Mr. Fox

It is strange that everything seems kosher for me on the 13900K and when I was running the 12900K on the Strix D4 mobo everything was fine on Linux. It is either the Z690-E or 12900KS, or both. I installed openSUSE last night with the ACPI=off kernel argument and fewer things are broken than before. I do not get the long list of errors when Linux is loading now and neofetch reports the correct clock speeds. CPU-X reports clocks close (100MHz below actual) but hardinfo still shows them way off (4100MHz instead of 5400 P cores and 4300 E cores). It could also be the Linux applications themselves that the developers are not updating. Even if the Linux kernel is providing the proper support, if the apps are not updated they may not interpret things correctly. This kind of thing is what holds Linux back from becoming a dominant force in the PC technology realm. Most people (me included) don't have the knowledge, desire or time to compile code and fix broken Linux code. Being "free" is both a strength and a major weakness. It is probably pretty safe to assume that most of the experienced Linux developers do not own cutting edge hardware, and they're only going to burn calories on hardware that matters to them.

It is strange that everything seems kosher for me on the 13900K and when I was running the 12900K on the Strix D4 mobo everything was fine on Linux. It is either the Z690-E or 12900KS, or both. I installed openSUSE last night with the ACPI=off kernel argument and fewer things are broken than before. I do not get the long list of errors when Linux is loading now and neofetch reports the correct clock speeds. CPU-X reports clocks close (100MHz below actual) but hardinfo still shows them way off (4100MHz instead of 5400 P cores and 4300 E cores). Edit: it could also be the Linux applications themselves that the developers are not updating. Even if the Linux kernel is providing the proper support, if the apps are not updated they may not interpret things correctly. This kind of thing is what holds Linux back from becoming a dominant force in the PC technology realm. Most people (me included) don't have the knowledge, desire or time to compile code and fix broken Linux code. Being "free" is both a strength and a major weakness. It is probably pretty safe to assume that most of the experienced Linux developers do not own cutting edge hardware, and they're only going to burn calories on hardware that matters to them.

This is the system I am struggling to make viable on Linux. I think it may be something bugged in the ACPI implementation by the dumb-dumbs at ASUS on the Strix ZX690-E. I cannot get the CPU turbo clocks to display correctly in any desirable monitoring tools. CPU-X shows the 12900KS clocks to a fixed 5.4GHz (correct) under load, still not correct at idle, but all of the "normal" things I use show either a fixed 3.4GHz (c-states disabled) or a fixed 4.1GHz (c-states enabled). I have tried KDE, POP!_OS and ZorinOS and all have the same issue. I have tried passing a variety of kernel parameters in GRUB. I have installed different packages intended for monitoring clock speeds and it is hit or miss. The couple that actually work correctly are worthless to me because they are CLI stuff I can't use the way I want to. I also wonder if it is something with the 12900KS not being recognized properly like a 12900K and 13900K. At any rate, this is a classic example of a thing that make noobs believe that Linux is not a viable replacement for Windows, and on this system it probably isn't solely for this reason. I would not embrace Linux if this were an example of normal, but I have used it enough to know it is an exception (albeit a more common problem than desired). Banshee The other two systems work exactly as they should... pleasure using Linux on both of them, in contrast. Unfortunate that the ASUS machine doesn't deliver the same quality of experience. Wraith Half-Breed

Oh, trust me... based on what I have seen in the past two years, the limits of human stupidity are beyond measure and incredibly stupefying. There are roughly the same number of imbeciles as there are people with common sense. Apparently, NVIDIA recognizes this and is willing to take advantage of the mentally handicapped shoppers. On another note... Wraith Half-Breed Banshee This is the system I am struggling to make viable on Linux. I think it may be something bugged in the ACPI implementation by the dumb-dumbs at ASUS on the Strix ZX690-E. I cannot get the CPU turbo clocks to display correctly in any desirable monitoring tools. CPU-X shows the 12900KS clocks to a fixed 5.4GHz (correct) under load, still not correct at idle, but all of the "normal" things I use show either a fixed 3.4GHz (c-states disabled) or a fixed 4.1GHz (c-states enabled). I have tried KDE, POP!_OS and ZorinOS and all have the same issue. I have tried passing a variety of kernel parameters in GRUB. I have installed different packages intended for monitoring clock speeds and it is hit or miss. The couple that actually work correctly are worthless to me because they are CLI stuff I can't use the way I want to. I also wonder if it is something with the 12900KS not being recognized properly like a 12900K and 13900K. At any rate, this is a classic example of a thing that make noobs believe that Linux is not a viable replacement for Windows, and on this system it probably isn't solely for this reason. I would not embrace Linux if this were an example of normal, but I have used it enough to know it is an exception (albeit a more common problem than desired).

