[SOLVED] What to expect with the 3800x OC'd using the Scythe Mugen 5 Rev.B Air Cooler

[Jason H.]

So recently I upgraded my cooler from the Wraith Prism to the Scythe Mugen 5 Rev.B.

I was looking online for info on this specific cooler on my specific CPU, the R7 3800x, but I couldnt find much, if any info on it.

So I decided to make a video doing a 5 minute pass on Cinebench (longer than 5 minutes doesnt seem to affect temps any longer) and post it on Youtube so anyone looking for what they can expect from the cooler can find something on it for this CPU.

This video is with a all core OC of 4.3ghz at 1.325v(idle) and something like 1.28v under load (watch video for exact number).

Cinebench is not the only tests Ive ran but its the only one I put on video as I was told running Prime95 and Aida64 will take a lot of lifespan off my cpu, so for simple everyday use "worst case scenario" or "gaming" type testing i just used Cinebench R20 for this video.

Max temp was 79.5c with 6 passes of Cinebench (85c with Prime95 Small FFT AVX disabled)

This is all with an ambient temp on average of 70 degrees F or, 21c. Internal case temp at max is 32c.

Gaming

Although not included in the video, while playing Apex Legends I seen a maximum temp of 75c (which must have been a spike during a loading section because I never visually seen it happen) with an Average temp of 58c via HWInfo. The temp I would normally see in game using RTSS was 61c with drops to 50c or spikes to 68c. This was at 40-55% CPU usage.

A well known game called "Rust", which is very CPU bound, had a max temp of 78c (again I didnt visually see it) with an Average temp of 61c. With RTSS in game I normally seen 62c.

Here is the Cinebench test using the 4.3ghz OC at 1.325v with the Scythe Mugen 5 Rev.B Air Cooler. Recommend FULLSCREEN for text clarity.

View: https://www.youtube.com/watch?v=5YVqvtWhQO4

Conclusion

In my personal opinion these temps seem perfectly fine with an all core OC. I was told they seem a bit high compared to the stock cooler but Ive also been told that it is on par for what the cooler is. Not all chips are made exactly the same so there can be some differentiating results when compared. This is just what I have came to find and your results may differ from my own.

With that said, if your not overclocking, I would stick with the stock cooler unless you can afford a better one.

PC SPECS

MOBO: MSI Tomahawk B450
CPU: Ryzen 7 3800x (4.3ghz 1.375v)
Cooler: Scythe Mugen 5 Rev.B
GPU: Gigabyte RTX 2060 Super OC 3x Windforce White (8pin only version)
MEMORY: HyperX RGB 2x8gb DDR4 3200mhz (OC'd)
Storage: Samsung 860 Evo 250gb / 500gb Seagate / 1tb Samsung 860 Qvo
PSU: Corsair cx550m
WIN: Windows 10 Pro
Case Cooling: Antec Prizm 3x 120mm rgb case fans with controller and 2 RGB strips / 1 default 120mm case fan
Case: Cougar MX330

 

[velocityg4]

It looks like the CPU fan was barely spinning up. I'd be curious what temps you experience with it set to 100% along with case fans.

 

[Jason H.]

It looks like the CPU fan was barely spinning up. I'd be curious what temps you experience with it set to 100% along with case fans.

The CPU fan has a low RPM as its supposed to stay quiet, but cool.

 

[Karadjgne]

Programs won't change the physical lifespan of your cpu. Not sure where you got that info, but it's not correct. It takes physical affects to physically affect a cpu, for example, running Prime95 with AVX 24/7/365 at 98°C consistent on an intel cpu isn't healthy, but it's the resultant heat/voltage that affects the cpu lifespan, not the program itself. With adequate cooling, it'd be no different than any other program.

With that Mugen, no worries running Prime for testing purposes.

 

[Jason H.]

Programs won't change the physical lifespan of your cpu. Not sure where you got that info, but it's not correct. It takes physical affects to physically affect a cpu, for example, running Prime95 with AVX 24/7/365 at 98°C consistent on an intel cpu isn't healthy, but it's the resultant heat/voltage that affects the cpu lifespan, not the program itself. With adequate cooling, it'd be no different than any other program.

With that Mugen, no worries running Prime for testing purposes.

A guy on this thread told me that info. https://forums.tomshardware.com/threads/scythe-mugen-5-rev-b-temps-on-a-3800x.3586159/

Another member (moderator) named Darkbreeze also told me that info about running Prime95 is, in his words, BS. lol. I was just going off what this guy I think named "Count Mike" on that thread told me.

