AMD Ryzen 5000 Builders Thread

[B-Riz]

	Cores Threads	Base Freq	Turbo Freq	L3 Cache	TDP	MSRP
Ryzen 9 5950X	16c/32t	3400	4900	64 MB	105 W	$799
Ryzen 9 5900X	12c/24t	3700	4800	64 MB	105 W	$549
Ryzen 7 5800X	8c/16t	3800	4700	32 MB	105 W	$449
Ryzen 5 5600X	6c/12t	3700	4600	32 MB	65 W	$299

5600X comes with cooler, the rest do not.

above from https://www.anandtech.com/show/1614...en-3-on-nov-5th-19-ipc-claims-best-gaming-cpu

 

[cytg111]

If it's the same case you posted earlier in the photo it looks pretty rooms too me. Does it have a HDD tray down under the bottom compartment?

Corsair 220t is a pretty tight fit for my backup rig....Maybe it's just the XFX that makes it look so? It had a HDD tray down under that I had to remove for some more space for cables.

View attachment 47639

The lower 140mm could be used as a GPU support if it was need. lol

View attachment 47640

Yes, exactly, HDD tray .. and its, was, bolted to the frame. So had to apply the good old equation of "violence * velocity = gtfo" to get the job done .. ripped it straight out, of cause it is pretty banged up down there now but hey, I am never gonna mount a spinner on the inside a case ever again. So meh.
Side note : That card is Yyyyuuuuge .. how long is it? It was actually one of the things I checked for when buying the case, space for gfx card.(shoulda checked PSU as well I guess...)

 

[cytg111]

Running Prime95 now.
I had some weird issue exiting a game and subsequent failures to fast-load windows.
Have begun to optimize memory etc in bios (had a few non starters as well, had to short clrtc)
Anything here catch your eye?


edit : stopped the test early, temps were reaching 90 deg C. Damn !!!

 

[Ajay]

Running Prime95 now.
I had some weird issue exiting a game and subsequent failures to fast-load windows.
Have begun to optimize memory etc in bios (had a few non starters as well, had to short clrtc)
Anything here catch your eye?
View attachment 47683

edit : stopped the test early, temps were reaching 90 deg C. Damn !!!

1. 5800X's run hot.
2. What is your room temp?
3. Could be poor seating of your HSF (did you use the pre-applied TIM, or use a better one?).

 

[cytg111]

1. 5800X's run hot.
2. What is your room temp?
3. Could be poor seating of your HSF (did you use the pre-applied TIM, or use a better one?).

- Ok.
- Its high ~25-28deg C.
- pre-applied. I think the seating should be ok, I've mushed it pretty good and is firmly screwed on without too much force.

 

[Ajay]

- Ok.
- Its high ~25-28deg C.
- pre-applied. I think the seating should be ok, I've mushed it pretty good and is firmly screwed on without too much force.

Then that seems reasonable to me. The 12cm Noctua's under perform the D-15s, obviously. Too bad Noctua doesn't include their own branded tube of paste anymore, it's worth a few degrees off.
Personally, I don't run any programs that stress out my CPU like Prime95 does - so don't stress out about your temps .

 

[Kenmitch]

Side note : That card is Yyyyuuuuge .. how long is it?

It's roughly 13" from the end of card to the bracket....Built like a tank!

Anything here catch your eye?

Effective clock max of 9,847.9MHz would be a new World Record? Obviously a glitch, but....

 

[Leeea]

Then that seems reasonable to me. The 12cm Noctua's under perform the D-15s, obviously. Too bad Noctua doesn't include their own branded tube of paste anymore, it's worth a few degrees off.

They did for me. But 1xnoctua tube did not seem like it was not enough. The tube is only like 1/4th full from Noctua.

I ended up finding a tube from a previous noctua heatsink I had ordered, and dumped both tubes on. The extra just squeezes out the sides anyway.

 

[Leeea]

- pre-applied. I think the seating should be ok, I've mushed it pretty good and is firmly screwed on without too much force.

I love the AM4 seating on the Noctua heatsinks, with the two spring driven screws. Self levels and floats with the spring tension pushing it down. Makes it nearly impossible to screw it up.

edit : stopped the test early, temps were reaching 90 deg C. Damn !!!

I have a somewhat "aggressive"* pbo2 "overclock" going, and I have never been anywhere near those temperatures**. Maybe pull your heatsink off and verify that the paste spread evenly over the entire CPU.

