For Science!

as I said, yes, there are differences, but if it is not something you (and I mean you, not somebody else) can perceive, then is it really a difference? Frankly at the moment until you are able to remove your GPU, this is a moot discussion point.

Last try. way too much effort that this deserves... You will see in this video that after I unlock the PCIe retention lever, the GPU is still stiffly bound enough that I can actually pull the entire PC on my table. Only when I put my finger on the CPU cooler to apply counter-force, does the GPU puil-out. You can see its a decent amount of force, given the way it yanks out, which is partly because of my one handed operation and lack of a PC case.

Theoretically maybe, but realistically for you, no. As you've stipulated in the other thread, its unlikely you would even notice if it was operating in PCIe 3.0 mode instead of 4.0, or at SATA speeds for that matter. So no, you aren't going to notice it for your use-case. I still think it makes most sense to install in the first slot, you've been trying to remove the graphics card for almost 2 years now. I think its time you learnt how to do it, and stop looking for excuses or workarounds to avoid learning how to do something as simple as removing a graphics card.

The 2nd slot always operates at PCIe 4.0 as it is using the chipset PCIe lanes and not the CPU PCIe lanes. But as the 3rd gen Ryzen supports PCIe 4.0, and so you will get 4.0 on essentially everything.

The retention lever is fine, you've already unlocked it. It won't move because it is open, this is not rocket science. You just need to pull the graphics card out of the PCIe slot. Stop getting hung up on the retention lever.... Just pull it.........!

The clip is fine where it is, it is already in the unlocked position. Looks like you just need more force in the right direction (you are moving the card up and down in the video, you need to pull it towards you) to pull the card out. And no, if you do it right, there is no damage to the card or motherboard. The force required to remove a graphics card, in my opinion is about the same as removing a EU plug. Or alternatively, a little bit less than prying two large Lego blocks apart and significantly less than prying two small Lego planks apart

Then honestly its unlikely to be that. Did you also rinse the radiators with hot water first? What kind of coolant are you using, and what are you diluting it with (if anything). It probably is just a matter of draining, cleaning, and re-filling, but good to have a think about what it was in the first place.

Do you think its plastic shavings? it looks a bit too much to be crap from the radiators. I guess you would know if you deburred the edges of your hardline, and didn't subsequently rinse out the tube which had flakes of plastic inside it.

This is hardline? looks like shaved off plastic to me that may have come from deburring the cut edges? Alternatively, the go-to answer would be crap from radiators.

Finally got a half decent print of a GPU vertical bracket that I designed and 3D printed. I couldnt make it stiff enough to counteract the sag, so I just tilted the whole thing so that after it sags its more or less straight. Not a great solution, but it looks okay. Designing and 3D printing seems to open new windows into keeping the hobby fresh.

For what its worth.... Put the case down on its back panel so that the GPU is up, unscrew the screws on the PCIe bracket, and then push down the PCIe lock and then pull the card up. Maybe not pulling horizontally and instead vertically will make it better for you. Otherwise as I have said 1 year ago, get a friend in to show you in person. There is no trick. Posting a video of exactly what you are doing may also let us know what youre doing wrong.

For reference to those uninitiated:

Don't forget about the two other screws on the I/O bracket

Please have a re-read of the sentence, what doesn't matter is where the radiator is, its location. That is not like saying you don't need a radiator.

efficiency and effectiveness are two different things. Sure, the radiator will be more efficient if the water temperature is 90 degrees over ambient, but as a cooling solution, it will not be as effective if the water temperature was 20 degrees over ambient. "The loop order does not matter" is a child of having enough flow rate, where with adequate flow, where the radiator is not important. As the flow rate goes to molasses rates, then it becomes a factor once again.

No, you are not missing some nuance, flow rate beyond molasses speed simply does not have a large impact on thermals.

Nvidia mods/mats is the program they are running, Tiny Linux is just a minimal OS that probably houses nothing but that program. But maybe you knew that already too.

If it won't eat away at the O-rings and can be handled safely and you have responsible disposal routes, then I guess it is fine. A bit like using engine coolant in PC loops.

The two pumps running in parallel was probably not a great idea, if one dies, the working pump will push water through the broken pump in reverse essentially creating a large bypass. So you don't get any redundancy from having the 2 pumps, and you don't gain head pressure. Assuming you change the pump layout to series instead, i gathered its a very specific use case distro-plate, which I think is fine.

As others have said before, looks like you’re going to have to hit rock bottom to learn your lesson. Clearly you just want some sympathy and make no effort in achieving anything worthwhile in overcoming your addiction. When you’re sad and hungry without a way to pay for next bill, then let’s hope that gaming laptop will at least keep you warm…

I'm going to assume that its a rhetorical question. When you buy a 1000 euro laptop for 800 euros, you haven't saved 200 euros, you have spent 800 euros. This is a good mentality for navigating around "great deals that you can't pass up" - you never save money by buying something that you don't need.

and that is why "hot air rises" is a non-concept when taking fans into account.... I mean CFM numbers are ideal case-scenarios with zero obstruction, so expect the real numbers to be lower, but the reality is probably not too far off. Whether you also successfully achieve a "new air in, old air out" scenario is also largely dependent on the layout. But yes, I would absolutely not be surprised that if every few seconds a case's air is displaced.

Hard to explain at a tech spec level, but I know that the A12x25 definitely pushes more air than either at the same noise level. Agreed on the iPPC being super loud, thats what got replaced for me.

If you can spend the money: https://noctua.at/de/nf-a12x25-pwm-chromax-black-swap But whether it justifies itself over the cost of Arctic P12s, hard to know. But after spending money on 6 of these, I convinced myself it was the best decision, and nobody can change my mind :P.

Same rule applies generally, good static pressure is desirable. Noise profile may be better if you have the fans in push compared to pull. Arctic P12s are a good choice, also Noctua A12x25. I don't have any RGB recommendations, but I vaguely remember P12s came in RGB...? Alternatively you can use Phanteks Halos to RGB-ify any fan.