Ryzen 5 1600 (non-"x") OC settings reality check

frizzlebyte

Honorable
Mar 21, 2013
26
0
10,530
Hey guys,

Just tried overclocking my Ryzen 5 1600 proc for the first time today. I am using an ASRock B350 Pro4 with the latest BIOS, and DDR-2400 RAM.

I have been deliberately conservative with my settings, and have achieved 3.7ghz on the default voltage setting, 1.2375v. I tested Prime95 for an hour, and things seem to be going well so far. Granted, I should run Prime95 for longer, but this is more stable than I expected for a first attempt.

My question is really related to the voltage, namely, does that seem like a reasonable overclock to get out of the default voltage? It surprised me that I didn't have to increase the voltage AT ALL to get a fairly stable system, and the highest temps in Prime95 were 77.4 degrees Celsius.

Is this reasonable, or is there something I should be concerned about? This is my first overclock ever, so I want to make sure I get some input from more experienced OC'ers before i commit to this.

Thanks, all.
 
Solution
Sounds like you have realistic expectations are are willing to do what is needed. It does take time to configure and validate a stable overclock. The majority of people don't bother. That's why they are typically back here in about six months whining about "why am I getting all these errors all of a sudden?", "I think my motherboard is bad", "I think my graphics card is failing?", "I think I got a crappy CPU".

No, you have an operating system full of micro-errors due to not taking the time actually required to validate an overclocked configuration.

But then they do a clean install of Windows and now they are sure it was just a bad update.

No, you most probably corrupted the operating system with errors. This happens with unstable CPU...

nobspls

Reputable
Mar 14, 2018
902
12
5,415
The Ryzen (favorite expletive here) is cheap. There is no need to baby them. If you fry 'em, it is just another excuse to move up to the latest version. Chances are the Ryzen's have their own internal safeties too, if you try to run too much voltage at too high a frequency, the thing will automatically go into the safe limp mode, it liimits it to something stupid like 1.8 Ghz and ignores all your voltage settings when you cross over the threshold.

I got a R5 1600 that I am stably running at 4.0 Ghz with a +0.1V for an effective cap of VDDR of 1.3375 on my Asus B350M-E Prime see:
http://www.userbenchmark.com/UserRun/7865794

The important thing is to us a good cooler unless you want go deaf. The cheap AMD box coolers become jet engines with a migraine inducing droning sound. I just slapped on a hacked together Cooler Master Hyper 212 plus ripped off from an old I5-2500K. Works great, basically silent, until under heavy load, and no drone even then and temps all within expectations.
 

Rogue Leader

It's a trap!
Moderator


Yeah telling people to do whatever there is a safety is BAD advice. We have had many users burn up recent Ryzen and Intel processors doing just this. You may think its cheap but not everyone can afford to burn up a $100-$400 processor, and the other damage that can go with it (ie take the motherboard too). Please avoid this type of advice in the future.

OP thats a great overclock to get on stock voltage, that just tells me you can probably conservatively add a little voltage and get some more out of it. Keep in mind your board is on the lower end so really you can only expect to get a bit more voltage out of it and still be stable, it has a small VRM setup.
 


WHICH Prime95 version did you test with?

WHICH Prime95 torture test did you run, and did you use custom settings?

Have you run Realbench? If not, THAT is what I would run FIRST. Prime95 is really only a useful tool for testing thermal compliance, while running Prime95 version 26.6 on the Small FFT option.

Large FFT is practically useless for thermal or stability testing of memory or CPU.

Blend is useful for testing stability, but either runs excessively high thermal conditions if using AVX versions of Prime, or does not use AVX instructions on versions 26.6 or older.

Realbench runs realistic AVX/Handbrake work loads and is a very good first step as a tool in verifying stability of CPU overclock settings. You should also perform ALL of your CPU overclock configuration and testing BEFORE setting memory XMP or custom values in the bios so that you will not get misleading instability indicators due to memory configuration.

