Question I **might** be back....

[rubix_1011]

Hi everyone.

Very long time custom watercooling guy here (23+ years). I also am a Senior Moderator on the forums and used to write CPU cooling articles for Tom's Hardware.

I am thinking about once again picking up my watercooling lab and testing environment as well as potentially doing photos, reviews, etc. - pump specs, blocks/radiatorss flow rates, restrictions, pressure drops...that sort of thing.

Maybe a standalone site like Martin's Liquid Lab and SkinneeLabs used to be, but not sure at this point...I haven't really started.

Mostly looking for this level of detailed interest here - does this appeal to anyone?

 

[Lutfij]

The platforms you've mentioned above are sorely missed and sadly people have resorted to using watercooling as a statement piece in their showcase builds which do cost a lot but don't hold a bang for the buck aspect which most people tend to look for given todays predicaments. Kind of like someone earning enough to keep a roof over their head, managing family expenses and food on the table, would look at a reliably built low upkeep car instead of a really high upkeep supercar.

Like all things, data is always a good thing. I myself have learned a lot from the wealth of data you'd compiled on both watercooling stickies before diving head first in my custom watercooling endeavors. By that I'm never saying no to any project you've done or going to do. It's super refreshing to see you pop back up in the forums though you need to be here more often! 🫡

Forza, my dude!
🌊 😎

 

[kanewolf]

Custom water cooling is a very niche topic. IMO, to be worth the expense and possible frustration, you have to have exhausted conventional cooling options. I don't chase the last 2% of performance, but some members do.

 

[rubix_1011]

Custom water cooling is a very niche topic. IMO, to be worth the expense and possible frustration, you have to have exhausted conventional cooling options. I don't chase the last 2% of performance, but some members do.

This is one of the most over-used and often misunderstood reasons why people watercool.

Most people don't watercool as a performance measure, most do it because it is enjoyable as a project and satisfying once completed. It does help when you do want to overclock, but for most, its is much more about customization and doing something different.

Most people don't modify their cars, but some do. It isn't always about crazy performance, but sometimes is.

To me, building a PC is rather boring when someone just wants the same parts that everyone else has, just with slightly different color or branding. Not everyone values RGB lighting - I quite like building data crunching servers with full watercooling builds that fit into a server rack or displayed like a piece of art.

 

[kanewolf]

This is one of the most over-used and often misunderstood reasons why people watercool.

Most people don't watercool as a performance measure, most do it because it is enjoyable as a project and satisfying once completed. It does help when you do want to overclock, but for most, its is much more about customization and doing something different.

Most people don't modify their cars, but some do. It isn't always about crazy performance, but sometimes is.

To me, building a PC is rather boring when someone just wants the same parts that everyone else has, just with slightly different color or branding. Not everyone values RGB lighting - I quite like building data crunching servers with full watercooling builds that fit into a server rack or displayed like a piece of art.

Maybe. But custom cars or having to show off has never been part of my personality.

I believe typical home PC builders/users priorities are:
Just work
Be reliable
Be affordable
Be quiet
Be cool/unique/eye catching/etc
...
Those priorities put custom liquid cooling at a lower priority, and IMO, therefore shrinks the audience. Do I spend budget on a better graphics card or custom water cooling for a lower end card? Those are the tradeoffs that MOST people have to decide.

I worked in high performance computing. I researched water cooling at the rack level and have seen HPC solutions from IBM and others with component level cooling. BUT, these really are for the last 2%. Liquid cooling has been a part of HPC since Seymore Cray and Flourinert -- And yes, I have been around liquid cooled Crays.
For most things, air cooling just works simpler and more reliably, IMO.

 

[USAFRet]

This is one of the most over-used and often misunderstood reasons why people watercool.

Most people don't watercool as a performance measure, most do it because it is enjoyable as a project and satisfying once completed. It does help when you do want to overclock, but for most, its is much more about customization and doing something different.

Most people don't modify their cars, but some do. It isn't always about crazy performance, but sometimes is.

To me, building a PC is rather boring when someone just wants the same parts that everyone else has, just with slightly different color or branding. Not everyone values RGB lighting - I quite like building data crunching servers with full watercooling builds that fit into a server rack or displayed like a piece of art.

Just like building the PC in general is a niche concept.
The VAST majority of people don't do that. And far too many glom on to a custom build as "cheaper". It usually isn't.

 

[Wolfshadw]

And far too many glom on to a custom build as "cheaper". It usually isn't.

Finding a pre-built with ALL your desired components is nigh impossible. Hence why the custom build/build your own market exists.

-Wolf sends

 

[Eximo]

A few issues:

Big downside to current water cooling is the vast increase in price for CPU and GPU blocks. $100 was a lot more palatable than the current 200+ for a decent GPU block. Now of course the GPUs are 2k+, but that is beside the point. Radiators got pretty cheap though.

And the higher end air cooled cards basically take care of themselves without the need for water cooling. With the locked down nature of CPU and GPU boost, there is a lot less overclocking to do to begin with.

Silent operation is no longer the realm of watercooling as much either.

I still do it for the fun of it, and because I have parts that still work. My next upgrade I plan to water cool, but probably not after that.

 

[rubix_1011]

Watercooling can be pretty quiet; it depends on how you design your loop. I run a watercooled 2990wx Threadripper workstation build in a server rack with a watercooled 2080 Super, can't hear it at about a meter away. The only fans I hear are the 2 drive bay intake fans. Runs at 28C CPU / 25C GPU all day.

 

[rubix_1011]

A few issues:

Big downside to current water cooling is the vast increase in price for CPU and GPU blocks. $100 was a lot more palatable than the current 200+ for a decent GPU block. Now of course the GPUs are 2k+, but that is beside the point. Radiators got pretty cheap though.

And the higher end air cooled cards basically take care of themselves without the need for water cooling. With the locked down nature of CPU and GPU boost, there is a lot less overclocking to do to begin with.

Silent operation is no longer the realm of watercooling as much either.

I still do it for the fun of it, and because I have parts that still work. My next upgrade I plan to water cool, but probably not after that.

Your post outlines it well - EVERYTHING costs more now, not just watercooling parts. Top of the line CPUs and GPUs from 10-15 years ago were under $300/CPU and $600/GPU. CPUs can still be found in that territory, but wanting to run top-spec GPU means missing a mortgage.