I know this is probably a false positive, but just wanted to put it here

LeshaI know this is probably a false positive, but just wanted to put it here

Yeah looks like a false positive in MemSpeed.exe.

What antivirus is that?

W1zzardYeah looks like a false positive in MemSpeed.exe.

What antivirus is that?

Judging by the style of the page - I'd guess Kaspersky.

DragonsmonkJudging by the style of the page - I'd guess Kaspersky.

W1zzardYeah looks like a false positive in MemSpeed.exe.

What antivirus is that?

Kaspersky.

Btw. I downloaded the Calculator from another website and did not get this warning.

/edit
And now it's fine here too.

Still no Hynix JJR support. The suggested timings are either too agressive or too loose.
geizhals.eu/kingston-hyperx-predator-rgb-dimm-kit-16gb-hx432c16pb3ak2-16-a1871590.html

Target is 3733Mhz. Im using two of these kits with X570 (Aorus Master, BIOS F12a)
Default speed is 3200 16-18-18-36

tRCDWR is suggested as 16. This is way too agressive as the default at 3200 is already 18. There is no way to boot with this value set at 16. Not even at XMP speeds.
tRDRDSD is suggested as 5 when stock is 4
tWRWRSD suggested as 7 when stock is 6

Also tRFC seems incorrect but i have not bothered to find stable value yet. Default is 560. Suggested is 504 but it seems too agressive as it won't boot with 504.

I appreciate the authors hard work. Just pointing out some problems with JJR.

LeshaAnd now it's fine here too.

Looks like they updated their detection, good that it happened so fast

TomorrowStill no Hynix JJR support. The suggested timings are either too agressive or too loose.
geizhals.eu/kingston-hyperx-predator-rgb-dimm-kit-16gb-hx432c16pb3ak2-16-a1871590.html

Target is 3733Mhz. Im using two of these kits with X570 (Aorus Master, BIOS F12a)
Default speed is 3200 16-18-18-36

tRCDWR is suggested as 16. This is way too agressive as the default at 3200 is already 18. There is no way to boot with this value set at 16. Not even at XMP speeds.
tRDRDSD is suggested as 5 when stock is 4
tWRWRSD suggested as 7 when stock is 6

Also tRFC seems incorrect but i have not bothered to find stable value yet. Default is 560. Suggested is 504 but it seems too agressive as it won't boot with 504.

I appreciate the authors hard work. Just pointing out some problems with JJR.

Someone mentioned that presets will be updated in 1.7.1 so I assume they weren't touched since the last release. I've had problems with getting any stable settings above 3466 with calculator before.

i really wish app like this would burn those settings to the bios or xmp profile itself, so you wouldn't have to bother putting everything manually.

hojnikbi really wish app like this would burn those settings to the bios or xmp profile itself, so you wouldn't have to bother putting everything manually.

Thing is, this is not automated process anyway as you still need to adjust some settings manually. Especially when you have an almost infinite amount of possibilities for build combinations.

I'm not sure how hard it can be, but I would like to be able to rename and rearrange the timing, so it would better match the way they are in your bios.

Guys, I own a Ryzen 3600, a MSI B450 Tomahawk Max and 16Gb of DDR 3200 CL16 Ballistix LT RAM.
Is this really worth doing, is it worth the trouble/work or are the gains going to be minimal?

Cheers

BadelhasGuys, I own a Ryzen 3600, a MSI B450 Tomahawk Max and 16Gb of DDR 3200 CL16 Ballistix LT RAM.
Is this really worth doing, is it worth the trouble/work or are the gains going to be minimal?

Cheers

same doubt here.

i have 3600x + 16gb ddr 2400 working nice here.

Cool. I checked the tab for DJR. I'm still a bit tighter than recommended or the frequency I'm at, so I guess that's good. Still can't get it to rated speed, but I'm sure it's my potato CPU holding me back.

Still cant read XMP on my G.Skill Hynix M-die.

