Mr. Fox

Let me know when you get ready to delid. You will need some QuickSilver and Flitz, as I don't have any extra on hand.

Dude... yes you do. You're just not remembering. You have a brand new tube of TF7 in the box with that brand new Thermalrite CPU frame that I gave you as a freebie with the mobo. It comes with the CPU frame. And, TF7 is actually a decent compound, so you should be set.

It looks like they are sold out already.

I have owned several TeamGroup Delta DDR5 kits. They make good products, and the Delta RAM is some of the best money can buy. Better than G.SKILL in my opinion. One word of caution about them. Look at how they are shaped. The weird appearance is not the issue, but clearance problems could be. They presented a fairly serious clearance issue for my 3090 backplate waterblock. I had to move the waterblock to a less favorable position in order to install the memory. Just something to be aware of, because the problem is not obvious until you encounter it. If you need to RMA TeamGroup products, you have to pay to ship them back to Taiwan on your own dime and wait for them to process it. So, don't expect anything to happen fast if you have a problem after the easy return/refund period passes with the vendor you buy it from. Awesome. Thanks for sharing this.

How do you know this? Is there a video somewhere I missed? I haven't looked for a couple of weeks, but last time I checked there was no verifiable info that I could find. Depends on what you are wanting to accomplish. In almost everything the performance using that memory kit overclocked to 6400 produced results that were roughly identical and indiscernably different than SK Hynix running 6800. AIDA64 benchmarks were better, but it was difficult to quantify improvement in most things. That said, going from 6400 to 7400 should provide a benefit. It will still be somewhat dependent on what you are doing whether it will be measurable or meaningful. Paying $400 for two 8GB memory sticks is pretty horrible. I bet you will be able to buy A-die for about half that much in 6 to 8 months. All of the early adopters are clamoring to get their hands on it and it is rare right now. It will probably soon become common and inexpensive, just as we saw with M-die. It started out rare and expensive too, 6 or 8 months ago.

That is an indication of an issue with a NewEgg employee(s) or crime somewhere else in the supply chain. Hard to know where, but probably someone that works there is stealing or people in charge of returns are not doing their job. Don't forget about the CPU that I bought from B&H photo that someone returned that was an older CPU with a new CPU IHS on it. I bought it as new, not open box. That doesn't mean B&H tried to pull a fast one on me, but it does say something about the QC and management. Good help is hard to find. I am not excusing it at all. It is inexcusable. I am only saying it is not necessarily an indication that NewEgg is a dishonest company that has a goal of defrauding customers as much as mismanaged or hiring sketchy employees that belong in jail or 6-feet under instead of working at NewEgg. If it is an employee problem, they do not deserve a job and the only real solution for that kind of problem is euthanasia. People that do things like that are sick puppies that do not deserve a place in our world, and do not deserve free (taxpayer paid) room and board in prison. People will stop using them when they have another option. I used to have a Fry's here. Not as good as Microcenter, but better than Best Buy. For a few things (not much) Best Buy is OK. They're not awesome and they don't sell 95% of what I want to buy. So, my options are NewEgg, Amazon, or pay more than something is worth somewhere else that also has a terrible returns policy and charges a lot for slow shipping. If I stop using NewEgg, that leaves me without an option in some cases. Their prices are often better than Amazon.

Yeah, it makes a huge difference, especially since Turing and Ampere have uncontrollable adverse behavior affected by temperature long before the GPUs actually get "hot" and need to throttle. This air cooled GPU is the only desktop GPU I have owned other than a 780 Ti and it is getting seriously hot when overclocked. I hate air cooling, LOL. It sucks. Here is another before and after comparison showing what a huge gap exists between 3060 and 3060 Ti... whodathunkit? That a huge leap between the two. https://hwbot.org/submission/5103306_mr._fox_catzilla___1080p_geforce_rtx_3060_(ga106)_31237_marks https://hwbot.org/submission/5103815_mr._fox_catzilla___1080p_geforce_rtx_3060_ti_(ga104)_40137_marks Do we already know that 13900K is exactly the same under the IHS, with no hidden surprises lurking there? I have not seen a delid video yet that shows us that a 12900K delid tool is safe. I have one you can borrow if you need to. I know it looks the same on the outside, but there could be some SMDs or other landmines. Unless I missed it, Roman hasn't popped the top on one yet.

