My 6700K would undervolt to a tremendous degree, just as it would run 3200-rated RAM very high. But, no matter how much stability testing I would do the machine never seemed to stabilize. Prime would run at the low voltage but there would be spurious problems with software. RAM would test with multiple instances of HCL, HCL with Prime, etc. But, again... random spurious problems. So, I went back to defaults.
It's a good thing this procedure happens automatically with lots of fast adjustments. I have given up on the coarse (human-derived) approach, both for undervolting and overclocking. I have decided it's simply a waste of time. I have never found true stability with anything but defaults and I have spent a huge amount of time with a variety of systems, almost finding stability but never getting it. As one ages one puts more value in one's time and so it had better be "automagic" for me going forward. Kudos to AMD if this works well.
I have been amazed by the auto OC on non-K chips achieved with the i7-10700, and a little upset as well as the chip outscores my OC i9-9900KF that is watercooled(Passmark CPU score for the 10700 beats the 9900KF by 8%). All I did on the 10th gen Asrock motherboard was tell it to treat the 10700 as a 135W CPU and it did all the rest(and I did enable XMP as well). I have the 10700 cooled by a Noctua NH-U14S and temps are in the mid 60s when loaded. So the auto OC magic is getting really good from my point of view, when looking between 9th gen intel and 10th gen. Can't speak of AMD side yet but there is less of a reason to individually tweak settings with this kind of magic (and on the Intel side I don't see a reason to get a K part when the non-K already have turbo freq set at the edge/top of the CPU capability.
What you describe is not overclocking in any way, it’s just throttling due to power limits less. It’s not going above its rated boost clocks. Not the same as AMD’s auto-overclock stuff which will go higher than its rated boost clock, and now apparently will also auto-undervolt.
Additionally, your i9-9900K only can be slower than an i7-10700 if the i9-9900K is throttling, which is pretty sad for an unlocked processor. It is 100 MHz faster all core and 300 MHz faster single core if it’s not throttling. And it can overclock to past 5 GHz (maybe with an auto-tuning thing in your bios if you didn’t want to much around) - but that’s useless if it’s throttling at stock like it is for you. If you actually wanted auto-overclocking capability an AMD chip would be a much better fit for you, as it does it without any user intervention once enabled (see the AMD 5000 series review here for the 5950X’s 5050 MHz clock)
It does not mean his 9900k is throttling, a 10700 can easily beat a 9900k with differences such as motherboard, memory, etc. and o/c potential. It's also a 65w processor that may have a lot more headroom than you realize, which shows in the passmark score and his own testing which you obviously haven't done. That being said, it's all going to be within a couple of percentage points overall, so nitpicking his comment doesn't really make any difference, nor does telling him to buy a processor you can't buy right now.
I dunno, it does sound like the 9900KF is being held back. Depending on how he's overclocking it, he (or the board) might be getting too frisky with voltage, causing it to bang into thermal limits sooner than a well-tuned OC. "Watercooled" doesn't really mean much either by itself - an NH-U14S with good case cooling can actually beat the snot out of your average CLWCs. Combined with the aforementioned potentially-poorly-tuned voltage and bam, you can't dump the heat fast enough to run at the chip's full potential.
Yeah I suppose he could be using much slower RAM, but you'd think he would have thought to mention that. Alternatively he's got some settings buggered up, like he did a BIOS update and forgot to re-enable XMP. A lot of fast RAM kits have really low JEDEC defaults. The motherboard doesn't make a ton of difference as long as they both have good power delivery, and I'm not sure what "etc" factors in to CPU performance. The only other thing I could think of is, did they manage to slip in any additional hardware security mitigations in the refresh?
The 9900KF of mine is got to be the worst k chip I have ever touched, it is barely stable on an all core 5GHz boost, but I need to set the voltage to 1.4+. XMP is enabled (Which I find helps little in regards to the passmark CPU score) and it is running much faster storage than the other computer (Optane vs SX 8200 pro). Its on a custom water loop with a 2x 480 radiators which also cool a GTX 1080, I assure you a much more capabile solution than the noctua and keeps the temps frigid(although the room heats up...). The mobo is a Z390 M Gaming from Gigabyte. I am planning on replacing it with a 10850k here shortly. I likely jsut got a lemon on the 9900KF and I am wholeheartedly impressed with the non-k chips on comet lake, so much so I would consider a plain old 10900 over the 10850k.
I feel pretty much the same on all counts. I've had a number of stable overclocked and/or undervolted systems over the years, but it's become an increasingly rare thing - progressively more squeezing required for less juice. This would be nice, if it can reliably produce a stable system.
I agree 100%. All the angst over CPU overclocking is (or should be) of concern to very few these days. I'm not going to spend hours of tweaking and testing for a small improvement in full-load performance on a machine that runs at 10% load 99.9% of the time.
