ฝัง
- เผยแพร่เมื่อ 29 ก.ย. 2024
- I wanted to make this test for the longest time since the last year, but my heart surgery made things complicated and I never come around to it till now. But, I'm glad I did. I have always had a feeling that distros are a little more of an artsy gimmick rather than a cooling entusisast device. Still, I find the result surprising.
Thank you to EK Team for providing reservoir samples for this and other experimentations.
Interesting findings. I've always wondered what the advantages were to distro plates. I now have my answers. None that I would personally care about. I'm all about efficiency - not bling, especially with all the associated hassles to this kind of setup...
Thanks for taking time to experiment for everyone! I no longer have to try it for myself!
Plot twist! Do what I did and use a distributor as a normal pump res combo. Why have extra tubing if you don’t have to? Best of both worlds
I actually think Distro plate is uglier than traditional pump +r reservoir, because every hard tube just 90% bend and all tubes connect to the plate. So it looks like just bunch of horizonal tubes, visually very boring.
DazMode, fantastic video. In the process of doing my custom loop (will be ordering some stuff from you soon) and have shied away from distribution plates and glad that I did. I have another idea for you for a video, could you compare doing a traditional serial loop vs. a parallel loop in terms of temps, performance, etc.? That would be awesome. Thanks.
Hate complex distros with like 10 ports. Forces too many parallel long tubes. And usually the tubes are forced to run over the ram sticks, blocking their view. Makes the case look like it belongs below a toilet.
hahaha
Distroplates are just for the optic use them with hardtubes and they look amazing with coloured fluid. If you want a watercooled pc and go for perfomance just do not use distroplates :)
Thank you for these testings. I was wondering about the distro plates. They always seem to be more trouble then there worth, and this is clear proof!
Same here. They seem to be more of a bottleneck and personal challenge.
Did he just talk shit about the youngsters working with screw drivers? 💀💀💀hahaha
Seated on lazy Susan is nice. I have a custom cart that’s a bit higher I use to spin mine. It has butcher block with 3” foam top
So I’ll add this…. I’ve used both types of systems. The cylinder pump style is easy like you say with flexible tubing. Once you use hard tubing, the distro becomes the easiest route as most of the tubes are straight and only a few 90’s. With the cylinder you have to get a bit more creative with the bends. For a noobs the soft tubing with a cylinder is the best bet. But for anyone looking to go hard tubing the distro makes life so much easier. Also if flow is a huge concern add another pump to the system and reduce both by half speed. Water cooling in general has diminishing returns after a certain point. An argument can be made with proper air-cooling you could see similar results albeit more loud.
Yes very easy, specially if you buy all the corresponding rads from EK etc that will all line up PERFECTLY. Then all you have to do is cut 4 perfect straight pipes for CPU and GPU.
Good work, Daz! The reason for the reduced flowrate (increased flow resistance) is right in front of our eyes. Each route through the plate adds two miter bends in close proximity. Each miter bend is very lossy, but the close proximity adds more loss. Also, we can't see the pump inlet geometry, but the standard pump tops have optimal shapes. The thin disto plate may have a small gap at the inlet. As the flow radius decreases going into the impeller, velocity increases. The flow resistance scales with velocity squared so... that feature could also hurt.
A much lower loss, simpler distro plate would be just a reservoir with a top inlet and a bottom outlet, plus the fill and drain ports. Use an optimal top on the pump. No vanes would be needed since the very flat shape will de-rotate the flow. The flow resistance would be comparable to the cylindrical reservoir. It could have constant velocity manifolds on the inlet and outlet to distribute the flow evenly in the lateral direction and look cool.
I was looking forward to geeky comments like yours 👌🏻
i believe the EK FLT pumps also impact its flow rate compare to normal tubular ones
is singularity cases just for looks?
@@allgamestotal6773 Daz: "the pros are only looks" He's always right, you know. Fortunately, liquid cooling systems often have significant margin. However, the hobbyist needs to consider cooling need for his intended use. With the latest gen CPU and GPU designs, many usage scenarios will require the absolute best cooling performance that can be packed into a case.