After your CPU overclock is finalized and has been verified as stable as much as is realistically possible, then it would be recommended to configure the memory XMP settings, or do any overclocking and/or tightening of memory timings, which will be a whole other procedure using Memtest86 and Prime95 version 26.6 Blend mode. I recommend clicking the Custom settings option and leaving everything except the amount of memory to use. Open system resource monitor or HWinfo, verify how much memory is currently unallocated/free, and input something relatively close to 75% of the free memory in the "amount of memory to use" field.

Personally, I wouldn't move on from the CPU stability test using Realbench until you are able to pass 8 hours of the stress test option. In the memory field of Realbench, choose the amount that is about half of your total installed memory capacity.

As Rogue leader has already explained to you, that is not a particularly desirable board for overclocking, so you may need to pay attention specifically to the thermals and clock frequencies to make sure no VRM or thermal throttling occurs. I recommend using HWinfo for monitoring of system sensors or CoreTemp for CPU core temp only monitoring.
 
I think if you check load voltage with cpu-z during stress testing you'll find the board will be running higher than your 1.2375v base voltage.
It has a doubled up 3 phase , which while absolutely fine for overclocking thermally can tend to overcompensate voltage wise.
 
Correct me if I'm wrong, but normally core voltage decreases under load, not the other way around, which is why there are load line calibration options.

Unless this is something new and specific to Ryzen, I've never seen core voltage increase under a load. Demand will certainly increase, but that is why LLC or offsets are so important in order to avoid instability. Or maybe I'm missing what you're getting at here?
 


No youre 100% right in general terms.

Im talking specifically about the b350 pro4.
It doesnt have any llc options configurable in bios at all,or any offset voltage settings.
You can leave the voltage on auto or set it manually & thats it.
While the board is fine for OC on a hardware component level imo its bios options are like going back in time 5 years.

Im assuming when OP said he left it 'default voltage' he means its set on auto (which it is at default out of the box)
If Im right that 1.2375v will be correct at low/idle loads but that particular board pumps excess voltage under load.


 
Then that would, in my opinion, definitely make for a board I would not want to overclock at all on, or recommend, or use. Automatic configurations are never something I trust much. IE, I would never use an automatic or desktop overclocking utility or bios preset, for the exact reason that it tends to overdo things in order to err to the side of caution like excess voltage, loose timings, etc.

It will definitely make it more difficult to be thermally compliant at any given OC setting, but not impossible, so long as you are realistic with your expectations.

Are all of the low budget B350 boards this way, or just ASRock? Personally, if the VRMs and other components on a given board cannot handle the inclusion of offset and LLC settings, I think it highly likely that overclocking capabilities should not have been included on the board at all, in the first place.

 
^ the vrm setup is exactly the same as the gaming k4.
While it's a cheat to advertise as a 6+3 phase when it's really (3+3)+3 it's still fairly capable on a hardware level.

The k4 has llc & offset voltage options, the pro 4 doesn't (It's had 9 bios revisions now so I have no idea why these options haven't been added)

I wouldn't Dismiss overclocking entirely , just proceed with caution simply because what you set in bios is not going to be what you get under load in windows) , load voltage needs to be checked with cpu-z to err on the side of caution.

I've had a 1700 At 4050mhz on the same board at 1.36v, solid as a rock , 2400mhz ram at 2933, & vrm temps around 60c (with an infrared thermostat) so good overclocks are eminently doable , just not as simple as they should be.

 

Rogue Leader

It's a trap!
Moderator


I agree these small VRM boards with Ryzen are not the same flaming disaster they were with FX chips. You can overclock lightly on them, just don't push it.
 
That falls just about in line with what I figured. Stupid that they don't include the option for LLC in the bios. Does the bios on that board offer something not present on other boards? I know that generally, when it comes to upgrading features in the bios, they are almost always limited by the actual capacity of the bios ROM, so maybe they cheaped out there or offer some other features that they don't want to do away with in order to add the support features for LLC.

For example, a while back Gigabyte removed the option for an F12 bios screenshot capture after about the fifth bios update. They said it was because they had to get rid of something in order to make other changes. Later, it was revised again and the F12 bios screenshot capture features were added back in.
 
I have a sneaking suspicion it was a purposeful decision so as not to impact sales of the b350 gaming k4 & x370 pro 4, both with identical power delivery setups but with better bios overclocing options, both $30 or 40 more expensive.
Why would you pay that extra just for sli support if even the bios options are identical.

