference in the fps... but they didn't... as the ram gets faster they effectively reduce the absolute latency when they drop the cas to 11. and that's where the fps increase came from.

T
Taddeljoy
Member
149
11-08-2016, 10:03 PM
#12
The overclocker comparison shows a rating of cl8 1600 versus 2400cl11
T
Taddeljoy
11-08-2016, 10:03 PM #12

The overclocker comparison shows a rating of cl8 1600 versus 2400cl11

L
LVK2003
Junior Member
2
11-09-2016, 05:58 AM
#13
yes 2400 is 1/3 faster than 1600
if the test was legitimate they would have applied cas 12 which is actually slower than cas 8, resulting in similar performance.
the improvement wasn't due to faster RAM speeds but rather from lower latency settings that shortened absolute memory access times.
as I mentioned before, it seems like a skewed test.
https://en.wikipedia.org/wiki/CAS_latency
you can check the CAS latency listings for cas to ramspeed and see ns timings.
it looks like you might need to study more or review the details.
also, reading through Reddit will confirm what I said.
L
LVK2003
11-09-2016, 05:58 AM #13

yes 2400 is 1/3 faster than 1600
if the test was legitimate they would have applied cas 12 which is actually slower than cas 8, resulting in similar performance.
the improvement wasn't due to faster RAM speeds but rather from lower latency settings that shortened absolute memory access times.
as I mentioned before, it seems like a skewed test.
https://en.wikipedia.org/wiki/CAS_latency
you can check the CAS latency listings for cas to ramspeed and see ns timings.
it looks like you might need to study more or review the details.
also, reading through Reddit will confirm what I said.

T
tamemarco
Senior Member
482
11-10-2016, 08:21 AM
#14
Try increasing it to cl 12. Do you have a 1 or a 2? Go to 2 if it's 1.
I believed cl 11 was quicker than 12? I'm currently at 11. Or am I misunderstanding?
T
tamemarco
11-10-2016, 08:21 AM #14

Try increasing it to cl 12. Do you have a 1 or a 2? Go to 2 if it's 1.
I believed cl 11 was quicker than 12? I'm currently at 11. Or am I misunderstanding?

Y
yClaaa_
Junior Member
7
11-13-2016, 10:07 AM
#15
11 performs better. 12 has a higher chance of success. The command rate should be adjusted to 2 changes instead of 1.
Y
yClaaa_
11-13-2016, 10:07 AM #15

11 performs better. 12 has a higher chance of success. The command rate should be adjusted to 2 changes instead of 1.

_
_LilacSoul
Member
183
11-14-2016, 07:28 AM
#16
Supahos :
11 is quicker. 12 works better. Same with command rate if it's at one change, use two. It actually lasted longer, though it still crashed (the game). In my bios settings I have to pick between cl 12 performance or compatibility, did I just pick perf? Does that matter?
_
_LilacSoul
11-14-2016, 07:28 AM #16

Supahos :
11 is quicker. 12 works better. Same with command rate if it's at one change, use two. It actually lasted longer, though it still crashed (the game). In my bios settings I have to pick between cl 12 performance or compatibility, did I just pick perf? Does that matter?

I
ItzDogeFTW_YT
Member
170
11-14-2016, 09:25 AM
#17
Not sure on that option. Try compatibility and see what happens. I suspect nothing different but worth a shot I suppose
I
ItzDogeFTW_YT
11-14-2016, 09:25 AM #17

Not sure on that option. Try compatibility and see what happens. I suspect nothing different but worth a shot I suppose

Pages (2): Previous 1 2