*:
TDP 185watts
TDC 125amps
EDC 170amps
curve optimizer -10
max boost over ride +200
pbo scaler x10

I have had no issues at all. No crashes, no high temps, nothing.
Just using the generic XMP memory profile for my memory, nothing special

**there was a program that came with my mainboard called "asus cpu-z". Under the bench tab there is an option to stress test the CPU. Using that stress test I end up at 70-71c as reported by "asus AI suite 3". Looks like it loads down all the cores, and the frequency runs about 4450 Mhz during the test with a cpu reported power consumption of 148-154 watts. I also left the Ethereum miner running during the test, so the GPU was a nice hot toaster right underneath the CPU heatsink.

My gut is telling me you did not apply enough paste .

 

[Dave3000]

Is the Corsair H100i iCUE Elite a better CPU cooler than the Noctua U12A? I currently have a Noctua U12A CPU cooler for my 5800x. My 5800x hit's 90C in CB20 multicore within 4-5 seconds at a room temperature of 78F and then throttles down to 4.4-4.45 GHz while at 90C near then end of one run and I tried the stock NT-H1 paste and now I'm using Corsair XTM50 paste and no change in CPU package temperatures as checked by HWinfo64. Also I have a Fractal Meshify-C case that can only support a 280 and 360mm AIO if those are front-mounted and if I want to top mount an AIO it can only support up to 240mm for that. I don't want to end up bringing hot air into my case from a radiator, so I was thinking that maybe I'm better of with a 240mm AIO and top mount it as exhaust in my case. Ideally I prefer it if my 5800x stayed under 85C in CB20 multicore and didn't throttle below 4.6 GHz all-core. Is it worth switching to a Corsair H100i iCUE Elite from a Noctua U12A?

Yesterday I reapplied the thermal paste using the 5-dot method and reinstalled the Noctua U12A. Idle CPU package temperature seems to have dropped 2-3 degrees but load temperatures are the same. I also made a mess of the thermal paste when applying the 5-dot method as the thermal paste would still keep leaking from the tube for a bit after making the each dot, and ended up with streaks of thermal paste between the outer dots and the center dot. Pull back on the syringe of the tube did not stop the leaking of the thermal paste from the tube but it eventually did stop. I'm not sure if those streaks of thermal paste between the dots are going to negatively affect the cooling performance.

 

[UsandThem]

Yesterday I reapplied the thermal paste using the 5-dot method and reinstalled the Noctua U12A. Idle CPU package temperature seems to have dropped 2-3 degrees but load temperatures are the same. I also made a mess of the thermal paste when applying the 5-dot method as the thermal paste would still keep leaking from the tube for a bit after making the each dot, and ended up with streaks of thermal paste between the outer dots and the center dot. Pull back on the syringe of the tube did not stop the leaking of the thermal paste from the tube but it eventually did stop. I'm not sure if those streaks of thermal paste between the dots are going to negatively affect the cooling performance.

You likely are using too much paste (and it's not spreading evenly). I know when I tried the usual "pea method" when I went from Intel to AMD, the paste did not cover the entire CPU.

Maybe try something like this to spread a very thin layer on the heat-spreader. You really don't need very much at all (just enough to fill the microscopic holes of the metal). If it's squeezing out of the sides, it's too much paste

https://www.amazon.com/ARCTIC-MX-4-Compound-Micro-particles-Durability/dp/B0795DP124/

 

[cytg111]

I love the AM4 seating on the Noctua heatsinks, with the two spring driven screws. Self levels and floats with the spring tension pushing it down. Makes it nearly impossible to screw it up.


I have a somewhat "aggressive"* pbo2 "overclock" going, and I have never been anywhere near those temperatures**. Maybe pull your heatsink off and verify that the paste spread evenly over the entire CPU.

*:
TDP 185watts
TDC 125amps
EDC 170amps
curve optimizer -10
max boost over ride +200
pbo scaler x10

I have had no issues at all. No crashes, no high temps, nothing.
Just using the generic XMP memory profile for my memory, nothing special

**there was a program that came with my mainboard called "asus cpu-z". Under the bench tab there is an option to stress test the CPU. Using that stress test I end up at 70-71c as reported by "asus AI suite 3". Looks like it loads down all the cores, and the frequency runs about 4450 Mhz during the test with a cpu reported power consumption of 148-154 watts. I also left the Ethereum miner running during the test, so the GPU was a nice hot toaster right underneath the CPU heatsink.

My gut is telling me you did not apply enough paste .

Prime95, IIRC, is particular nasty cause it will burn avx circuitry to max on top, and that burns calories. But yea, I might give it another look.
After a heavy gaming session it picked at 80 degrees on one core, but thats it.

