Mr. Fox

Well, the new 3DMark Solar Bay benchmark is quite the joke. Another sloppy mess from the clowns at UL. It is designed for laptops, so maybe the thought process is that it is a toy, not a tool, and accuracy doesn't matter. It seems very unpredictable and unreliable. Very wild extremes in scoring between runs. In less than a dozen runs on the same system I am seeing a spread of more than 10,000 3DMarks. Some people are posting illogically high benchmark scores without GPU overclocking and I think there are a lot of bugged submissions. It does pull some hefty watts with the ray tracing in spite of the erratic results. There is no CPU scoring and changes to CPU clocks doesn't seem to affect much. Example: all over the board, erratic results (same computer, only minutes apart) and the same erratic and inaccurate scoring is visible looking at details in the 3DMark Hall of Fame. I find all of this puzzling and unamusing. I've never looked to LTT for valuable information. It has always been entertainment with a PC tech theme, and I don't really understand why Steve or anyone else gives a rat's butt about the flawed information from LTT. He could point out that the results are inconsistent with his own testing and the testing of others without starting a schoolgirl cat fight. The kiddos that believe everything they hear or watch on YouTube aren't the sharpest tools in the shed. If not Linus, they'll believe some other source of nonsense that entertains them. It wouldn't surprise me to see a lot of erroneous information from LTT and other YouTubers, but I doubt the mistakes and random manifestations of incompetence are deliberate or with nefarious intent. That only helps Gamers Nexus stand out as being more accurate. All too stupid for me to want to burn any calories caring about it, or how it ends.

I might have missed something. Busy day today. Is there a video that you are referring to brother?

A lot of people don't know this about me because of the things I do that they see causes them to assume that I constantly abuse my components, but most of the time I pamper and baby my hardware. I almost never run long and extended strenuous stress tests on CPUs or video cards. Yes, I do overclock the crap out of them but usually only for short bursts, never extended stress. Everything is always overclocked but rarely subjected to extended stress. I prefer drag racing over gran prix and I would prefer landing a single death blow over winning a fight. In and out, maximum damage with minimal effort. Take a bazooka to a knife fight.

Yes, we live in sad days and we are surrounded by compromised trash made with a Dell mentality. I have to chuckle a little bit about the Dell hate we both have because it hits too close to home for their betrayal of Alienware customers. The underlying truth is that all laptops are now equally trashy piles of feces and no more or less trashy that Dell's broken garbage. Pretty much similar dung when we look at the big box prebuilt desktops from HP, Lenovo and Acer. All proprietary trash boxes like the Dell desktops. They are all worthy of our hate and condemnation at this point. We expected more from Dell, once being the best, and we expected nothing less than great things from Cablemod. Until now, we haven't seen anything bad from Cablemod. I try to keep things in perspective and I don't want Cablemod to be crucified for a single mistake. They have been noted for producing quality products for a very long time and only this adapter for the 12VHPWR abortion hasn't turned out good. Nobody is perfect and the fact that they are paying for the mistake whenever something unfortunate happens is deserving of recognition. Their custom cable kits are excellent and nobody would be having anything but a trouble-free experience had the Green Goblin not invented this ridiculous, crappy, unwanted, unnecessary and unimportant 12VHPWR connector. It would be business as usual for Cablemod and just another day in paradise for owner's of the most overpriced graphics cards the world has ever seen. We live in a world where basically everything sucks now. From the food we eat, to the cars we drive, to the idiotic political agendas driven by the most worthless, feckless, evil and corrupt national leaders, to the anti-family, self-absorbed and godless society that caters to the world's lowest common denominators and scummiest, repulsive and most misguided wastes of DNA among us. The human race is circling the drain due to stupidity, ignorance, selfishness, greed, corruption and an overwhelming perversion of everything good, right and decent.

