• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

AMD Ryzen 3000 Builders Thread

Page 123 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

MalVeauX

Senior member
Dec 19, 2008
592
136
116
I went with a 3700x this morning from Newegg. It's $279 there right now too. I have been collecting parts since May for my build, and got tired of waiting for the 5000 series to be available. The way it looks, the 5700X if it's ever released will be 6 months away. Also picked up 3600 B-die, Asrock B550, and a Samsung 980 pro.

I see the 3700's selling for more than $280 used on Ebay, so I figure I won't lose much if I decide it's worth upgrading later. my only fear was that the 3700x will drop below $200 when the 5000 series is readily available. But with all the PC part shortages this year, it probably won't get any cheaper.

My thought anyways.
Thanks; just poking around and looking at supply stuff, it seems realistically the 5000 series isn't going to be in my hands any time soon. Prices are all over. Stock is empty. I'm not desperate enough to sit and camp a website with the add/buy button over and over and hope to get in line with a sale. I may just end up getting a 3700X and use that until the 5000 series is really mature and available and then go for a 5900X or something, if not a 5600X after I see the real performance in my application needs of the 3700X.

I think I'll wait through this weekend to see if any real sales happen that make a pathway choice a lot easier. It's silly that the 3700X is $280 and the 5600X is $300, but doesn't matter what the price is if you cannot buy it. And every 3700X I see on the `bay is being bid on, so worse case scenario, looks like I could sell the 3700X later on if I had to, under-cut it for a fast sale, when a 5000 series is readily available. Or, just use the 3700X in a second build that I will eventually need anyways, for a mid-tier couch coop gaming PC (steam box).

Edit: wow, 3700X on amazon just went to like January/Feb for shipping lol. Guess I won't be upgrading at all any time soon. Jeez.

Was thinking:

3700X
TeamGroup 3600Mhz CL16 32Gb DDR (8x4, only because of reading that using 4 sticks is better than 2 sticks?)
Asus TUF Gaming B550-Plus MB (flash BIOS with USB)

Very best,
 
Last edited:

MalVeauX

Senior member
Dec 19, 2008
592
136
116
Well, was going to do 3700x but they're all sold out and/or the same price, or more, than a 5600.

I guess I just haven't tried to upgrade during a market scenario like this. This is really disappointing for the big sales and holiday type stuff.

Very best,
 

EXCellR8

Diamond Member
Sep 1, 2010
3,887
788
136
I saw some 3700X stock yesterday on Newegg for under $300... probably long gone by now though.

AMD did have the 3950X available to purchase direct, earlier today.
 

MalVeauX

Senior member
Dec 19, 2008
592
136
116
I saw some 3700X stock yesterday on Newegg for under $300... probably long gone by now though.

AMD did have the 3950X available to purchase direct, earlier today.
I think I'll just let the dust settle over the holidays and into first quarter. I'm on an 8 year old FX8350 right now and I'm beyond ready to finally upgrade. So naturally its near impossible to get anything without over-spending by a lot (I wouldn't bother getting a 3700x for over $300, when I can get a 3900X for $450 new), or basically nothing is in stock anywhere. I've picked out a B550 board and some memory, and I have the rest of the system already, just need the new MB, CPU and RAM and the CPU is just not available or not appropriately priced. So I guess if I just wait a bit more, what's a few months after 8 years, right? I think everyone has convinced me at this point to not buy the 3700x for $280~300 when the 5600x is $300. So I'll just watch for stock I suppose.

Very best,
 

WilliamM2

Golden Member
Jun 14, 2012
1,759
102
106
So I got the Asrock B550 Velocita, Ryzen 3700X CPU, and some 3600 CL16 ram yesterday.

The only issue I'm having is that the ram will not run stable when I set XMP. It crashed memtest in 20 minutes, and locked windows once or twice. The only way to boot again was to clear cmos. So I set it to 3400mhz, CL 16 and it ran memetest for 8 hours without issue. seems stable so far, but I want 3600.