@@MrKentaroMotoPI Thanks am considering buying one of spectre cases. coz i like the looks. but lets hope it can reduce the heat also.:🙂
Thank you for the video, the point is the most distro plates are designed for only two radiators except the o11 xl one, if you want the most cooling power in a smaller o11 case or say NZXT H9, distro plate limits your cooling power. Secondly, the bistro plates are designed for the showpiece, however, everyone is using the same case also using the same distro plate, in consequence everyone's custom loop looks the same, which makes your loop is not real customized, because you share the same recipe.
Случайно попадаю на твое видео спустя, сколько 7 - 10 лет? Каждый раз улыбаюсь от четкого и чистого произношения =DDD
I want looks, low temps. Can be done with a well designed, larger case and buckets of money. In earlier years, my cycle was cpu upgrade every 5 years, gpu upgrade every 3 years. Then the crazy pricing hit. My 3080ti cost 1200 $, my 9900k cpu $500. I expect my system to serve me 8 years. So I'll spend $1400 for a reusable distroplate water cool system
With a distroplate, the case really needs 2 full size radiators and then temps can be favorably low. And many cases, cant effectively handle 2 full size rads and a distro. I had to try 3 different cases that each vendor said would be ok. Two phantek, one a no go. And 2 full size rads vs 1 rad, lowered my temps 4 degrees from very hot to ok temp with both gpu and cpu water cooled
You always can spend more money to compensate for poor design to favouryour desired looks.
Luv this video
Really disappointed you did record temps and compare. Yes different sytems but same case gpu,cpu,rad so we would know how much lower temp we loose with a distro
Check Jaz2cents video, flow rate relate to fast/slow the loop to certain temp, but the final temp will NOT change! Final temp depend to RAD capacity only, So this guy is making some mistake.
Distroplates always looked like more hassle than it is worth in my opinion.
Nothing wrong with either choice. I've built systems for performance only and systems that I like to look at every day. Different goals in each case.
Defo
I made a slim custom distro plate on my CNC so I had enough space to put a monterous radiator in my case lmao :D
What one is better for performance
I really like your quick disconnects. What brand are they, and what size are you using in this build. Also the tubing. I have the 011D w/Alphacool tube res and a D5 pump. These would make it so much easier to service my system.
I swear distro plates were made just to sell more fittings.
Very interesting! For me as "showcase" modder I still try to avoid distro plates, the disadvantages are to big and much. I try to build nice loops with reservoir mods.
I dont know any technices with use distro plates... only saw it in gaming section for the RGB kiddies
I have a full tower like yours and have a smaller reservoir, but have three rads (a triple rad, a single 280mm, a single 140mm) so there is plenty of distilled water moving around.
Temps are very good. But its an older PC (i7-3770k and dual gtx-680 on Win7 pro) that has served me well.
Am moving to a mini-itx MB in CoolerMaster NR200 with Ryzen7 and RX 3070-ti
It is very tight in there.
Am hoping to squeeze in two rads but a reservoir will be tricky or impossible.
A small Distribution plate will take the place of the reservoir, without using much space.
But I wouldn't otherwise do it in a full size case.
I am happy to hear you are well after heart surgery!
I realize these surgeries are more common-place now-a-days....but recovery from major surgery can be very difficult on one's body.
Welcome back!
It would be very interesting to see the differences between the two pump tops. There can be huge performance losses if the clearances between the pump impeller and housings don't meet pump manufacture specs.
Great idea to try this test. Thanks Daz.
You are the second person that points to pump top clearance as possible issue
"Really have to love this configuration, extra cost and extra hassle" 😂🤣😂 Like you didn't know us open loop guys are masochist.....
LOL, true.
Daz and I both cut back on computer related videos about the same time. But daz always does the best " use this knowledge " kind of video. Computers were always a fun, expensive hobby but then crazy pricing and disappointing marginal cpu performance started . But daz and his real world useful stuff is exceptional. There are only so many times I can watch master jaytwocents, master steve, master lionus do the same stuff. Although I luv watching lionus add technology stuff to his new house, daz is relevant and never disappoints
My flow rate it a LOT lower with my distro. I mean it's borderline..bad. my runs are so nice looking. I could add a second pump. I swapped put the magnitude for a monoblock and added active backplate to the gpu. I also have 3 rads so it's basically the max that will fit in a case. My flow rate is 140l/h at 100% but it's too loud so I keep it to 110l/h(0.5 gallons per minute) then ramp it up when under load. Basically 80% at idle and 100% at load because I don't want to hear it. My temps are great and adding another pump would be a big undertaking so idk. Let me say one thing though..every time I turn it on..it just looks so damn good.