AMD is offering exactly what I want in this and every other regard. The fact that an Intel chip may "OC better" is laughably irrelevant, and fixating on it only serves to confuse newcomers.
A few years back when I (regrettably) rebuilt my PC around an i7-8700 and an Asus motherboard only because at the time there were no decent ITX solutions for the Ryzen, I noticed insanely bad temps for this 65w CPU compared to the 84w or whatever my previous i7-4770 was. This is a basic closed loop setup in an ITX chassis but it should have handled 65w without a problem.
After some digging, this was a common issue on 8th-9th gen Intel CPU's due to BIOSes overvolting by default. My board had the CPU at "initial voltage 1.3v" so naturally it would spike above 80c under load. So I got to playing and had no trouble running 1.15-1.2v (the stock baseline) which effectively resolved all my problems (under 24 hour stress it wouldn't crack 75c.)
Some more playing revealed I could run the CPU even lower. I've had it at 1.1v (some people have reported BSODs this low) for 2 years without an issue and it rarely cracks 70c under load and generally idles around 30c, 3c above room temperature.
I'm not sure why Intel partners are screwing this up so bad.
Because Intel allows them to do so even in a default, non-OC-enabled state. Most of the time people are testing desktop performance in a larger chassis or on a bench. Imagine if you are the only motherboard manufacturer that runs at "Intel default" settings, and you are getting consistently murdered in benchmarks all over the internet.
It's the reason Intel chips have increasingly run roughshod over TDPs to the point Intel just throws them out the window, even at completely default settings. You can't blame "partners". Intel absolutely could (and should) restrict default behavior, requiring an OC flag to be set to run wild like that. But since they don't, this has even gradually bled over into AMD's strategy. Notice the newer Ryzens, while still better behaved than many of their Intel counterparts, do consume more power at defaults than their "same TDP" previous gen brethren.
I managed a similar undervolt with the same CPU, but it would occasionally (maybe once a fortnight) crash at that level whilst under low loads. A solution that prevents that from happening is definitely interesting to me.
The 5600x in my PC would like to introduce itself to you... FWIW, I walked into my local Microcenter last week and bought it, and there were at least 10 in the cabinet with it...
It's a high yield processor because the chiplet doesn't have to bin well to meet the spec. The enthusiast grade ones line the 5900x and 5950x require lower yield, higher binned chiplets and therefore they can't produce as many of them. My microcenter has plenty of the 5600x's as well. They are abundant. Sadly the 5900x that I want is basically non-existent. I have waited in the cold for 5 days in a row outside microcenter for them to open up and be told they didn't get any in.
I think it's the other way around, most of the US population is in the cities or suburbs. That number is way off, but it's more like 2% of the US population doesn't have a Microcenter nearby.
2 days ago I've seen the first 5900X in stock in a a pretty good price comparison engine searching several European countries. It was "just" 900€ instead of the AMD recommendation of 550€. Already sold out today. No 5950X and 5600X, just a few 5800X at ~650€ (AMD: 450€). Yes, they exist, but it's frustrating.
1usmus has a similar tool already out that can be used under windows for all the AMD CPUs from 3000 series on, he also developed the Ram optimiser etc its called Ryzen CTR and it would not surprise me if AMD didnt bring him onboard to do the same tweeks via BIOs. Am surprised(hmm) its not been reported on here already like it was on other independent review sites
The Curve Optimization tool will be part of AMD’s Precision Boost Overdrive toolkit, meaning that using it will invalidate the warranty on the hardware
Don't you mean it will NOT invalidate the warranty?
Absolutely no reason for such paranoia...;) Even if AMD could prove that a user had invalidated his warranty by running overclocked--even *if*--denying the RMA based on that would be foolish for a number of practical reasons. For one thing, the cost of defending itself in a warranty denial case would amount to 5x-10x what replacement silicon would cost AMD *just to file the paper work* to defend such a case...;) They're going to send the replacement CPU, no question. The only ime I might imagine them denying warranty replacement is if there's obvious physical damage done to the CPU like shorting or broken pins, or some other damage not related to a manufacturing defect. They aren't "spying on you or me" and probably no one at AMD even knows we're alive, let alone addresses and phone numbers. Not even our email, unless we give it to them. We just aren't that interesting, believe it or not...;)
This is a smart way to undervolt. I always wondered why the setting for a constant offset was there for CPUs with boost anyway, as it just doesn't fit over a larger dynamic range.
However, I do not yet understand how this works in practice. Will the software automatically find the lowest voltage curve for all cores individually? Based on which criterion does it decide / test the minimum voltage? How long does this test take? Or do I have to test manually?
@AMD: why do I have to activate PBO to use this and void my warrenty? Undervolting can in no way damage the CPU, right? Compromise: by default undervolting with PBO2 can be used. The warrenty is voided only if one explicitly agrees to unlock the overclocking settings.
