- Joined
- May 23, 2014
- Location
- Miami, FL.
Thanks Roger I got, will post soon
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
FEATURED Summer CPU BBQ
- Thread starter Johan45
- Start date
- Joined
- Aug 14, 2006
MetalRacer/ Intel / 8 / Pifast / 16.21s
MetalRacer/ Intel / 4 / SPi 32m / 7m 37.596s = 457.596
MetalRacer/ Intel / 8 / WPrime 1024 / 89.391s x 8 = 715.128
MetalRacer / Intel / 8 / GPUpi (CPU) 1B / 3m 25.210s x 8 = 1641.68
Total 2830.614
MetalRacer/ Intel / 4 / SPi 32m / 7m 37.596s = 457.596
MetalRacer/ Intel / 8 / WPrime 1024 / 89.391s x 8 = 715.128
MetalRacer / Intel / 8 / GPUpi (CPU) 1B / 3m 25.210s x 8 = 1641.68
Total 2830.614
Last edited:
- Joined
- Jun 28, 2012
ATMINSIDE / Intel / 4C8T / PiFast / 16.22s
ATMINSIDE / Intel / 4C8T / SuperPi 32M / 0h 7m 38.657s = 458.657s
ATMINSIDE / Intel / 4C8T / wPrime 1024M / 180.495s x4 = 721.98s
ATMINSIDE / Intel / 4C8T / GPUPi 1B / 0h 6m 57.042s x4 = 1668.168s
ATMINSIDE Total = 2865.025
ATMINSIDE / Intel / 4C8T / SuperPi 32M / 0h 7m 38.657s = 458.657s
ATMINSIDE / Intel / 4C8T / wPrime 1024M / 180.495s x4 = 721.98s
ATMINSIDE / Intel / 4C8T / GPUPi 1B / 0h 6m 57.042s x4 = 1668.168s
ATMINSIDE Total = 2865.025
- Joined
- Jan 14, 2006
None of the pictures work for me...
Also, they are all PNG files. Save them as JPG and you (all) may not run into the pictures being too big.
OK aLL FIXED sry
- Joined
- Aug 14, 2006
There is something going on with the last slot on the left ( A1). CPUZ shows 16GB and all four slots as being populated but running triple channel, XTU shows 12GB with three channels.
I switched to a different BIOS and now CPUZ shows Quad channel with 16GB but XTU and Mem Tweak show 12GB running three channels.
- Joined
- Aug 14, 2006
Still getting triple channel even with a fresh BIOS flash and all default settings. Maybe the SS head is too tight and its dropping a channel like the socket 1366 used to do.
- Joined
- Jun 28, 2012
Why not? It doesn't matter how many cores you have.
- Joined
- Jan 12, 2015
Just a quick message, to let you guys know that I will be down (offline) for awhile. I just got some new parts for my gaming rig and it's also that time to clean it
I used to be able to do this stuff in a single day but now it can take me up to a week to complete the task
This is also MY only link to the internet but I did finish my benching rig Not sure if you can see in the picture but it's not linked so I don't have a AV or FW in the background taking up resources
Thank You For Your Time
In Reading This Post
I used to be able to do this stuff in a single day but now it can take me up to a week to complete the task
This is also MY only link to the internet but I did finish my benching rig
Not sure if you can see in the picture but it's not linked so I don't have a AV or FW in the background taking up resources
Thank You For Your Time
In Reading This Post
- Joined
- Aug 14, 2006
Pulled all of the ram out and tried booting with one stick at a time in slot A1 and only one out of four worked. So it's back to running quad channel like it should.
I did some research and apparently if a channel fails training on boot up that channel is intentionally dropped.
You can force additional training which helps ( simply enable everything with dram training ) but can cause some issues at higher clocks ( 3300+ ).
I think it's a board's fault as the same was happening on every DDR4 I had just on some IC it's at higher clock. At least on current IC optimal performance is at ~3000 and tight timings.
On the forums are many users who are showing how high they set their DDR4 but many of these results are in 2-3 channels instead of 4.
I think it's a board's fault as the same was happening on every DDR4 I had just on some IC it's at higher clock. At least on current IC optimal performance is at ~3000 and tight timings.
On the forums are many users who are showing how high they set their DDR4 but many of these results are in 2-3 channels instead of 4.
- Joined
- Aug 16, 2009
OK so I am completely new to fiddling with ram so bear with me.
Picked up some new corsair dominator platinum 2133 cl 8 10 10 27 kit and plugged them in last night. Got them running 8 9 9 25 1T 1 hour AIDA and 7 9 9 9 27 kinda stable. Now I didn't expect a huge difference but found no difference in scores from my previous and gamer sticks @ 10 11 11 30. Am I missing something thought there would be at least a little better showing. CPU/NB was 2600 ish.
Any thoughts!
Picked up some new corsair dominator platinum 2133 cl 8 10 10 27 kit and plugged them in last night. Got them running 8 9 9 25 1T 1 hour AIDA and 7 9 9 9 27 kinda stable. Now I didn't expect a huge difference but found no difference in scores from my previous and gamer sticks @ 10 11 11 30. Am I missing something thought there would be at least a little better showing. CPU/NB was 2600 ish.
Any thoughts!
OP
- Joined
- Dec 19, 2012
- Thread Starter
- #158
Low main timing performance can be offset by less than optimal subtimings. So if all other timings are left on auto with "ram training" the board will loosen the subtiming to allow the main timings to run at lower latency. That's why 10-11-11 and 7-12-7 can produce the same results at 2400. This was something I noticed when setting some of the profiles on the board for ram. The PSC profile set the tighter main timings but really loosened up everything else.
- Joined
- Aug 16, 2009
I was working on my AMD system last night...... I know my Intel board has ram training but my Crosshair? I did try the kit at 2400mhz but it wouldn't boot. But again I only had a bit of time and really wanted to see if the Corsair ram would play nice with my FX rig, which it did. I need to swap around some drives and re instal windows and make a benching drive before I get serious on this comp though.
@Blaylock...... kinda stable........ I only had about 2 hours free time last night so what I meant by that I never had a chance to test ram stability. I think that was my last run I did.
@Blaylock...... kinda stable........ I only had about 2 hours free time last night so what I meant by that I never had a chance to test ram stability. I think that was my last run I did.
OP
- Joined
- Dec 19, 2012
- Thread Starter
- #160
Try using Aida64 to test ram/NB performance bassnut. Sometimes just adding .02 v to the ram can make all the difference. It's a relay race inside those sticks and if the timing /frequency/volts aren't in sync then someone drops the ball and corrections are made which hurts performance.
Similar threads
![(G{in}[AK)TION]](/forums/data/avatars/s/116/116613.jpg?1469679448){kind=avatar}