-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
5800X Boost set to 5050Mhz all cores with AMD Curve Optimizer
- Thread starter gerardfraser
- Start date
I get 4.5ghz all core boost and about 4.8 single core boost. Think i’ll just leave defaults for now as it doesnt seem like there are much gains from tweaking
Thanks - makes sense. Areeeeee, I see your single core boost is to 5050Mhz, not 5050Mhz on all cores!! I wasnt trying to push mine to 5ghz all cores
I managed 4.9ghz with 1.475 on all cores but only 1 Cinebench R20 runNope, all core hits around 1.325v VID when all cores loaded at 4.5Ghz
- Joined
- 27 Apr 2014
- Posts
- 859
You are confusing me. This is exactly how the AMD Ryzen CPU is supposed to work. I set AMD Curve OPtimizer to boost to 5050Mhz on any core when boost,it is not supposed to be all at the same time. That is a different form of overclocking and has nothing to do with this thread.
Although AMD 5000 CPU's are a little different on voltages from previous AMD Ryzen CPU's UK_SPAWN is correct the Ryzen CPU voltage range at default is 0.200v-1.5v and depending on PBO and agesa BIOS can go up to 1.55v
I am not trying to say anything here if you want ,I suggest maybe check out some more Ryzen stuff and get more familiar with Ryzen the way it works. It may or may not help but you already have the fastest CPU out of the box if you just let it do its own thing with boosting and running.
Like I said in this thread a few times,if you like big numbers then a thread like this is good, there is no real gains in performance from default and you even mention it in the above post.The AMD Ryzen CPU is still as fast as any Intel CPU @ 5300Mhzin PC gaming/applications. Do not sweat the littles things that do not mean crap and is all in your head with having the absolute best score in everything,your not missing a thing with your AMD Ryzen setup.
Leave it default and enjoy your CPU
Soldato
Yeah we hear you and yes I will probably leave it at default but you can't dangle an overclock suggestion such as this over our heads and then say dont swet it when we give it a try and it doesnt happen on our kit !
Altering the PBO settings like you suggested on my M/B did sweet franny adams apart from it lowered the CPU vcore which is really weird and not what I expected. So its not quite as straightforward on my M/B as it is with yours. It would be nice to see if I could get a similiar result, simply because I can (thats what overclocking is all about) but even if I did manage over 5ghz on one of the cores I would probably change it back to default anyway for the long term.
I've ordered a 240mm AIO so I might have another play (because I can) when I've fitted it.
Sorry if i confused you. I confused myself as thought it was 5050Mhz all core (all at the same time) boost. I now know I just read it wrong.
Thanks
- Joined
- 27 Apr 2014
- Posts
- 859
No it is OK ,just had to read it a few times,I am not that swift to get it first time.
LOL I made the thread to let people know what is possible, so everyone can try. If they do not get it or their motherboard does not want to do it then default is perfectly fine in my opinion.Yeah we hear you and yes I will probably leave it at default but you can't dangle an overclock suggestion such as this over our heads and then say dont swet it when we give it a try and it doesnt happen on our kit !
Altering the PBO settings like you suggested on my M/B did sweet franny adams apart from it lowered the CPU vcore which is really weird and not what I expected. So its not quite as straightforward on my M/B as it is with yours. It would be nice to see if I could get a similiar result, simply because I can (thats what overclocking is all about) but even if I did manage over 5ghz on one of the cores I would probably change it back to default anyway for the long term.
I've ordered a 240mm AIO so I might have another play (because I can) when I've fitted it.
It is overclocking after all fun to play with.
@gerardfraser I managed a 641 R20 single core run with +150Mhz PBO. Not too bad, thanks for your help
Out of interest, are people running the R20 bench with other things overclocked too? i.e. RAM. Not sure if it has any effect or not but im running 3600 CAS 16 memory @CAS14 which gets 53.9 ns memory latency
Out of interest, are people running the R20 bench with other things overclocked too? i.e. RAM. Not sure if it has any effect or not but im running 3600 CAS 16 memory @CAS14 which gets 53.9 ns memory latency
Last edited:
- Joined
- 27 Apr 2014
- Posts
- 859
Ram has no effect in Cinebench but you can always set 2133Mhz and test
What settings you have for your ram to run @cl14?
- Joined
- 27 Apr 2014
- Posts
- 859
You could always put your ram to 2133Mhz for a day and then you would know for sure.
Soldato
According to Custom PC Mag when they benched the 5xxx series the speed of RAM did make a difference. I am running 3000mhz and Guest2 is running 3600mhz so that may make sense why his results in Cinebench R20 are higher than mine ?
Caporegime
It makes very little difference from what I have seen, but it doesn’t hurt to have faster memory and adds a few extra points on.