FYI you don't have to connect the ports you can just skip them
If you use D5 it should be barely audible. Noise means either impeller failure or trapped air
Hi Are you from Russia, comrade?
Thank you for the testing results sir! Sidenote: I'm so sick of 011 dynamic case Being everywhere. I tried to like it but I ended up returning it. The case is so overrated
which one did you get instead?
Indeed.
I bought the Core P8 instead. So much versatility with that Core P8 case. Going to hang it on the wall to show off my custom loop.
@@dartleg13 Thermaltake level 20 XT
It's weird how everyone is raving about the 011. It looks like a fish tank ffs. There is the bonus of being able to fit an extra 360 rad but that fish tank also has quite the footprint so you better get something extra out of it.
Great video. Thank you for doing the test and showing the results.
Given the length of certain GPU’s these days their is now a practical reason to own this distro plate in the o11 XL. Weird saying that.
A short reservoir can be used, but you are right.
Weird question maybe, but where do you get the T-shirts you wearing? Thank you.
Enjoy: www.threadless.com/search/?sort=popular&departments=mens&style=t-shirt&type=regular
Would you be confident to say that with 2 EK P360M radiators, (EK Mana G2 PC-011D distro plate) I can get the same cooling performance as with a classic reservoir? I already got all the parts to start building my pc. It won't be for gaming purposes...Thank you.
I'd take the cooling capacity of 2x 400-500W radiators and the straight lines of hard tubing all day .. over some flow rate in oversized fat tubing that's pinched off at every fitting. Not to mention I would software control all those fans and the pump. To each their own.
Exactly the testing I needed. Almost suckered into buying a distro plate...but at 9:40, THIS would have been my instant regret as well. Well done!
Had a Distroplate and it was terrible :/
As my first Pump.
So after years I consider Watercooling again (with Quickconnects)
The advantage of the front distro plat like that is that your fill and drain holes are in the front already. The looks are better and the system fill is easier. It's also easier to setup with hard tubing when EK brings all of it's Matrix 7 product line to the market.
However, I like you will stay with my light house pump reservoir combo. Though I like to call it a warp reactor.
Your issue with the bottom rad is kinda weird because I can't cool my machine with a single 360 and no one with modern hardware that would be buying 2K worth of water cooling parts is going to get by a single 360.
As far as flow rate goes the D5 has so much flow losing even 20 percent isn't really tha big of a deal when it comes to temps, but I am using a dual and maybe a single rad it's a bigger deal, and I'm not using very thick rads either so I think flow would make a bigger difference with thicker rads. With all of the extra 90 degree turns the water has to make it is not a surprise the flow is less. I think there is a dual D5 distro plate.
You really should check out the EK matrix concept, it's pretty cool.
btw, I really like your vids Das. Thank you
I am sooo gratefull for your video. It has helped me a lot! If I may ask for one thing: please put more light on the shots of the build. It is difficult to see the details of your hard work, especially when there is tubing and fittings involved. Thank you and god bless you!
You are simply THE BEST. This is a video I was looking for. 5 STARS !!!!!!!!!!!
Thank you for this video! I just started planning my first custom loop and was up until now aiming for a distro plate but after this video I’m definitely not picking it. I feel that looks are important but that amount of loss is too much for me. Have you got any experience with a flat reservoir-pump combo? Curious to find out if it also means the same decrease in flow rate (because I still like the look of a flat pump/reservoir combo over the cylinder). Thank you again!
Great video. Which model EK tubular reservoir / pump is that?
Apart from my test bench (which uses 16mm ZMT soft tubing), I aways build hardline. Soft (especially black) tubing with that EK distribution plate looks like you're housing a spaghetti monster in your case.
I have the Lian Li O11D XL and will be using a Singularity Computers 150mm D5 pump/res (waiting on the Alder Lake release). Sure the distribution plate looks cool but I'm populating the O11D XL with 3 x 360 radiators and that (I believe) prevents you from fitting a front distribution plate in the case. That said, I still use smaller distribution plates - I have a build in a Phanteks Enthoo 719 case using the Phanteks D140 plate and the Lian Li build I have planned will use of the Phanteks D120 plate. Some people like the look of having many tubes running parallel from one side of the case to the other which is necessary for a front or side distribution plate; not me... I even minimise it further by using cross-flow radiators where appropriate.