BadelhasGuys, I own a Ryzen 3600, a MSI B450 Tomahawk Max and 16Gb of DDR 3200 CL16 Ballistix LT RAM.
Is this really worth doing, is it worth the trouble/work or are the gains going to be minimal?

Cheers

CounterSpellsame doubt here.

i have 3600x + 16gb ddr 2400 working nice here.

So the bump from xmp/loose timings but at 3200 won't yeld that much compared to bumping from 2400 ---> 3-ish clocks --- this due to the memory controller bound to infinity fabric (which is responsible for all the CCX communication). And at 3200 it is already at 1600, where the theoretical max is 1800 (I believe). Afterwards dividers are used.
As I remember correctly -- Hardware Unboxed did a test with 3200 standard and optimized timings --- you might look at their YT channel and even at 3200 it yielded performance increases.

I would still give it a shot... even when running already 3200 stock.
And DEFINITELY give it a go running @2400 memory.

Oh and btw --- for memory stability testing use either the integrated tools OR --- TM5 with the custom profile by 1usmus himself (I use it to torture test the memory stability)->
www.overclock.net/forum/27937684-post4314.html

BadelhasGuys, I own a Ryzen 3600, a MSI B450 Tomahawk Max and 16Gb of DDR 3200 CL16 Ballistix LT RAM.
Is this really worth doing, is it worth the trouble/work or are the gains going to be minimal?

Cheers

Less than minimal. Also, the "peculiarities" of individual motherboards BIOS'es/UEFI implementations much less differences of "doing things" at low level between manufacturers make the numbers spewed out by the app just about pointless. For example: You update the BIOS and all of the sudden the CAS is too tight and you can't boot. You update the BIOS again few weeks later, CAS is now "fine" but AIDA64 is reporting bizzare latency fluctuations. Thats just an example of ONE setting on ONE motherboard. Too much variation on all fronts to account for makes the numbers this app generates somekind of loose suggestive guidline at best, at very best.

Also, shows you how much overclocking has been depriciated and how low we have fallen. In order to compete, AMD has pretty much overclocked their CPUs to the max limit so we are left to fiddle with obscure memory timings like its 2003 (in a bad way) in order to get 0.5% performance in a random benchmark, maybe. (Ryzen owner here, also owner of every CPU from Zilog Z80 and on, before the fanboys get heated)

...
..
.

AddSubLess than minimal. Also, the "peculiarities" of individual motherboards BIOS'es/UEFI implementations much less differences of "doing things" at low level between manufacturers make the numbers spewed out by the app just about pointless. For example: You update the BIOS and all of the sudden the CAS is too tight and you can't boot. You update the BIOS again few weeks later, CAS is now "fine" but AIDA64 is reporting bizzare latency fluctuations. Thats just an example of ONE setting on ONE motherboard. Too much variation on all fronts to account for makes the numbers this app generates somekind of loose suggestive guidline at best, at very best.

Also, shows you how much overclocking has been depriciated and how low we have fallen. In order to compete, AMD has pretty much overclocked their CPUs to the max limit so we are left to fiddle with obscure memory timings like its 2003 (in a bad way) in order to get 0.5% performance in a random benchmark, maybe. (Ryzen owner here, also owner of every CPU from Zilog Z80 and on, before the fanboys get heated)

...
..
.

That's exactly what I thought, thanks for your honest opinion. My last CPU before I this one was a Intel sandy bridge 2500k and that was an overclockable CPU. From day one I overclocked from 3.3 to 4.8ghz. That's a 45% increase! Last great CPU Intel released, in my humble opinion.
Cheers

Used new calculator to get 1866fclk/3733ram stable with no audio issues on a 3600x x570 build. Before I couldn't go past 1800/3600 without audio issues. Yes it is worth it imo.

oobymachUsed new calculator to get 1866fclk/3733ram stable with no audio issues on a 3600x x570 build. Before I couldn't go past 1800/3600 without audio issues. Yes it is worth it imo.

How much time did you spend doing it and how much performance gains did you get?
Cheers

Sweet