-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
*** Official Ryzen Owners Thread ***
- Thread starter Frozennova
- Start date
Hello Guys
Can anyone help
I have a MSI x370 Gaming Plus https://www.msi.com/Motherboard/X370-GAMING-PLUS.html#productFeature-section
I bought corsair CMK16GX4M2B3200C16 16 GB (2 x 8 GB) Vengeance LPX DDR4 (which happens to be v5.39 hynix and single ranked) to go with it
and i'm never hitting my rated speeds dispute 1 official release version and 3 beta version of the bios 1.0.0.6 agesa.
I have tried everything including MSI overlcoking guides.
So i'm giving up and returning the ram.
I'm just about to purchase the Team Group Dark Pro 8pack 3200Mhz 16GB
Does this work well on the MSI x370 Motherboards
K
Can anyone help
I have a MSI x370 Gaming Plus https://www.msi.com/Motherboard/X370-GAMING-PLUS.html#productFeature-section
I bought corsair CMK16GX4M2B3200C16 16 GB (2 x 8 GB) Vengeance LPX DDR4 (which happens to be v5.39 hynix and single ranked) to go with it
and i'm never hitting my rated speeds dispute 1 official release version and 3 beta version of the bios 1.0.0.6 agesa.
I have tried everything including MSI overlcoking guides.
So i'm giving up and returning the ram.
I'm just about to purchase the Team Group Dark Pro 8pack 3200Mhz 16GB
Does this work well on the MSI x370 Motherboards
K
What speed were you hitting, seems like you were doing something else wrong really...
Try dropping the CPU overclock and see how far you can get your memory. There has to be a balance of CPU overclock and memory overclock.
Soooooo finally got round to putting the parts together on my new Ryzen rig.
R5 1600
Gigabyte Gaming 5
16gb Corsair lpx 3200
Just in the process of leak testing then I'll shift my ssd and gpu over and attach rad fans. In the meantime just trying to work out how to reinstall windows 10!
R5 1600
Gigabyte Gaming 5
16gb Corsair lpx 3200
Just in the process of leak testing then I'll shift my ssd and gpu over and attach rad fans. In the meantime just trying to work out how to reinstall windows 10!
Would be even fancier if I could find the fan mounting screws for the rad. Can't for the life of me remember where I put them! Lol
Many months back (early May before AGESA 1006), had written that DR RAM showed more performance on Ryzen CPUs than SR RAM at same speeds, roughly 10% more.
Some called it "anecdotal evidence" and that I was crazy writing it.
Yesterday's AMD blog does a thorough analysis on RAM, timings and optimal settings for better performance using ASUS Crosshair VI and AMD Ryzen 1700 (stock speeds).
https://community.amd.com/community...emory-oc-showdown-frequency-vs-memory-timings
However that's not the whole picture. Manually setting subtimings and the rest of the settings on both SR & DR ram, results to this chart.
Unfortunately they don't have DR ram with manual subtimings.
Also if you see that performance just between 266mhz (3200-3466) that is 10%+ in some cases, the reviewers who still use their March/April charts, with 2333/2400 ram, are way off than just 10%.....
Some called it "anecdotal evidence" and that I was crazy writing it.
Yesterday's AMD blog does a thorough analysis on RAM, timings and optimal settings for better performance using ASUS Crosshair VI and AMD Ryzen 1700 (stock speeds).
https://community.amd.com/community...emory-oc-showdown-frequency-vs-memory-timings
However that's not the whole picture. Manually setting subtimings and the rest of the settings on both SR & DR ram, results to this chart.
Unfortunately they don't have DR ram with manual subtimings.
Also if you see that performance just between 266mhz (3200-3466) that is 10%+ in some cases, the reviewers who still use their March/April charts, with 2333/2400 ram, are way off than just 10%.....
Many months back (early May before AGESA 1006), had written that DR RAM showed more performance on Ryzen CPUs than SR RAM at same speeds, roughly 10% more.
Some called it "anecdotal evidence" and that I was crazy writing it.
Yesterday's AMD blog does a thorough analysis on RAM, timings and optimal settings for better performance using ASUS Crosshair VI and AMD Ryzen 1700 (stock speeds).
https://community.amd.com/community...emory-oc-showdown-frequency-vs-memory-timings
However that's not the whole picture. Manually setting subtimings and the rest of the settings on both SR & DR ram, results to this chart.
Unfortunately they don't have DR ram with manual subtimings.
Also if you see that performance just between 266mhz (3200-3466) that is 10%+ in some cases, the reviewers who still use their March/April charts, with 2333/2400 ram, are way off than just 10%.....