I am still planning to try to fix the 2080 Ti FTW3. That GPU has been too wonerful to throw away. I just haven't had time to do it. The 3060 Ti never would have been purchased otherwise, and I chose it because (a) it was the cheapest half-decent option available, and (b) Intel's Arc A770 wasn't available. That A770 would have been my first choice for a low cost GPU, and the performance is very similar to the 3060. This purchase was based on business need as it involved the computer I use for work and I use my triple-monitor setup to the point of being crippled without it. I could only use two with the Intel iGFX and using a USB Type C docking station did not support the third monitor. My original plan was to spend nothing and just use the iGPU until I figured out whether or not I could repair the 2080 Ti, but that did not work out. Overclocking doesn't require the newest or most expensive hardware to be fun. Wanting the lastest and greatest is a first-world hangup for me and most of the people I know and like the most, LOL. Even if I hated the 3060 Ti and viewed it as a piece of garbage I would still overclock it and push it to the edge of functionality, just because I like doing that. Even BGA trash that can be overclocked is fun to play with to some degree. It's all fun and games until someone loses an eye.

It does seems weird for sure. But, we are also talking about my "work" computer not my benching beast. And, yeah... totally economic. For economic reasons I may be hanging it up as a matter of having priorities straight. It has never been inexpensive, but I'm just not OK with the absurd and abusive extortion-level pricing of desirable tech, and every time I say yes to overpriced stuff because I want it, I become part of the problem and a direct contributor to their criminal behavior.

True, but if it runs long enough to let you pass benchmarks with a higher score then it just doesn't neeed to be daily driver stable. You can do daily driver mediocrity when you're not benching. (This is obviously a statement for the benefit of readers, not you, LOL.)

Yup, a nice bump for $20 less, LOL. And, this is with it running crazy hot using a Strix vBIOS with a 65W higher power limit. I guess I am going to have to shunt mod this little scrapper after the water block gets here. Before/After links... https://www.3dmark.com/compare/spy/31679598/spy/31408485# https://www.3dmark.com/compare/fs/28627838/fs/28566515 https://www.3dmark.com/compare/sw/38525/sw/13298

Well, I got a GPU upgrade for less than the Zotac 3060. I grabbed a 3060 Ti FTW3 on sale from EVGA for about $20 less than the low budget Zotac. Zotac is on its way back to Amazon for a refund. I found out after it was too late that the Zotac 3060 (maybe all of them) have a different bolt pattern around the core and nobody makes a block for it. I have a used EK block incoming for the FTW3. All I should need to do is remove the coupling in the line and screw the compression fittings into the block. (The G1/4 female-to-female coupling is in there now between the compression fittings because I had to remove the 2080 Ti from the loop and that required less effort than messing with tubing.)

Ummm, how 'bout no. No workey. Option to buy online and pickup at the store like Best Buy doesn't seem to be an option for Microcenter. Yeah, totally. It's a good business model even though it is of zero benefit to me. I don't blame them for running things that way. It's really smart on their part.

I am considerably more jealous about the access to a Microcenter that I do not have and may never have than I am the 13900K. That's just wrong, LOL. 🤣 I would buy things from them online, but 100% of the time 100% of anything I would buy is "in store only" so they may as well not exist for me. I am not sure they actually sell anything that involves shipping. This is good for them and good for the people that have access to their stores. It only sucks for the people that don't have access. It looks like you can maybe complete a purchase online, but still have to pick it up at the store.