Here is the ram, Samsung B-Die:

How do I get this to run at rated speeds? I read that ram has to be tweaked on AMD. I have not ran an AMD chip since my Athlon XP2400, probably 18 years ago.

Any help would be appreciated.

Also, the proc seems to run at 3.6 or 4.4 ghz, never lower (all stock settings so far). Don't these have lower states for when the system is just idling?
 

sdifox

No Lifer
Sep 30, 2005
83,284
8,591
126
So I got the Asrock B550 Velocita, Ryzen 3700X CPU, and some 3600 CL16 ram yesterday.

The only issue I'm having is that the ram will not run stable when I set XMP. It crashed memtest in 20 minutes, and locked windows once or twice. The only way to boot again was to clear cmos. So I set it to 3400mhz, CL 16 and it ran memetest for 8 hours without issue. seems stable so far, but I want 3600.

Here is the ram, Samsung B-Die:

How do I get this to run at rated speeds? I read that ram has to be tweaked on AMD. I have not ran an AMD chip since my Athlon XP2400, probably 18 years ago.

Any help would be appreciated.

Also, the proc seems to run at 3.6 or 4.4 ghz, never lower (all stock settings so far). Don't these have lower states for when the system is just idling?
did you update your bios?
 

WilliamM2

Golden Member
Jun 14, 2012
1,759
102
106
are you able to set the values manually?
I am, that's how I set it to 3400. When set to XMP, all the settings I know of seem correct. But that's just the CL16 etc. There are a dozen settings that I don't have specs for.
 

WilliamM2

Golden Member
Jun 14, 2012
1,759
102
106
Use 1usmus' DRAM Calculator for Ryzen to get the remaining values.
I did try that, but it asks you to fill out several fields I don't know, like the pcb stepping.

It's a moot point now for the time being. I rebooted the computer to change the fan curve in the bois, never got there, it hung on error "4D". No amount of resetting cmos or swapping out memory (2400 mhz) altogether would make it boot again. Internet searches tell me it's a bad board. I sure hope it's not the cpu.

Spent 4 hours putting this together yesterday in a new R6 case. Now I had to spend another 3 hours tonight troubleshooting, and swapping back in my trusty 6600K.

At least Newegg is replacing it, but it will take at least a week. So far I am not liking AMD at the moment.
 
Last edited:

JimKiler

Diamond Member
Oct 10, 2002
3,434
108
106
I think I'll just let the dust settle over the holidays and into first quarter. I'm on an 8 year old FX8350 right now and I'm beyond ready to finally upgrade. So naturally its near impossible to get anything without over-spending by a lot (I wouldn't bother getting a 3700x for over $300, when I can get a 3900X for $450 new), or basically nothing is in stock anywhere. I've picked out a B550 board and some memory, and I have the rest of the system already, just need the new MB, CPU and RAM and the CPU is just not available or not appropriately priced. So I guess if I just wait a bit more, what's a few months after 8 years, right? I think everyone has convinced me at this point to not buy the 3700x for $280~300 when the 5600x is $300. So I'll just watch for stock I suppose.

Very best,
Why not get a 3600 or 3600X to save money. It will be cheaper than a 3700X or 3800X and then you can upgrade to a 5000 series in 6 months when stock is up and prices come down. A 3600 would still be faster performing than your old bulldozer. I used to buy cheap and upgrade later but now i prefer to buy what I want now. However in your shoes even a hexacore 3000 series would be a huge boost in performance and cheap enough that an upgrade in 6-12 months is justified.

Otherwise maybe you can get a deal on eBay for someone who did score a 5000 series chip and no longer needs a 3000 seris.
 

