Mr. Fox

I still prefer doing more with more, versus more with less. More watts is MOAR better. I still miss seeing 1.2kW+ getting sucked from the wall with my 7980XE in Cinebench. Fun times. I probably should upload one of my 3090 KPE with the XOC vBIOS pulling 1100W+ in 3DMark 11.

I am nearly 60 years old and I don't remember ever seeing things as messed up as they are now. Sleepy Joe and his insane clown posse have really done a number on us. It's going to be difficult to recover from the multi-faceted damage. So yeah, $100 USD buys less than ever.

In my case the load temperatures were about 6-8°C cooler with the OptimusPC block versus the Raystorm Pro. Idle temperatures were affected less, maybe 2-3°C. That was with an Intel 10th Gen setup, so it may not translate the same with AM4. Right now the best IC for DDR5 overclocking is SK Hynix. That was definitely not true with DDR4. Samsung B-die was, by far, the best IC on DDR4 and SK Hynix was not good.

100%. Bare die has been amazing in my experience as well. Nothing is better, even for a run-of-the-mill loop or an AIO cooling solution. Delid (except for the 5950X) with LM has also been a close second place enhancement. And, as well all know, all it takes sometimes is a 2-3°C improvement to mean the difference between throttling/malfunction when you are pushing everything to the edge of stability. All it takes is a couple of points on a benchmark run to move a score up in rank.

I think it happened to fit just right. Temps were identical to the Foundation block and it worked surprisingly well. That was the only monoblock I had ever owned and I do not intend to buy another because mounting from the back of the motherboard and the extra screws for the VRM mounting are just too much of a pain in the butt to deal with. I take things apart frequently. That kind of design is far too much hassle to deal with no matter how well it works for a person that doesn't assemble their PC and leave things alone.

I think a better water block is definitely worth considering. I had an OptimusPC foundation and an EK monoblock on the 5950X. They both worked well and produced equivalent results. The monoblock was gorgeous to look at, but I tinker with things often enough that I found it inconvenient. For a person that does not do a lot of messing with the hardware, the monoblock is probably the best bet because of enhanced aesthetics with no loss in cooling performance. I have been a fan of OptimusPC blocks since they were first released. The Signature and Foundation blocks are excellent. I was using a Raystorm Pro block on my work desktop for a long time. I recently replaced it with a Signature block and saw immediate improvement in CPU load temps. Having owned 3 Intel (Signature and Foundation) and 1 AMD (Foundation) CPU blocks, I don't want to use anything else now. I actually prefer the Foundation block with the clear top so I can see if/when cleaning the jet plate is needed. The Signature block looks awesome, but it is solid metal and you cannot tell if the jet plate is collecting sediment or debris without taking it apart for inspection. Here is a photo of the EK monoblock. I loved how it looked.

Probably so. I was more concerned at the time about recovering what little thermal headroom there was back to at least stock level so I could move on to other things. At that point I was also disappointed to the extent that I was ready to get rid of everything so I could offset the cost of going back to Intel. So, my interest in trying to make it something worth keeping was starting to wane. The WHEA errors, USB drop-outs, modest CPU overclocking capabilities, FCLK limitations and crappy memory performance had pretty much ruined the experience to the point that reigniting my interest was probably going to be impossible.

I would encourage you not to. Delidding my 5950X provided no benefit. It turned out to be a waste of time and money. I never figured out a bare die solution, and that probably would have helped. When I first got it back from RockitCool temps were much worse than stock. Uneven chiplet heights seem poorly suited for liquid metal. Using normal paste instead put me back to stock temps with the copper IHS. Net benefit was zero. Net loss was what I spent on it.

I see RockitCool has the delid kit back in stock again, but I am waiting until the bare die kit is available. Delid will certainly be better than stock, but I have run all my Intel CPUs bare die since 7960X/7980XE and that is definitely the way to go. I wish Intel would sell all of their CPUs with no IHS. All laptops are bare die and it is kind of stupid that desktop CPUs are not. The IHS is just another stupid piece of crap to have to deal with.

