Це відео не доступне.
Перепрошуємо.
AMD Ryzen Heatpipe Orientation Benchmark & CPU Cooler Myths
Вставка
- Опубліковано 3 сер 2024
- In this video, we benchmark a series of myths relating to CPU cooler orientation on AMD Ryzen CPUs, featuring Noctua air coolers.
Sponsor: MSI GTX 1660 Super Gaming X (Amazon geni.us/kPp0kmu)
Article: ready to publish, will go live after video
Noctua NH-U14S on Amazon: geni.us/tXr5
Noctua NH-D15 on Amazon: geni.us/b5E93vO
A while back, people talked about how heatpipe orientation might affect CPU thermal performance for asymmetrical die layouts on non-monolithic CPUs. Given that AMD Ryzen takes the multi-chip approach to CPUs ("chiplets"), and that its lower-end SKUs only have a single chiplet and single I/O die, the heat sources for these chips would be off-center and imbalanced toward one side. That's where the initial idea arose. The question is whether that has any relevance once you get to the heatsink level, or whether any differences are meaningful (if even measurable).
In this benchmark, we'll be looking at how the Noctua NH-U12S, Noctua NH-U14S, and Noctua NH-D15 air coolers perform on AMD Ryzen chips when rotated in their two different directions. The GPU was accounted for and eliminated as a variable, as discussed in the content, and airflow was run from the EPS side in to the GPU side out (with the GPU in the lowest possible slot and as a half-height card).
Buy the Disappointment PC Shirt in cotton (geni.us/izO3) or Tri-Blend (geni.us/rw8QX) to support us, our intros/in-depth testing, and to commemorate a year of disappointment! These will be a limited run due to the nature of the shirt.
We have a new GN store: store.gamersnexus.net/
Like our content? Please consider becoming our Patron to support us: / gamersnexus
** Please like, comment, and subscribe for more! **
Follow us in these locations for more gaming and hardware updates:
t: / gamersnexus
f: / gamersnexus
w: www.gamersnexus.net/
Host, Editorial, Test Lead: Steve Burke
Testing, Additional Reporting: Jake Henderson
Video: Josh Svoboda, Andrew Coleman
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video ("this video is brought to you by") and above the fold in the description. We do not ever produce paid content or "sponsored content" (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
We did in fact test the D15 (with two fans), not the D15S, contrary to an earlier comment that incorrectly "corrected" it. Removed the comment to reduce confusion so that people don't see it and get confused. We filmed this one like a month ago, but finally got around to publishing it. Article is ready and will go live after the video has some time to breathe.
Check out the Disappointment PC build! ua-cam.com/video/YO_xDzDOua4/v-deo.html
Buy the Disappointment PC Shirt in cotton (geni.us/izO3) or Tri-Blend (geni.us/rw8QX) to support us, our intros/in-depth testing, and to commemorate a year of disappointment! These will be a limited run due to the nature of the shirt.
Great video!! Now, to take this a step further, would it matter for coolers with direct contact heat pipes (for example the hyper 212 Evo)? I thought maybe a cooler like this would be represented, due to the fact that there was a cooler on the table/workbench which had direct contact heat pipes. Since the coolers tested here all have nickel plated cold plates(I believe that's the proper term... But essentially there is an extra layer of metal between the IHS and the heat pipes in the coolers tested in this video, which something like a 212 Evo does not). Just a thought since the orientation may make more of a difference due to the direct contact nature. (I hope this makes sense)
So what you're telling me, is that other random people on the internet may be giving out bad information? Next thing you will be telling me is that the Earth isn't flat...
Good video. I know this is a little impractical but, I would be interested to see how well a cooler works upside down vs right side up. I have read that different types of heat pipe wicks (grooved, wire mesh, sintered, etc.) are affected differently by gravity. Some are not affected much, others (like grooved) are affected much more. I believe most (if not all) CPU coolers use sintered metal (which works very well against gravity) but, I wonder if certain wick styles are more cost effective and if that would lead some cheaper CPU cooler manufactures to use grooved or mesh wicks. If they do, I wonder to what extent those coolers would be affected by gravity when mounted in a more gravity challenging position.
The efficiency claims would have potential basis when it comes to heatsinks with direct contact heatpipes. In practice, you're looking at a 2-3 degrees difference.
In non direct contact implementations there's no practical difference really.
