Yes, I'm afraid you seem to misunderstood the article. The fact that your CPU can hit advertised boost clocks on any core does not change the fact that some can't (or hit them on even any single core).
Also, in your video I'm only seeing 1 core hit 4525 MHz. In the description you say "Set Override in Bios", do you mean you turned on PBO?
Only one core on our Ryzen 5 3600X processor will hit AMD's rated boost frequency. AMD confirmed some cores in Ryzen 3000-series processors are faster than others, which is denoted in Ryzen Master. That means that not all cores on can hit the single-core turbo frequencies. Instead, there are a mix of fast and slow cores.
Well I think the article is misleading and the real truth is BIOS are not done correctly and that is the real problem.
All I can say I have 2 x 3600X and both of them can surpass Max boost written on the box.The video I showed ,is that 3600X is cable of surpassing the max boost.
Also some BIOS PBO override is a thing.
Here is another screen of all at cores 4400Mhz with boost 4500Mhz
Really? Where in the video is that shown? The only thing I saw was him talking about the BIOS and doing some benchmarks. The closest I could find was the Userbench results which showed a Turbo average of 4.15GHz
Just so you don't say I am lying:
Thats from 8:16 in the video.
I would guess that the most expensive X570 motherboard, designed specifically for Ryzen 3000, would have the best chance at getting the best performance out of it. but hey maybe I am wrong.
Because tomshardware used to be good, less bias in the past.
unfortunately its still searched a lot, and tomshardware seems to be going for clickbait rather than anything else.
I assume you have proof of this "Intel Bias"? I swear every time an article comes out for AMD that isn't stating they are a gift from God they are being Intel or nVidia biased.
AMD is a company. Companies make mistakes. No AMD does not have your best interests at heart. Yes they want to make as much money as possible. Deal with it.
Only one core on our Ryzen 5 3600X processor will hit AMD's rated boost frequency. AMD confirmed some cores in Ryzen 3000-series processors are faster than others, which is denoted in Ryzen Master. That means that not all cores on can hit the single-core turbo frequencies. Instead, there are a mix of fast and slow cores.
Well I think the article is misleading and the real truth is BIOS are not done correctly and that is the real problem.
All I can say I have 2 x 3600X and both of them can surpass Max boost written on the box.The video I showed ,is that 3600X is cable of surpassing the max boost.
Also some BIOS PBO override is a thing.
Here is another screen of all at cores 4400Mhz with boost 4500Mhz
It seems to me they're talking about their 3600X, not all Ryzen 3K. There is also a typo/poor grammar in the 2nd last sentence "not all cores on can hit the single-core turbo". I agree it would probably help if they corrected it to say "not all cores on every [Ryzen 3K] CPU can hit the single-core turbo".
Do you have any proof that this is just a BIOS issue? Seems like we couldn't know that until if/when they release a BIOS update that corrects it.
I'm still don't understand what this "override" is that you're talking about. Are you referring to turning on PBO, i.e. essentially overclocking the CPU?
What motherboard (and BIOS rev), what cooler, and what version of Windows are you using?
I fully expected there would be some disappointments with this year's release and had no intention of buying until next year or the year after, when both Intel & AMD should be up to relative parity in technology and fighting each other on price instead. The only good buy this year is last year's Ryzen if you're looking to build a workstation or server, or the i9-9900K if you're building a gaming rig, but again, I'm waiting til they redesign & move past the disappointments and of course cut prices. 7nm & price competition will be standardized going forward and we have AMD to thank for that
It seems to me they're talking about their 3600X, not all Ryzen 3K. There is also a typo/poor grammar in the 2nd last sentence "not all cores on can hit the single-core turbo". I agree it would probably help if they corrected it to say "not all cores on every [Ryzen 3K] CPU can hit the single-core turbo".
Do you have any proof that this is just a BIOS issue? Seems like we couldn't know that until if/when they release a BIOS update that corrects it.
I'm still don't understand what this "override" is that you're talking about. Are you referring to turning on PBO, i.e. essentially overclocking the CPU?
What motherboard (and BIOS rev), what cooler, and what version of Windows are you using?
