thesmokingmanWhat is most watercooling setups? It won't come close to my loops. If you mean an AIO is most watercooling loops then that's not much of an ask. AIO are not anything to write home about yet there is a market for those looking for water w/o the expense of a full loop.

An regarding the quote that triggered you both, give it up. His point is actually correct. A fullcover block will wick away a lot more heat than a heatsink and fan will. His post also implies a supporting loop.

I'm sorry, but this begs to be paraphrased for clarity. What you're saying here is essentially "Who cares if I was wrong? If we change the premise of the discussion to something I like, you'll see that I'm right." If you go back and actually read my post before the one that "triggered" us (nice and subtle use of derogatory language there, I'm sure that felt good - you really got us there! Zing!), you'll see that it makes the same false assumption that you are doing here. I never mentioned full cover water blocks, I said "most water cooled GPUs". You misreading that isn't on me. And besides, with my clarification in the response to that misunderstanding, the point stands even for custom loops - just not overblown ones. Of course there's massive variance within custom loops, so nothing will apply universally there, but that's also why "most water cooled GPUs" (in addition to sheer sales volumes, of course) are AIO-cooled.

The response to my first statement, as well as your continuing attempts at moving the goal posts to make it correct, is like someone saying "this coffee mug can hold as much as most water bottles" and you responding with "no it can't, look at this 10-liter jug, it can hold tons more!" It's misconstruing what was said to begin with, and then getting defensive and trying to redefine the discussion in order to make yourself right. That's not a discussion anyone gets anything of any value out of. So maybe let's leave it at that, yeah?

ValantarI'm sorry, but this begs to be paraphrased for clarity. What you're saying here is essentially "Who cares if I was wrong? If we change the premise of the discussion to something I like, you'll see that I'm right." If you go back and actually read my post before the one that "triggered" us (nice and subtle use of derogatory language there, I'm sure that felt good - you really got us there! Zing!), you'll see that it makes the same false assumption that you are doing here. I never mentioned full cover water blocks, I said "most water cooled GPUs". You misreading that isn't on me. And besides, with my clarification in the response to that misunderstanding, the point stands even for custom loops - just not overblown ones. Of course there's massive variance within custom loops, so nothing will apply universally there, but that's also why "most water cooled GPUs" (in addition to sheer sales volumes, of course) are AIO-cooled.

The response to my first statement, as well as your continuing attempts at moving the goal posts to make it correct, is like someone saying "this coffee mug can hold as much as most water bottles" and you responding with "no it can't, look at this 10-liter jug, it can hold tons more!" It's misconstruing what was said to begin with, and then getting defensive and trying to redefine the discussion in order to make yourself right. That's not a discussion anyone gets anything of any value out of. So maybe let's leave it at that, yeah?

No, you guys are wrong. A fullcover block is best, period end of story. A fullcover block implies a supporting loop obviously. You replied to a poster who was talking about a fullcover block for christsake.

Then you link an AIO with a 240mm rad and a cpu block added in and suggest that that's not great or some such nonsense whilst it still outperformed your setup. That should make it obvious that there can be a huge variance with water. And as I stated a loop is only as good as you make it. Then another chimes in and goes on a tangent of cost. And then ya both circle back to AIO etc etc and some more circling around and around. :rolleyes:

ValantarThe most common mounting location for radiators is as a front intake, so not really.

ValantarYou seem to be starting from some implicit assumptions of worse operating conditions for this card compared to your theoretical water loop, which would of course make it a biased comparison to begin with.

Not talking about "common", but "well thought". People often use WD40 penetrant as a bike chain grease but it doesn't mean it's proper, much less good.

The first place to mount a LC radiator is on the exhaust so You don't push hot air inside the case. And I just pointed out that putting a high cubature air cooler on the same level of performance and noise as a fullcover block supported with a well designed loop is a mistake - the former won't even come close on core/memory/VRM temps, while keeping similar noise level.

Not trying to undermine Your own setup - if it does it's job, than all is good. But in this case (as well as in many others), words like "most people", "common" etc. don't have to go side to side with "quality" and "performance". And I think most of us enthusiasts won't be satisfied with a s***ty cooling design. :)

thesmokingmanNo, you guys are wrong. A fullcover block is best, period end of story. A fullcover block implies a supporting loop obviously. You replied to a poster who was talking about a fullcover block for christsake.

What? I didn't reply to them, they replied to me. Christ, how hard is this to grasp? I responded to someone entirely else commenting on the sheer thickness of this card (with no mention of water cooling or anything), where I said in effect "sure, it's thick, but I also expect it to rival most water cooled GPUs". @Dammeron then responded to that. Get your facts straight.