I tend to agree with that. It is generally the approach I have taken. CPUs are a whole lot easier to kill than they used to be, but still not as easy as some might think. I generally find a reason to upgrade before they degrade, but we do have to be a little more careful than we used to be. Voltage plus severe heat also seems riskier than voltage plus cold. If I can't overclock the crap out of a CPU or GPU, then I have no interesting in owning it.

There are two ways to do it. One way is to just grab it and rip the band-aid off. It is quick and it only hurts for a few seconds. The band-aid is still sticky. You could stick it back on again if you wanted to. The other way is to get the band-aid wet. The adhesive will soften and after a while it will just fall off. It does not stick to anything and it is totally worthless. You can't change your mind, and you would not want to because it is disgusting. This way takes longer, but does not involve any anxiety or discomfort. I am taking the latter approach. ----------- Edit ----------- Well, we can add Call of Duty: Infinite Warfare and Call of Duty: WWI to the Linux play list.

Based on my contempt for what a mess Winduhz has become in the past couple of years, I have really stepped up my efforts to replace it with Linux. It isn't 100% yet, but getting closer. For those that enjoy gaming, things are better than they ever have been. I am finding many "Windows-only" titles play very well under Linux using Steam Proton. I recently tested a random assortment from my Steam Library and the only one I did not get working was Gears 5. That kind of makes sense since it is a Micro$lop Studios title and they would have a vested interest in making sure it doesn't work on Linux, LOL. Quake 2 RTX Bright Memory Infinite Crysis Remastered Death Stranding Doom Eternal Halo: Combat Evolved Halo 4 Metro Exodus Enhanced Edition Rise of the Tomb Raider Shadow of the Tomb Raider

I can hardly wait to see what kind of astronomically ludicrous price tag one of those with a 1600W capacity is going to carry. I anticipate that they'll be much prouder of it than they should be. Hopefully, they will keep the new standard in place for at least another 20+ years again rather than inventing new ways to milk us for more money sooner for something we don't actually need apart from the fabrication of need.

I was super excited about the 5950X build... like a kid at Christmas. I had been anti-AMD, and for good reason, for about a decade after some major product failures and disappointments with their brand. Once upon a time, I was a team red fanboy. That was all I owned for a time. But, that ended with a long string of product failures and disappointments that spanned a few years, mostly with defective GPUs, (about 11 in total,) buggy drivers, and lackluster overclocking. With a heavy heart at the time, I gave up on the brand. I shook the dust off of my feet, went back to Intel and NVIDIA, and never looked back. I typically don't care about brand and I only practice brand loyalty to the extent that it serves my interests. My natural tendency is to blacklist brands that disappoint me. I am lenient to the extent I give second chances, but beyond that it becomes difficult for a brand to redeem themselves. Ryzen 9 got them off the perma-ban list with me... at least intially. I started noticing signs of improvement early on with Ryzen development. It was encouraging. Not enough to make me care at first, but things were trending positively on the CPU side of things. Then here came the impressive and beastly Ryzen 9 5950X benchmark scores. I drooled over the amazing Cinebench runs I saw posted on HWBOT and impressive 3DMark physics scores. After a lot of hemming and hawing I decided to bury the hatchet, forgive the past and dive back into the red pool. I thought my overclocking experience would allow me to improve on the results I was seeing from most 5950X owners and jumped in head-first with great optimism and high expectations. I was very disappointed with my purchase. The system I had suffered from the USB dropout problem and none of the fixes fixed anything. About 90% of the time, none of my connected USB peripherals worked without unplugging and reconnecting them (dozens of times every day) and WHEA errors numbered in the thousands. I frequently had to power off the PSU to clear the "00" POST code on a cold boot. Overclocking abilities were lackluster (about 500MHz less than the 7960X and 7980XE processors I loved owning). I found AMD's approach to overclocking and the PBO thing to be an absolute joke. The AMD UEFI firmware is a mess, and having to use software in Windows to improve results simply added insult to injury. (I appreciate the efforts by 1usmus to try to remedy the situation using his software. It helps a little bit.) Thermal management of 5950X was more difficult than Intel, despite having the same core count as the 7960X. Running under chilled water helped, but it didn't appreciably enhance overclocking headroom. Delidding made it worse, not better. Memory latency was totally absurd. After about 90 days of struggle and disappointment, more time wasted with troubleshooting than having fun, and failing to identify a reason I should like it other than a few impressive benchmark scores, I sold the 5950X, Crosshair and EK Quantum monoblock at a huge loss on eBay and went back to Intel. I tried very hard to like it and failed. The monetary loss was painful.