Thank you. I appreciate the kind words. The first issue I would have with that is him. He would not be down with that. He's almost 30 so he's not a little boy that I can push around. The second issue is that it is a laptop and, while I have one, they're not very interesting to me anymore. I have some fond memories of better days and they will be with me forever. But, even if it were built with current generation desktop CPU and MXM graphics, that ship has sailed and I chose to get off of it while it was still docked. I saw the storm on the horizon. I wouldn't even want it at this point. Desktops are too much better for me to embrace the idea. I came to peace with the idea that laptops are dead to me. The ease of rejecting the X170 was the last litmus test. It just felt good and natural to say no to it. Your comments made me feel happy though. Thank you.

Dang. That makes me want one real bad now, LOL. I guess this is a rare situation that it is good I don't have a Microcenter here. I'd be buying it tomorrow at that price instead of waiting for KS. You would be surprised how many places that you would think have it do not. Some larger cities have very limited options for ISPs and if they do have an option for it they get raped on the price due to zero competition. Where I live you have the option of grossly overpriced Gigabit with Cox cable or crappy Centurylink DSL with a max speed of 20 Mbps. TMobile 5G is blistering fast a half mile away and totally unusable at my residence. Due to the lack of competition, I pay significantly more for my Gigabit service than Cox customers do that live in Phoenix and other areas serviced by Cox where they have to compete for business against another ISP with equivalent service. Random info... My youngest son has an Alienware M17xR2 that I gave him about 10 years ago. He lost the AC adapter 6 years ago and hadn't used it since. He decided he wanted to use it and I happened to have a spare AC adapter. He brought it to me to clean up, so I replaced the anemic HDD with a 1TB SSD and did a clean OS install. I had an unopened Alienware Windows 7 Ultimate DVD I used for the OS. It runs like a champ... 4780 Crossfried still working nice. Temps were OK, but it was filthy dirty inside. So, I took it apart for cleaning and thought I might as well repaste and repad. It was sobering experience to be reminded of what an awesome laptop used to be built like. I had forgotten what amazing quality Alienware used to have before they started churning out chintzy garbage. It was a real pleasure to work on it. Weirdly enough, it still had the original phase change TIM paste and original thermal pads and the temps were totally decent before I took it apart. It had never been taken apart... still a virgin. I bought it almost new from @Batboy at NBR as a gift for him. The heat sink radiators and fans were filled with a bunch of fuzz and crud. The thermal pads were still like new... soft and sticky, so I left them. After a thorough cleaning and putting it back together with KPX on the CPU and both GPUs and unobstructed radiators the temps were just a little bit better. Remarkable in every way. Amazing build quality, gorgeous aesthetically, as heavy as a Buick, LOL. But, VERY old and slow by today's pathetic and trash crap gaming notebook standards. It brought back many fond memories of traveling with one of these beasts and staying up late at night talking with Brother @johnkssson the phone and benching it on top of a hotel room AC unit.

I am not 100% sure we are talking about the same thing, but I do see low temp throttling in practice as well. So annoying. In case the GPU is lightly loaded, say < 20% the clocks dial down to sometimes as low as 50-60% of max. This is despite max performance power mode being set in both Windows and Nvidia control panel. If anyone is aware of a workaround, I would be grateful. The fact that Jensen said it makes it hard to believe for me. He's not someone that I would consider a reliable source of information. I don't think EVGA is going anywhere, just excused themselves from the GPU business. I think EVGA's statement on the matter is the one worth believing. You can circumvent the silly dynamic clocking baloney using EVGA Precision X1. Click the button to lock in full boost clock regardless of GPU load. I think it is better than Afterburner for most things.