As for the need to join the ports from the plate's bottom radiator inlet back to the main distribution manifold... you're not the first TH-camr to fall for this _rather obvious_ oversight.
just looking at both i can say that the distro plate has a higher risk of leaking while reservoir doesn’t.
I went with looks, so a distro plate. Good to know what the reduction in flow rate is.
I found you channel just yesterday and I like it already. You test real world problems. I have only one question. Isn't the flow rate reduced because the system gets separate loops for the CPU and the GPU. Depending on where is the flowmeter mounted, but if the GPU is getting 4 LPH and the CPU is getting 4 LPH, doesn't it means that the flow actually improved? Also CPU and GPU gets fresh water from the radiator, not preheated from CPU or GPU depending on the loop order ? Thanks.
Great Test! I was thinking about Destro plate as it looks nice but now when i have seen the resoluts i will pass ;) Keep next vids comming :)
Is there a real difference in temperature? I've must have missed those numbers. I would agree the loss of gpm/lpm is brow raising but I'd question, given that coolant velocity is lowered did it in fact hamper, help or have no affect on the removal of thermal energy from the load? Lower velocity would also indicate more time spent in the heat-exchangers.
EDIT: Why wouldn't the Disto-plate be designed to also be a heat-exchanger made from aluminum/copper tubes and fins? Seems a waste of space when the name of the game is to transfer thermal energy.
Distro plates, in my opinion, are for looks. Depending on the compter case chosen, sometimes the build is easier with less abrupt turns and corners. But, I've found a build with distro plates alway harder, always way more expensive. More fittings, more tubing, reservoirs less expensive than distroplates.
Thank you for a wonderful video. Well explained.
Very informative thanks
Thank you for this video!
Very entertaining! And helpful.
Lovely video. Thanks for the data!
you can just setup as a reservoir, block all those fittings, then you got less tubes.
That's not a bad idea actually. That way you'd be saving space instead of wasting it. Still expensive though.
First of all i want to say your channel is the best waterccoling channel on TH-cam imo.
Guys i hope you can help me with my loop, i have a D5 and 4 rads, 2 480mm*60mm and 2 360mm*60mm, my temps are great and my pwm pump is running 2800rpm when under full load. Do i need another pump? If yes do i need 2 separate loops?
Keep as is.
@@DazModeWatercooling Thanks for the reply! I noticed that temps kept improving when raising pump speed so i was a little hesitant.
I don't understand you say you're upset with flow rate but in your other video you prove flow rate doesn't really matter. Is that the case?
I never said flow does not matter. It does for me.
@@DazModeWatercooling what I'm saying is that you're other video show that unless you really lower the flow rate it diesnt really matter. Meaning if it's above a certain l/H the results are similar. Does that make sense and do agree?
You want me to come over and install hard tubing?
😂 stopping by will-be enough
Awesome 🤟
I bought some fan shrouds from you and they are really great. You don't actually see many available for sale these days and they can be very effective.
100% true
Thanks
Luv this video. All water cooled sytems are for looks. Years ago, air cooled sytems could not give the low temperature advantage of a water cooled system. Now, there is no advantage. Equal temps. Perhaps air cooled is louder but perhaps if well choosen maybe same.I have not kept up with todays best designed air coolers. Years ago. Had to actually bend my motherboard pressing so hard to install the air cooler. Then there were very large heat exchange towers hanging there. But, a water sytem is way more expensive, collects sediment, needs water changes, and rads are heavy and collecting dust
Which flow meter is that?
Koolance fm-16 with display adapter fm-01
Sweet vid
21:22 😂❤
This was very interesting to watch . Ok you lose flow rate with all the bends and if you measure the loop length out you would have at least doubled it over the first loop. My question is does the cooling performance suffer much in a load test? Does the pump make more noise/ or much higher rpm to function the same.
That monster air lock you had is a real worry because those new to water cooling may not know how to fix that before they wreck the pump or overheat the system.
I have a CNC res on my rig but the port lay out is much different to the EK you have there. I did mine for looks mostly and yes for maintenance it's a bit of a pain over a tube res.
My res combo allows for a parallel loop as my res is a dual chamber type from Radikult customs. The dual chamber is basically a return/tank one side and the chamber on the other side of the D5 pump has an upper and lower outlet port for the cpu and gpu loops. I keep an eye on temps and I've had no issues.