Sorry not trying to confuse and I think you see where I disagree with article.When I have 2x3600X
Yes I have proof of the BIOS being the issue ,because I have different BIOS that run differently where boost is concerned .
AMD ComboPI1.0.0.3 in use
AMD ComboPI1.0.0.3 AB not in use
All Motherboard vendor call there stuff different and on my motherboard ,the cheapest MSI X470 Gaming Plus it is called,well screenshot and all Motherboards will /should/possibly have this in one form or another.think i covered my bases there.
I disagree with the article not being 100% accurate,that's all.
Hmm. I have a 3600X bought launch day. It's supposed to have a 4.4GHz turbo but according to Ryzen Master it never goes above 4.225GHz. All 6 cores can hit that speed, but none of them ever go faster. Oddly, though, sometimes HWInfo64 will report 4.35+ speeds, so something's weird.
As the article mentions, these processors can adjust their clock speed hundreds of times per second, but if a utility were to actually poll the processor for its clock speed hundreds of times per second, that would negatively impact performance, and could actually even affect those clock rates. So, the utilities are likely only updating their readings about once per second and perhaps averaging results, so any quick bursts to the full clock rate would likely not get detected. Even the test setup for this article only polled the processor's clock rate 10 times per second, which is still a fairly course measurement relative to the rate at which the clocks can change on these processors.
In any case, that wasn't the point of the article. The point was that there's a variance in boost capabilities between individual cores. However, that doesn't really matter in practice, outside of affecting overclocking headroom. While it may technically be possible for other processors to run all cores at their maximum boost clocks, doing so is considered overclocking, and voids your warranty. At stock clocks, these other processors will also only reach full boost clocks on up to one core at a time, even if that boosted core may get passed around to different physical cores. What's going on with these processors simply comes down to them being binned more accurately on a per-core basis, allowing the processors to operate closer to their limits at stock, providing additional out-of-box performance and largely negating the benefits from overclocking.
I did find the article interesting, but ultimately this performance difference between the fastest and slowest cores only came down to about a 2% difference in clock rates in their testing, something they should have pointed out, rather than only referring to that difference in less-clear terms like "~75 - 100MHz". The updated Windows scheduler will keep lightly-threaded tasks on the fastest cores for these processors, so in practice this doesn't really affect anything, unless you are operating the hardware out of spec through overclocking. And the lack of overclocking headroom was already well known, and was something that affected previous-generation Ryzen processors as well, or at least the higher-clocked versions of the chips, which similarly operated near their limits.
But this does open up another line of questions. How does AMD decide what the minimum bar for a 'slow' core is? Logic dictates that would be the base frequency of the chip, but that would theoretically mean some cores on a 3600X would only operate at 3.8 GHz.
And then there's a few parts of the article like this that come across as needlessly alarmist, since we have never seen Ryzen processors limited to base clocks for their all-core boost, so logic would certainly not dictate that AMD is using that as the minimum bar for a core's clocks. While it would be nice to have an all-core boost frequency target listed in the specs, unless you encounter processors that show a substantial difference between the best and worst-clocked cores, the possibility of multi-core boost clocks being substantially lower than normal is little more than baseless speculation. The processor Tom's acquired at retail only showed a 100Mhz difference in boost clocks between its fastest and slowest cores, and its performance seemed about in line with other sites that have tested retail samples as well. I would be more concerned about trusting the results of review samples provided by a manufacturer, as it's very possible that they could be sending out specially-binned samples to reviewers.
As for what happens to processors with certain cores that can't keep up, those cores likely end up disabled, such as in these 6-core parts that make use of an 8-core chiplet. And those that don't have enough good cores to meet the standards of those chips will probably end up in that rumored Ryzen 2500, which I suspect will be a quad-core part with half the chiplet disabled.
Man have I got news for you. There are countries where something as little as "advertised clock speed" not meeting would be a big problem. Where consumers have vastly more rights than corporations.
It is an issue. Is it a major one? No. But everyone who buys a CPU buys on reviews and stated specifications. What it they stated 8 cores but it was really 6 cores yet performed just as well? Would you still be OK that you got a 6 core CPU that was claimed to be 8 cores?