 

[Karadjgne]

Ahh ok, no he wasn't wrong, he just worded it awkwardly. He was referring to your OC + Prime95 resulting in 102°C. Not that Prime95 is bad per se, just that running it with a large core cpu, will full cores, OC and inadequate cooling isn't healthy.

In my experience, and thats 40 years now of messing around in pc guts, both professionally and amateur, there's only 2 programs I rely on. Prime95 for temps only and Asus RealBench for stability.

Prime95 small fft w/out AVX enabled drops the absolute cleanest 100% cpu load of any program. No bounces, no mix of ram to speak of, so any temp you end up with is as good as it gets for maximum normal gaming abuse. There are some games that do make use of AVX but it's minimal. So p95 small fft is great for seeing if your cooling is acceptable.

Asus RealBench tests everything. Cpu, gpu, ram, storage, pcie bus, USB, everything in combinations. Something p95 can't do. So it's good for a multipurpose 'let's trip up the cpu' test for stability.

I'm sure others have other opinions, those are mine.

 

[Jason H.]

Ahh ok, no he wasn't wrong, he just worded it awkwardly. He was referring to your OC + Prime95 resulting in 102°C. Not that Prime95 is bad per se, just that running it with a large core cpu, will full cores, OC and inadequate cooling isn't healthy.

In my experience, and thats 40 years now of messing around in pc guts, both professionally and amateur, there's only 2 programs I rely on. Prime95 for temps only and Asus RealBench for stability.

Prime95 small fft w/out AVX enabled drops the absolute cleanest 100% cpu load of any program. No bounces, no mix of ram to speak of, so any temp you end up with is as good as it gets for maximum normal gaming abuse. There are some games that do make use of AVX but it's minimal. So p95 small fft is great for seeing if your cooling is acceptable.

Asus RealBench tests everything. Cpu, gpu, ram, storage, pcie bus, USB, everything in combinations. Something p95 can't do. So it's good for a multipurpose 'let's trip up the cpu' test for stability.

I'm sure others have other opinions, those are mine.

OK ty for that info!

Something very weird happened today. Someone suggested on my video to turn on PBO as their Cinebench score on a 3700x with no OC was higher than my 3800x with the manual oc. So I turned on PBO with the manual OC and Im seeing a max temp of 75c with a score about 200cb higher than the one in my video....

This is very strange to me.

Not sure if this is because I havent used the PC all day so Im seeing a lower temp which is giving me a higher score, or if the PBO is actually doing something that is helping have a lower temp with a higher score..

 

[Karadjgne]

Pbo optimizes voltages and best speeds. So as long as temps are decent, you'll get the most out of the cpu. Not hard to see lowered temps, many times even stock voltages are considerably higher than is necessary for a cpu, with or without OC, and then ppl OC the cpu and add even more unnecessary voltages, while forgetting or not realizing other aspects of OC which would enhance stability at a lower voltage.

 

[Jason H.]

Pbo optimizes voltages and best speeds. So as long as temps are decent, you'll get the most out of the cpu. Not hard to see lowered temps, many times even stock voltages are considerably higher than is necessary for a cpu, with or without OC, and then ppl OC the cpu and add even more unnecessary voltages, while forgetting or not realizing other aspects of OC which would enhance stability at a lower voltage.

Well, does PBO still work if I have the clock and voltage manually set?

Thats my thing, with PBO enabled, WITH the manual OC, the temp is lower, and the CB score is 200 points higher.

And if it DOESNT work with the manual OC enabled, then idk how I have a lower temp/higher score today, vs the video lol.

 

[Karadjgne]

You have manual OC set, as far as you can see. Which might be a few user settings. What you don't see is the multiple cpu set settings in the bios, which are changed by pbo, windows, other settings. There's a lot more to OC than just changing the multiplier and vcore and llc. Things like min/max load amperages, turbo amperage timers, vid, ring voltages, package voltages, phase locks etc and some of those Are accessible on some bios like used in the asus ROG extreme and msi Godlike. But even then, 98/100 ppl will not change them from auto or optimized settings, only the pros and record seekers have a clue what half that stuff does.

But pbo does too.

 

[Jason H.]

