ZXRaziel:

I was waiting to say this for ages ... Finally it's over 9000 !

this is so underrated

D1stRU3T0R:

U gud m8? Lately AMD doesn't have those listed bugs, and EVEN Those there were only because of faulty PCIe4 on older motherboards with old chipset. No matter (almost) how good the 10000x series will be, it will be almost impossible to win against 3950X

You are misinformed. WHEA errors and file corruption happened on x570 chipset when PCIE Gen 3devices installed and have their own non windows driver: Nvidia card, Samsung SSD etc AMD admitted to this and have a blog post about it, their "solution" is to mask WHEA errors so they wont appear, thats what it says. As far as file corruption, most people [ more like 99.9%] very rarely check their windows for system file corruption and majority never opened Event Viewer in their life. Turbo Boost is a known issue, none of AMD CPUs can hit advertised turbo on single core, and when they do, it happens for second [whats called a spike] and then drops down to low speed, such turbo has no value. Compare this vs how turbo should work, Intel CPUs, Ryzen 1000/2000 and youll understand

The "new" Intel cpu should be named "Cascade Fail-X-Times fixing old bugs", and to @MegaFalloutFan, the problem about the boostclocks on the Ryzen 3xxx(x) series, seems to be how the motherboard manufacturers implemented stuff, not AMD's fault, but that might be a problem that you can't understand, as the Intel-freak you are.

nizzen:

/guru3d 2019 Miss guru3d forums before 2010... Now it's 90% toxic, flaming and blaming.

I think you're exaggerating, but you'd think Intel would be on 10nm by now. I still think what they're trying to do is cool, but I'm clearly not the target audience.

MegaFalloutFan:

You are misinformed. WHEA errors and file corruption happened on x570 chipset when PCIE Gen 3devices installed and have their own non windows driver: Nvidia card, Samsung SSD etc AMD admitted to this and have a blog post about it, their "solution" is to mask WHEA errors so they wont appear, thats what it says. As far as file corruption, most people [ more like 99.9%] very rarely check their windows for system file corruption and majority never opened Event Viewer in their life. Turbo Boost is a known issue, none of AMD CPUs can hit advertised turbo on single core, and when they do, it happens for second [whats called a spike] and then drops down to low speed, such turbo has no value. Compare this vs how turbo should work, Intel CPUs, Ryzen 1000/2000 and youll understand

There was no corruption. People were just running sfc and assuming the repairs were due to the WHEA errors. I've gotten dozen of those errors on both my X570 motherboards and my SFC/DISM shows zero corruption. This was debunked the first day it was found.

patteSatan:

The "new" Intel cpu should be named "Cascade Fail-X-Times fixing old bugs", and to @MegaFalloutFan, the problem about the boostclocks on the Ryzen 3xxx(x) series, seems to be how the motherboard manufacturers implemented stuff, not AMD's fault, but that might be a problem that you can't understand, as the Intel-freak you are.

An ASUS employee said AMD lowered the frequency to assist in long term reliability. Even if it's not AMD's fault AMD should be working with their partners to ensure the products they are selling are performing to what they are advertising. AMD should have cleared everything up by issuing a statement on it - something they haven't done yet. That's not to mention the fact that nearly every board, especially the review boards, boosted higher with the initial BIOS than they do with the current one. If Intel had done something akin to this people would be having seizures on their keyboard.

RzrTrek:

I think you're exaggerating, but you'd think Intel would be on 10nm by now. I still think what they're trying to do is cool, but I'm clearly not the target audience.

It's obviously hyperbole but there is a serious increase in people either being intentional obtuse or incredibly ignorant in a lot of these threads. It's especially frustrating because it's the same people that have been doing it for years. They often contribute nothing useful to the discussion and just repeat the same nonsense week after week.

have a 9900k and I already said, Intel HEDT is game over, I crafted 2 wishlist on my regular webshop and intel prices are crazy they have to adapt or the market % are going to flip around I also made my own tests with my 9900k and while I enjoy even 5.1Ghz today and yes the amd are slower, the thing is for the speed they have they do more, just test it yourself downclock your intel to 4.2Ghz then compare cinebench etc...you'll see amd gets way higher score clock for clock so yes it's slower but does more with it it's not a 1:1 clockspeed/performance ratio technically they are better, amd has a margin to improve intel doesn't I don't understand fans, if I do one thing and you can prove me there's a better way I'll instantly switch I won't try to find a way to diminish your way.

