Mr. Fox

Fans wear out, but 2 years isn't very long.

It works well for laptops with poorly fitting heat sinks. I tried it on a water cooled desktop and it was inferior to most of the popular thermal pastes. I do not think it is great in terms of thermal conductivity, but it is good at filling gaps

The renaming is only necessary for recovery flashback using the button on the rear I/O. You can flash it inside the BIOS with the original file name. I really do not like how ASUS handles their firmware. It is the worst. The dual BIOS is a joke. It is not a real dual BIOS. That is why when you switch BIOS using that button it goes through a reflashing sequence. If it were a proper dual BIOS it would have a switch that disables one chip and enables another and they would function independently of one another. Because of how they did it, it is not brick-proof. And, one BIOS position is not completely agnostic of the other. It is not a very smart way of doing things. The only right way to do it is how EVGA did it, and how it is done on GPUs. You slide the switch and the other one no longer exists. The ME and EC are not shared from common NVRAM. ASUS used to do it that way. I am not sure why they thought it would be better to do it this way now, but it sucks. The way it was on the X299 Dark was the best, having at least one SOP8 chip in a socket so you could swap out spare chips and manual flash them with an SPI programmer was so sweet. I put one of those sockets on a couple of laptops and it worked flawlessly. That is a truly brick-proof system. Solder once and you're forever free to make your own decisions about what firmware to use.

https://hwbot.org/submission/5520984_ Completed all 4 passes with CPU in parallel mode. Parallel mode stresses the memory with all cores simultaneously rather than rotating amongst the cores.

Yeah, I have historically not liked them at all, but the past couple of GPUs were well made (including my A770). I had a Thai Chi (Chai Tea) motherboard several generations back and is was a sucky, flimsy piece of trash. I do not like their warranty policy. You have to first try to go through the place that you bought it from. Sounds like a slow process, especially if the place you bought it from is a bunch of losers. I think (don't hold me to it) that their warranty is only to the original buyer. If you buy one used in warranty, you're not covered. It doesn't follow the part like most other brands do. If they're open to change and improvement, the former EVGA employees might be exactly what they need to go to the next level. Here's hoping they give the dimwits at ASUS a run for their money.

Either they were not thinking, or they were and it was a deliberately nefarious decision.

That is an amazing first PC. What a blessing that so many parts were donated.

I hope that both of you get back to feeling good again. Seaonal allergies are bothering me right now, but compared to what is bothering both of you that is hardly worth mentioning.

Same with both of my systems. It's not actually "Auto" as much as what they decide to set as the default value so that the most horrible CPU silicon will run without lockups and blue screens. It's not "intelligent" at all. They call it "Auto" and what it really means is "ASUS BIOS Default" value. It has to work for 99% which means it's way more than necessary for probably at least half of the systems, maybe even more than necessary for +80% considering some people are content with 6800 or 7200. The lowest common denominator gets accounted for. If I set mine to 1.250V under load the actual IMC VDD is 1.225V. I haven't tested this on the 24GB A-die modules, only the 24GB M-dies. The M-die uses lower voltage all the way around. Totally stable with the TeamGroup Xtreem 8200 bumped to 8400 and 1.435V. CL38-49-49-84 2T tRFC 700 tREFI 131071.

I'm having issues with having so many bios options set to auto. Can be perfectly stable for a while. Then if I power the system off and leave it for a while and then come back to it and turn it back on it's no longer stable and I have to go back to my old manual overclock settings. I've had this problem off and on over the years with vatious systems, which is why I usually set all values manually and use static voltage values. Then my systems behaves the same all the time even if it might be getting a little bit more voltage and drawing more power than what it needs it behaves more reliably. In some respects it is the same, or the opposite, as pushing a system to the edge of stability on the overclock, or pushing the undervolt to the edge of stability. In either circumstance the edge of what works and what does not moves based on temperature, and unless "auto" works 100% of the time, (regardless of environmental changes,) there will be times it does not work.

Thanks for the suggestion. I did that. It seems to run and perform exactly the same except for only the two best cores boosting to the higher clock. Perhaps in a situation where only two cores are used (maybe in a game or something) it would be more stable that way?

I have 15 Thermalright fans in my systems. They work well and are priced the way all 120MM should be priced. The only thing that would make me like those fans more is if the ARGB was just plain white LED. Then I could just plug in the power and enjoy the white color that I prefer. I refuse to spend a lot of money on fans. To me they are just not worth paying a lot for. All of my systems have so many fans in them that it would cost way too much. Including the MO-RA 360, my system with the white Apex has 27 fans, LOL. I actually loaded one of those AIOs in my shopping cart because it was so inexpensive. Then I stopped and asked myself what I was doing and why I would buy something I have no use for just because it was priced so nicely. Just for giggles, last night I started goofing around to see how high I could coax Cinebench R23 by using as minimal overclock as possible. With BIOS defaults it is over 42K. By increasing the all P-core clock 100Mhz and all E-core clock by 300Hz and disabling cache downclock it is almost 44K. Load voltage is like 1.217V. Most of the BIOS settings are optimized defaults. I set the Adaptive Voltage value stated on the VF curve page with adaptive voltage as the Additional Turbo Voltage value, positive offset and offset amount left on Auto. Memory is XMP Tweaked, Mode 2. But, nothing has been tuned otherwise. Idle voltage is high, but load voltage and temperatures, and power draw, are all lower. Interesting.