I got a 5800X pretty much at list price and been pushing it through its paces since.
While I set it up using a cheap AMD stock cooler left over from a Kaveri A10, I switched to a beQuiet top blower, once it had passed initial quals and I was sure it's a keeper. That Dark Rock TF is supposedly able to blow away 200 Watts, but it turns out that at around 110 Watts there is just no going forward, because the CPU will steady the clocks to not exceed 90°C. I guess the surface area of the single CCD is just not enough to transfer more to the heat spreader, even soldered.
It did 4850 MHz single and 4450 MHz multi on things like Cinebench 20/23, Prime95 in "mean" settings might even lower it to just below 4 GHz all core.
Automated overclocking will move both 100 MHz upward, but I couldn't get mine to break the 5 GHz "sound barrier" (well, I never tried doing anything crazy).
Honestly, I don't see much upward headroom; evidently they bin higher voltage but otherwise defect-free chips as 5800X, while they'll put defective cores into 5700X and 5900X and reserve the best voltage bins free of defects for 5950X: true unicorns, as I've never seen one in the wild.
What I wonder is this: Do the settings, profile (standard, precision, manual) and the settings (validated presets) then get pushed into the BIOS so they remain static and independent of the OS, or are they just persisted in the registry, requiring a reload of RyzenMaster and activation of the profile to become active?
I'm planning to switch between various OS and use cases on the machine and sure won't always want "game mode", while actually running a Linux server (which may actually run a Windows VM with GPU pass-through). So will settings made and tested persist into Linux or will that always revert back to stock?
What's all this with the warranty?
Do they burn a fuse the minute you say "Ok" on RyzenMaster launch?
Or do they just say we won't swap the CPU in case it constantly crashes on fringe settings?
I can't quite believe the first case would stand up in court, not that I've ever managed to kill a CPU just yet (nor wanted to).
Been on Intel-only since my last Kaveri, and it's nice to see just how unspectacular operating Zen 3 seems to have become.
Swapped out a Z170 board with an i7700k and 32GB of DDR3-2400 for an X570 with 64GB DDR3-3200 ECC and Windows just wanted an extra boot.
CentOS7 crashed immediately, but CentOS 8 just filed a formal complaint about this being an untested setup.
BTW: HWinfo seems to have a hard time reading the sensors. Where RyzenMaster will report really low clocks on idle cores, HWinfo keeps things stuck at 3.3 GHz or higher, which had me nervously hunting for background loads...
Very nice feature. I miss the days when computers had hardware "turbo" button. They could implement something similar now, OS independent on-case switch with 3 modes (low, normal, high), connected to 3 bios profiles (so you don`t have to go to bios or switch in OS).
My Pentium 166 had one of those buttons, along with a led displaying some numbers. On my computer it never did anything though, not being connected to the mb. I doubt whether the board even supported any of this.
Reminds me of an MSI Ryzen 1/+ gimmick in their bioses, if they still do it. It was something called "Game mode" and you turned it on or off in the bios--all it did was overclock the Ryzen 1/+ CPUs by 200Mhz. You could do it yourself--I managed much better than that manually with my Ryzen 1. Such gimmicks. "Turbo" = "Super-Duper Hot Rod 12-cylinder Jet-Engine Boost mode", imo...;)
I get why overclocking (and specifically, overvolting) would, and arguably, *should*, void warranty. After all, this can brick a processor.
Why on earth would I accept undervolting as a reason for warranty lost? Lower voltage means lower electronmigration, after all. It's a bit weird to me.
At first i was very scared to contact Digitaltechhacker at g mail com. Entrusting your personal details with a total stranger isn't that easy and also paying before service but i'm happy i did. This guy is amazing, contact him i vouch for him.
I wish they did this earlier, as I see finding the limit of how far you can lower the voltage for a given clock more beneficial than just finding the highest boost speed on whatever thermal headroom you have.
Thermal headroom typically isn't a problem, nor was it ever for enthusiasts.
"What's all this with the warranty? Do they burn a fuse the minute you say 'Ok' on RyzenMaster launch?"
A "modern CPU comes with an additional 'black box' spyware CPU. I thought that was common knowledge in enthusiast circles?
Everyone's doing it. Apple has T2 built into M1. AMD has one. Intel has one. I'm sure IBM has one in Power. The only wildcard is probably RISC-V, where the different companies can add whatever one in they have. I don't know if there is a standardized black box spyware CPU for RISC-V in the specs. Unlikely, since keeping all that info 'trade secret' for only the corporation and its government seems to be the thing.
The temptation to spy is too great to pass by. And, government intel wants it. I believe I read that AMD's Zen 1, when it was implemented by China, required that AMD strip their black box CPU out (no doubt to be replaced by China's).
