8 PACK MEMORY RANGE GROWING: SAY HELLO TO 8 PACK RIPPED EDITION & 32GB KITS!!!

michty_me said:
I'm on 1440p so probably no discernible different running at that.
I'll just try tighten a bit further and see how I go.
When I first got my 5800X, I did just bang it up to 3800/1900 and left the settings the same. It ran for maybe 30 minutes gaming then started to bug out. So I would hope I could make it stable.
Click to expand...

Here's my experience of overclocking new parts.

The 5800X, MSI MEG Unify X570 and 32GB RAM arrived around the same time as my 3080FE.
I stability tested (with benchmarking) the CPU, board and RAM at stock speeds, RAM set to XMP which is 3600MHz, CL16. All good
3080FE was stability tested which was also good.

I then left the board, ram and CPU at stock and overclocked the GPU. It got a few extra points in 3dmark and other tests but then crashed mid Warzone. I tried undervolting too but that was similar, benchmarking was fine but at that crucial moment of a game my whole system reset probably because it couldnt deliver the volts the GPU needed due to being limited. I went back to stock settings but set a fan curve to the fans were not spinning up and down all the time. Temps were nice and low, low noise and games and benchmarks were all completely stable with max temps only 65c when gaming.

Clocking the CPU was next and I tried all manner of options including full manual overclocking, curve optimiser, finding the gold and second best cores in AMD optimizer and many other CPU overclocking options. Again, stable for benchmarking unless going crazy but a CPU intensive game would cause a crash. I decided to let PBO (AMD default overclocking) do its own thing which works perfectly.

The RAM I could have left at XMP but when I tried lowering CAS timings I suddently noticed even the Windows desktop became slightly faster at navigating through folders. Maybe it was in my head but it just felt a bit nippier.
I explored RAM overclocking more and set 3800Mhz up from 3600MHz and bumped voltages. I was going into dedicated RAM stability tests to make sure there were no errors then tried benchmarks. Benchmarks were stable with improved minimum FPS, then onto game testing which were also stable and felt improved over the XMP settings.

On the face of it overclocking RAM would be much harder as there are many many timings, voltages and options to set. There are some quick easy tests though which uncover errors much quicker than CPU and GPU testing without the need to power the machine off or run through a full long test in something like Cinebench or 3dmark. A quick AIDA or Ryzen DRAM Calc would be sufficient, then a longer test with TestMem5 using the extreme anta config.

Lots of useful stuff in this thread for RAM overclocking (not just page 47) https://forums.overclockers.co.uk/t...ack-ripped-edition-32gb-kits.18885405/page-47

Edit - 1440p there may be a good improvement. I'm on 3440 x 1440
 
Guest2 said:
Here's my experience of overclocking new parts.

The 5800X, MSI MEG Unify X570 and 32GB RAM arrived around the same time as my 3080FE.
I stability tested (with benchmarking) the CPU, board and RAM at stock speeds, RAM set to XMP which is 3600MHz, CL16. All good
3080FE was stability tested which was also good.

I then left the board, ram and CPU at stock and overclocked the GPU. It got a few extra points in 3dmark and other tests but then crashed mid Warzone. I tried undervolting too but that was similar, benchmarking was fine but at that crucial moment of a game my whole system reset probably because it couldnt deliver the volts the GPU needed due to being limited. I went back to stock settings but set a fan curve to the fans were not spinning up and down all the time. Temps were nice and low, low noise and games and benchmarks were all completely stable with max temps only 65c when gaming.

Clocking the CPU was next and I tried all manner of options including full manual overclocking, curve optimiser, finding the gold and second best cores in AMD optimizer and many other CPU overclocking options. Again, stable for benchmarking unless going crazy but a CPU intensive game would cause a crash. I decided to let PBO (AMD default overclocking) do its own thing which works perfectly.

The RAM I could have left at XMP but when I tried lowering CAS timings I suddently noticed even the Windows desktop became slightly faster at navigating through folders. Maybe it was in my head but it just felt a bit nippier.
I explored RAM overclocking more and set 3800Mhz up from 3600MHz and bumped voltages. I was going into dedicated RAM stability tests to make sure there were no errors then tried benchmarks. Benchmarks were stable with improved minimum FPS, then onto game testing which were also stable and felt improved over the XMP settings.

On the face of it overclocking RAM would be much harder as there are many many timings, voltages and options to set. There are some quick easy tests though which uncover errors much quicker than CPU and GPU testing without the need to power the machine off or run through a full long test in something like Cinebench or 3dmark. A quick AIDA or Ryzen DRAM Calc would be sufficient, then a longer test with TestMem5 using the extreme anta config.

Lots of useful stuff in this thread for RAM overclocking (not just page 47) https://forums.overclockers.co.uk/t...ack-ripped-edition-32gb-kits.18885405/page-47

Edit - 1440p there may be a good improvement. I'm on 3440 x 1440
Click to expand...