Nothing is going to help the trolls on reddit and I'd prefer to spend my time on something else. Whining on reddit won't fix anything and I don't have a vendetta against Cablemod. I appreciate that they are taking care of the people who have had their GPU connector melted while using their adapters. That negates any legitimate incentive for class action and is a demonstration of uncommon valor in a space usually devoid of companies that do the right thing. We have seen way too many examples of companies that do nothing, say nothing, and don't care, and those are the ones that deserve to be financially damaged, or exterminated, in lawsuits. The only winners in a class action are the attorneys that become richer of off the misfortune of others. The victims just get screwed a second time. They just have to wait a long time to get re-screwed.

I try not to be associated with reddit. I hate that place and generally steer clear of it. Way too many emotards and I don't have time for their nonsense. Right up there with Facepoot and Tik Tok on the scale of morbid stupidity.

Maybe I should sell mine on eBay. It has been sitting on my desk since I went back to the open bench configuration. I have a couple of aftermarket 12VHPWR cables I could sell as well. I got the invitation email, but I don't need an aftermarket part with the stock ATX 3.0 12VHPWR cable. I actually like the stock flat Corsair PSU cables design better than aftermarket individually sleeved cables anyway. I also think they look better than individually sleeved cable because they are all bonded together in a flat side-by-side ribbon design. No need for combs and other silly stuff to have neat and tidy wiring. The only one that I think has some room for aesthetic improvement is the 24-pin main motherboard cable. It looks a little bit messy at the motherboard connector because some of the wires have to cross one another out of sequence with the ribbon order in a few places for the correct pin-out. (They could have done that on the hidden PSU end of the cable.)

100% load at over 300W sustained for 5 minutes is more severe than real-world testing. Most workloads, and especially on a "gaming PC" scenario, do not come anywhere close to that. This is also why stressing a CPU for hours is kind of stupid and pointless. All one is accomplishing in doing so is potential fatigue of the silicon. It is neither helping nor proving anything. Demonstrating the thermal qualities of one particular case and fan configuration is outside the scope of testing a cooling solution and would merely convolute the results, especially comparing against an external radiator. That would be useful to demonstrate the superiority of an open bench or external radiator, but not a useful test otherwise. It might be more relevant inside of a case with a custom loop and 360mm radiator of similar dimensions. Again, this was a test of the AIO and not an opportunity to showcase the benefit of a custom loop and open bench instead of an AIO. The more relevant data was the comparison of the air cooler to the AIO. That is where the real world benefit would be most easily relatable and recognized by the average gamerboy.

Correct. You're exactly right. Everything matters. Fit of the cold plate, design of the cold plate, the TIM, water moving through the jet plate fast enough, but not too fast, or with excessive turbulence, and the ability of the radiator to pull the water temperature down as close to ambient as possible. If you get one of those things right and miss the mark elsewhere, results will be diminished compared to what they would be if everything worked together for an optimal outcome. And, if everything works as optimally as possible then you're still limited by ambient temperatures. It is all very simple to understand, and the science is very simple, but the execution sometimes isn't great. If you're running your PC in a hot environment, even with a perfect cooling system, you're going to have a sucky experience in terms of thermals. It is easy to get sidetracked by some specs as well. We would all choose higher speed when making a selection. But, pump speed doesn't actually guarantee flow rate just as fan speed doesn't guarantee CFM or static pressure. The destination matters more than the journey and the outcome more than process.