I already had doubts about AMD's intelligence and that POS 5800X3D threw gasoline on the fire. I still haven't healed from my deep regrets over the money I flushed down the toilet on an X570/5950X turd system. I only fault myself for that though... should have listened to that still small voice that was telling me "no, don't do it" and I would have come out of it about $1,300 richer. I enjoyed the nice Cinebench scores for a few minutes, but the overall lackluster nature of the product as a whole and the defects it carried was a situation that I found impossible to forgive.

I'm way too lazy to stop and figure it out, especially when it's not something I am curious about. If I had to guess, it would probably be close to the same in terms of power, but probably significantly more performant. Plus, I wouldn't care too much if the power consumption increased. That has never been something that mattered to me other than my belief that more is better. I've never been a believer in the silly notion of doing more with less... always more with more. I never cared that my 7980XE pulled 1100W from the wall in Cinebench when pushed to 53x on all cores. It was actually pretty awesome. The cynical part of my personality expects Intel to do something like that as a middle finger salute to 12900K/KF owners to show their (lack of) appreciation for buying a flagship product that gets moved to second fiddle in the same product generation. Just like the Green Goblin and 3090 Ti. "Thanks for stepping up and buying our best. Enjoy this ice pick in your back, fella." Intel kind of already is doing it with the 12900KS, but that's more of a binning thing than a unexpected surprise flagship replacement.

Thank you. It is pretty amazing that only an 8 core hyperthreaded CPU with a extra 8 garbage cores can run this fast and be so crazy powerful. It would have been especially nice to see a 12- or 16-core version without the wimpy baby-girl e-cores. Overall, I really like the platform, but I have yet to embrace the concept of the crappy core architecture. That part aside, I am glad I spent the money on it. Linux still has some broken functionality with Alder Lake. Hopefully full support won't be too long coming. Some of the software I use on Linux (such as cpufrquency-gui) refuse to launch and CPU-X reports erroneous information. Until it affects more Linux-only software developers it probably won't get fixed because it doesn't matter so much to them.

I pretty much butcher my Windows installations so they are totally FUBAR'd. I don't really give a rat's butt about security. I run for years with no updates and no antivirus software on the OSes I care about. I run scripts that physically remove Defender, disable updates, delete "essential" (essentially worthless) Micro$lop Services using Nsudo, etc. Then I have what I call a disposable crash dummy (stock) Windows installation that I leave exposed to the cancer and when it gets too bogged down with their worthless filth I restore a Macrium Reflect image on that drive and let the losers start their cycle of digital "update" scum all over again. I figure that the only way my spitefulness, hate and disrespect for their broken filth can be validated as having merit is by allowing them space to do their worst without them having the ability to affect me in a way that actually matters. That way, their stupidity and incompetence never creates an impediment or inconvenience for me. Both of my desktops are multiboot. Both have the disposable crash dummy OS, plus 2 versions of LTSC, Windows 7, Linux, and one of the crash dummy trash buckets is Winduhz 11. When I render a verdict of guilty and pass judgement on their crimes it is based on my experience as a victor that has triumphed in spite of their plan to make me a victim.

Thank you. No, but 32K would be nice. Already broke 31.5K, but not on Linux using wine at only 5.2GHz. /img/logo.png Mr. Fox`s Cinebench - R23 Multi Core with BenchMate score: 31537 cb with a Core i9 12900K HWBOT.ORG The Core i9 12900K @ 5454MHzscores getScoreFormatted in the Cinebench - R23 Multi Core with BenchMate...