I am surprised it made it that high before the BSOD. Usually see errors start to occur about 8-10°C sooner than that. I have found that a memory overclock that is unstable at 45°C might run error-free at 35°C, and vice versa. It kind of makes sense considering the clock speeds and voltage though. You can't do that without producing heat. That's why using a fan or putting them on water is recommended. Taking the stock heatsinks beautifying heating blankets off also helps, but not doing it the right way can rip the chips off the PCB because the imbeciles that manufacture RAM use adhesives that are stronger than the solder. (They don't need to use any adhesives and it is ridiculous that they do.)

They received a lot of hate and crap for the stupidity, and 11th Gen was a failure because of it. Nobody wanted it. Chances they will do that again might be pretty slim. Time will tell if they are dumb enough to shoot themselves in the foot financially a second time. Nothing ever surprises me though. Intel, AMD, NVIDIA, Micro$lop and crApple are run from the top down by fricken idiots. So, I will not have a surprised look on my face the next time that any of them prove my point about their abject idiocy. It is not a question of if, but when, with the retarded tech giants

I haven't seen one yet that lacks it. It would be very unusual for a desktop motherboard intended for gamers and enthusiasts that isn't a pile of crap (i.e. Dell, HP, Lenovo, etc.) to be missing CSM. CSM is an absolute necessity for a variety of reasons, including OSes that need it and diagnostic/repair utilities that require it. Any garbage desktops and laptops that do not have CSM are severely limited and crippled. I rejected the opportunity to purchase the X170 that I used for development purposes for zTecpc because Clevo lacked the common sense to include it. According to @Premathere was no way he could add CSM support, so rather than being an admirable alternative to the turdbook cess pool, it instantly became a worthless object of contempt to me. The aftermarket desktop mobo manufacturers do not lack the basic common sense that their loser laptop and "big brand" OEMs do.

100%... I have never experienced any real issues with either of them. They look better, work better, perform better than anything that came after them. I'm not sure they are committed to anything--certainly not excellence or the sheeple flock they think they own--but, the people responsible for their atrocities are such demented losers that they probably should "be committed" in a different use of that verbiage. Best solution is to not utilize any of their worthless, buggy "security" trash. Remove Defender.7z

“And no one puts new wine into old wineskins. If he does, the wine will burst the skins—and the wine is destroyed, and so are the skins. But new wine is for fresh wineskins.” - Saint Mark, the Evangelist But, hey... it fits the 4-year old socket. Must be good to go. Some features are not working, but that's OK. I saved money. Meanwhile, back at the ranch, I decided to play with the Sabrent M-die some more. Not my best, but still very good.

Yup, exactly. And, their X3D "upgrade" CPU was castrated. Nice upgrade joke. Cookie-cutter garbage for the gamer kiddos. Everyone gets a trophy for participating. One size fits all. Yay. Utterly worthless for anyone interested in overclocking.

It has been a selling point for the underdog, but you're correct. It is kind of a gimmick. As AMD grows toward achieving parity with Intel, I expect that to become a thing of the past. They can't compete on performance without competing on the financial side. It takes money, and that model is a money-loser for AMD and for motherboard manufacturers. Without support from motherboard manufacturers (as we have all seen, they treat AMD less favorably) they will have difficulty gaining parity on performance or financial strength.

I guess the degree to which it matters really depends on what you find the most value in. AM4 kept the same socket for a long time, and so did AM3, but those that kept the same motherboard didn't get the full benefit of generational architectural changes by remaining on an old chipset. That is great for everyone that found value in that. Those that wanted to enjoy the emerging technology to its fullest were in no better place than the typical 2-generation socket pattern Intel owners enjoy. For the most part, the only people complaining about Intel only using a socket for 2 generations are AMD owners. Most folks that prefer Intel have nothing to complain about and don't.