So your saying these countries now have class action suites against intel for having to disable hyper-threading to mitigate a design flaw? Funny, I haven't heard of any.
Are there any cases in which the AMD CPU's are not hitting their advertised base clock? Funny, I haven't heard of any.
Please show me the federal standards (any country) that has laws in place for how long or how many cores CPU must maintain it's boost clock. (again, there is none).
Needless to say your mention of big problems and class action suits is nothing more than useless drivel.
So, again I say .......YAWN....... these CPU's are still performing exactly as we have been shown they should. It's no different then me buying a new car and suing Ford/GM because my car is not hitting the advertised MPG in the city. (your mileage may very)
Would the CPU's be any less impressive if AMD had labelled the boost clocks lower, but performed the same? No, they would not, and perhaps, even more impressive.
So your saying these countries now have class action suites against intel for having to disable hyper-threading to mitigate a design flaw? Funny, I haven't heard of any.
Are there any cases in which the AMD CPU's are not hitting their advertised base clock? Funny, I haven't heard of any.
Please show me the federal standards (any country) that has laws in place for how long or how many cores CPU must maintain it's boost clock. (again, there is none).
Needless to say your mention of big problems and class action suits is nothing more than useless drivel.
So, again I say .......YAWN....... these CPU's are still performing exactly as we have been shown they should. It's no different then me buying a new car and suing Ford/GM because my car is not hitting the advertised MPG in the city. (your mileage may very)
Would the CPU's be any less impressive if AMD had labelled the boost clocks lower, but performed the same? No, they would not, and perhaps, even more impressive.
Preliminary approval was granted to a class action settlement, resolving allegations that GM duped consumers into believing certain SUVs had better MPG.
topclassactions.com
Doesn't hurt to do a little research first.
If AMD lowered the advertised boost rates and met them or exceeded them it would have been better. Absolutely. Much like the Intel Q6600 G0. Rated for 2.4GHz and almost every single one would overclock to 3GHz on stock or lower voltage. It was never advertised as such but the enthusiast community knew it did and jumped on it. Probably one of Intels best selling Core based CPUs.
I said I really liked the article, but I'm reading totally unrelated comments. I get my boost clocks, and I get my base clocks. All the article is saying is that you most likely can't have all core boost at the boost clock frequency, which is fine, and was never promised by AMD anyways. In the end, I like the performance. A 3600 is an 8700k for the most part for half the price. The best CPU ever.
Something I just noticed. This article's forum thread is called...
"AMD's New Binning Strategy on Ryzen 3000: Core-by-Core Turbo Analysis"
...which seems like a pretty neutral and unbiased title. However, the title that currently appears as the page title for the article has a far more negative tone...
"Zen 2's 7nm Complications: Why not All Ryzen 3000 Cores Are Created Equal"
And then there's also a shorter title appearing as...
"Our Tests Show Not All Ryzen 3000 Cores Are Created Equal"
So, was the original title not considered click-baity enough? This gives me the impression that there might be some questionable practices going on here by someone, probably not the article's author, trying to put a negative spin on what appears to simply be a more thorough binning process allowing processors to operate closer to their limits.
Sorry not trying to confuse and I think you see where I disagree with article.When I have 2x3600X
Yes I have proof of the BIOS being the issue ,because I have different BIOS that run differently where boost is concerned .
AMD ComboPI1.0.0.3 in use
AMD ComboPI1.0.0.3 AB not in use
All Motherboard vendor call there stuff different and on my motherboard ,the cheapest MSI X470 Gaming Plus it is called,well screenshot and all Motherboards will /should/possibly have this in one form or another.think i covered my bases there.
I disagree with the article not being 100% accurate,that's all.
Yes better performance on BIOS AMD ComboPI1.0.0.3
If I disable all PBO settings my Ryzen 3600X only Max boost on all cores to 4400Mhz what is written on the box.
Who didnt know this already just look at what you hold in your hand android phones not sure about apple do this already the main core will always turbo to the said GHZ. Obviously reviews had no problems with this new zen and who knows where AMD takes this.