Nice Cinebench run. I like that it runs on Linux, too. Not sure how much more nonsense I can tolerate from the Redmond Reprobates.

100%. I have been a huge fan of liquid metal for roughly a decade and frequently see a ~20°C improvement in load temps. And, nothing beats bare die + liquid metal. On my 7980XE it meant passing Cinebench R20 at 5.3GHz all core with no AVX offset and no thermal throttling versus thermal shutdown.

I don't have a good way to cool them as I did the DDR4 modules. They do get hot under certain circumstances (I have seen a max of about 52°C, which is very warm for memory) and start throwing TM5 errors due to the thermals. The same was true of my DDR4 modules. If they got hotter than about 40-45°C it would produce errors in TM5. Until heavy duty DDR5 heat sink jackets (like those used for water cooling) are available, I have a 120MM fan blowing on the modules to try to keep them cool. I may try some of my DDR4 heat sinks on these modules, but the jury is still out on whether they actually fit correctly. I tend to trust Igor more than a reddit users. DDR4 RAM cooler to DDR5? Not a good idea! The differences in detail including 3D scan and instructions WWW.IGORSLAB.DE Today it's about non-existing coolers for unavailable RAM, so once again pure unicorn cinema for all who are only missing the RAM for the new Intel system with

I also use an OptimusPC water block. I think they are by far the best. The Signature and Foundation water blocks are built extremely well in addition to working superbly. My radiator is a MORA 360, but I don't use any kind of radiator when I do extensive benching because the radiator conflicts with the efforts of the chiller and prevents the water from getting cold enough. I simply reroute the lines with QDC fittings and bypass the radiator. I'm also planning to delid and go bare die. I have been running bare die since 7th Gen and I wish Intel would offer a no IHS option for their processors. Everything would be so much easier that way. Some people say bare die is dangerous, but laptops have almost always been bare die. Even the mobile Extreme CPUs were.

Yes that is Cinebench stable with 0 AVX offset. This is my first MSI motherboard and I've not become intimately familiar with all of the settings as I have with Asus and EVGA firmware. The LLC is set on 6 but that's not the same as 6 on an Asus board. It's probably equal to about 5 on Asus. But I'm just guessing LOL. EVGA does LLC settings the best by expressing the values as a percentage rather than a meaningless number. This firmware seems like it is geared more toward tickling the aesthetic focus of a gamer than a no-nonsense overclocker the way things are laid out. And of course, as usual, some of the terminology is the MSI version of something Asus, EVGA, or another brand call something else. I wish all brands would adhere to stricter nomenclature standards. My video card still runs at x16 even with all of the m.2 slots filled and even with a sixth NVMe SSD inserted into the open X4 slot on an add-in card. I do have the onboard Wi-Fi/BT and second NIC disabled in the BIOS since I have no need for any of them and that may be allowing it to run at x16. It is unfortunate that all of the enthusiast motherboards are using UWP trash control center software. MSI Center is not as bloated and life-sucking as Armory Crate, but it is still rubbish. The UWP version of Mystic Light is every bit as trashy as Armory Create.

Thank you for your words of wisdom and concern. I sincerely appreciate them. I am not sure why Benchmate showed 1.51V unless it was an overshoot issue, and I will monitor that closer. I had the BIOS set at 1.475V override. I definitely don't want to kill this excellent CPU sample. It runs 52/41 on cores and 43x cache at 1.275V and 53/41 at 1.295V. My bigger concern right now relating to my daily driver clock speeds and not benching is the PCH temperature. Doing nothing (just idle) the PCH hovers around 80-85°C and I have seen it at 90°C. I have all five NVMe slots filled. If I remove a couple of them the temperatures drop about 5-8°C. I guess I am going to have to do some kind of hardware modification to keep the PCH from committing suicide. The PCH voltage is set to 0.850V (BIOS default value) but instead of "Auto" I switched it to override in an effort to try to control the high temps, since I don't know that "Auto" is not higher than default.