Denial:

There was no corruption. People were just running sfc and assuming the repairs were due to the WHEA errors. I've gotten dozen of those errors on both my X570 motherboards and my SFC/DISM shows zero corruption. This was debunked the first day it was found.

Tell that to people with actual corruption computer freezes and broken windows on this forum. YOU have zero corruption, plenty others have a lot.

patteSatan:

The "new" Intel cpu should be named "Cascade Fail-X-Times fixing old bugs", and to @MegaFalloutFan, the problem about the boostclocks on the Ryzen 3xxx(x) series, seems to be how the motherboard manufacturers implemented stuff, not AMD's fault, but that might be a problem that you can't understand, as the Intel-freak you are.

Sure its the motherboard manufacturers "fault" its all them, AMD is fine, the house is not on fire.

kakiharaFRS:

have a 9900k and I already said, Intel HEDT is game over, I crafted 2 wishlist on my regular webshop and intel prices are crazy they have to adapt or the market % are going to flip around I also made my own tests with my 9900k and while I enjoy even 5.1Ghz today and yes the amd are slower, the thing is for the speed they have they do more, just test it yourself downclock your intel to 4.2Ghz then compare cinebench etc...you'll see amd gets way higher score clock for clock so yes it's slower but does more with it it's not a 1:1 clockspeed/performance ratio technically they are better, amd has a margin to improve intel doesn't I don't understand fans, if I do one thing and you can prove me there's a better way I'll instantly switch I won't try to find a way to diminish your way.

Cinebench doesn't represent real life performance, most of us are gamers here so here is a nice link with benchmarks, pay attention to the 1% FPS, how 3900x and 3600x most of the time are close to each other while one is twice cheaper and how Intel is faster anywhere up to 40FPS in 1% category. When Ryzen 3000 benchmarks came out, the initial wave had just average FPS, and it looked in the same area of Intel results and people got hyped, but now that people did deeper benchmarks, Ryzen 3000 has lower 0.1% and 1% results compared to intel in 1080p and even 1440p, even in GPU limited 4K it still slower compared to Intel in plenty games. Since the day of Ryzen 1000/2000 and to this day ryzen has slower NVMe performance compared to Intel. Even AMDs own first Gen PCIE GEN4 NVMe drives [AMD said they helped develop the first gen PCIe Gen 4 based NVMe controller to speed things up and have the SSDs ready for release] perform better on Intel PCIe Gen 3.0 in every benchmark except synthetic sequential Read. The situation is not one sided as Fanbois pretend it is: "AMD is cheaper and faster", it isnt cheaper if you move from Intel and dont want to buy outdated chipsets that are not made anymore and were manufactured by assmedia, a company most people hate for poor quality and under-performing controllers. AMD is not faster then Intel if you a Gamer first, if you need fast AVX2 for work, its not faster if you need AVX512, its much slower if you into Emulation, especially Advanced console emulation like PS3, Xbox360, even Gamecube, its not faster if you need TSX [ for example PS3 emulator uses it to speed up emulation], its not fast if you relay on lowest RAM to CPU latency, its not faster if you relay on fast Cache and low Cache latency, Ryzen lacks any graphics chip and i only mention it because Intel CPU has QuickSync so if Ryzen is faster vs Intel head to head in Video encoding, it loses to Intel the moment you enable quicksync. My2c I game in 4K so i dont care about gaming FPS that much, im not fanboy i had 1090T, i had AMD Athlon CPU back during Single Core and Dual core era, and im interested in doing RAID0 with two NVme drives, which cant be done on Intel mainstream motherboard or AMD pre x570 motherboard, it can be done on x570. I always wanted 16+ core Server/Workstation/Gaming Monster with tons of NAND, Opatne and RAM, Intel [for now] its too expensive, old TR is not good for gaming [maybe TR3 will perform as fast as Ryzen 3000 in gaming, that will be good], so im "stuck" with 3950x With all my dislike for AMDs cheapness and low performance in Gaming, its the only 16 core option if i dont want to spend insane money

MegaFalloutFan:

Sure its the motherboard manufacturers "fault" its all them, AMD is fine, the house is not on fire.