Lot of stupid people out there. Too many, actually. Idiocy has flourished over the past few years and become a force to be reckoned with. It may soon be illegal to be normal, express dissent with stupidity or demonstrate common sense in public. And, if you teach your kids what's right and what's wrong, it is domestic terrorism. Maybe. Probably so. Some Threadripper owners were not so lucky, and I recall at lot of AM4 owners being livid about AM5, in spite of the obvious improvement in the design of everything about it, including the socket itself. Unfortunately, PC tech is frequently a monkey-see, monkey-do realm, especially when it comes to schemes and scams that increase revenue. The more progress (and money) AMD makes, the less likely they will be to do what is right for their customers. Only time will tell.

Spend the extra $100 if you are not planning an upgrade for a few years. If you divide the extra cost out over the time you plan to use it, the extra cost is minuscule, but the extra performance is beneficial. If you change your mind later it will cost you the full price of the better CPU instead of the small price difference today. Plus, if AMD's price drop is temporary you'll lose even more money buying the cheaper CPU.

Thank you. The main reason I kept it is because I can't really do much more with the 2080 Ti FTW3 than what I have already done. I will be able to grab a few extra points using different hardware. The 2080 Ti is definitely a superior GPU overall. My top Fire Strike score with the 2080 Ti is substantially higher. But, yes... for 3060 Ti this is very good. Much better than before the shunt mod. It is still gimped on voltage, but at least the power limits are gone now. The 2080 Ti with the shunt mod, high voltage and 2000W vBIOS clocks quite a bit higher on core and little higher on memory. https://www.3dmark.com/compare/fs/19438829/fs/28914049

Shunt mod FTW! https://hwbot.org/submission/5130408_ | https://www.3dmark.com/3dm/83902699

Yes, that would be correct. Improved IPC and significantly higher clock speeds is too much for the 5950X to keep up with. DDR5 is better. The new Ryzen platform also handles memory a whole lot better in general as well. Not nearly as many functionality and stability issues as Ryzen 9.

Well, I am glad that the only thing messed up is your car.

Sorry to hear that you were in an accident. I hope you were not hurt, and I hope the person that did it has insurance.

In case anyone is interested...

Price reduced to $425 + shipping for members of this forum. I just listed it on Mercari, Nextdoor and Craigslist as well (at a higher price).

That really is dependent on us. As long as we (gamers and enthusiasts) continue to pay asinine prices for their products, they will continue to be asinine in their pricing of the products they sell us. At least 50% (probably closer to 100% if we are honest with ourselves) is our fault for being compliant with our molester. They can't charge inflated prices for products if nobody is willing to buy them. I am still very keen on the idea of settling for less performance and supporting Intel's venture into the GPU space. I hope they end up being wildly successful and extremely disruptive in this segment. If they can keep their foot in the doorway and end up forcing it open wide, I think there is potential for severe distrubance that will be very harmful to NVIDIA's position of dominance, and cause AMD to have to work overtime just to remain relevant.

I still think there is a lot of truth to the old cliche that "clock is king" but one answer to that is more cores, even when the cores are slower. Higher clocks is like a small group of snipers that quickly and discretely take out their adversaries with a single headshot, whereas more cores is like a platoon of foot soldiers spraying a bazillion bullets every direction with a wide swath of destruction. They can't run very fast and they are not precise, but a swarm of bullets is a force to be reckoned with.

Yes, totally agree with that. Even though I don't like the fact that some of them are wussy Atom cores, I can't find any fault in the results. Fantastic performance. But, there is still a part of me that wonders how much performance is being forfeited by their substitution in place of more P-cores. I'm guessing they are cheap to produce and this is a way of delivering a level of performance that stays a step ahead of AMD while not making the product too expensive.

I did that and found the overall experience regrettable. Cinebench scores were nice, but that is probably the only kind thing I can find to say about my misguided endeavor. It was probably the worst waste of money on an upgrade that proved to be a downgrade that I have ever experienced. Apart from my magnetic attraction to mediocre silicon quality, (seldom lucky, unfortunately,) I am very happy with the experience I have had with 10900K, 12900K/KS and 13900K. These are excellent processors, but I still feel like I am getting screwed because of the stupid Atom core crap. Even though performance is stellar, it would be better if they ditched the chintzy E-core crap and made them all P-cores. I think the issue with NVIDIA is that they are an inherently dishonest company that values profit more than doing the right thing, and the issue with AMD is ignorance from the leadership level down. They are essentially posers in the enthusiast space, kind of like the McDonald's of the technology realm.