MalVeauX

Senior member
Dec 19, 2008
592
136
116
Why not get a 3600 or 3600X to save money. It will be cheaper than a 3700X or 3800X and then you can upgrade to a 5000 series in 6 months when stock is up and prices come down. A 3600 would still be faster performing than your old bulldozer. I used to buy cheap and upgrade later but now i prefer to buy what I want now. However in your shoes even a hexacore 3000 series would be a huge boost in performance and cheap enough that an upgrade in 6-12 months is justified.

Otherwise maybe you can get a deal on eBay for someone who did score a 5000 series chip and no longer needs a 3000 seris.
I went ahead and decided to skip all the 3000 and just go straight for the CPU I want in reality (5900x), so I'll be waiting a while. I picked up an Athlon 3000G APU to hold me over the next few weeks or months until a 5900x becomes an option for MSRP.

Very best,
 

JimKiler

Diamond Member
Oct 10, 2002
3,434
108
106
I went ahead and decided to skip all the 3000 and just go straight for the CPU I want in reality (5900x), so I'll be waiting a while. I picked up an Athlon 3000G APU to hold me over the next few weeks or months until a 5900x becomes an option for MSRP.

Very best,
I hope you still noticed a speed increase from your old bulldozer system.
 

MalVeauX

Senior member
Dec 19, 2008
592
136
116
I hope you still noticed a speed increase from your old bulldozer system.
Yeap, in my routines, my FX8350 would finished up a routine in 19 minutes. This Athlon 3000G does it in 15 minutes. 26.7% time saved. Equates to a few hours total. It's a big deal for me. I know a Ryzen 1600 does it in 6 minutes, same file. So if the 5900x can drop this to under 4 minutes, it will be a massive time saver for me. So I'm not losing anything frankly going from FX8350 to the Athlon 3000G which is kind of sad, but 8 year tech gap makes a big impact. Just waiting on that 5900x, maybe in 6 months sigh.

Very best,
 

CB434

Junior Member
Sep 11, 2009
4
2
81
Safe voltage was a big issue in the first few months of launch last year.. in terms of manual OC vs boosting, and a lot of talk about safe (long term) voltages. It's harder to find talk about this 12 months on. Does anyone have any thoughts of 1.15V vs 1.25V vs 1.3V in terms of a safe long term voltage?

I'm currently testing a 3700X (a launch CPU that sat in the box a long time) that's in between 4.2Ghz and 4.3Ghz, depending on how much voltage I want to run. I haven't tried 4.4 yet because I don't want to go any higher on the voltage. Whenever I see 1.35 or 1.4V manual OC I cringe.

But yeah, I'm looking for a cool and quiet rendering system but so far the numbers for 4.3Ghz are nice if I can keep them. Currently using 1.28V that droops down to 1.26V under load. I can't remember the voltages for 4.2Ghz but it's be something like 1.22V or less (with nice temps). I've also been playing around with per CCX OC, but yeah.. the voltage is all in the same ballpark. So far my initial thoughts are that running different frequencies 43.25 and 42.75 might be more unstable than just doing 43 across all cores. Need more testing for that but that is an initial impression. The benefit if it works is a little extra single core boost.

Does anyone have any thoughts about degredation, 12 months on from the initial talk about it? I ran it stock and loaded up the intense tests like Prime and it dropped to 1.325V. I guess this is called the FIT voltage. So I'm already way under that, but there is no clock stretching with manual OC.. so it seems hard to predict what the degradation could be. I'm running 3600 Cl16 at decent voltages (1.36V), so that might help the stability a bit.. rather than 3800CL16 for example. This system is trying to be a 24/7 render machine, rather than a gaming high fps machine.

In terms of trying to buy them at the moment.. and the stock available and prices, it's the wildest time I've ever seen. The local shops here in Australia have nothing and the prices are all over the place. The prices seem backwards, forwards and all over the place. And a lot of scalping. Hopefully it settles down eventually.
 