It does,come,across as annoying but it is an $80 board , the only sub $100 board with twin m2 slots, easily the best vrm setup in its price bracket so something had to give.

Msi have done exactly the same with the b350 pc mate compared to the gaming 3,5,plus range.
 
All generations of PC mate are on my "do not recommend" list anyhow. Looks like this board will be as well. True, it's cheap, and what do you want for the price. Fine for a user not intending to overclock. But as an overclocking option, not going to be recommending it at all. Not that I probably would have anyway, but in some cases...........
 

nobspls

Reputable
Mar 14, 2018
902
12
5,415


Who in their right mind is going to pay $400 for a Ryzen? The whole point is they are cheap an expendable. $200 max. You can get a R7 1800x for $200 see:

http://www.microcenter.com/product/485483/ryzen-7-1800x-36-ghz-8-core-am4-boxed-processor

It is really effectively $170 because you get $30 off on the mobo. The 2700x is ripoff and the 1800x is basically par with a 2700x once overclocked. But the whole point of being cheap and expendable is why it is worth the effort and risk for going AMD. Cook it and cook it good! Just like good cooking, you OC just right and you won't be making char coal.

The biggest hurdle is probably the VRMs on cheap mobos will fry long before you Ryzen destroys itself. You can see my efforts to overclock my R7 1700 here:
http://www.tomshardware.com/answers/id-3701338/am4-1700-vrm-mos-temps-high.html

Most likely the board will crash under prime95 AVX stress test because the VRMs hit there temp limits.
 
If there was ever a candidate for being instantly banned after only one "incident", this one ^^^^would be it.

This information is almost, and maybe just as, stupid as the guy who swore there were 7 core Intel processors, and a lot more dangerous.

Not-sure-if-trolling-or-just-stupid.jpg
 

Rogue Leader

It's a trap!
Moderator


Who would? Me. I paid $500 for an 1800X on launch day and I don't regret it for one second.

The 1800X is discontinued, sure you may have been able to get one for $200, but not anymore as its sold out. As someone who actually owns an 1800X and can speak to its abilities, it IS great,. but that doesn't change the fact that stock vs stock the 2700X is faster, and its an optimized sample and has overclocked higher. Is it worth the money if you can find an 1800X for 200 bucks? No. Except you can't anymore.

They may be expendable for you, but for many of our members they are not. Right now their 2 highest line processors are over $200 (2600X and 2700X). Either way overclocking it until it beats the crap out of your board is not how to do it either. I read your thread, and thats all you're doing is beating on stuff until it breaks. You don't have experience doing it properly and have gotten lucky so far.

You should not be giving any advice to anyone on overclocking. Learn to do it properly or keep your methods to yourself. You will not be damaging other users equipment with your bad advice here.
 


In my experience, and in the experience that has been passed on to me from some folks around here who have probably been overclocking longer by far than I have, and longer than most of our members including you, have been alive, running AVX versions of Prime for the purpose of thermal OR stability testing would show a clear lack of understanding of the underlying principles involved AND would fly directly in the face of practically any of the overclocking communities best practices.

Non-AVX versions, or modification of the "local.txt" file in the Prime folder for AVX versions of Prime95 are what should be used for steady state thermal testing and other applications such as Realbench which runs realistic AVX workloads should be used for stability testing of the CPU. Prime95 non-AVX Blend mode is however a great metric for testing memory configurations when XMP or custom memory settings are in play.
 

nobspls

Reputable
Mar 14, 2018
902
12
5,415


Yep telling people to overpay for stuff is good advice. If anyone that is no qualified it should be you.



It is not discontinued, and I drive to the store and pick right up as the link I've shown you. No lies. Less than $200 effectively.



It is a $60 board, and a $140 processor. The whole point is to get it cheap and beat the crap out of them. Overpaying for that extra maybe 5%-7% for the latest 2600x/2700x, at nearly 45% more money is simply insanity. If a $200 processor is not expendable then over paying for 260x/2700x is a total nonstarter.