Everyone wins. Corporations can invalidate warranties for undervolting (which is completely ridiculous anti-consumer behavior) and governments can feel they know something.
And one can rest assured that if an ARM or RISC-V design doesn't have a black box spyware CPU piggybacked then it will appear on your motherboard or some peripheral (or both).
And, don't expect all of those groovy AI pathways in CPUs to be particularly passive, particularly going forward. They're not being put there just to make your photos look faker.
Pretty soon, the CPU will "likely know" you're going to overclock before you even open the red button screen. That way we can implement a former president's plan for indefinite preventative detention. The computer will know what crimes you will commit in the future and lock you up accordingly. (That is, if you're not a domestic corporate executive. In that case, the company will be fined instead.)
In this case, it will be preventative indefinite warranty invalidation.
So they are acknowledging the work of the guy who made the auto oc and undervolting tool, I do not remember the name, it is the same guy who made the dram calculator
Fine tuning the optimizations voltage/power wise with PBO2 also will help these 7nm cpu's last longer (when overclocking) as there's talk of how fast these chips degrade with even mild overclocking vs older higher mass silicon.
Have seen absolutely no evidence of degradation with either Ryzen 1 or a Zen 2 3600X or 3900X. However, overvolting any CPU can cause electromigration--generally the smaller the process node the more vulnerable a given CPU is. I first read about it with the advent of Intel's first 130nm CPUs (The guy at [H]--brain freeze on his name!--talked about putting his $1K 130nm Intel CPU on his key ring after he trashed it--discussed what happened in detail, IIRC) by overvolting , and then 90nm, etc. 14 nm & 7nm would indeed be susceptible--especially since Intel continues with old-style monolithic designs to date. So I'm not a fan of pumping a lot more voltage into a Zen 2 CPU to achieve an all-core overclock of some note. You can achieve better multicore benchmark results, certainly, but at the expense of single core boost performance, as Zen 2 disables boost when manually overclocking. After a lot of testing I found it better to simply let the CPU run automatically as programmed--as I got nearly the same performance overall.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
59 Comments
Back to Article
Oxford Guy - Monday, November 23, 2020 - link
My 6700K would undervolt to a tremendous degree, just as it would run 3200-rated RAM very high. But, no matter how much stability testing I would do the machine never seemed to stabilize. Prime would run at the low voltage but there would be spurious problems with software. RAM would test with multiple instances of HCL, HCL with Prime, etc. But, again... random spurious problems. So, I went back to defaults.It's a good thing this procedure happens automatically with lots of fast adjustments. I have given up on the coarse (human-derived) approach, both for undervolting and overclocking. I have decided it's simply a waste of time. I have never found true stability with anything but defaults and I have spent a huge amount of time with a variety of systems, almost finding stability but never getting it. As one ages one puts more value in one's time and so it had better be "automagic" for me going forward. Kudos to AMD if this works well.
cyrusfox - Monday, November 23, 2020 - link
I have been amazed by the auto OC on non-K chips achieved with the i7-10700, and a little upset as well as the chip outscores my OC i9-9900KF that is watercooled(Passmark CPU score for the 10700 beats the 9900KF by 8%). All I did on the 10th gen Asrock motherboard was tell it to treat the 10700 as a 135W CPU and it did all the rest(and I did enable XMP as well). I have the 10700 cooled by a Noctua NH-U14S and temps are in the mid 60s when loaded. So the auto OC magic is getting really good from my point of view, when looking between 9th gen intel and 10th gen. Can't speak of AMD side yet but there is less of a reason to individually tweak settings with this kind of magic (and on the Intel side I don't see a reason to get a K part when the non-K already have turbo freq set at the edge/top of the CPU capability.genekellyjr - Monday, November 23, 2020 - link
What you describe is not overclocking in any way, it’s just throttling due to power limits less. It’s not going above its rated boost clocks. Not the same as AMD’s auto-overclock stuff which will go higher than its rated boost clock, and now apparently will also auto-undervolt.Additionally, your i9-9900K only can be slower than an i7-10700 if the i9-9900K is throttling, which is pretty sad for an unlocked processor. It is 100 MHz faster all core and 300 MHz faster single core if it’s not throttling. And it can overclock to past 5 GHz (maybe with an auto-tuning thing in your bios if you didn’t want to much around) - but that’s useless if it’s throttling at stock like it is for you. If you actually wanted auto-overclocking capability an AMD chip would be a much better fit for you, as it does it without any user intervention once enabled (see the AMD 5000 series review here for the 5950X’s 5050 MHz clock)
Dug - Tuesday, November 24, 2020 - link
It does not mean his 9900k is throttling, a 10700 can easily beat a 9900k with differences such as motherboard, memory, etc. and o/c potential.It's also a 65w processor that may have a lot more headroom than you realize, which shows in the passmark score and his own testing which you obviously haven't done.