DrMrLordX

Lifer
Apr 27, 2000
16,902
5,849
136
Safe voltage was a big issue in the first few months of launch last year.. in terms of manual OC vs boosting, and a lot of talk about safe (long term) voltages. It's harder to find talk about this 12 months on. Does anyone have any thoughts of 1.15V vs 1.25V vs 1.3V in terms of a safe long term voltage?
It's all about the current draw.

For anything that isn't PrimeGRID, Prime95 smallFFTs, or similar (and honestly, how many workloads are similar to THAT?), I would not eclipse somewhere in the ballpark of 1.325v-1.344v for workloads that tax all available cores. That's for one of the 105W TDP CPUs like a 3900x, 3800x, and 3950X. For a lower-TDP CPU like the 3700x . . . well actually, it shouldn't matter, if you're going for a static overclock.

That's the most voltage I would throw at a Matisse running something like CBR20/CBR23, Blender, any kind of x264/x265 encoding, or what have you. For the hardcore AVX2-instense crap (like certain PrimeGRID workloads) you might want to stick to 1.2v or maybe even lower?

Take a look at how much current those workloads (CBR20, Blender, etc.) run @ stock (and current draw WILL track upwards if voltage increases) and then compare your other workloads under the same conditions, and make your best guess. You can maybe afford higher voltage if you're absolutely sure your current draw (relatively speaking) will remain low. OC headroom will not increase much with higher voltage, though, beyond a certain point. I would be surprised if you could get an all-core static OC of 4.4 GHz on a 3700x. It's rare enough on a 3900x like mine.
 
  • Like
Reactions: CB434

CB434

Junior Member
Sep 11, 2009
4
2
81
Your first two paragraphs summarized my problem nicely. I wish there was AVX offset in the BIOS. Because it needs probably 2x less multiplier for those workloads (ballpark figure). But why compromise everything else.. for workloads that you will probably never use? This current Intel system I'm typing on is six years old, I doubt it's Prime 95 stable.. and it hasn't had one BSOD all year. It can encode x264 all it wants.. no problems. The only BSOD it ever had was IMC/memory related. Once upgrading memory, no problems at all for regular use or even days or weeks of encoding.

Your last paragraph, I will return to that and use it as a reference. To be honest, I'm surprised I could get 4.3. Even if it could get to 4.4, I don't have the cooling for it.. I want it to stay between 65-75C. But yea, I was expecting the worst (it's a launch CPU) after watching a few YT videos of manual OC, and reading around the internet in the last 6-12 months. I don't know if this is a good chip or a bad chip, but at least it's not a dud. I actually have another 3700X new in the box (a very recent one), unopened.. and I might have to return it (have around 20 days to return it), because this one seems like it's a keeper.

It's a long story why I have two.. I have an order for 5900X that's already payed for but who knows if or when it'll arrive. Others who ordered on the same day as me, they got it 3 days later.. while I've been waiting a month. But it's not all bad. This is good practice for understanding how Ryzen works.. and also.. the AGESA for Zen 3 have some bugs and problems.. with WHEA issues, USB and low IF speeds. In some ways, waiting is a positive thing.

Anyways, thanks for your advice, I will have a look at the current.
 

DrMrLordX

Lifer
Apr 27, 2000
16,902
5,849
136
@CB434

AMD isn't using an AVX offset specifically because they wanted to reach clocks based on current draw rather than arbitrary criteria. A hypothetical workload that uses AVX2 that only spawns one thread could still boost all the way to your max ST boost, theoretically, while on an Intel CPU you automatically take a clockspeed hit just for using AVX2. Or if you've got multiple workloads running and one of them happens to fire off some clusters of AVX2 instructions every now and then, you don't want every core immediately downclocking even if current draw still stays relatively low. Look at what happens to Intel CPUs running multiple VMs when one of the VMs starts up AVX2 or (even worse) AVX512 instructions. It's actually pretty bad.