There could be a candle lit in the house and that'd be enough for you to say the whole house is on fire.

MegaFalloutFan:

Tell that to people with actual corruption computer freezes and broken windows on this forum. YOU have zero corruption, plenty others have a lot.

Yeah I don't believe it - sorry man. Show me one thread where there is corruption and "broken windows" directly linked to the WHEA errors. I personally know 3 people with 3000 chips - none of us have problems with corruption in Windows, all of us have WHEA errors. When the whole corruption thing started I was one of the first people to post about it on this forum - for the next 2-3 days I followed it on reddit and various other forums, no one was able to confidently say it was coming from the WHEA errors. It was just random people running DISM/SFC and saying that it found problems - some of those people actively admitted switching from Intel to AMD and not reinstalling Windows and/or having the same Windows install for years. Some of these people weren't even getting the WHEA errors with their hard drives but with video cards and still claiming it corrupted files because SFC said so. You need to actually link the WHEA error with corruption and prove it came from it. If I went to work right now and ran SFC on every machine like 40% of them would report some kind of file being replaced - we don't have AMD. Should I proclaim Intel has corruption issues? No. Same way you shouldn't here.

MegaFalloutFan:

You are misinformed. WHEA errors and file corruption happened on x570 chipset when PCIE Gen 3devices installed and have their own non windows driver: Nvidia card, Samsung SSD etc AMD admitted to this and have a blog post about it, their "solution" is to mask WHEA errors so they wont appear, thats what it says. As far as file corruption, most people [ more like 99.9%] very rarely check their windows for system file corruption and majority never opened Event Viewer in their life. Turbo Boost is a known issue, none of AMD CPUs can hit advertised turbo on single core, and when they do, it happens for second [whats called a spike] and then drops down to low speed, such turbo has no value. Compare this vs how turbo should work, Intel CPUs, Ryzen 1000/2000 and youll understand

It does exist that term "how it should" AMD said the CPU is boosting UP TO X GHz (let's say 4.5). If that cpu ks boosting to 4.475, thats okey, CUZ they are saying UP TO. If you cry "oh but why they changed it right now", well, because. AMD is "lying" about turbo, Intel lies about TDP. Well, noone lies, they just measure it differently. About corruption, just because 5% of users had it (maybe running old faulty bios + pie 4 with x470 etc) it doesn't mean its a common problem lol. I rarely saw someone crying about it. About whea errors and what amd said, well this isn't their particular fault. So again, what is really their fault in here? All of these seems user related problems. (imho the only problem is that they restricted gen 4 on older chipset, but alot disagree on this topic, and they are somehow right in șlem ways... If they would let them enabled, more non-tech users would get more problems like you said thus making bad image for AMD.

D1stRU3T0R:

AMD said the CPU is boosting UP TO X GHz (let's say 4.5). If that cpu ks boosting to 4.475, thats okey, CUZ they are saying UP TO.

No it doesn't: https://www.amd.com/en/products/cpu/amd-ryzen-9-3900x It says the Max Turbo Boost is 4.6Ghz, not up to 4.6Ghz, just 4.6ghz. 4.6Ghz is something I've never seen on this processor at all across two different $300+ motherboards on liquid cooling. So if I'm not hitting it who is? That's not to mention that they had their chief marketing guy post a youtube video claiming you could see higher than 4.6 with the right cooling/power. This is clearly not the case. It's misleading and either AMD should make a definitive statement on it or change the max frequency. It's worse when you consider the fact that the original/review BIOS's allowed much higher frequency than current ones. The fact that people are trying to defend this is ridiculous to me but not surprising - AMD constantly gets away with false marketing because they are the "underdog". It needs to end or they'll keep doing it.