- Joined
- 27 Apr 2014
- Posts
- 859
Great thing about the internet you can always find what your looking for. Let me go another step further and blow your mind. AMD Ryzen FCLK speed is a myth. Tune your ram best as you can and there is no real performance difference same as in Cinebench ,now I can not say all applications because I did not test every one.
I could fill up the thread with lots of benches I did but no point really.Well here is one for fun in PC games.
The reason why I say this,I see people going out and buying expensive ram for no real gains.If you have a specific need for faster ram ,then I am no ram destroyer ,just sharing an opinion. I am an idiot,I have no less than 6 different Ryzen CPU's ,4/5 sets of ram/10+ gpus/NVME/SSD/Hard drives . I like playing with gear and playing PC games.
AMD Ryzen 1467Mhz (DDR4 2933Mhz) vs Fabric Clock 1933Mhz (DDR4 3866Mhz)
BF5 GTAV 1920x1080
@gerardfraser
I have a favour to ask.
If i am not mistaken, from what i saw in your videos, your VDDP voltage is pretty low @0.7v?
I dont have a sample to test it, but i think that some of voltages left on auto(board just overvolts) are making 5800x ''overheat''. My bet is that its VDDP, IOD or CCD one, or maybe even all 3. Last bet is LLC left on AUTO, or being 1. It all depends board from board.
So my favour is if you can test all of them on minimum by CRT for your ram speed, and then lets say all on 1.05v, and then compare temps. If they are the same, then i am wrong, its much bigger issue then i thought.
If i am right, high temps that are reported by many can be fixed in like 10 minutes of testing, if even that much.
I have a favour to ask.
If i am not mistaken, from what i saw in your videos, your VDDP voltage is pretty low @0.7v?
I dont have a sample to test it, but i think that some of voltages left on auto(board just overvolts) are making 5800x ''overheat''. My bet is that its VDDP, IOD or CCD one, or maybe even all 3. Last bet is LLC left on AUTO, or being 1. It all depends board from board.
So my favour is if you can test all of them on minimum by CRT for your ram speed, and then lets say all on 1.05v, and then compare temps. If they are the same, then i am wrong, its much bigger issue then i thought.
If i am right, high temps that are reported by many can be fixed in like 10 minutes of testing, if even that much.
- Joined
- 27 Apr 2014
- Posts
- 859
Well I only use auto voltages for all settings when I am posting anything related to PBO,AMD Curve Optimize even all core overclocks. LLC is all auto everything auto except on all core overclock I have to change CPU voltage to 1.3v for 4800Mhz.
Yes I agree with you about some motherboards overvolting on AMD CPU 5800X but some people use the same motherboard and still overheat. I am going to assume you responded to the thread in Techpowerup on the two video I show voltages and Temperatures. I have no problem testing on minimum at all and I would only get lower temperatures but currently testing a set of Ram for errors for someone. It is at 1 hour in the test and I usually do the test for 10 hours ,then it is Saturday I have begun drinking some beers and after the 10hr test,this fat cat will be too drunk to test minimums . So I say it would be tomorrow evening at minimum before I could do the test.For now three video all auto voltages,no LLC with Hwinfoinfo64 on.
4800Mhz all core CPU voltage- 1.29v Temperature- 72.5 ℃ HWinfo64 on and all voltages auto except CPU voltage- 1.29v
Effective Clock 5800X at 5100 Settings voltages are all auto voltages and on low side by the looks of it.
AMD 5800X 5050Mhz AMD Curve Optimizer Cinebench R23 -This ran for 30 minutes and hit 79.8 ℃ all auto voltages
Yes I agree with you about some motherboards overvolting on AMD CPU 5800X but some people use the same motherboard and still overheat. I am going to assume you responded to the thread in Techpowerup on the two video I show voltages and Temperatures. I have no problem testing on minimum at all and I would only get lower temperatures but currently testing a set of Ram for errors for someone. It is at 1 hour in the test and I usually do the test for 10 hours ,then it is Saturday I have begun drinking some beers and after the 10hr test,this fat cat will be too drunk to test minimums . So I say it would be tomorrow evening at minimum before I could do the test.For now three video all auto voltages,no LLC with Hwinfoinfo64 on.
4800Mhz all core CPU voltage- 1.29v Temperature- 72.5 ℃ HWinfo64 on and all voltages auto except CPU voltage- 1.29v
Effective Clock 5800X at 5100 Settings voltages are all auto voltages and on low side by the looks of it.
AMD 5800X 5050Mhz AMD Curve Optimizer Cinebench R23 -This ran for 30 minutes and hit 79.8 ℃ all auto voltages
- Joined
- 27 Apr 2014
- Posts
- 859
Of course but remember a few pages back where I showed clock stretching like this. Post #90 but try it anyway to test yourself.
Missed that post, there are so many
So clock stretching is just running at a manual low voltage like 1.3 V instead of default which can ramp up to 1.5 v automatically?