I do not know where you get the idea that you are the only authority to overclocking, and only your can give advice. Overclocking is all about getting lucky. It is the playing the chip lottery. Everyone knows this is how you play, you have to gamble to win. The risk is low, the costs of going cheap on the mobo and getting cheap ryzen is how to mitigate your risks and not overcommit. And if they crap out, return the board within 30days of purchase, if the CPU does not overclock you liking exchange in the 30 day period. Got to beat the hell out of them early so you can return/exchange the flakes fast. Rinse and repeat. The hassle is real, but the payoffs is worth it. This is how to maximize your dollars.

Protecting the pocket book is far more important worrying about wearing out the gear at a faster rate. You over dramatize calling it "damaging". Heck I still got my Sandybridge from 2011 overclocked to 4.2 Ghz. Stuff still kicking and stable as it ever was, and my R5 1600 an R7 1700 are 100% stable, the normal load operating temps, and voltages are within reason and expectations.

I've done this overclocking stuff since the days of the of my first PC, an 80286 and over clocked the thing to 12mhz from 8mhz, you know that famous turbo button put it to real use. Sure the memory was cooked after 4 years, but I was ready to move on to the 386 by then. The strategy is simple, buy the previous year stuff or whenever they are cheap, overclock the hell out of them, and profit.
 

Rogue Leader

It's a trap!
Moderator


Maybe you need to go back and read that again, I never told anyone to "overpay" for anything, I was responding to your ridiculous claim that Ryzen is a cheap sub $200 processsor and anyone whom spends more is dumb. You can't get a 2700X for any less than $279 right now.



Maybe you don't understand what discontinued means but the 1800X is NO LONGER BEING MADE. AMD stopped producing them when they started making the 2xxx Processors. Just because your Microcenter has some leftover ones in stock does not make it not discontinued. I clicked the link, the Microcenter by MY house is sold out. They won't be getting more, because its DISCONTINUED. It was being sold that cheap because it was discontinued.



You missed the part where not everyone can burn $200 like you can. I don't know where you are getting your % numbers but they don't make sense. The 2700X has 2 extra cores, the 2600X while it has the same number of course again the base clocks out of the box are faster. If you can still get a 1600 today for a good price gerat, for many they can not because again, its discontinued. On the overclocking side the 2xxx processors have proven to be able to overclock better, the process has been refined, the yields are better. AMD is just plain putting out better chips. I totally agree if you can get one of the 1st gen chips for the prices they go for right now its a no brainer, but they will be dried up shortly.



No its not ALL about getting lucky. That sure is part of it. But you can properly overclock ALMOST any chip without turning it into a smoking burning heap using proper overclocking methods. Even using a cheap motherboard. But sure advocate burning up the VRMs on a board and returning it, and if the shop you gor it from doesn't look then maybe you'll get away with it. Or you could just do it the right way and not need to do that.

Your horrid advice is not how to maximize your dollars, its how to waste both time and money, oh but wait its not yours because you're just burning other people's boards with your bad advice, or the stores since you'll return it. People like you are why prices go up.




You may have done it a long time it doesn't mean you are doing it right or you haven't just gotten lucky. Well clearly you did burn up some stuff.

I'm not the overclocking authority, theres many more here far more qualified than me to give overclocking advice. What I am is a moderator and my authority is to stop bad and dangerous advice from being dispensed to our users. Beat on your own equipment all you want, don't tell others whom are coming here for safe/good advice to do it too.
 

nobspls

Reputable
Mar 14, 2018
902
12
5,415


The whole idea is "discontinued" is marketing lies. It is really only discontinued when AMD quits supporting them. and there is no supply out there. As it is the prices are low because there is a glut of them. This is the market reality.



Burning them up over 4 years is well worth it, even if it did drive to thermal migration/avalanche over that time period. But seriously not since my 286, I've never see it happen, and even then it was just random panics and my cousin still wanted that machine when I was done with it. And you don't know where I get my numbers, is because you are too focused on the board, chips themselves, and missing the market opportunities. Heck the 2600x is going for $190 with $30 mobo discount so effectively $160 right now:
http://www.microcenter.com/product/505629/ryzen-5-2600x-36ghz-6-core-am4-boxed-processor-with-wraith-spire-cooler

There is not good reason to overpay. And I got my R5 1600 for effectively $100, and my R7 1700 for $140, because AMD had $30 rebates on top of Microcenter's mobo discounts. You just got to shop better.