Sounds exactly like an ASUS move. Let Intel pay for the CPUs their motherboards kill. At least it was nice for a while. Now EVGA have joined the asshat club that all of the technology leaders hold a charter membership to. I agree. I want only bad things to happen to crApple and Micro$lop. They both suck real bad and deserve bad things to happen to them. Their products are crap and their approach to business is even crappier.

Yup. Exactly. What I said a minute ago.

Intel is the only option. I don't like saying never, but I cannot imagine any scenario that I would consider purchasing another AMD product again. I've never been satisifed with anything I have owned that they made and I don't like the company in general. I've given them plenty of opportunities to change my mind and that has never happened. It would be foolish for me to think that they ever could at this point. I have friends and family that like their AMD products and I am happy they do. I'm not really open to the prospect of dealing with the disappointment all over again. What we are seeing with the crappy 14th Gen silicon samples is a symptom of the world we live in. Everything sucks now no matter what name is on the label. Crappy is the new normal for almost everything, including lots of people.

I've tried using them and they are not good. Neither are the graphite pads. You can expect your CPU load temps to increase substantially using that. They're not even as good as using a sheet of 0.1mm indium, which isn't very good. Better off using paste or liquid metal. Those fittings are nice, but they will not fit well on every system. It depends on how your lines are route. With both fittings being the same height you might encounter contact interference. If I have both lines attached on the same side of the GPU, I have to use a spacer on the inlet port to avoid having the compression fittings and tubing routed the way I want the because the outlet port gets in the way of the inlet. If you have the CPU block and GPU block mounted in series and the lines are oriented at 90° angles to one another the lines would collide, but then the elbow is pointed toward your motherboard. Indeed, they certainly could have. Although, they might end up having too few samples that anyone would be willing to buy as things are looking now with how few good 14900KS samples exist. If I had to take a wild guess based on anecdotal bits and pieces, my guess would be that maybe 1 out of 20 would be good enough silicon not to RMA. I'm not sure why that is. Maybe their binning process is flawed and they pulled the wrong samples for KS branding. That's the main reason I sold my Z690 Dark. The product itself was amazing, like everything else always was... and with unrivaled support and warranty service. But I could see the writing on the wall. When they stopped making GPUs they were doomed for destruction. If you aren't making money, you can't afford to provide good product support and you can't do world-leading product development. You can only provide service if you can afford to pay the employees that provide it.

Looks like Dufus drew the same conclusions most of us have. The 14900KS is merely a scummy scam cash grab designed to trick overclockers into wasting money on an "upgrade" to what will ultimately prove to be an inferior product more often than it proves to be a micro-upgrade. If you are extra-lucky you might get something better, but probably will go the other direction.

I've had that happen a couple of times as well. I bought something on NewEgg that was from a marketplace seller. It wasn't sold and shipped by NewEgg. If memory serves me correctly it was an NVMe but it was a couple of years ago and I don't remember now what it was. The seller used FedEx and it was delivered and signed for by "Dave at Front Desk" (or something similar) in Mesa, AZ. That's about 15 or 20 miles away and a different city. I had to do the same thing. They were saying it was delivered and not looking at the destination being wrong. I had to actually speak to a person to get them to pull their head out of their butt and pay attention. My name is not "Dave" (or whatever the name was) My residence does not have a "front desk" My city is not Mesa.

Something is really goofed up with logistics in general. It definitely seems worse with USPS, but I have been experiencing crappy service from FedEx and UPS as well. That chiller that I ordered that we chatted about got delivered by FedEx to some address in Indianapolis, Indiana. So, now I have to request and wait for a refund from the seller. It wasn't the seller's fault. FedEx got their wires crossed somehow. What's really frustrating about it is the order originated in Chino, CA and had a stop in Phoenix (on day #2) on the way to Indy.

So, it looks like they are taking a unified hardware mounting approach. If what I read is correct, you could have a Thermalright air cooler, remove it while leaving the brackets installed on the motherboard and simply bolt on the AIO pump in place of the air cooler without having to change the brackets on the motherboard. That's brilliant if it works well. That would be especially nice on an open test bench scenario. So far everything I have purchased from Thermalright has been decent quality. CPU mount frames, liquid metal, thermal paste, fans and other things... no complaints.

That is a fantastic price. I wonder if it can be purchased for that price or whether it will be jacked up by greedy retailers? Very interesting mounting design. Looks more like an AM5 mount than Intel. It should make getting even pressure on the IHS easier. Edit: for now at least... if it gets more popular the price may go up. https://www.amazon.com/dp/B0C7Q3F42D/?tag=wccftech0a9-20&th=1

I just finished testing the official 2101 release on the white Apex and it seems good so far. Nothing got broken and it does not seem to have impaired performance. Still need to do more testing to confirm, but initial impression is that it is OK. I had to tighten tWR to 12 to avoid TM5 error #10, but it has been the same on other BIOS versions. Left on Auto, tWR is too loose and causes errors.

Mine changes often as well. It is often not the same between BIOS version changes on either of the Apex. It can vary anywhere from 1 to 5 points up or down. It is often 1 or 2 points lower the second time if I run it twice in a row.

I wouldn't mind getting my hands on one. From what I read they are not going to release them in European or American markets, only Asian distributions. If that is true it really sucks.