 

[Dave3000]

You likely are using too much paste (and it's not spreading evenly). I know when I tried the usual "pea method" when I went from Intel to AMD, the paste did not cover the entire CPU.

Maybe try something like this to spread a very thin layer on the heat-spreader. You really don't need very much at all (just enough to fill the microscopic holes of the metal). If it's squeezing out of the sides, it's too much paste

https://www.amazon.com/ARCTIC-MX-4-Compound-Micro-particles-Durability/dp/B0795DP124/

The thermal paste spread evenly on the whole CPU on the previous application of this Corsair thermal paste upon inspection before I went to reapply that paste using the 5-dot method. However, previously, I did create a bigger blob in the middle than a pea size that time and probably too big because after removing the heatsink off the CPU, I noticed some of the Corsair thermal paste went on my motherboard and the side of my CPU and the side of the CPU socket and ended up cleaning it up with alcohol prep pads and it's now mostly cleaned off except the hard to reach edges, like where the motherboard meets the edge of the CPU socket and some of the side of the CPU. It tried using the corner of an old credit card and covering the corner with the alcohol prep pad but I still could not clean off the slight line of thermal paste residue right next to the CPU socket. I hope I don't short something while my PC is turned on due to not doing a perfect clean job of this thermal compound, but this is non-conductive and and non-volatile according to Corsair's description of this paste.

 

[Leeea]

I hope I don't short something while my PC is turned on due to not doing a perfect clean job of this thermal compound, but this is non-conductive and and non-volatile according to Corsair's description of this paste.

Nearly all heat sink paste is non-conductive and can be pasted all over the place without harming anything.

The exceptions are boutique pastes that will let you know they are conductive, or liquid metal foolishness.

 

[Leeea]

Yesterday I reapplied the thermal paste using the 5-dot method and reinstalled the Noctua U12A. Idle CPU package temperature seems to have dropped 2-3 degrees but load temperatures are the same. I also made a mess of the thermal paste when applying the 5-dot method as the thermal paste would still keep leaking from the tube for a bit after making the each dot, and ended up with streaks of thermal paste between the outer dots and the center dot. Pull back on the syringe of the tube did not stop the leaking of the thermal paste from the tube but it eventually did stop. I'm not sure if those streaks of thermal paste between the dots are going to negatively affect the cooling performance.

They will not. There is no such thing as to much paste:

We've all seen people say "that's too much thermal paste." Most the time, it turns out, they're wrong. This test benchmarks thermal paste application methods.

The difference between perfect and way to much is insignificant for cooling performance. The only penalty is if you do not put enough on.

 

[cytg111]

Getting these a lot, either hard freezes or reboots .. typically in context of a gaming session.


And others just like it but

Processor APIC ID: 10
Error Type: Cache Hierarchy Error

Not good eh?
I am building two of these rigs, minus the 3080ti, but got other spare gfx units lying around... But I got another 5800x, should I give that a try?

 

[Kenmitch]

Getting these a lot, either hard freezes or reboots .. typically in context of a gaming session.

View attachment 47734
And others just like it but

Processor APIC ID: 10
Error Type: Cache Hierarchy Error

Not good eh?
I am building two of these rigs, minus the 3080ti, but got other spare gfx units lying around... But I got another 5800x, should I give that a try?

That's not good

Are you running stock or did you monkey around with PBO and offset voltages?

 

[cytg111]

That's not good

Are you running stock or did you monkey around with PBO and offset voltages?

I set the memory to 3600 everything else auto. I did have to clear the bios one time cause it wouldt post after I put in the actual latencies that the ram supports.

 

[bba-tcg]

Getting these a lot, either hard freezes or reboots .. typically in context of a gaming session.

View attachment 47734
And others just like it but

Processor APIC ID: 10
Error Type: Cache Hierarchy Error

Not good eh?
I am building two of these rigs, minus the 3080ti, but got other spare gfx units lying around... But I got another 5800x, should I give that a try?

Swapping the CPU would be the place to start, since you have another to try.

 

[Asterox]

Mindfactory CPU sales numbers, more or less expected no doubt.

https://twitter.com/x/status/1419280190761586689

 

[Leeea]

Getting these a lot, either hard freezes or reboots .. typically in context of a gaming session.

View attachment 47734
And others just like it but

Processor APIC ID: 10
Error Type: Cache Hierarchy Error

Not good eh?
I am building two of these rigs, minus the 3080ti, but got other spare gfx units lying around... But I got another 5800x, should I give that a try?