EVGA 850W PSU that is a bit more than a year old is being replaced under warranty. It is overheating and shutting down because the fan stopped working. Placing a 120MM fan in front of the exhaust vent corrects the problem I contacted them Monday afternoon and it is out for delivery today. There is a pre-paid return label in the box for me to ship back the original. This is how you treat customers if you want them to choose your brand first when the option exists. MSI, ASUS, Gigabutt, ASSRock don't care... they suck. They make you pay to ship them the part, then wait for them to verify that you're not a liar, (a week or longer,) and with no sense of urgency or gratitude for your patronage, they send out a replacement when it is convenient for them, the slowest and cheapest way possible. Every time my ASUS products have been replaced under under warranty it takes 3 to 5 weeks to resolve. Nothing they say or do demonstrates that they give a rat's butt about the people that buy their products, because they do not.

They are doing what is best for them because too many people are stupid with their money. If new GPUs like the 4090 Suprim Liquid X and ROG Strix were being treated with a reception as chilly as the new AM5 platform, we would not see them or ASUS behaving in a manner so worthy of disrespect. They would be more conservative with their pricing because overpriced garbage sitting on store shelves hurts them badly. Making a small profit is better than no profit or operating at a loss. They would stop producing products that they are unable to turn a profit on. Prices follow what the market will bear, and as long as the market is composed primarily of retards, prices will remain retarded. On top of this fact, we already know that ASUS and MSI are both dishonest companies run by people that don't care about their customers. This is evident in the way they treat people after they get their money when they sell them broken garbage. So, MSI is being more transparent by showing their true colors before the sale instead of saving it as an icky surprise after the sale.

Yes. And, it it not a new issue, unfortunately. It is an unresolved issue that will never get fixed because it is deliberate and intentional.

It's not entirely accurate though. It starts about 45°C from my vantage point. Boost clock goes up more than it does above 45°C, but people with air cooled cards and normal water cooling don't know that and they can't see it because they cannot achieve or maintain temperatures that low. I don't know at what temperature it does not occur or where going colder no longer increases boost clock, or what temperature would have to be maintained for the GPU to reach max boost clocks and remain there because chilled water isn't cold enough to show me or prevent "thermal throttling" LOL. What I can tell you with certainty is that I really loathe this kind of dynamic clocking crap more than I can say without resorting to use of vulgar language. It is totally stupid and unnecessary and the existence of the phenomenon is a symptom of wickedness and an underlying darkness of heart.

Yes, that is my preferred way of dealing with it. Just kill it and move on. End of the drama forever with Defender Antivirus. Remove Defender.zip

I think it can vary by GPU model and generation, but Turing and Ampere are when they went full spaz on us. It is roughly a 15MHz drop for every 5°C increment over 45°C if memory serves me correctly. TechPowerUp had a piece on this but I am not sure if NVIDIA has publicly posted technical details on how their firmware cancer algorithm operates. Some people believe Ampere is less aggressive about it than Turing, but I think it is not so. What changed is the clock range seems to hold better with Ampere between 45° and 70°C than Turing did, but with Ampere more than Turing, the boost clocks remain higher below 45°C. So, people that are not running an Ampere GPU below 45°C are unable to identify that, and that means basically nobody with an air cooled Turing or Ampere GPU can see the beneficial effect of staying below 45°C https://www.techpowerup.com/review/nvidia-geforce-rtx-3080-founders-edition/39.html With mining the issue is the GPU memory. The GPU core doesn't really get used. You can overclock the crap out of the core and the hashrate doesn't change much. The memory clock is what matters and it is the memory that gets cooked. What is the most goofy about any of this is the core is what "thermal" throttles at low temperatures, even though there is no legitimate reason for it do because the thermals are excellent. The memory, on the other hand, can be overclocked until it totally burns itself to death and there is no protection mechanism for it.

Chiller + radiator keeps the temps from going low enough for condensation, but keeps the GPU cold enough to avoid room temperature thermal throttling that degrades performance even though the GPU is not hot enough to need to be throttled. You start losing clock speeds rapidly and performance begins to decline around 45-48°C and progressively worsens as the temperatures rise. It is unfortunate that Turing and Ampere technology began this aggressive psychotic behavior. There is no reason for the GPU core clocks to move down from maximum boost clock until core temps start pushing past ~75°C.