Ryzen 3 is out soon, it will be interesting to see which reviewers redo all their Ryzen results and which ones stick with the old results adding Ryzen 3 results to that and which ones continue to say "oh we only used 2400Mhz Ram on all out Ryzen CPU's because we only used 2400Mhz on our Intel CPU's so that's fair, just ignore the fact that before Ryzen we used 3200Mhz ram on our Intel CPU's"
If they are going to compare overclocked results, and they always do, then they have no excuse not setting the XMP profiles to at least 3200Mhz.
Last edited:
But you can see it right now. Some are testing the SkylakeX with 3200Mhz ram, while the Ryzen CPU numbers are stuck as 3 months ago with 2400/2667 ram
Last edited:
Who are they?
Lets start.
Anandtech.
7900X
http://www.anandtech.com/show/11550...core-i9-7900x-i7-7820x-and-i7-7800x-tested/11
1800X (Ram used running at 2400Mhz on the test)
http://www.anandtech.com/show/11170...review-a-deep-dive-on-1800x-1700x-and-1700/18
Tweaktown is running on the 7900X benchmark, the Ryzen @ 2400 (3000Mhz ram module as with Anandtech), having the 7900X both with 2666 and 2133.
Both have used the same if not worse (Anandtech) numbers on the 7900X benchmark compared to the April review.
Can go through Guru3d etc, but after I come back from the cinema.
I never did like Anand, at least not for a few years now... they have during this time been very long winded and high on detail but very low on substance or anything tangible. a lot of it is just a vast collection of slides for the sake of packing your review full of slides, most on those slides are meaningless.
To sum up Anand in a way that is relevant to this, its actually impossible to know for certain which CPU or platform is doing what with what, all they did in the hardware listing....
For a start Anadnd's reviews with a mixture of hardware performance settings like that are Apples for Oranges, but more importantly they don't say which hardware is paired with which hardware of that inconsistent mixture.
All that for me renders Anands entire review unreliable, untrustworthy and irrelevant.
Anand are another one of those reviewers who used to be good but are these days junk living off the name they once had..
I mean look at that, they are pretending different ram from different vendors running at different speeds are all the same in performance results, of course every noob knows they are not, especially not when it comes to Ryzen.
Following on my previous post I think I have a good chip, or I've missed something. (Entirely possible)
I was rather put off by the technical shennigans people were going through. All I did was increase the multiplier (by steps) with volts at Auto. DOCP 3200 for the RAM.
I had to boost the fan profile a little, but I have 3.9 under a H105 never breaks 75c on test and 62c with handbrake.
Cinebench gives 1712.
Does this sound OK ?
I was rather put off by the technical shennigans people were going through. All I did was increase the multiplier (by steps) with volts at Auto. DOCP 3200 for the RAM.
I had to boost the fan profile a little, but I have 3.9 under a H105 never breaks 75c on test and 62c with handbrake.
Cinebench gives 1712.
Does this sound OK ?
Sounds very OK to me. Now download OCCT and run the Linpac tab with AVX enabled for 8hrs, then you will know for shure your setup is stable.
Don't know what CineBench score is OC'ed but a stock 1800X scores just over 1600, so just over 1700 @ 3.9Ghz sounds about right, its about a 6% performance boost over a stock 1800X.
Temps also look perfectly fine to me.
Its an instruction set coded into the cpu. If you look at the cpu tab on CPUID you will see AVX and AVX2 listed as part of the instruction set.
If you make shure that you tick the AVX box before running OCCT Linpac it will test the cpu much harder than not ticking the box. If OCCT completes an 8hr run with AVX enabled you can be as certain as can be that you have a stable setup.
Its an extension or instruction set on the CPU that some applications use, stress testing with this extension is known to put a lot of stress on the CPU and there in heat.
HandBreak, for example, uses it, you already tested HandBreak and i'm assuming you use it? if you don't have any problems with that i wouldn't worry, what is stable and isn't IMO is a meaningless argument, Synthetic stress testing like OCCT stress the CPU far more than you ever will in everyday use.
If the system is stable for what you use it for it is stable. if what you want is 'piece of mind' a much higher level of stability testing might be something you want to do, but its not something you need.
My i5 is not linepac AVX stable for 8 hours, or even 2, and yet many many hours of Handbreak have passed through it without issue.
If i wanted it "linepac AVX stable for 8 hours, or even 2" i could get it that with more volts, but for what? the extra heat? no thanks.... my system is perfectly balanced for my needs.