Thanks for taking the time to make this video Daz !
Noise levels were about the same on both models. Higher flow spells lower temp as i showed in other vid.
Well
You did not do the most important test wich is temperature test !! Hmm i wonder why ??
All u did was flowrate testing
Maybe because There is almost no difference in temps mybe 1 or 2 degrees
Thats why it doesnt matter that much if u use this or that
Otherwise please show proof with temp tests
Lower flow rates increase temperature, that's just basic physics, plus there are videos about it out there already.
If you keep the computer running 5+ hours daily it becomes a huge problem, because you overcome the water thermal capacity to quickly.
Nope
Not in every case
There are hardware constelations where higer flow rate does not lower temps
Also it depends how much radiators snd wich thikness is beeing used
If u have a very big loop ( multiple thick rads) tben flow rate should matter
But not with 1 or 2 slim 360 rads
I tested this on my stuff and i do not have a single degree difference in temps
And even if you have differences they are very minimal
Not worth running a pump with 100% wich also shortens the lifespan of it
Nobody runs his pump with full speed
If u have a massive temp increase when u lower your pump speed then u might have a very big loop or just a crap pump
Im only speaking of proper ddc or d5 pumps
Other pumps are not woth using
@@JJ-gk5tn one question for you. Do you think the order of components in a loop matters? Do you think a parallel loop is better than a linear loop?
@@Jarb2104 loop order makes no difference
But your components does
What i am trying to say is when you have low heat components you have no benefits from higher flow rate
And again its so sad that he didnt made any temp test because the differnce in the results are very minimal
But hey if u like to sit next to a jet engine then run all your stuff at full speed
@@Jarb2104 or another example fir you
The type of fans wich you are using on your radators do have way more impact on cooling power then flow rate
You only mention efficiency, but you didn't mention which lowers temperature better. When you answer that, only then will I decide what to buy.
You reservoir doesn't lower any temps. That is based on the strength of your pump and the quality of your blocks and radiator. At the end of the day, it's gonna come down to preference of looks.
th-cam.com/video/oA0u1KJEjM4/w-d-xo.html
@@DazModeWatercooling Thank you, this is what I needed!! 👍
@@DazModeWatercooling flowrate does not mean lower temps in every constelation
For example
I got the same temps with my 5600x no matter wich rpm i run my ddc
Tested with 1200 rpm min and 4500 rpm max
Same temps !
@@JJ-gk5tn means you already over capacity for what you have
Correct me if i'm wrong, but doesn't 1L →2L double the amount of heat/energy the system can store?
Yes takes longer to boil 2 L , after that no difference
Wouldn’t more liquid mean more heat could be absorbed?
yes, for the first 5 min, then no diffrence
Great work, but don't get me wrong if I would say that you have rushed with your conclusion for the distribution plate, that particular plate (EK-Quantum Reflection) needs two radiators in order to get effective thermal distribution across the plat surface (this internal wide x-y plane)! That is, to properly utilize the wide surface for effective thermal evaporation (i.e., effective cooling), otherwise no point of having this design, don't you think? Geat work overall. Thanks.
Reservoirs have zero positive effect on cooling efficiency.
lol noob.
Like you. This guy has been doing this for over a decade buddy.
You totally neglect the fact that bigger distribution plates are easier to refill and fill after installing. They provide more liquid which is better for performance cause the more volume you have the more it takes the system to heat up. Also distro plates are actually better fitting in some cases and dont take up interior place. My D5 pump sists in the middle next to my PSU where as my distro plate can be easily mounted like a 360 radiator. Also distro plates are made for easily changing loops and therefore when you drain your system often, I think it is a QoL thing. Just my 2 cents.
"better for performance cause the more volume you have the more it takes the system to heat up" ... Nope, doesn't work that way at all, mate.
@@iAmCodeMonkey yes, that is exactly how it works, my wording is just bad and you just proved you dont understand thermo-dynamics. The more volume u have the LONGER it takes the system to reach its highest temp. Sure in the long run, when the system is at its working temperature it will not make a difference, but that was not my point, was it?
I think if you got a distro plate with pumps inside of it it would have helped your tests
He did.
Nice to watch another one of your videos. This was an interesting look at the differences between the 2 products.
Thanks for sharing this with us and showing the results. 👍
thanks for hard work on comparison