You have manual OC set, as far as you can see. Which might be a few user settings. What you don't see is the multiple cpu set settings in the bios, which are changed by pbo, windows, other settings. There's a lot more to OC than just changing the multiplier and vcore and llc. Things like min/max load amperages, turbo amperage timers, vid, ring voltages, package voltages, phase locks etc and some of those Are accessible on some bios like used in the asus ROG extreme and msi Godlike. But even then, 98/100 ppl will not change them from auto or optimized settings, only the pros and record seekers have a clue what half that stuff does.

But pbo does too.

Hm, that makes sense.

I dont suppose going for an all core 4.4ghz then would be better than 4.3 with the PBO on lol

 

[Karadjgne]

Well the thing about pbo is it'll only boost what it has to, within voltage and temp restrictions. It's like a 'smart turbo' in that if temps are getting up there, it'll relax a little on the boost. And it'll do that for as many cores as necessary, and park the rest. With an all core 4.4GHz, you are locking the cores at 4.4GHz, all of them, so they'll be running at 4.4GHz even when not really doing anything. Which does nothing more than add to temp values.

For the 100MHz, honestly it's not worth it imho, you get almost identical performance, better temps, less stress on the system, the VRM's, less noise, nothing but bonus's by staying with pbo.

 

[Jason H.]

Well the thing about pbo is it'll only boost what it has to, within voltage and temp restrictions. It's like a 'smart turbo' in that if temps are getting up there, it'll relax a little on the boost. And it'll do that for as many cores as necessary, and park the rest. With an all core 4.4GHz, you are locking the cores at 4.4GHz, all of them, so they'll be running at 4.4GHz even when not really doing anything. Which does nothing more than add to temp values.

For the 100MHz, honestly it's not worth it imho, you get almost identical performance, better temps, less stress on the system, the VRM's, less noise, nothing but bonus's by staying with pbo.

On my toubleshooting thread about this issue someone suggested to turn off the OC as PBO would likely hit 4.4ghz with better temps and the same, if not better cinebench score.

With PBO ENABLED, no OC (settings on Auto) it only boosts to 4.15 in cinebench, and my score is almost 300 points lower. I hit 5111 yesterday with my OC and today, with how he told me to set it up, is 4815 and is hotter by 2c (77c) because it uses a higher voltage to achieve the 4.15 (1.44-1.5v idle, 1.31-1.33v under load) than I use to achieve 4.3 steadily accross the board (1.325v idle, 1.26-1.28v under load)

Just tested it just now.

SO how is it not boosting higher than 4.15? Whats wrong... Something HAS to be wrong... lol I havent messed with any other options besides fan curves and set xmp on the memory so reverting bios to stock doesnt seem worth while either.

 

[Karadjgne]

Welcome to the wonderfully confusing world of OC. My i5-3570k at a 4.3GHz OC ran cooler and lower voltages than the stock settings at 3.4GHz

The 3800x should boost to @ 4.5GHz on a couple of cores, and then that'll drop some with additional core brought online. With the pbo and OC, sounds like you were setting a minimum base clock, and then allowing pbo to work from that with the boost. Returning to stock values, you wipe out that minimum value and replace it with a stock value, while simultaneously bumping voltages as OC you tailored the voltages, stock won't, it'll apply a higher than necessary voltage to guarantee stability, which raises temps, and at higher boost throttles back sooner.

Sounds like your best performance will be a mix of the pbo applied on top of a decent OC, something that drops the voltage ranges and consequent temps, allowing for higher boosts.

 

[Jason H.]

Welcome to the wonderfully confusing world of OC. My i5-3570k at a 4.3GHz OC ran cooler and lower voltages than the stock settings at 3.4GHz

The 3800x should boost to @ 4.5GHz on a couple of cores, and then that'll drop some with additional core brought online. With the pbo and OC, sounds like you were setting a minimum base clock, and then allowing pbo to work from that with the boost. Returning to stock values, you wipe out that minimum value and replace it with a stock value, while simultaneously bumping voltages as OC you tailored the voltages, stock won't, it'll apply a higher than necessary voltage to guarantee stability, which raises temps, and at higher boost throttles back sooner.

Sounds like your best performance will be a mix of the pbo applied on top of a decent OC, something that drops the voltage ranges and consequent temps, allowing for higher boosts.

I managed to get 4.4ghz at 1.28v to run stable with a max temp of 81c after like 10 runs of cinebench with a score of 5207.