I think you might be jumping to conclusions and assuming too much. That's a common problem with readers of product reviews and we are all guilty of it to some degree, myself included. The testing isn't complicated and the more complicated you make it the less accurate the results will be, and the easier it will be to confuse yourself and anyone reading the review. It's actually very simple. The water can never be colder than the ambient temperature unless you use an artificial means like a water cooler. If the transfer of heat from the CPU to the water is roughly equal and the water carries the heat away at an equal rate then you're going to see the same result as long as the coolant temperatures are equal. It's that simple. Pump speed and flow rate matter, but they only need to be adequate to accomplish the purpose of carrying the heat away in the water. At a certain point increasing either one does nothing to change the outcome. I think it's even possible that increasing the flow rate too much might even slightly hinder the results, but certainly won't help if the flow rate is enough to remove the heat from the CPU cold plate at the same rate. The outcome is changed by how effectively the heat exchanger (radiator) removes the heat from the water to equalize it with the ambient temperature. It's also possible for the flow rate to be so high that the water doesn't remain in the radiator long enough to be cooled as effectively. When everything is working efficiently and effectively the difference in temperature of the CPU resides largely in the temperature of the water. Being bigger and more expensive and more complex doesn't help if the water is not cooler. What my testing didn't show, and did not need to show, is how much faster the water temperature would increase in an AIO with such a tiny amount of water in it. That's why my stress test was limited to 5 minutes. Had I run it for 15 to 30 minutes the scales probably would have been heavily tilted in favor of the custom loop that holds probably 10 times as much water and has roughly 300% more heat dissipating surface area to prevent the water from heating up. The intent of the review was to show the function and performance of the AIO, not highlight how an AIO is inferior to a large custom loop with much higher coolant capacity and flow rate. That's also why I tested it on the open bench and not inside of an enclosed case. Testing it on the open bench helps normalized the results. Testing inside of a closed case would produce more erroneous results that would vary widely from one system to the next depending upon the case. The open bench tests the cooler, not the case. This is a common and accepted testing methodology among people doing product reviews. You don't want something like a case being a wild card to skew results. The only way testing inside of a case would be reliable is if everyone purchasing the product uses the same case and the same fans, runs the fans at the same speed, in the same environment. The other element that your statement doesn't take into account is that if you're only going to cool a CPU and you have a really nice, cool working environment with pleasant ambient temperatures a good AIO that can minimize or prevent the water from growing warmer during use then it's probably enough to get by. If you're going to add a GPU and/or memory into the loop the AIO is not even an option.

New single from Dokken... While I am not a big fan of Country Music, the message is right and there's no way I can't support such righteous lyrics. The fact that the woke left is having an emotional breakdown over it is just icing on the cake. I hope he wins the lawsuit he started. Pretty cool song.

That's good. Glad they did that. As long as everything remains consistent in the testing, the other conditions are not that important if the intent is to compare them to one another. I took a similar approach to try to replicate the miserable state of tuning that would be likely on the average gamer noob setup. My office is seldom that cool. I wish I could keep it at 23°C or lower, but it is nearly impossible to get it that cool in the summer months. Usually 25-26°C (78-80°F) is about is cool as I can get it with the AC set to 69-70°F. I need to do something like add a booster fan in my central AC duct or something because the rest of the house stays nice and cool. Even in the dead of winter it is difficult to get it down to 69-70°F (20-21°C). I think the main take-away from my cooler testing was confirmation of what all of us already knew. If you're using ambient cooling for air or water, you will never be able to be cooler than the environment, even with a zero degree delta. The fact that the EK AIO and my custom loop had nearly identical temperatures was a direct reflection of that law of physics.

https://hwbot.org/submission/5329954_ | https://www.3dmark.com/3dm/98374064 Same run above GPU score only: https://hwbot.org/submission/5329955_ https://hwbot.org/submission/5329959_ | https://www.3dmark.com/spy/40812238 https://hwbot.org/submission/5329976_ | https://valid.x86.fr/6fjmjp CPU-Z bench single: 1025.3 | 18714.8 That's all for today. My ambient temperatures have increased and now the chiller can't get as cold as it was when I started. GPU boost won't go high enough.

I was editing my post when my power went out because I tripped the circuit breaker while benching. I was going to change it to ask if they controlled the ambient temperatures in comparing the AIO results. If they did not disclose that and take the steps necessary to do that then the results may not be accurate for the comparison between any of them. Hopefully they did. I have not looked at the review. Yeah that's extremely retarded about the caps around the perimeter of the IHS. I guess I would have to break out my soldering station to take the caps out of the way to do the delid and then solder them back again. I'm sure that someone like Roman will figure a way to do it because they did the same idiotic moron thing on the latest AMD processors with caps all around the perimeter. Where there's a will there's a way, but it's still idiotic that Intel or AMD would build a processor that way. Totally sucks. I'm probably not going to upgrade anything for a long time so they'll have time to figure that out before I do, if I ever do. The way things are trending I can see myself losing interest in the hobby, and especially losing interest in wasting so much money on overpriced garbage.