Considering the tiny delta between 2600x at $160 vs the 2700x see:
http://cpu.userbenchmark.com/Compare/AMD-Ryzen-7-2700X-vs-AMD-Ryzen-5-2600X/3958vs3956

There is no rational justification for overpaying for the 2700x. And the 1800x has just as many cores and is effectively priced at $170. See:
http://cpu.userbenchmark.com/Compare/AMD-Ryzen-7-2700X-vs-AMD-Ryzen-7-1800X/3958vs3916

Unless the 2700x drops to $200 it is massively overpriced.


It is about getting lucky, and too date I've not seen anything actually release smoke or even smell burnt. There is no smoking heap. I'll admit I wrecked a $70 Athlon T-bird once beause I removed the HSF to prove a point to friend, he lost the bet I got my $70 for it.



I've not wasted my money. Time maybe because I'm perhaps a bit to OCD about things. And the prices have not gone up. People like me is what holds these companies accountable. Microcenter actually like me visting them plenty, I'd rather be more anonymous, but they actually seem to recognize me. They are happy to work with me on the returns and exchanges. And they gotten plenty of money from me. And same can be said for newegg, Best buy etc. But when Asus, ASRock, MSI, Gigabyte, etc. make claims about their board's OC capability they better be able to deliver. If their VRMs suck they need to be called out. They can eat their own crap boards, if they fail to deliver.




When stuff get old enough, I overclock them even more and cook the last bits out of them. Even then there is no smoke or burning heap, just instability. I just play abit aggressive, but I don't consider myself a crazy aggressive overclocker. None of the overclocks are about getting world records, and my voltage values are always on the safe side anyways. I just want cheap and easy. With PC tech stuff where they are now, where you don't need to set jumpers for voltage dividers or clock multipliers, and they got all these safeties built-in into the bios, this should be minimum people can do and easily do.

 

frizzlebyte

Honorable
Mar 21, 2013
26
0
10,530


Thanks to all of you for the response!

I'm using Prime95 v.29.4, build 8, on blend, no custom settings.

With regard to the voltage being "default", there is no "auto" option for voltage once you select custom from the BIOS menu. 1.2375v and a 3200mhz clock speed is what the board menu defaults to when you select custom on the menu. I didn't increase or decrease the voltage from that setting, left it as-is.

So far as seeing a voltage drop under load, that's precisely right, to the tune of 1.20v when under load. Not sure what that means in terms of vDroop. When I tested Realbench briefly (I'm back at stock until I have a bit more time to play with this next weekend), I noticed the voltage dropping to 1.19 and 1.18 at times, but there was no instability. The clock speeds were also reporting the same 3700mhz. What would voltage throttling look like in HWinfo? Temps at the time were in the 65-67 range, but as I said, I only tested briefly to see how Realbench works. I'll be checking this out more later.

Thanks for the help, and let me know what you think. I'm aware that this board isn't a beast, and the VRMs might not be particularly robust. That's one reason I'm not planning to push my overclock to insane levels. If I can get XFR-level speeds (3.7ghz on the Ryzen 5 1600) across all cores, I'll consider myself one lucky clam.
 
Sounds like you have realistic expectations are are willing to do what is needed. It does take time to configure and validate a stable overclock. The majority of people don't bother. That's why they are typically back here in about six months whining about "why am I getting all these errors all of a sudden?", "I think my motherboard is bad", "I think my graphics card is failing?", "I think I got a crappy CPU".

No, you have an operating system full of micro-errors due to not taking the time actually required to validate an overclocked configuration.

But then they do a clean install of Windows and now they are sure it was just a bad update.

No, you most probably corrupted the operating system with errors. This happens with unstable CPU configurations more than often enough to mention. And if you fiddle with your memory configuration, even setting it to XMP settings, regardless that a lot of people think you don't need to take the time to check stability and run tests afterwards, you do, and you should.

