Search results

I need one or two more rads to finish an external rad box project I've started, but the XE is EOL and the RX360 is OOS everywhere. Willing to pay retail for very good/mint used radiators (~$90 for RX360, ~120 for XE), price goes down from there based on condition, but as long as the core is in...

Yeah. GTs are still good, but there are better options out there. Note that the NF-A14 is NOT the same design as the NF-A12x25 and the NF-A14 isn't actually all that good.

It's because AMD is supply constrained. They don't have to lower prices on their desktop parts because they're going to sell every single piece of silicon they produce. Even if demand falls off in the face of ADL (which does appear to be happening, just very slowly), they can just repurpose...

Sounds like you have a bad QUADRO.

Make sure you haven't offset the connector by a pin or something (I've done this before) and if that's not an issue, try connecting it to a motherboard fan header to see if you can read the RPM or control the pump that way. If neither works, your pump may be bad.

You're connecting it to a fan header (all the QUADRO has), so it shows up under fans. Just rename the channel so you know what it is and control it from there.

We get it, you don't like it. You're not going to convince anyone by repeating the same thing with more BOLD CAPITAL LETTERS.

*complains about vertical resolution* *turns around and says resolution isn't actually important*

In other words, nothing we didn't already know. (Also, lol at "poor image quality.") Technically the only requirement is Shader Model 6.4 support, so the 900 series should be fine in that respect. Of course, if this were an NVIDIA product, those cards would have been locked for no other reason...

The Hardware Unboxed video linked above has a few comparisons in motion.

The interpretation of the launch day performance and quality of FSR makes it really easy to pick out the people who were never going to give it a chance in the first place. My thoughts: * It looks better than I thought it would at Ultra Quality and Quality modes and appears to be a fine...

I don't think anyone in this thread is "lapping it up," but it does have one indisputable advantage for people running Pascal cards: it's better than anything NVIDIA offers them.

Not the NF-A12x25. The NF-A12x15 has been out for some time, but that's not a good fan for anything other than applications where you can't fit a 25mm fan.

Meanwhile, I'm still painting NF-A12x25s for people almost two years since they said the chromax version was supposed to come out.

Numbers change very little at the low end because you're restricted in how much heat you can remove from the loop for a given delta by the low airflow. As for your claim about decreasing difference between inlet and outlet temperatures, you're still talking about a constant heat load, which is...

Correct (at constant temperature, anyhow.) Incorrect. At greater flow rates, you can dissipate more heat at the same delta as you can with a lower flow rate. Say you dissipate 250 W at 1.0 GPM for a 10 C delta, you might be able to dissipate 300 W at 1.5 GPM while maintaining that 10 C delta...

Yes, but you can dissipate MORE heat at the same temperature (so still in equilibrium) at higher flow rates. This isn't controversial.

This isn't really true. At equal temperature, a loop in equilibrium is dissipating more heat at higher flow rates than lower because the mass flow rate is higher.

It is true of every radiator, but not necessarily to the same degree. XtremeRigs testing back in the day did a pretty good job of fully characterizing the performance of a radiator across a range of liquid flow rate, airflow, and static pressure.

This isn't really true in general. At lower fan speeds even thick, low FPI radiators can struggle against good thin radiators like the HWLabs Nemesis GTS or XSPC TX. You tend so sacrifice other things with these sorts of radiators - flow restriction in the case of the GTS due to the thin tubes...

Nice! I secured a 6700 XT for my wife before doubling back to try and grab one of these but they were out of stock by then. Love the black cooler.

They're far from flat, at least in the case of the two cold plates I've had. Both of them were visibly wavy.

Just for the record, my 6800 XT direct from AMD was packaged acceptably.

It's not bad it's just the worst. :D

I don't really agree with that; its price-per-core is worse than every other CPU in the lineup, the gaming performance is indistinguishable from the 5600X and lags behind the 5900X in certain scenarios, and it runs hotter than either of those two chips owing to the fact that it's driving two...

The coping and rationalization from some people around the web about this review is fascinating; from magical BIOS fixes, to memory speeds, to cooling, to the fact that the motherboard used is under NDA when the CPU isn't, there's nothing that isn't suspect.

I wonder if this one will have more local dimming zones than the original? Might finally make it worth a purchase.

Oh, maybe that would explain it; they were GSs. Brain fart.

I don't know what constitutes crappy mouse feet, but I always used quality mice from brands like Logitech.

Man, I wore two Ratpadz XTs smooth in a matter of months. I don't know how people use them for years.

The Q&A on the product page says they recommend staying below 54 W total, or about 4.5 A.

The splitting and combining essentially cancel each other out in terms of pressure. As for the slower flow through each radiator, the reduction in flow rate is actually what causes the pressure drop to be lower, so it's kind of two sides of the same coin.

He's actually right as far as the large pressure drop goes, if not the reason. The flow rate is the same because it's an incompressible fluid in a sealed loop, but the pressure drop is pretty large across a MO-RA. The pressure drop is actually reduced by its use of round tubing, but it's still...

Your first picture looks like a bad mount; the paste spread isn't even at all and the paste on the left looks like there isn't good pressure there. Do as Pabaisa says and re-paste then remount the block, paying special attention to the way you tighten it down. I don't know if the EK AIO mount...

From Francois, the same guy who thinks "race to sleep" is only a thing when Intel does it.

I didn't actually. Maybe you're thinking of someone else? This I agree with, and it's something Intel has a problem with currently (and will until Alder Lake is released.) This ranges from debatable to outright FUD.

"IPC per clock." They're already matching or surpassing Intel and Rocket Lake most likely isn't going to be worth buying even if they claw back a 5-10% single threaded performance lead. That performance lead will shrink as you pile on the threads, and it will run out of threads long before...

They could with hand-tuning of the design, but why bother? Not worth the investment.

Make sure you're on the latest version of Thaiphoon Burner; the data is updated pretty regularly. I have a set of Micron E-Die sticks that it couldn't identify using a version just two minor versions out of date that it fully identifies with the newest version. Even if that doesn't work though...

Those sticks are on the QVL, but not in a 4-DIMM configuration. Download Thaiphoon burner, confirm the ICs used on those sticks (should be Micron E-Die), and then use Ryzen DRAM Calculator (read the guide here) to get something approaching proper timings and voltages required to get this...