That is just so nuts to me compared to with only PBO, only boosting to 4.15ghz at 1.31v, getting a max temp of 81c and only a 4807cb score.

I am so beyond confused with these new results lol.

Maybe its my motherboard. I dont really have cash for a new one and lots of people say the Tomahawk b450 is good and has good VRMs. Although I did have to update the BIOS for the new R7 chip. But even that doesnt explain the boosting issue unless the BIOS is the problem... So weird lol.

 

[Karadjgne]

Did you update the bios in steps according to directions? Or just skip to the latest?

 

[Jason H.]

Did you update the bios in steps according to directions? Or just skip to the latest?

All of the directions I have seen or read just shows them going and downloading the newest BIOS and installing it.

Previously when I was running my 2600, I decided to update the BIOS to the latest version as I had planned on getting the R7 and just wanted to be prepared for when I got it. But the new BIOS (same one I have now) ONLY while running with the 2600, would not come out of sleep mode. I could restart the PC just fine though but anything that made the PC lose all power, it would not come back on without resetting the cmos jumper. I have no idea why... but so when I went to downgrade the BIOS back to the known working BIOS for my 2600 it wouldnt allow me to install the ORIGINAL BIOS that came with my MOBO because the BIOS's UI had changed and idk it jsut wouldnt let me. SO I had to install the NEXT BIOS (the one released after the original BIOS) and that one worked fine with the 2600, reverted back to the older UI (so much better). Then when I got my 3800x, I again installed the latest BIOS. I did not install each BIOS release up until the latest one.

With that said, while IN GAME, the cpu DOES boost up to 4.3ghz (never boyond) and this is at 58-65c. Someone told me once temps go above 70c PBO shuts off, hence why my clocks are staying stuck at the low 4.15 while running cinebench at 77c. So maybe thats it?

It just boggles my mind how they advertise 4.5ghz boost clock, but you need like a dry ice cooling configuration to even get the CPU cool enough to boost past 4.15ghz like mine lol. I mean this cooler was like 50$ I would expect it could have at least got to 4.3ghz while being stress tested at stock settings.

 

[Karadjgne]

Ahh damn. Depending on the original bios version, updating straight to the newest only applies to brand new made motherboards and Intels, not amd.

With Amd bios needs to be stepped on almost every version except the MAX. When the factory installs the bios, it's a complete bios with only 1 thing, the bios info. Updates that follow are different as they'll contain not only any fixes to the bios, but also the instruction sets on what to fix. Unfortunately the bios chip is only so big and can contain only so much at any given point.

So you'd have a 2nd Gen, B450 bios with Agesa 1.04 and ram specs XYZ and all of the ami bios code necessary for your mobo. The latest update will include all the info for Agesa 1.06, all the ram specs ABC and the instructions to remove lines aa11, bb22, cc33, add all the 3rd gen cpu info, change ram specs 123 to ram specs ABC etc. Because of size limitations, it doesn't include Agesa 1.05, ram specs 123, instructions to change ram specs XYZ to 123 etc etc.

Consequently, you get Agesa 1.06, there's nothing to change ram specs from XYZ, so those remain intact, the changes from ram specs 123 don't happen because there is no 123 to change and you added all of the 3rd gen info which really squeezes out much 1st gen and some 2nd Gen info to make room for it.

End result is you get a 1st gen bios, with a 3rd gen tag, and 2nd Gen ram characteristics and cpu abilities on a 3rd gen cpu.

If you can revert the bios, I'd do so, or if applicable swap to 2nd bios and take note of the bios version number. Then go to msi/support/bios and find that version, look at the next version newer to see if there are any important steps to take first.

So if you had 7C02v11 you'd need to update the driver package first, then update to 7C02v12 etc.

It'll all depend on the actual manufacture date as to which bios version you had, the date you bought the board is irrelevant.

 

[Jason H.]

Ahh damn. Depending on the original bios version, updating straight to the newest only applies to brand new made motherboards and Intels, not amd.

With Amd bios needs to be stepped on almost every version except the MAX. When the factory installs the bios, it's a complete bios with only 1 thing, the bios info. Updates that follow are different as they'll contain not only any fixes to the bios, but also the instruction sets on what to fix. Unfortunately the bios chip is only so big and can contain only so much at any given point.