Denial:

Yeah I don't believe it - sorry man. Show me one thread where there is corruption and "broken windows" directly linked to the WHEA errors. I personally know 3 people with 3000 chips - none of us have problems with corruption in Windows, all of us have WHEA errors. When the whole corruption thing started I was one of the first people to post about it on this forum - for the next 2-3 days I followed it on reddit and various other forums, no one was able to confidently say it was coming from the WHEA errors. It was just random people running DISM/SFC and saying that it found problems - some of those people actively admitted switching from Intel to AMD and not reinstalling Windows and/or having the same Windows install for years. Some of these people weren't even getting the WHEA errors with their hard drives but with video cards and still claiming it corrupted files because SFC said so. You need to actually link the WHEA error with corruption and prove it came from it. If I went to work right now and ran SFC on every machine like 40% of them would report some kind of file being replaced - we don't have AMD. Should I proclaim Intel has corruption issues? No. Same way you shouldn't here.

You can read the whole thread, bunch of people complained, but here on this page one specific user https://forums.guru3d.com/threads/beware-of-whea-errors-on-x570-platform-pcie-data-coruption.427789/page-3

D1stRU3T0R:

It does exist that term "how it should" AMD said the CPU is boosting UP TO X GHz (let's say 4.5). If that cpu ks boosting to 4.475, thats okey, CUZ they are saying UP TO. If you cry "oh but why they changed it right now", well, because. AMD is "lying" about turbo, Intel lies about TDP. Well, noone lies, they just measure it differently. About corruption, just because 5% of users had it (maybe running old faulty bios + pie 4 with x470 etc) it doesn't mean its a common problem lol. I rarely saw someone crying about it. About whea errors and what amd said, well this isn't their particular fault. So again, what is really their fault in here? All of these seems user related problems. (imho the only problem is that they restricted gen 4 on older chipset, but alot disagree on this topic, and they are somehow right in șlem ways... If they would let them enabled, more non-tech users would get more problems like you said thus making bad image for AMD.

Ahem? There is no such thing as up to, they could avoid this by writing lower Turbo boost on the box, one that actually works and thats it, they wanted to write big number to compete with Intel. Lets not forget that Turbo boost worked fine on Ryzen 1000/2000, it performed as it should, no one had such issues, so this is totally new thing AMD did. Intel never lied about TDP, some mobos had this option unlocked by default, against Intel spec. And its not the same situation, one is lower performance and another is giving more performance by increasing the TDP. The WHEA errors never happened due to user error, it happened because of how AMD created the chipset. If Samsung SSD works fine on everything on this planet except x570 chipset, then its not the user to blame, but the chipset creator.

This is my MAX boost with Asrock Taichi x570 and 3900x on chilled water 🙂 https://www.diskusjon.no/uploads/monthly_08_2019/post-42975-0-63474500-1566768158_thumb.png Ps: It did't help the max boost with chilled water, same as indoor ambient water.

MegaFalloutFan:

WHEA errors and file corruption happened on x570 chipset when PCIE Gen 3devices installed and have their own non windows driver: Nvidia card, Samsung SSD etc

Oddly enough, the same happened on both my Intel 9900K/Z390/970 Evo and my 7700K/Z270/960 Evo systems with Windows 10 1903 repeatedly when Samsung NVMe 3.1 drivers were installed (the drivers didn't have any problems on Windows 10 1809, which they have been created for to fix some issues). It took me quite a while to figure out what was causing it. I got rid of the Samsung NVMe drivers and am using Windows 10's native ones and there hasn't been any file corruption ever since. So at least in my cases, it wasn't the x570 chipset itself causing it as I'm using Intel platforms.

that was fixed

Denial:

No it doesn't: https://www.amd.com/en/products/cpu/amd-ryzen-9-3900x It says the Max Turbo Boost is 4.6Ghz, not up to 4.6Ghz, just 4.6ghz. 4.6Ghz is something I've never seen on this processor at all across two different $300+ motherboards on liquid cooling. So if I'm not hitting it who is? That's not to mention that they had their chief marketing guy post a youtube video claiming you could see higher than 4.6 with the right cooling/power. This is clearly not the case. It's misleading and either AMD should make a definitive statement on it or change the max frequency. It's worse when you consider the fact that the original/review BIOS's allowed much higher frequency than current ones. The fact that people are trying to defend this is ridiculous to me but not surprising - AMD constantly gets away with false marketing because they are the "underdog". It needs to end or they'll keep doing it.

Yes. MAXIMUM. Almost like UP TO. Almost the same. If it would boost to 4.605 then the claim would be false, right ? So That's why they say MAXIMUM turbo boost. I don't see the problem here tbh. It's not false marketing, it's just "physical" marketing for people who can't really read. That's it. Not the best marketing but hey, it works.

D1stRU3T0R:

Yes. MAXIMUM. Almost like UP TO. Almost the same. If it would boost to 4.605 then the claim would be false, right ? So That's why they say MAXIMUM turbo boost.

No, then the claim would be accurate because in the PBO video that AMD released he specifically says that you'll hit what's printed on the box and potentially exceed it if your power and cooling is adequate enough. You're also still ignoring the fact that since the reviews AMD lowered the maximum clocks. Your argument is setting a precedent where it's fine for companies to overclock the chips to unobtainable numbers for reviews then hide behind "up to" marketing (which doesn't exist here) for the release. Anyway this is an Intel thread so I'm done dragging it off topic.

nizzen:

This is my MAX boost with Asrock Taichi x570 and 3900x on chilled water 🙂 https://www.diskusjon.no/uploads/monthly_08_2019/post-42975-0-63474500-1566768158_thumb.png Ps: It did't help the max boost with chilled water, same as indoor ambient water.

Im sorry, image too small.

Cave Waverider:

Oddly enough, the same happened on both my Intel 9900K/Z390/970 Evo and my 7700K/Z270/960 Evo systems with Windows 10 1903 repeatedly when Samsung NVMe 3.1 drivers were installed (the drivers didn't have any problems on Windows 10 1809, which they have been created for to fix some issues). It took me quite a while to figure out what was causing it. I got rid of the Samsung NVMe drivers and am using Windows 10's native ones and there hasn't been any file corruption ever since. So at least in my cases, it wasn't the x570 chipset itself causing it as I'm using Intel platforms.

The issue you describing is a known issue due to newer version of windows demanding nee type of drivers, it happened to everyone who kept using old drivers, especially on Intel IGPU was the worst, its a driver issue due to MS updating the drivers scheme. I upgraded all the drivers before updating windows.

MegaFalloutFan:

Im sorry, image too small.

https://www.diskusjon.no/uploads/monthly_08_2019/post-42975-0-63474500-1566768158.png (for future reference, I got that URL by opening the image in a new tab and removed "_thumb" at the end of the filename)

Denial:

No, then the claim would be accurate because in the PBO video that AMD released he specifically says that you'll hit what's printed on the box and potentially exceed it if your power and cooling is adequate enough. You're also still ignoring the fact that since the reviews AMD lowered the maximum clocks. Your argument is setting a precedent where it's fine for companies to overclock the chips to unobtainable numbers for reviews then hide behind "up to" marketing (which doesn't exist here) for the release. Anyway this is an Intel thread so I'm done dragging it off topic.

The only thing i want to add, is that AMD never told us about maximum all core boost either, they hide this stat. Ryzen 1000/2000 worked like Intel, all identical the CPUs had identical all core boost and single core boost but now on 3000, one guy gets on 3900x 4650Mhz another 4600Mhz and another 4455. Issues like this shouldn't be spoiled by fanboy wars, everyone should work together against the Corporations, in this case if we stay silent then ll future CPU will have "Up To" or "Maximum Possible" turbo boost and 5 people with same CPU in hand will get different speed.