Thank you, brother. The question I would have is were the ambient temperatures at least 5°C lower than mine? I would guess my ambient temperature was at least 5°C higher. If that's true, maybe it's not better. did they disclose the ambient temperatures when testing each AIO and run the comparison with each system in the same operating conditions? If they did not then the results may not be accurate. If they did and did not disclose it, they should so that question does not go unanswered. I haven't read the review(s).

The mirrored video image is a little bit annoying, but otherwise this is pretty crazy stuff. Incredible that it actually worked. Brother @ryanI bet you could get back into desktops without a huge investment doing it this way.

Wow. Just. Wow. I had not heard about this until just now. This is totally Satanic crap. I don't want to stop using Chrome, but this makes me want to see very, very horrible bad things happen to the bastards at Google. This wicked Nazi dictator behavior is inexcusable and it is deserving of death for the company.

Sometimes Bill's videos can be too painful and boring to watch from end to end. The guy is super smart, but there is often too much rambling about stuff I don't care about. The problem is if you skip through it and don't pay close attention you can miss something really important because he sprinkles valuable little nuggets here and there among the ramblings. So, it is best to endure the rambling if you want to extract useful information. I've gotten a lot of useful information from his videos, but I don't watch them if my time is limited or my attention is divided. I do enjoy his smugness and sarcasm, and I think I would get along with him well at a personal level. I suspect that he has less experience with AMD memory overclocking. Most of his videos relates to GPUs and Intel-powered systems. The concepts are similar but the difference in working values and settings are often misleading or not understood. Some settings on one platform do not exist on the other, or they are obscured by different terminology and naming conventions. Beyond benchmark scores it can be difficult to compare Intel and AMD overclocking. What works for one often will not work for the other. At the end of the day, the only things that matters are the experience and the benchmark scores. If you're not happy with both, or frustrated with the experience, then you're on the wrong path and need to do something else.

It's all good brother @jaybee83. Kind of funny really when you think about it because I have played games more on the 3060 TI and the 6900XT that replaced it than I have with my incredibly more expensive, powerful and capable 4090. Couldn't tell you why since the systems are within an arm's length of one another. When I'm using the benching rig it's generally at a time that my focus is on benching, but otherwise I have no explanation for why it has worked out that way. Just random. But what it does show is how capable at gaming both of the inferior GPUs actually are. I wouldn't have gamed on them if they did not perform admirably.

Here it is, bro. 👍

Sorry if I wasn't clear by what I said. I never mentioned gaming. "Outperform" doesn't mean "gaming" to me. "Gaming" means gaming, and it's something I seldom do. Overclocked CPU benching and CPU-intensive tasks, where core count and clock speed does matter, is what I meant about outperforming. And, you are correct that the benefit of more cache when more cache is helpful is brand agnostic. Because I am not a "gamer" per se, I generally do not think in terms of the effect on PC gaming. Even so, what matters to me most generally (except for the above) benefits gaming as well despite the fact that it probably wasn't contemplated in my thought process.

That will be a super nice upgrade to the 2700X. The 10850K should outperform it having more cores/threads, higher stock turbo clock and great overclocking capabilities, but it would be interesting to see your comparison.

It booted. Must be good to go.

Well, 87.1 ns is the crap you can expect from using an XMP profile, so I guess we know where his weakness is when it comes to tuning. Thanks. So, now we have grandsons that are named Magnus, Sigurd and Bjorn. That means they can't stop until they give us an Ivar, Hvitserk and Ubbe. (I'm not sure those last two names aren't too weird for the 21st Century. But, maybe they aren't strange in another part of the world. Not sure.) https://screenrant.com/vikings-sons-of-ragnar-power-ranked/#magnus