, with water cooling my "base" frequency seems to be 4.2Ghz on all cores (I have never seen 3.8Ghz or under 4.2Ghz under load). I can get it to 4.4Ghz on one core with overclocking (PBO settings and Auto-OC) but the increased voltage and power applied just does not seem worth the extra 100Mhz. Odd that i have to take it "out of warranty" to get what should have been ca
The article is misleading and really not 100% accurate for 3600X.
I have a 3600X which can hit 4525Mhz on all cores and up to 3CCX or 6 cores same time.
Hell I did a short single thread run and 3600X Hitting 4500Mhz Cinebench 20 and over 4400Mhz Max boost on 3CCX in a minute video.
I could be wrong or misunderstood the article or did not read what it was trying to say but I disagree with some areas where it is not true for me.
Interesting results here. That is above the rated clock speed of the chip, which would indicate some type of overclocking. The VID peaks at ~1.531. Are you using PBO and/or AOC? What are the power settings? Can you post a screenshot of your Ryzen Master home screen with the same settings?
Interesting results here. That is above the rated clock speed of the chip, which would indicate some type of overclocking. The VID peaks at ~1.531. Are you using PBO and/or AOC? What are the power settings? Can you post a screenshot of your Ryzen Master home screen with the same settings?
Well I am not using Ryzen Master at this moment but it is a very nice tool I am able to use it to change CPU voltage and CPU clocks on the fly ,without restart.I can install if you want and give you a screenshot.
I will list videos with setting of Ryzen Master /HWinfo64 I uploaded already.Any question or scenario I can try it out.
What I am doing is using what AMD provide within the BIOS.
I am using at the moment .Not trying to trick anyone.I will give as much information below so forgive spoilers.
Any chipset driver and I have tried many combinations.With everything is set to auto in BIOS,along with AMD Ryzen™ Power Saver.pow, I use this plan for Higher boost than what is written on the box.
Link to AMD Ryzen™ Power Saver.pow I have imported
AMD ComboPI1.0.0.3 I am Still using Beta BIOS and not the new BIOS with AMD ComboPI1.0.0.3ab on MSI X470 Gaming Plus
AMD Chipset Driver 1.07.29 Released July 30 2019
AMD Ryzen™ Power Saver.pow which I imported after new chipset drivers do not contain this plan.
This screen shot was BIOS AMD ComboPI1.0.0.3 set to Enhanced Mode 4
5 CCX @ 4400Mhz
1 CCX @ 4450Mhz
BIOS AMD ComboPI1.0.0.3 screen shot of Precision Boost Overdrive Modes and Max CPU Boost Clock Override
Ryzen Power plans effect CPU voltage on my setup.I am only using AMD Ryzen™ Power Saver.pow for maximum boost for Ryzen 3600X
BIOS AMD ComboPI1.0.0.3 depending on which mode I use in BIOS .
CPU voltage up to 1.57v in idle and light loads/gaming/Cinebench Single -Max Boost 4525Mhz
Heavy loads Intel Burn Test/Cinebench Multi ,CPU adjusts clock speed and CPU voltage to proper levels to avoid overheating or to high temperatures.Depending on what I set in BIOS with AMD Ryzen™ Power Saver.pow ,CPU Clocks and CPU Voltage varies a little but only hit around 80C on CPU
Ryzen Master Change Settings On The Fly Test Cinebench 20 with HWinfo64 on.
I changed all CCX to 4250Mhz and CPU voltage 1.325v and ran Cinebench first part of video.While still recording I changed 3 x CCX to 4350Mhz and 3x CCX to 4300 and clanged CPU voltage to 1.37v and ran Cinebench 20
Preliminary approval was granted to a class action settlement, resolving allegations that GM duped consumers into believing certain SUVs had better MPG.
topclassactions.com
Doesn't hurt to do a little research first.
If AMD lowered the advertised boost rates and met them or exceeded them it would have been better. Absolutely. Much like the Intel Q6600 G0. Rated for 2.4GHz and almost every single one would overclock to 3GHz on stock or lower voltage. It was never advertised as such but the enthusiast community knew it did and jumped on it. Probably one of Intels best selling Core based CPUs.