That being said, it's all going to be within a couple of percentage points overall, so nitpicking his comment doesn't really make any difference, nor does telling him to buy a processor you can't buy right now.
Alexvrb - Thursday, November 26, 2020 - link
I dunno, it does sound like the 9900KF is being held back. Depending on how he's overclocking it, he (or the board) might be getting too frisky with voltage, causing it to bang into thermal limits sooner than a well-tuned OC. "Watercooled" doesn't really mean much either by itself - an NH-U14S with good case cooling can actually beat the snot out of your average CLWCs. Combined with the aforementioned potentially-poorly-tuned voltage and bam, you can't dump the heat fast enough to run at the chip's full potential.Yeah I suppose he could be using much slower RAM, but you'd think he would have thought to mention that. Alternatively he's got some settings buggered up, like he did a BIOS update and forgot to re-enable XMP. A lot of fast RAM kits have really low JEDEC defaults. The motherboard doesn't make a ton of difference as long as they both have good power delivery, and I'm not sure what "etc" factors in to CPU performance. The only other thing I could think of is, did they manage to slip in any additional hardware security mitigations in the refresh?
cyrusfox - Monday, November 30, 2020 - link
The 9900KF of mine is got to be the worst k chip I have ever touched, it is barely stable on an all core 5GHz boost, but I need to set the voltage to 1.4+. XMP is enabled (Which I find helps little in regards to the passmark CPU score) and it is running much faster storage than the other computer (Optane vs SX 8200 pro). Its on a custom water loop with a 2x 480 radiators which also cool a GTX 1080, I assure you a much more capabile solution than the noctua and keeps the temps frigid(although the room heats up...). The mobo is a Z390 M Gaming from Gigabyte. I am planning on replacing it with a 10850k here shortly. I likely jsut got a lemon on the 9900KF and I am wholeheartedly impressed with the non-k chips on comet lake, so much so I would consider a plain old 10900 over the 10850k.Spunjji - Monday, November 23, 2020 - link
I feel pretty much the same on all counts. I've had a number of stable overclocked and/or undervolted systems over the years, but it's become an increasingly rare thing - progressively more squeezing required for less juice. This would be nice, if it can reliably produce a stable system.Arbie - Monday, November 23, 2020 - link
I agree 100%. All the angst over CPU overclocking is (or should be) of concern to very few these days. I'm not going to spend hours of tweaking and testing for a small improvement in full-load performance on a machine that runs at 10% load 99.9% of the time.AMD is offering exactly what I want in this and every other regard. The fact that an Intel chip may "OC better" is laughably irrelevant, and fixating on it only serves to confuse newcomers.
Samus - Tuesday, November 24, 2020 - link
A few years back when I (regrettably) rebuilt my PC around an i7-8700 and an Asus motherboard only because at the time there were no decent ITX solutions for the Ryzen, I noticed insanely bad temps for this 65w CPU compared to the 84w or whatever my previous i7-4770 was. This is a basic closed loop setup in an ITX chassis but it should have handled 65w without a problem.After some digging, this was a common issue on 8th-9th gen Intel CPU's due to BIOSes overvolting by default. My board had the CPU at "initial voltage 1.3v" so naturally it would spike above 80c under load. So I got to playing and had no trouble running 1.15-1.2v (the stock baseline) which effectively resolved all my problems (under 24 hour stress it wouldn't crack 75c.)
Some more playing revealed I could run the CPU even lower. I've had it at 1.1v (some people have reported BSODs this low) for 2 years without an issue and it rarely cracks 70c under load and generally idles around 30c, 3c above room temperature.
I'm not sure why Intel partners are screwing this up so bad.
Alexvrb - Thursday, November 26, 2020 - link
Because Intel allows them to do so even in a default, non-OC-enabled state. Most of the time people are testing desktop performance in a larger chassis or on a bench. Imagine if you are the only motherboard manufacturer that runs at "Intel default" settings, and you are getting consistently murdered in benchmarks all over the internet.It's the reason Intel chips have increasingly run roughshod over TDPs to the point Intel just throws them out the window, even at completely default settings. You can't blame "partners". Intel absolutely could (and should) restrict default behavior, requiring an OC flag to be set to run wild like that. But since they don't, this has even gradually bled over into AMD's strategy. Notice the newer Ryzens, while still better behaved than many of their Intel counterparts, do consume more power at defaults than their "same TDP" previous gen brethren.
SirMaster - Monday, November 23, 2020 - link
Yeah I have been using this Curve Optimizer on my MSI x570 Tomahawk since day 1 on AGESA 1.1.0.0 Patch C and it’s been working great.Compared to stock, it has increased my CPU-Z benchmark from 675 single core to 700.
And my multi-core from 9400 to 10000.