I myself stick to around 1.344v which gets me 4.4 GHz on a 3900x in just about anything except Cinebench (4375 Mhz) or Blender Benchmark (4350 MHz). For crazy stuff like Prime95 SmallFFTs I just revert to defaults to let the boost algo handle things on its own. Ryzen Master makes this sort of thing easy.
 

CB434

Junior Member
Sep 11, 2009
4
2
81
@DrMrLordX

Taking your advice on looking at the current side of things, I've logged a lot of data in the last few days. Hopefully it doesn't crowd this thread too much, if it does, maybe I could make a new one. I've been reading a lot about Ryzen in the last 12 months, but only finally got around to building my system in the last week or two. It's been fun just trying to figure out how it all works. And it's interesting to see how these Zen chips boost more like GPU's these days. But at the same time, I miss some aspects of the old Intel manual OC.. with the different frequencies for how many cores are loaded. And a bit of headroom to play with. Not sure how Intel does it these days, as my last Intel system was 4790K, which I'm typing this message on at the moment. My main complaint about stock Ryzen boost is the midrange temps and wattage. I like the idle and the full load, but don't like how jumpy and hot it gets in smaller tasks while idling. I also prefer to prioritize multi core over single core.

I found this picture below in a recent motherboard review from this site, which gives a sort of baseline for 3700X clocks and voltages.. the AGESA is pretty recent as well. I have to say I haven't tried the 'actual' manual OC setting so far. I've just been using per CCX, but using 42x 42x if I want 4200mhz for example. I'm not sure if it makes a difference, I just find it simple and easy to work with. The board is ASUS Crosshair Hero VI.

ASUS ROG Strix B550-F Gaming Overclocking Performance_575px.JPG

It seems my chip is better than this one at manual OC, but who knows. There's so many factors, maybe if I used that board with that BIOS, my numbers would be worse.

3700X_ManualOC.jpg

The wattage is measured from the wall for the entire system. And those voltage numbers were going for the lowest voltage possible to complete many CB R20 runs. To make it stable for linpack AVX or AVX2, it needs another .1 volts or more. For example IntelBurnTestV2 at very high for 10 runs. The easiest way is to make it stable for 4300mhz in CB R20, and then drop the frequency back to 4250 or 4200. For example, it can be stable at 4300mhz at pretty much anything at 1.286V under load (using 4350 voltages from above).

I've discovered two stress programs I didn't know about in the last few days. OCCT V7.2.0 and Realbench (great for running GPU and CPU at 100% at the same time, and it uses X264 which is great). The great thing about OCCT is HWINFO is a bit buggy with the sensor chip on my board (freezes a lot) but OCCT has the same data and it never freezes, so I can finally monitor all the Ryzen specific stuff while it is at 100% load. The great thing about OCCT is that it has all the Prime type stuff and the IntelBurnTest type stuff, and also Memtest type stuff all in one program and it monitors at the same time. And can automatically stop the test as soon as it detects an error.

Above 3300mhz, it gets a bit tough and the voltages and temps skyrocket. CB R20 can be stable but other stuff starts to get too hard to cool down. I didn't try 4400mhz, it felt too far out of reach without putting the voltages really high. The sweet spot is around 4250 - 4350mhz depending on how much wattage/temps you prefer. 4200mhz would be easy to run with low voltages and temps, but the SC score suffers quite a lot.

I recorded some current (A) numbers back at stock in OCCT and compared them to manual OC numbers. The stock numbers for my 65W TDP chip are..

PPT - 88W
TDC - 60A
EDC - 90A

Through all of my manual OC'ing.. the EDC stayed locked at 90A, not sure if there's a way to unlock that or if it's hardwired in. Or if that would be a good or bad thing. I mean, outside of the firmware isn't this just a poorly binned 3800X? The 3800X can do 140A EDC. I'm unclear about this. The EDC is supposed to be a VRM rating, but if I put in a 3800X, magically my boards VRM can handle an extra 50A?