I did spend quite a bit of time fine tuning my current ram settings but that was done using a 3600. I've since swapped to a 5800X which the settings run perfectly fine on.
Maybe its time for just a fresh setup and see where I can get to with it.
 
michty_me said:
I did spend quite a bit of time fine tuning my current ram settings but that was done using a 3600. I've since swapped to a 5800X which the settings run perfectly fine on.
Maybe its time for just a fresh setup and see where I can get to with it.
Click to expand...
Did you change anything / fresh setup?

Has anyone tried running the 8 pack 3600CL16 at 4000/FCLK2000? I’m tempted to update BIOS to AGESA 1.2.0.5 and try 2000IF
 
Guest2 said:
Did you change anything / fresh setup?

Has anyone tried running the 8 pack 3600CL16 at 4000/FCLK2000? I’m tempted to update BIOS to AGESA 1.2.0.5 and try 2000IF
Click to expand...

I haven't mate. I'm still at work and not due home until the 6th so may start looking at tightening them after that.
 
Minstadave said:
Just noticed Corsair have a 3600mhz 14-16-16-36 kit. I assume that must be B Die to be that tight at 3600mhz?

CMT32GX4M2Z3600C14
Click to expand...
According to the bdie finder web page they are B-die. May take a look into those actually. Thanks!!

Looks like I can get that RAM for £170 with the vouchers from my work. May order that tonight.
 
Last edited:
Minstadave said:
I don't think so, they did some really fast 8GB modules but nothing beyond 3600mhz C16 in 16GB modules. The Corsair one I've mentioned above is in stock and OK priced.
Click to expand...
Thanks. The Corsair ones sound quite good for the price, not that I would change from my current 8 pack kit
 
michty_me said:
In gaming performance at 1440P, Would there be any benefit for changing? The more I read up on it, The general consensus seems to be a couple of FPS at best.
Click to expand...

You may see a slightly better average usually achieved from bringing up the bottom end which can make a difference to fluidity but unlikely to worth the cost to change out RAM. It's often game specific as well so you may not see anything at all.
 
Joe Jones said:
You may see a slightly better average usually achieved from bringing up the bottom end which can make a difference to fluidity but unlikely to worth the cost to change out RAM. It's often game specific as well so you may not see anything at all.
Click to expand...

That's what I figured really and echos others thoughts else where. Maybe not worth the expense or time to manually tune them.
 
michty_me said:
That's what I figured really and echos others thoughts else where. Maybe not worth the expense or time to manually tune them.
Click to expand...
Personally its a hobby upgrading and tweaking, if its £100 imo worth it plus you get to have ‘fun’ tweaking voltages and timings :D
I found it improved minimum fps in benchmarks and gaming. My minimums were all over before i upgraded RAM and CPU. Improved minimum fps is a winner, even if its a few fps.
What else is there to upgrade, GPU. Good luck getting a 40XX series card!

Also, I know Heaven 4 is ancient now but my minimums around 60fps.
People have said it’s due to Heaven bugs but I i’ll be hitting 60fps minimums most of the time.
Other peoples minimums range from 8-55fps minimum. Below 60 isnt really acceptible in such an old benchmark. Single figures is shocking and would indicate system instability to me, not a ‘bug’
 
Last edited:
Yeah I get what you are saying. I think with selling my old kit it would be around a £70 upgrade.
If I were guaranteed a 10-20fps increase then they would have been fitted already.
As much as I love tinkering about with the PC and using it daily, I'm in a bit of a battle with myself on if its worth it or not, Currently for me I'm saying no. My £70 upgrade cash may be better going towards further audio upgrades or a DAC for my new amp.
 
michty_me said:
Yeah I get what you are saying. I think with selling my old kit it would be around a £70 upgrade.
If I were guaranteed a 10-20fps increase then they would have been fitted already.
As much as I love tinkering about with the PC and using it daily, I'm in a bit of a battle with myself on if its worth it or not, Currently for me I'm saying no. My £70 upgrade cash may be better going towards further audio upgrades or a DAC for my new amp.
Click to expand...
What other specs do you have?

But yes, probably better spent on something like this - https://uk.creative.com/p/b-stock/sound-blasterx-g6-b-stock
Out of interest, what 1080p minimums are you seeing on Heaven 4?

I've done another few runs and getting around 65fps minimums

hI4jcjJ.png
 
Guest2 said:
What other specs do you have?

But yes, probably better spent on something like this - https://uk.creative.com/p/b-stock/sound-blasterx-g6-b-stock
Out of interest, what 1080p minimums are you seeing on Heaven 4?

I've done another few runs and getting around 65fps minimums

hI4jcjJ.png
Click to expand...

I have a 5800X with a 3080FE.
Audio setup is a G6 (Used as a DAC feeding a Schiit Asgard 3 amplifier)
I'm not sure, I'll need to carry out a benchmark test when I get home. I'm due home this week so I'll download and run that as a comparison.
It must be a similar setup to you then just different RAM.
 
You must log in or register to reply here.
Back
Top Bottom