The Worst Product We’ve Tried in YEARS!

[Spotty]

23 minutes ago, AnonymousGuy said:

I get that you have a ton of hardware laying around, but you don't need to build actual systems to do this.  Make a box with some threaded ports to attach 500W immersion heaters and there's your heat load test rig for watercooling.  Or you can get a point-of-use water heater that already has the tank attached to it.  I use one of these to make a water-heated oven that's driven by a D5 pump.  Everything scales linearly so if it's a delta T of 20C with 1500W then it'll be 10C at 750W...no need to build an actual 750W test.

 

And after about 5 minutes of the thing not priming itself I would have attached an external reservoir to it with some rotary couplers.  Also first filled it without a whole system attached, just the inlet and outlet looped on each other.  When you attached the whole system to it you create a giant airbubble worth of pressure that the pump had to try working against...which obviously didn't work.

I don't think the point would have been to find a way to cool 2000W worth of heat, but just to test the manufacturers claims about this product. There's probably a dozen better ways to dissipate 2000W of heat.

Edit: My mistake, I completely misread your post and thought you were talking about immersion cooling - not ways to generate the heat load. Just woke up.

 

20 minutes ago, AndreiArgeanu said:

"We would have made it better than this"

You tried to before, and you rushed it, and it was arguably much worse as a result. At least none of the fittings or tubes leaked here.

That was the joke.

Edited April 21, 2023 by Spotty

[AnonymousGuy]

26 minutes ago, Spotty said:

I don't think the point would have been to find a way to cool 2000W worth of heat, but just to test the manufacturers claims about this product. There's probably a dozen better ways to dissipate 2000W of heat

Yeah, what I'm getting at is just using pure resistive loads instead of building a whole system just to generate heat.  It's more reproductible too.

 

Having dealt with watercooling 2000W, eyeballing that radiator it could do it but the fans would be annoying loud if it's in the same room.  I have a 120x2 rad cooling 500W on my lan rig but it's a 20C water temp rise, which is about the max you really want to deal with.  Tubing starts getting funky to work with at 45C water temp.

[tkitch]

3 hours ago, Spotty said:

I'm curious what the video was going to be about if you managed to fill the loop without issue? Did you have a concept that was scrapped because it wasn't working or was it always scripted to be about not being able to fill the loop? I'm sure you had something in mind when you ordered it.

 

I ask because it seemed like the video just ended after you managed to fill it and get it working and there wasn't really any objective or plan beyond that. 

I imagine if it worked correctly, they'd have checked temps, etc.  and said "This is a solid product, and it should be able to cool both a high-end CPU and GPU at once!"

But it clearly didn't work well enough to say anything of the sort. 

[tkitch]

2 hours ago, AnonymousGuy said:

Yeah, what I'm getting at is just using pure resistive loads instead of building a whole system just to generate heat.  It's more reproductible too.

 

Having dealt with watercooling 2000W, eyeballing that radiator it could do it but the fans would be annoying loud if it's in the same room.  I have a 120x2 rad cooling 500W on my lan rig but it's a 20C water temp rise, which is about the max you really want to deal with.  Tubing starts getting funky to work with at 45C water temp.

synthetic loads are great, but also not real world.  Which is why GN has both a synthetic load test, as well as a CPU test.

[motomat86]

8 hours ago, swimtome said:

wtf is with the babies? Did I miss a meme?

came here for same question

[Stahlmann]

10 hours ago, Caroline said:

Funnels exist Linus. Seriously.

That wouldn't have fixed the issue of the water coming back out, which was the main issue.

 

9 hours ago, Radium_Angel said:

New channel name:

"We build computers BADLY, so you don't have to."

 

Seriously, have you never tried to work with a water cooling system before? That amount of carelessness, mixed with electrical current, can kill.

And nothing screams "professional" youtuber, like plastic babies in your build.

 

Goddamn your videos never cease to amaze me with how shoddy they are.

At some point they gave up on the product and just winged it for entertainment. It was completely intentional and very, very obvious.

 

5 hours ago, motomat86 said:

came here for same question

I think their point was that no moving part in the main PC means you can put whatever you want in it and it won't be cought in fans.

[JeffLo]

the Bykski 1080 external rad comes with a remote why didnt you guys test it out?
 

[Rex Hite]

They're reviewing what amounts to a toy attached to another toy.

Plastic babies are entirely appropriate.

 

 

 

 

[Russ T.]

I watched this and made my comments on Youtube as well. 
This video should have been a quick exterior, quick interior tour, a quick fill, and then the rest of the fun of watching Alex try hooking up whatever thermal load systems he was going to test. 

Given every chance you've had, you built radiators and jank but workable setups; the Car Radiator build; the Rusty Old Steam Radiator build, the One House Water-cooling Solution, I figured you'd have some history under you and some ideas of what you were doing. 

1) You put quick connects on but didn't polarize them. You could have a quick connect loopback while you filled your radiator... Oh wait how ever will you find a 4 pin Molex 12/5v power supply cable.  I dunno, every enthusiast spec modular power supply has a little Water Pump Jumper block, or you could have sent Alex to go find a paperclip to use on a Test Bench power supply. 
2) "Even bad videos are content" is a horrible takeaway from this. I understand it's absolutely on brand to tune in on a weekly basis to "Watch Linus Drop Stuff" and see what's cooking in Alex's test kitchen, but this was just bad all around. 

LTTStore run out of funnels? 

[dan42c]

On 4/20/2023 at 7:32 PM, AlexTheGreatish said:

Plan was to hoop up 3 super high powered systems to test their 2000W cooling claims, but the filling went so terribly and took so long we just gave up on that

I know when making a video about this kind of thing the entertainment value is in the fills and spills, but you should consider getting a negative pressure circulator for labs: https://chilldyne.com/cooling-distribution-unit/

[themrsbusta]

The problem is the PC industry or Nvidia?

[JeffLo]

its all for the show, its not about tech anymore hahahaha entertainment youtube rev

[Joe5985]

Fun fact: How BY(Bykski) reacts to this video and an other video in bilibili.com

 

For this video (LTT authorized translated and post)

www.bilibili.com/video/BV1Mc411n7tc/

Original text:

Quote

感谢用户的批评与指正，视频中测评的产品给用户带来困扰，我们深感抱歉。直面问题，敢于担当，在这里我们深深地感谢测评人及用户对我们的监督与鞭策，让我们意识到我们的不足，在收到负面反馈后，为了给用户更好的使用体验以及减少因为安装所带来的风险，我们第一时间对此款产品进行升级，现我们已增加一个可拓展注水口，用以保证水路板腔体内部气压平衡，达到泄压作用。同时视频中所描述的10天有效期，并不意味着10天的保证，用户可在10天内验收产品是否存在产品错误，零件缺失或产品损坏等问题，Bykski的产品保修期为3年。再次为这次不愉快的经历感到抱歉！Bykski将秉持初心，不断开发多元化产品，突破传统思维的束缚，给用户带来更好的使用体验！

 

 

 

Here is the reaction for a testing video with a BY product, the video was made and posted by a chinese account '老弟一号'. ATM, the orginal video has been removed.

www.bilibili.com/opus/820410986306469961

Due to the orignal source, unable to copy&paste plain text atm.