I really like it so far.
yeeeeman - Monday, November 23, 2020 - link
6700hq user here. Without undervolt it hovers at 40w, with at 25-28w.Spunjji - Tuesday, November 24, 2020 - link
I managed a similar undervolt with the same CPU, but it would occasionally (maybe once a fortnight) crash at that level whilst under low loads. A solution that prevents that from happening is definitely interesting to me.Smell This - Monday, November 23, 2020 - link
Go clock, go . . .
C'mon AGESA 1180 firmware updates!
Machinus - Monday, November 23, 2020 - link
What is a Ryzen 5000? I've never seen one for sale.Luminar - Monday, November 23, 2020 - link
It's AMD's answer to Sandy Bridge.catavalon21 - Monday, November 23, 2020 - link
I see what you did there. Nice.nathanddrews - Monday, November 23, 2020 - link
boom roastedSamus - Tuesday, November 24, 2020 - link
LOLBlazingDragon - Monday, November 23, 2020 - link
The 5600x in my PC would like to introduce itself to you...FWIW, I walked into my local Microcenter last week and bought it, and there were at least 10 in the cabinet with it...
oRAirwolf - Monday, November 23, 2020 - link
It's a high yield processor because the chiplet doesn't have to bin well to meet the spec. The enthusiast grade ones line the 5900x and 5950x require lower yield, higher binned chiplets and therefore they can't produce as many of them. My microcenter has plenty of the 5600x's as well. They are abundant. Sadly the 5900x that I want is basically non-existent. I have waited in the cold for 5 days in a row outside microcenter for them to open up and be told they didn't get any in.Dug - Tuesday, November 24, 2020 - link
And 98% of the US population is too far away to get to a Microcenter. Especially when only 4 stores have them in stock.Flying Aardvark - Saturday, November 28, 2020 - link
I think it's the other way around, most of the US population is in the cities or suburbs. That number is way off, but it's more like 2% of the US population doesn't have a Microcenter nearby.Beaver M. - Thursday, November 26, 2020 - link
True, those are the only ones available, because nobody wants them. And yet they still cost as much as the 5900X MSRP.MrSpadge - Monday, November 23, 2020 - link
2 days ago I've seen the first 5900X in stock in a a pretty good price comparison engine searching several European countries. It was "just" 900€ instead of the AMD recommendation of 550€. Already sold out today. No 5950X and 5600X, just a few 5800X at ~650€ (AMD: 450€). Yes, they exist, but it's frustrating.evilpaul666 - Tuesday, November 24, 2020 - link
Local MC had a bunch of 5800s yesterday and 5600s this morning. Not really worth upgrading to from a 9700K overclocked and used mostly for gaming.Crazyeyeskillah - Tuesday, November 24, 2020 - link
I own one :)5600x "65w"
the undervolting on it is rediculous.
BlazingDragon - Monday, November 23, 2020 - link
I’m glad that it appear that this will be able to be enabled via BIOS, so that other (non-Windows) OSS’s can take advantage...alufan - Monday, November 23, 2020 - link
1usmus has a similar tool already out that can be used under windows for all the AMD CPUs from 3000 series on, he also developed the Ram optimiser etc its called Ryzen CTR and it would not surprise me if AMD didnt bring him onboard to do the same tweeks via BIOs.Am surprised(hmm) its not been reported on here already like it was on other independent review sites
Gigaplex - Monday, November 23, 2020 - link
CTR doesn't do the same thing as this.Sharma_Ji - Tuesday, November 24, 2020 - link
Yeah LTT demonstrated.Redstorm - Monday, November 23, 2020 - link
The Curve Optimization tool will be part of AMD’s Precision Boost Overdrive toolkit, meaning that using it will invalidate the warranty on the hardwareDon't you mean it will NOT invalidate the warranty?
Gigaplex - Monday, November 23, 2020 - link
PBO does invalidate the warranty.oRAirwolf - Monday, November 23, 2020 - link
How can they tell? Does it burn out a resistor on the cpu or something?sld - Tuesday, November 24, 2020 - link
PBO Auto aka Core Boost keeps the warranty.PBO Enabled, which raises all the wattage, current, temp limits past stock, invalidates the warranty.
Oxford Guy - Wednesday, November 25, 2020 - link
It seems you are unfamiliar with the fact that CPUs have 'black box' secondary CPUs lurking in them.Spyware has a multitude of uses, including warranty invalidation.
Invalidating a warranty for undervolting is completely absurd and should be called out by Anandtech.