So you'd have a 2nd Gen, B450 bios with Agesa 1.04 and ram specs XYZ and all of the ami bios code necessary for your mobo. The latest update will include all the info for Agesa 1.06, all the ram specs ABC and the instructions to remove lines aa11, bb22, cc33, add all the 3rd gen cpu info, change ram specs 123 to ram specs ABC etc. Because of size limitations, it doesn't include Agesa 1.05, ram specs 123, instructions to change ram specs XYZ to 123 etc etc.

Consequently, you get Agesa 1.06, there's nothing to change ram specs from XYZ, so those remain intact, the changes from ram specs 123 don't happen because there is no 123 to change and you added all of the 3rd gen info which really squeezes out much 1st gen and some 2nd Gen info to make room for it.

End result is you get a 1st gen bios, with a 3rd gen tag, and 2nd Gen ram characteristics and cpu abilities on a 3rd gen cpu.

If you can revert the bios, I'd do so, or if applicable swap to 2nd bios and take note of the bios version number. Then go to msi/support/bios and find that version, look at the next version newer to see if there are any important steps to take first.

So if you had 7C02v11 you'd need to update the driver package first, then update to 7C02v12 etc.

It'll all depend on the actual manufacture date as to which bios version you had, the date you bought the board is irrelevant.

Well I followed the directions directly from MSI (they made a video about how to flash their BIOS's) and they did not even come close to mentioning that I had to do it in steps.

That is some very useful info for the future.

I dont think Ill revert to an older BIOS and do it in steps, and heres why.

On my troubleshooting thread we have found out that since my PBO is boosting to 4.3ghz while gaming at 58-65c, that PBO IS IN FACT, working properly.

The whole time they forgot to mention that PBO turns off past 70c, and that I will NEVER be likely to see a high boost clock while stress testing using things like Prime95, Aida, Cinebench, ETC. with the cooler I am using.

This all came about because I was told 81c at stock settings, with this cooler, is still too high as the stock cooler was maxing out at about 85-86c.

I bought this cooler via a suggestion from a Moderator, and my budget wasnt very high. All in all it dropped temps by about 8-10c on average so I mean it does its job as a 50$ cooler. But I wish I would have known ALLLLLLLLLL of this info before purchasing the cooler as I paid 61$ for it to get it off Ebay because Amazon was delaying shipments, so I paid a premium to get it faster. I could have just saved another 60$ and got a much better water cooling setup.

 

[Karadjgne]

The Mugen is a 200w+ TDP cooler, it'll run neck and neck with the better 240mm aios in performance mode (noisy) or a decent 280mm in balanced mode. So 'better water cooling setup' isn't really a thing. That cooler will easily keep up with a NH-D15 at anything less than @ 200w, and even with pbo you'll only be pushing @ 100w ish at best.

So 'better water cooling setup' would have run you @ $150-$200ish or so, and your temps would be within @ 2°C.



The mod who suggested the cooler (Darkbreeze?) really didn't stear you wrong, the scythe is a great cooler choice for an air cooler, but do understand a 3800x isn't exactly a cool running cpu once all cores are brutalized in stress tests which in some will run upto 130% virtual stress temps with AVX enabled

 

[Jason H.]

The Mugen is a 200w+ TDP cooler, it'll run neck and neck with the better 240mm aios in performance mode (noisy) or a decent 280mm in balanced mode. So 'better water cooling setup' isn't really a thing. That cooler will easily keep up with a NH-D15 at anything less than @ 200w, and even with pbo you'll only be pushing @ 100w ish at best.

So 'better water cooling setup' would have run you @ $150-$200ish or so, and your temps would be within @ 2°C.



The mod who suggested the cooler (Darkbreeze?) really didn't stear you wrong, the scythe is a great cooler choice for an air cooler, but do understand a 3800x isn't exactly a cool running cpu once all cores are brutalized in stress tests which in some will run upto 130% virtual stress temps with AVX enabled

Oh yea Im not saying he made the wrong suggestion. I just wish I would have known all of the info I know now about PBO and boosting before I bought it as I would have just gave him a higher budget to make a suggestion from. I had only told him 50-70$ so I think he did good in those regards!

I dont see the H100i on that list, would that be a good cooler to upgrade to? or is there any "worth" upgrading to within 150$?

Of course Im aware I wont get a "150$ improvement" but even if it dropped the temps by like another 5-10c I would be satisfied finally lol.

It would be a shame if I could really only get a 2c improvement from 150$ lol. I have always thought water cooling was some mystical legendary stuff that I just never got into lol.