I have never, ever, said that a full cover water block isn't "the best" - there's a reason I have one myself, after all. So please quit your straw man argumentation and actually read. Please.

thesmokingmanThen you link an AIO with a 240mm rad and a cpu block added in and suggest that that's not great or some such nonsense whilst it still outperformed your setup. That should make it obvious that there can be a huge variance with water. And as I stated a loop is only as good as you make it. Then another chimes in and goes on a tangent of cost. And then ya both circle back to AIO etc etc and some more circling around and around.

Again, what are you on about? An AIO with a CPU block added in? What? Where have I ever mentioned that? Are you talking about the Igor's lab test setup? Yes, the rad used by Igor's lab is from an Alphacool GPU "AIO" - an AIO consisting of purely custom loop parts, with quick disconnects. Sadly their test setup isn't fully detailed, but given the inclusion of a GPU AIO in the list the safe assumption is that it's copy-pasted in, and that the setup for this custom loop card was different from their standard setup. My assumption is that the cooling described there is for the CPU alone - which is also corroborrated by their statements of ~35°C loop water temperatures - there's no way to maintain those water temps with a 240mm rad, a 6900 XT LDU, and a CPU. Given that they also use a chiller for testing, I find it safe and reasonable to assume that they're not sticking the GPU into the same loop as the CPU with just a single 240mm rad - this isn't some borderline technically incompetent wannabe reviewer, after all.

What I have mentioned, in the context of this discussion, is this: I have linked test results for a common AIO-cooled GPU, as well as test results for a GPU with a full cover block to put some perspective on my own loop, its thermals, and the prioritizations made in that build. Context matters, you know!

And yes, there is huge variance with water! As I argued for throughout this discussion! Yet you are here arguing that "water is the best", period, bar none, with no reservations. If water "is best", but "has huge variance", are you sure there are no situations where an over-the-top air cooler wouldn't beat a ho-hum water cooler? Because that is what you are arguing here. You see the contradiction, right? The premise of my statement that started this whole discussion was:

ValantarI would expect this to rival most water cooled GPUs in thermals and noise though

So, let's pick that apart, as it's apparently needed:
-"most" does not mean "all". It literally means "more than half". And given custom loop costs and the prevalence of AIO-cooled GPUs for at least three generations, I would bet (likely far) more than half of water cooled GPUs are AIO-cooled.
-"rival" does not mean "beat", it means to be in the same ballpark, beating some, losing to others.
-"in thermals and noise" means combined performance. How one chooses to weight thermals vs. noise is of course up for debate.
I'm truly sorry that we had to get to this point, but apparently this is necessary in order to have a discussion where what is discussed is the actual statements made and not some random misinterpretation of them. You're consistently making straw man arguments, arguing against made-up claims nobody has presented. So, please take a step back, look at the posts you are responding to, and perhaps ask yourself "am I responding to what they are saying here, or am I changing things in order to make my argument work?"

DammeronNot talking about "common", but "well thought". People often use WD40 penetrant as a bike chain grease but it doesn't mean it's proper, much less good.

But ... you responded to my post, where I said "most water cooled GPUs". If you're suddenly not talking about most (or "common") setups, but "well thought out custom loops", then you're not talking about the same thing whatsoever. Which begs the question of why you responded to my post in the first place.

DammeronThe first place to mount a LC radiator is on the exhaust so You don't push hot air inside the case. And I just pointed out that putting a high cubature air cooler on the same level of performance and noise as a fullcover block supported with a well designed loop is a mistake - the former won't even come close on core/memory/VRM temps, while keeping similar noise level.

... sigh. Again: I have never said that. See above, as I really don't want to keep repeating myself.
Also, "cubature"? Do you mean volume? I don't mean this as snark, but that is not a word I have ever heard before.

DammeronNot trying to undermine Your own setup - if it does it's job, than all is good. But in this case (as well as in many others), words like "most people", "common" etc. don't have to go side to side with "quality" and "performance". And I think most of us enthusiasts won't be satisfied with a s***ty cooling design. :)

... but you responded to a post where I said "most water cooled GPUs". So, either you misread my post, misunderstood what it said, interpreted it as saying something that it didn't, and went on a tangent from that, or ... I frankly don't see any other options. Either way, you're arguing against things nobody has claimed.

This thing is so quiet you can hear the coil whine , temps in the 60~ish , hotspots 72c ... all according to der8auer , it’s a winner imo

noctua.at/en/asus-geforce-rtx-3070-noctua-edition-graphics-card

Wonder what the load is in their testing but 60* at 13dB, 50* at 33dB looks pretty great.