Considering the issues you have already had this with this mainboard, I suspect the mainboard is the core of your problems. But the CPU swap is what you have, so why not try it.

I set the memory to 3600 everything else auto. I did have to clear the bios one time cause it wouldt post after I put in the actual latencies that the ram supports.

Odd choice? -> I am under the impression the correct way is to set the XMP profile, although on AMD mainboards it is called a DOCP profile . This normally sets all the sub timings correct.

Your previous grounding issues makes me wonder if one of the mainboards layers is slightly miss-etched.

 

[JoeRambo]

Mindfactory CPU sales numbers, more or less expected no doubt.

Interesting that ZEN3 has completely killed HEDT sales from both Intel and AMD. I wonder is that due to supply issues with Zen2 ThreadRippers, or just noone is really into rendering anymore and 16C is more than plenty?

 

[Dave3000]

I was checking the individual core temperatures of my 5800x in the latest version of OCCT and during idle they are within 1-2 degrees C of each other but during full load like running an OCCT Linepack 5-minute run there was a 9C difference in temperature between the coolest core and the hottest core, the coolest core got up to 81C and the hottest core got up to 90C. 2 cores hit 90C. Is this temperature difference normal for a 5800x or is this a sign that the thermal compound didn't spread evenly upon seating the heatsink on my CPU? Ideally should it be 81C-83C, and not 90C if I have an even spread of the thermal compound?

 

[UsandThem]

I was checking the individual core temperatures of my 5800x in the latest version of OCCT and during idle they are within 1-2 degrees C of each other but during full load like running an OCCT Linepack 5-minute run there was a 9C difference in temperature between the coolest core and the hottest core, the coolest core got up to 81C and the hottest core got up to 90C. 2 cores hit 90C. Is this temperature difference normal for a 5800x or is this a sign that the thermal compound didn't spread evenly upon seating the heatsink on my CPU? Ideally should it be 81C-83C, and not 90C if I have an even spread of the thermal compound?

What RPMs are your fans spinning at on your NH-U12A? Those fans go up to 2000 RPM, so maybe you need to increase your fan curve in your UEFI (or set it to performance).

Outside of the NH-D15, your cooler is the highest performing Noctua cooler, so something's not right if you're seeing 90c (with good case airflow and good ambient temperatures).



Edit:

I also came across this previous discussion about the 5800X and the Noctua NH-U12S. Maybe that's a normal temperature when running synthetic benchmarks which push your CPU.

https://forums.anandtech.com/thread...-thread-reviews-begin-page-39.2585890/page-48

Also of note are the last two processors[5600X and 5800X] – both processors are reporting 4450 MHz all-core turbo frequency, however the 5800X is doing it with 14.55 W per core, but the 5600X can do it with only 10.20 W per core. In this instance, this seems that the voltage of the 5800X is a lot higher than the other processors, and this is forcing higher thermals – we were measuring 90ºC at full load after 30 seconds (compared to 73ºC on the 5600X or 64ºC on the 5950X), which might be stunting the frequency here. The motherboard might be over-egging the voltage a little here, going way above what is actually required for the core.

 

[DrMrLordX]

Error Type: Cache Hierarchy Error

Hmm. Try reading through this:

5950x Cache Hierarchy Error

I have a new system built and running seemingly well, apart from this "fatal hardware error" which has popped-up at least twice (maybe more). Admittedly, I have another PSU coming from Corsair that will hopefully fix a 5% to 10% reproducible startup issue where the PSU has to turned off and...

community.amd.com

I haven't gotten to the end of it, but it looks like a microcode issue to me.

Odd choice? -> I am under the impression the correct way is to set the XMP profile, although on AMD mainboards it is called a DOCP profile . This normally sets all the sub timings correct.

XMP profiles do some crazy . . . stuff. Anyone who can commit the time should really use the Ryzen Timing Calculator instead.

 

[Noid]

XMP profiles do some crazy . . . stuff.

PBO is no different ...
.. they are both subject to BIOS programming.

I found the " Dram Calculator " is good for getting you ball park settings.
( but I dont recommend the " Tuner " it has )

My shutdown problem .... still exists.

And, I am able to boot PBO disabled now with static my O/C.
But, Prime95 still draws ~220w PPT ( still looking to control PPT )

As, for my shutdown problem ---
I just re-seated everything, and going to re-install that instant shutdown game I have.
... If i still crash ---- > Im going to put my Nvidia 2070 back in ...
( maybe my 3 month old $900 6700XT is failing --- > I'm in denial -- lol )