Anything above what the CPU and motherboard supports by default is still considered an overclock and the timings chosen by the DRAM manufacturer are no different than the settings used by motherboard manufacturers for automatic CPU overclock profiles. Yes, in both cases they WILL probably work. Also yes, there is every possibility that from module to module and motherboard to motherboard, there might be something that does not agree with the memory controller or any of ten other potential factors that could cause what seems to be a perfectly normal XMP overclock to be creating miniscule errors, a zero here where a one should be, two ones where there should have been a zero.

Before long at all you have a system that will begin giving you problems. It might not be blue screening, and it might, but there will be indicators of some kind. Simpler just to validate your memory XMP settings with one time runs (Hopefully) of Memtest86 and Prime95 blend mode. If you want to be even more thorough, run Memtest86 a second time after running Blend, and choose only test 6 and 13. This will check for row hammer which can be a common, but relatively unheard of issue on modern memory modules, especially at very high frequencies and very low latencies.

Run Realbench for 8 hours. If there are errors, the system is not stable and you need to make a change. It might be reducing the multiplier. It might be increasing the voltage. It might be that you enabled the XMP profile and your error is not actually due to the CPU overclock, but due to something memory related. Always leave the memory at the default configuration until you are completely through validating your CPU overclock settings and testing.

Also, ditch that version of Prime95. Download Prime95 version 26.6.

Run version 26.6 and choose the Small FFT option. If it remains within the thermal specifications for your CPU for 15 minutes, then it is thermally compliant at it's current overclock configuration. Do not confuse thermally compliant with stable. They are different. Thermally compliant just means that your CPU, motherboard and cooling system are capable of withstanding the voltage and multiplier that are currently assigned and that you are not going to have to worry about thermal conditions unless you are running an unrealistic AVX workload like in versions of Prime95 newer than version 26.6. Actual applications and utilities that use AVX instruction sets do not use them in the same way that Prime versions that use them do.

Prime versions newer than 26.6 do not present realistic workloads. You would never see those temps in any practical application or game environment, ever. Trust the thermal compliance check using version 26.6 for 15 minutes.

So, Prime for 15 minutes. Thermally compliant, for THAT particular setting. If you change the multiplier or voltage, you need to do this step all over again, and you will likely do that during the process of determining where your overclock has reached a plateau.

Then, Realbench on the stress test using the setting for 50% of your RAM, for 8 hours. Do not cut corners here. If you cannot run Realbench for 8 hours, your overclock is unstable. There is no two ways about this.

Next, when you have determined that everything is good, THEN, and only THEN, at least in my opinion, should you enable your memory XMP settings or do any custom configuration of memory speed, voltage and/or timings. Once you begin doing that, you need to run Memtest86 for at least two passes after ANY change you make to the memory configuration to establish that the settings are at least basically stable. Follow that up by running Prime95 version 26.6 and choose the Blend option or the Custom option. I prefer the Custom option. If you choose the custom option, input approximately 75% of the amount of RAM that is currently NOT in use, and change the Minimum FFT size to 512(k).

Run that, either on Blend or on the custom option, for a minimum of 8 hours. If there were problems passing the Memtest validation then there is a problem with either the memory itself, or the configuration. You can try adding a small, minimal amount of additional voltage if they refuse to run at XMP settings. If you cannot gain stability at XMP settings without adding more than .020v to the stock XMP voltage value, then I'd look at whether there are issues with the modules or not.
 
Solution
^ If you maintain stability after darkbreeze's info I would say you have a blinder of a chip there - that's 'golden sample' result wise.

Vdroop honestly does not matter a jot if you're 100% stable .

Honestly though I've used that board half a dozen times & it's always overcompensate voltage wise.

I'm running a way older bios though.
 

Rogue Leader

It's a trap!
Moderator


Discontinued is not marketing, discontinued is no longer being manufactured, thats the definition. Literally... in the dictionary

https://dictionary.cambridge.org/us/dictionary/english/discontinued

AMD still supports FX processors and Phenoms with patch files. They are very much discontinued.

Supply is low, they are no longer being made, it will run out.

As for the rest of your post I am not going to continues this argument as its not helping the OP one bit, and he is getting help from some of our most knowledgeable overclockers. He has seen where the right information comes from and thats all that matters. Good luck.