Thanks Steve, but these myths are basically exercise in stupidity. If even once those redditors or basically anyone spreading those myths performed at least a heat simulation I'm sure difference would be negligible.
If only there was some thing on your CPU that helps spreading heat from those dies across wider area...
I got it, I made one! I'll call it a heat spreader and make billions.
It also has a benefit of protecting the die from excessive or uneven pressure that could crack it.
@@Jaker788 that sounds implausible at best! Rabble rabble!
Need to delid and repeat test to eliminate this effect. At same time use direct contact heat pipes. So many variables to consider.
@@jamesthomson8659 Pretty sure this whole conversation is about CPUs that haven't been delided though
@@jamesthomson8659 I hope that was a joke. It really doesn't matter what the orientation is. The wattage is low. If a few pipes are used it still spreads enough over the fins and other pipes touching each other.
In the end, it's never going to effect your performance in any normal scenario. If you're in a rare scenario, you should be using LN2 or a water block anyway.
Always front to back to avoid infections, not the opposite.
Very clever!! :))
Always front to back so i dont have to smell when my cat farts again.
Only if you're a woman
Ok, Coach Mann.
pfft LOL
i love how you always explain everything about the test, so professional and reflect how much u care for detail, the time and effort in the making of this video.
I absolutely bloody love it when someone knows the difference between correlation and causation. So so much misinformation gets spread because of people being purposely ignorant to this and twisting the data to suite their opinion/agenda.
braces cause puberty!
No wait, they just correlate in terms of time.
@@HappyBeezerStudios exactly! Ha ha ha
'Purposely' ignorant?
*"cool it on the sensationalism..."*
Sorry, Steve. I can't "cool it" because my heat pipe is contacting me in the wrong direction.
No! If we are renaming months I propose Burketober
Julinus
I know a lot of this stuff is boring for them to go over again and again but I really appreciate the fact they keep taking the time to do it.
Gosh,,, why do you always guys make the perfect match. The best possible solution to any of my weird search queries always leads to your channel!
Tomorrow is the winter equinox, in accordance with ancient druidic traditions ALL CPU coolers should be rotated 90 degrees anti clockwise until 20th March next year(Spring equinox). For the most part you wont notice any effect at all, but you'll FEEL it's right and have the KNOWING.
Cosmic power will cool down the CPUs!
Also, healing crystals can be used to dynamically cool the system down via frequency resonance.
Learn more at snakeoilmerchant123.scam
@@diatomsaus That's just ridiculous. Everyone in certain internet circles knows crystals can only be used to levitate pyramid blocks into position(like minecraft in reverse), and cure hemorrhoids. Cosmic power probably wont help much, look what it did for Captain Marvel and Rey in the new Star Wars: it just Mary Sue'd them, and definitely didn't make them remotely "cool" lol.
Remember to forward this comment to 20 of your contacts or your CPU will throttle for 6 months!!
Burn a few enemies in a wicker man is good for additional 10 degrees of cooling.
Rotate it 4x 90° for 4x the effect!
The whole point of the integrated heat spreader was to allow heat to be dissipated more evenly!
Which works great when they're connected by some material that is great at transfering heat. Materials like solder.
Paste on the other hand is less ideal.
Try telling that to the reddit users with a lukewarm iq and lower..
@@Scootermagoo lol They obviously live by the saying the internet said it was right it has to be true.
That and cracked or chipped dies originally due to warranty issues. This boomer chipped and cracked many a die with high pressure, heavier than hell solid copper coolers, and induced ear bleeding fans.
I just built my new AMD rig, and was wondering this exact question. Ultimately, rather the searching the internet, I just put the damn cooler on. Good to know I made the right choice!
Indeed!
It's more important that there is a cooler in the first place.
What could be interesting is if the orientation of the cooler in regards to case fans does have an impact.
For example, in my case I got room for 2x 140mm fans in the top and only 1x 120mm in the back, so I let the CPU cooler blow upwards where I got more fan to take the heat away.
@@HappyBeezerStudios When doing the orientation with the fan facing the GPU, the distance from the GPU back (plate) becomes a big influence, then add heat coming off that back. I've seen GPU/cooler combos that place the fan less than half an inch a part, especially since we all know how mobo manufacturers follow the "CPU boundary" rules.