During Manual OC, PPT went from 88W to 108W at the max, and current would sometimes go to 68A or 70A. So that's 10A more than stock. Is this unsafe or bad? That was mainly in linpack AVX. In regular tasks or even CB R20.. or X264 encoding, it stays around 58A to 62A. In Prime.. I don't even run Prime (small FFT). Nothing I use in regular real life usage comes close to it, AVX is about the most I ever do. Any run like that with manual OC isn't worth it. It feels like over stressing the chip for no reason.

The stock current numbers in OCCT 7.2.0 were -

Linpack 2012 (AVX) -

CPU Current - 59A
EDC - 82A
PPT - 88W
Vcore under load - 1.22V
Multiplier - 39.3X
Temp - 65C

Linpack 2019 (AVX2) -

CPU Current - 56A
EDC - 89A
PPT - 89W
Vcore under load - 1.29V
Multiplier - 40X
Temp - 69C

OCCT Small data set (AVX2) - (Prime 95 Small FFT type test)

CPU Current - 62A <--- (exceeding stock spec)
EDC - 89A
PPT - 89W
Vcore under load - 1.13V <--- wow much different than 1.325V
Multiplier - 38.25X <--- behaving almost like a built in offset
Temp - 67C

Manual OC - 3300mhz at 1.286V

CPU Current - 64A <--- (is this too high?)
EDC - 90A
PPT - 106W
Vcore under load - 1.264V
Temp - 72C

The setting above can be slightly unstable in RealBench after an hour or two if the temperatures are high. If I increase voltage even more.. it becomes super stable even over 80C.. and the current goes down a bit but the temperature and wattages goes up a bit more.

Manual OC - 3300mhz at 1.303V

CPU Current - 70A
EDC - 90A
PPT - 110W
Vcore under load - 1.286V
Temp - 74C

I'm not sure what is better, if if the current goes down or if the voltages goes down? The temps are between 70-75C. It only goes towards 80C in a stress test like RealBench after 30 mins when the GPU is at 100% load, at the same time the CPU is at 100% load. I've adjusted case fans to help reduce that and stop the motherboard/VRM from soaking up too much heat. The VRM max temp so far has been 52C.

I guess this would be safe? As long as I don't run any of those Prime small FFT type loads? Cooler is Noctua D15S with two fans and in CPU heavy stuff it is always below 75C. In CB R20 it's 69C. I was hoping to use voltage under load of 1.22-1.29 because those are the stock Vcore for Linpack AVX and AVX2 but yeah. The temps are ok and the voltage isn't exceeding 1.3V? The main thing is that under some loads, the current is 10A higher than usual.
 

MalVeauX

Senior member
Dec 19, 2008
592
136
116
Ahh heck, I caved.

When I realized the little $50 Athlon 3000G wouldn't work in my ASUS TUF x570 Gaming Plus WiFi motherboard (yet the Ryzen 3000 series APU's work), I realized I wasn't going to be just waiting for a 5k series Ryzen, but I'd be waiting for any damned CPU at this point since everything is obliterated for stock. I almost just threw it all away and went Intel.

But, I caved and picked up a Ryzen 3700X. Overpriced a bit, but it will end up being used until I can get a 5900X and then it will migrate into my 2nd machine which needs an upgrade anyways.

So, currently will be on:

Ryzen 3700X + Hyper 212 Black Edition
TeamForce 3200Mhz CL16 32Gb DDR4
Asus TUF Gaming Plus x570 Wifi AM4
Crucial NVMe M2 SSD
Curicial SSD SATA

Sigh.

Very best,
 
  • Like
Reactions: Tlh97 and scannall

richierich1212

Platinum Member
Jul 5, 2002
2,741
357
126
Walmart allows scalpers.
If it's not shipped by Walmart, it's not Walmart.
However, Walmart does the refunds.
( It's complicated, and I found out the hard way )
So, your warranty may be questionable also.
Not sold by Walmart, I edited my post
 
Last edited:

ASK THE COMMUNITY