WaltC - Saturday, November 28, 2020 - link
Absolutely no reason for such paranoia...;) Even if AMD could prove that a user had invalidated his warranty by running overclocked--even *if*--denying the RMA based on that would be foolish for a number of practical reasons. For one thing, the cost of defending itself in a warranty denial case would amount to 5x-10x what replacement silicon would cost AMD *just to file the paper work* to defend such a case...;) They're going to send the replacement CPU, no question. The only ime I might imagine them denying warranty replacement is if there's obvious physical damage done to the CPU like shorting or broken pins, or some other damage not related to a manufacturing defect. They aren't "spying on you or me" and probably no one at AMD even knows we're alive, let alone addresses and phone numbers. Not even our email, unless we give it to them. We just aren't that interesting, believe it or not...;)jerrylzy - Thursday, December 31, 2020 - link
Still makes no sense to void warranty for undervolting. If they void warranty for overvolting, I can understand the reasons behind it.MrSpadge - Monday, November 23, 2020 - link
This is a smart way to undervolt. I always wondered why the setting for a constant offset was there for CPUs with boost anyway, as it just doesn't fit over a larger dynamic range.However, I do not yet understand how this works in practice. Will the software automatically find the lowest voltage curve for all cores individually? Based on which criterion does it decide / test the minimum voltage? How long does this test take? Or do I have to test manually?
@AMD: why do I have to activate PBO to use this and void my warrenty? Undervolting can in no way damage the CPU, right? Compromise: by default undervolting with PBO2 can be used. The warrenty is voided only if one explicitly agrees to unlock the overclocking settings.
ChaosFenix - Monday, November 23, 2020 - link
Because AMD needed an even bigger lead in multi threaded workloads.svan1971 - Monday, November 23, 2020 - link
awesome feature, only problem you need a CPU you can't buy to use it.abufrejoval - Monday, November 23, 2020 - link
I got a 5800X pretty much at list price and been pushing it through its paces since.While I set it up using a cheap AMD stock cooler left over from a Kaveri A10, I switched to a beQuiet top blower, once it had passed initial quals and I was sure it's a keeper. That Dark Rock TF is supposedly able to blow away 200 Watts, but it turns out that at around 110 Watts there is just no going forward, because the CPU will steady the clocks to not exceed 90°C. I guess the surface area of the single CCD is just not enough to transfer more to the heat spreader, even soldered.
It did 4850 MHz single and 4450 MHz multi on things like Cinebench 20/23, Prime95 in "mean" settings might even lower it to just below 4 GHz all core.
Automated overclocking will move both 100 MHz upward, but I couldn't get mine to break the 5 GHz "sound barrier" (well, I never tried doing anything crazy).
Honestly, I don't see much upward headroom; evidently they bin higher voltage but otherwise defect-free chips as 5800X, while they'll put defective cores into 5700X and 5900X and reserve the best voltage bins free of defects for 5950X: true unicorns, as I've never seen one in the wild.
What I wonder is this: Do the settings, profile (standard, precision, manual) and the settings (validated presets) then get pushed into the BIOS so they remain static and independent of the OS, or are they just persisted in the registry, requiring a reload of RyzenMaster and activation of the profile to become active?
I'm planning to switch between various OS and use cases on the machine and sure won't always want "game mode", while actually running a Linux server (which may actually run a Windows VM with GPU pass-through). So will settings made and tested persist into Linux or will that always revert back to stock?
What's all this with the warranty?
Do they burn a fuse the minute you say "Ok" on RyzenMaster launch?
Or do they just say we won't swap the CPU in case it constantly crashes on fringe settings?
I can't quite believe the first case would stand up in court, not that I've ever managed to kill a CPU just yet (nor wanted to).
Been on Intel-only since my last Kaveri, and it's nice to see just how unspectacular operating Zen 3 seems to have become.
Swapped out a Z170 board with an i7700k and 32GB of DDR3-2400 for an X570 with 64GB DDR3-3200 ECC and Windows just wanted an extra boot.
CentOS7 crashed immediately, but CentOS 8 just filed a formal complaint about this being an untested setup.
BTW: HWinfo seems to have a hard time reading the sensors. Where RyzenMaster will report really low clocks on idle cores, HWinfo keeps things stuck at 3.3 GHz or higher, which had me nervously hunting for background loads...
Kuhar - Tuesday, November 24, 2020 - link
Very nice feature. I miss the days when computers had hardware "turbo" button. They could implement something similar now, OS independent on-case switch with 3 modes (low, normal, high), connected to 3 bios profiles (so you don`t have to go to bios or switch in OS).GeoffreyA - Thursday, November 26, 2020 - link
My Pentium 166 had one of those buttons, along with a led displaying some numbers. On my computer it never did anything though, not being connected to the mb. I doubt whether the board even supported any of this.WaltC - Saturday, November 28, 2020 - link
Reminds me of an MSI Ryzen 1/+ gimmick in their bioses, if they still do it. It was something called "Game mode" and you turned it on or off in the bios--all it did was overclock the Ryzen 1/+ CPUs by 200Mhz. You could do it yourself--I managed much better than that manually with my Ryzen 1. Such gimmicks. "Turbo" = "Super-Duper Hot Rod 12-cylinder Jet-Engine Boost mode", imo...;)GeoffreyA - Thursday, December 3, 2020 - link
Checked it on my B450 Tomahawk now and found the setting. It's called "Game Boost," so I'd guess they've got it on their 500-series boards too.jvl - Tuesday, November 24, 2020 - link
I get why overclocking (and specifically, overvolting) would, and arguably, *should*, void warranty. After all, this can brick a processor.Why on earth would I accept undervolting as a reason for warranty lost? Lower voltage means lower electronmigration, after all. It's a bit weird to me.