Reading comprehension, use it. I said countries, not frivolous lawsuits brought forward by individuals. Furthermore, these lawsuits are not because of lack of performance, resulting from the disabling of hyperthreading, they are the result of the security risks involved with the bugs themselves. Please stay on topic. What we are talking about is the need for a party to show a loss because of what you are saying is false advertising, Case law 101, there has to be a loss. So, again the question is, are the CPU's not performing as advertised? Why yes spanky, they are.
But no you say, they can't reach there full boost clock on all cores. Newsflash, they don't have to. If one core can reach that speed for 1 nana-second, lawsuit over.
Reading comprehension, use it. I said countries, not frivolous lawsuits brought forward by individuals. Furthermore, these lawsuits are not because of lack of performance, resulting from the disabling of hyperthreading, they are the result of the security risks involved with the bugs themselves. Please stay on topic. What we are talking about is the need for a party to show a loss because of what you are saying is false advertising, Case law 101, there has to be a loss. So, again the question is, are the CPU's not performing as advertised? Why yes spanky, they are.
But no you say, they can't reach there full boost clock on all cores. Newsflash, they don't have to. If one core can reach that speed for 1 nana-second, lawsuit over.
You seem to be assuming I want a suit to happen. I do not. And I was uinable to show a country but there are plenty of countries with heavy consumer laws, some even force companies to refund or fix past stated warranties etc. I am not saying it will happen just that there could be a case depending on the law in specific countries.
Just an example, you know that big fine Microsoft was charged by the EU? Do you know why that fine was even levied? Because they included WMP in Windows. Now what loss did the consumer have for Microsoft including a basic media player that could be removed and replaced with anything a person wanted?
And I was just responding to you. You picked a bad comparison.
Now could this issue be resolved with a BIOS update? Possibly. Although in my life I have never seen a BIOS update increase clock speed unless there was already an issue to be resolved such as the BIOS not allowing it to run properly in the first place.
Interesting research, I have been doing a lot of testing with my 1600x and have observed similar behavior. In addition to the latest windows version the latest chipset driver is needed, in my case I have an asus ch7, so x470, I pulled the driver directly from AMD site, this added the amd performance power plan.
Also using PBO to override the default CBS behaviour, I have 1 core that will hit 4.49 if temps are low enough, 2 v m core that hit 4.46 1 core that hits 4.41 and 2 that max out at 4.36. Temps are a major factor though, if I set my rad fans to full it will idle as low as 24c but as soon as cinibench starts it jumps to the low 50's and ramps up, in the 50's all cores run at 4.3 -4.2 once temps reach low 60's core clock are in the 4.1x range, over 65 and they drop to the 4.0x range.
So main findings, to hit 4.4, PBO, performance power plan, as much cooling as you can throw at it.
I did find the article interesting, but ultimately this performance difference between the fastest and slowest cores only came down to about a 2% difference in clock rates in their testing, something they should have pointed out, rather than only referring to that difference in less-clear terms like "~75 - 100MHz". The updated Windows scheduler will keep lightly-threaded tasks on the fastest cores for these processors, so in practice this doesn't really affect anything, unless you are operating the hardware out of spec through overclocking. And the lack of overclocking headroom was already well known, and was something that affected previous-generation Ryzen processors as well, or at least the higher-clocked versions of the chips, which similarly operated near their limits.
That is a good point, it is very minimal change to the CPU otherwise no two reviews of Ryzen CPU's would be the same and we could start claiming AMD is cherry picking their demo units when they show us the benchmarks.
That is a good point, it is very minimal change to the CPU otherwise no two reviews of Ryzen CPU's would be the same and we could start claiming AMD is cherry picking their demo units when they show us the benchmarks.
I would be surprised if all the hardware manufactures didn't cherry pick review samples. They used to do ES samples which were never indicative of end products but that seems to have changed recently.
I know some PSU manufactures have been called out for sending samples that were vastly better than end units.
You seem to be assuming I want a suit to happen. I do not. And I was uinable to show a country but there are plenty of countries with heavy consumer laws, some even force companies to refund or fix past stated warranties etc. I am not saying it will happen just that there could be a case depending on the law in specific countries.
Just an example, you know that big fine Microsoft was charged by the EU? Do you know why that fine was even levied? Because they included WMP in Windows. Now what loss did the consumer have for Microsoft including a basic media player that could be removed and replaced with anything a person wanted?
And I was just responding to you. You picked a bad comparison.
Now could this issue be resolved with a BIOS update? Possibly. Although in my life I have never seen a BIOS update increase clock speed unless there was already an issue to be resolved such as the BIOS not allowing it to run properly in the first place.
Just an example, you know that big fine Microsoft was charged by the EU? Do you know why that fine was even levied? Because they included WMP in Windows. Now what loss did the consumer have for Microsoft including a basic media player that could be removed and replaced with anything a person wanted?
LOL, talk about your bad examples. That is so completely unrelated to this I don't even begin to know where to start. They brought suit against microsoft for having a monopoly on the market place. When they do that to AMD, please let me know. You might have well have said "there are examples of countries bringing suits against companies". It would be of equal relevance.
We are talking about 100 to maybe 200 Mhz here. What is this, roughly 1-2% ? Care to tell me about the margin of error of the testing software? Good luck with your lawsuit.
None of any of this changes the benchmarks we have seen for these processors. So where exactly is the problem?
Just an example, you know that big fine Microsoft was charged by the EU? Do you know why that fine was even levied? Because they included WMP in Windows. Now what loss did the consumer have for Microsoft including a basic media player that could be removed and replaced with anything a person wanted?
LOL, talk about your bad examples. That is so completely unrelated to this I don't even begin to know where to start. They brought suit against microsoft for having a monopoly on the market place. When they do that to AMD, please let me know. You might have well have said "there are examples of countries bringing suits against companies". It would be of equal relevance.
We are talking about 100 to maybe 200 Mhz here. What is this, roughly 1-2% ? Care to tell me about the margin of error of the testing software? Good luck with your lawsuit.
None of any of this changes the benchmarks we have seen for these processors. So where exactly is the problem?
Fine. Want related? What about the lawsuit against the FX series CPUs? While I don't agree with it, the suit is happening due to AMD marketing the CPU as a 8 core part and depending on how the court defines a core it may or may not win. The biggest issue is that a dual core section of an FX CPU is not equal to what Intel or current AMD CPUs.
This is similar in that AMD lists the CPU as a turbo clock of say 4.6GHz. Product never hits 4.6GHz.
Again I am not saying it is something that anyone should sue over or that any lawsuit should be filed. I find the majority of them, my Microsoft one included, stupid. I got a settlement claim paperwork for one against Microsoft and threw it away.
What I am saying is that depending on the country and the consumer laws it is possible AMD might face a lawsuit due to this. Do it mean they should? No. Is it still possible? Of course it is as people sue over everything these days even if it doesn't directly affect them.
This is a small issue that could potentially cause AMD legal problems depending on the country and consumer laws.
Also margin of error is different than advertised specifications.
Super great detailed work. I tried Precision Boost and Auto OC in Ryzen Master by merely turning it on. Those two things with the same RAM setting gave me little or not indication there was a huge difference over neither being on. I see my Ryzen 9 3900X indicate
the Full Monty of advertized CPU speed in CPUZHWMONitor.
I have run the Arithmetic SANDRA test and hit 100% Load and acheived an excellent score, but SANDRA limits the CPU to 4.4Ghz,
so I am of the opinion that a whole lot of work is going on from AMD and WIndows and improvements are being made. I upped the Chipset today and found it did the same thing as pushing the OC "wipe button thing" and I lost the RAM setting. I was using 3333Mhz to the best result so far, and lost some Geek Bench scores at 3466Mhz as well......so, I am not really worrying about it.
HOWEVER I and so glad you are keeping a close eye on it all!! Very very neat work.
AMD---thanks for fixing D2. It is a wonderful game I enjoy, but doing Tales from the Borderland has me all ready for the Next Bordelands....that is a real charming game to. BTW the 3900X is a cool runner in both game...and I mean in the 50C range.
Facebook
Twitter
Instagram