caliva002 - Tuesday, November 24, 2020 - link
At first i was very scared to contact Digitaltechhacker at g mail com. Entrusting your personal details with a total stranger isn't that easy and also paying before service but i'm happy i did. This guy is amazing, contact him i vouch for him.jtd871 - Tuesday, November 24, 2020 - link
"Spam, spam, spam, spam...."xenol - Tuesday, November 24, 2020 - link
I wish they did this earlier, as I see finding the limit of how far you can lower the voltage for a given clock more beneficial than just finding the highest boost speed on whatever thermal headroom you have.Thermal headroom typically isn't a problem, nor was it ever for enthusiasts.
Oxford Guy - Wednesday, November 25, 2020 - link
"What's all this with the warranty? Do they burn a fuse the minute you say 'Ok' on RyzenMaster launch?"A "modern CPU comes with an additional 'black box' spyware CPU. I thought that was common knowledge in enthusiast circles?
Everyone's doing it. Apple has T2 built into M1. AMD has one. Intel has one. I'm sure IBM has one in Power. The only wildcard is probably RISC-V, where the different companies can add whatever one in they have. I don't know if there is a standardized black box spyware CPU for RISC-V in the specs. Unlikely, since keeping all that info 'trade secret' for only the corporation and its government seems to be the thing.
The temptation to spy is too great to pass by. And, government intel wants it. I believe I read that AMD's Zen 1, when it was implemented by China, required that AMD strip their black box CPU out (no doubt to be replaced by China's).
Everyone wins. Corporations can invalidate warranties for undervolting (which is completely ridiculous anti-consumer behavior) and governments can feel they know something.
Oxford Guy - Wednesday, November 25, 2020 - link
And one can rest assured that if an ARM or RISC-V design doesn't have a black box spyware CPU piggybacked then it will appear on your motherboard or some peripheral (or both).Oxford Guy - Wednesday, November 25, 2020 - link
And, don't expect all of those groovy AI pathways in CPUs to be particularly passive, particularly going forward. They're not being put there just to make your photos look faker.Pretty soon, the CPU will "likely know" you're going to overclock before you even open the red button screen. That way we can implement a former president's plan for indefinite preventative detention. The computer will know what crimes you will commit in the future and lock you up accordingly. (That is, if you're not a domestic corporate executive. In that case, the company will be fined instead.)
In this case, it will be preventative indefinite warranty invalidation.
GeoffreyA - Thursday, November 26, 2020 - link
Sort of like a home, miniature version of Minority Report's technology.umano - Thursday, November 26, 2020 - link
So they are acknowledging the work of the guy who made the auto oc and undervolting tool, I do not remember the name, it is the same guy who made the dram calculatorrdkone - Thursday, November 26, 2020 - link
Fine tuning the optimizations voltage/power wise with PBO2 also will help these 7nm cpu's last longer (when overclocking) as there's talk of how fast these chips degrade with even mild overclocking vs older higher mass silicon.WaltC - Saturday, November 28, 2020 - link
Have seen absolutely no evidence of degradation with either Ryzen 1 or a Zen 2 3600X or 3900X. However, overvolting any CPU can cause electromigration--generally the smaller the process node the more vulnerable a given CPU is. I first read about it with the advent of Intel's first 130nm CPUs (The guy at [H]--brain freeze on his name!--talked about putting his $1K 130nm Intel CPU on his key ring after he trashed it--discussed what happened in detail, IIRC) by overvolting , and then 90nm, etc. 14 nm & 7nm would indeed be susceptible--especially since Intel continues with old-style monolithic designs to date. So I'm not a fan of pumping a lot more voltage into a Zen 2 CPU to achieve an all-core overclock of some note. You can achieve better multicore benchmark results, certainly, but at the expense of single core boost performance, as Zen 2 disables boost when manually overclocking. After a lot of testing I found it better to simply let the CPU run automatically as programmed--as I got nearly the same performance overall.GeoffreyA - Thursday, December 3, 2020 - link
"First read about it with theadvent of Intel's first 130nm CPUs"
I believe early Northwood steppings had a problem when overclocked, where they'd degrade rapidly.
Oxford Guy - Wednesday, December 2, 2020 - link
Surprised no one said "I see what AMD did there" with that first chart. Paging Elon Musk...