You know Steve? You heal my broken heart with all the testing and bullshit busting. My sciency brain can't thank you enough. It's so good to listen you you explain data charts or nuances of something that is my hobby :) feels like having an older brother teaching me. Never had one as I am the oldest. Thanks and Mahalo my dude.
You've probably answered this multiple times, but a cursory search didn't find anything: Where'd you get those screwdriver holders for your pegboard?
Great video! Always appreciate how much care you put in to controlling different variables.
Thank you for testing and debunking some myths. Great work as always.
Happy holidays!
Thank you for exploring this and clearing CPU myths
Thanks for your time testing and reporting all this things. Amazing job.
Replication studies are important. You guys are doing gods' work.
Excellent work and data as always. Thank you.
Thanks for testing this as it is interesting from a personal interest and as a general knowledge question.
Thank you for your time. Great info. 👍👍
Love this type of content of the stupidest but important things.
Great video. What I respect the most here, is that you don't try to be sensationalist, use facts and figures, and are wise in your choice of words that there might be other unknown influences/variables by other authors (on reddit), what you can't reproduce, because that's the way it is.
I'm just here to nudge the algorithm, also the video was great. Thanks Steve
How funny that you did this test! My noctua fan keeps slipping down about once a month and i was about to flip it 90° so that the bad clip stays on with gravity. I can live with 2 degrees difference if it means avoiding an eventual short.
I love when redditors are proven wrong
"Just put the damn cooler on the die"
Absolutely f**kin love this content.... 👍👍👍
Thank you tech sensei, this clearing up something in my mine.
Brutal Analysis, as always!
das war noch nichtmal brutal ich hab noch nichtmal hochgedreht bei den 9600 kbrüchte 256 mehr ram dann wer ich auf 6.4 gekommen oder so im turbo so bleibt er nur auf 26666 so lecherlich was alles geht
trotzdem fährt er damit schinmal symulirt 5,1 auf lüfter
I love your niche stuff !!!
This is the content i crave. thanks
Great content, as always! Thank you for all the hard work. I recently switched to Ryzen 3600 form the previous gen (1600, then 2700), and I use Alpenföhn Brocken 2 with 5 direct contact heatpipes for cooling. I exactly thinking on the cooler and heatpipe orientation when I found this, so you spare some time for me. But still remains 3 questions for me:
1), Ist there any difference in cooling with the 3. gen compared to other cpu-s with the same heatload (For example: a 9900k with 200w and a 3800x with 200w)?
2), Aren"t the AIO coolers are more effective, not because the higher cooling capacity, but the coldplate (since it watercooled, the entrie coldplate cooling capacity the same, right)?
3), Is there any difference between a non direct and a direct touch cooler?
Good job on this video.
I was wondering about this when the 3900x came out. Glad to know there's not a noticable difference!
Merry Christmas!
I think this video signals the RIGHT moment to create a "How/Why do we Test the way we do" sort of video explaining the "what and whys" of the testing methodology. I'm sure you are aware the technical understanding of the audience varies a lot and even interpreting results is not always completely straight forward it takes understanding some other aspects and terminology.
I think it would be beneficial to the whole tech youtube community a video of this sort. Explaining for example. Why we need to control variables? why things can't be set auto (even though is how a user would experience it), why room temperature and deltas over that temp are important, how do you determine margin of error (this is a huge sore spot on the youtube tech crowd), what are the std deviation results and how do you determine what the resulting value means, stuff of this sort.
The audience knowledge is all over the place and I think this video could serve as a reference for the whole UA-cam tech community and even as advice to some tech youtubers that while having good intentions make some errors in testing
Cool video, and expected results.
That said, I was under the impression that the heat pipe orientation was only a concern on CPUs with two chiplets, and it wasn't about rotating the cooler so so fans went top to bottom, but rather it was that if you buy a cooler with pipes that run vertically in a back to front fan orientation, you effectively only have one heat pipe running over both chiplets, where if you buy a cooler where the pipes run horizontally across the IHS each chiplet would be under a different heat pipe.
I don't see how orientation on a single chiplet CPU would make any difference either way.
Take a vacation steve! You sound better much better tho
Filmed awhile ago, unfortunately, but I did avoid filming for a day to recover my voice, so it's much better!
@@GamersNexus make sure you're taking time for yourself when you need to! You may be Tech Jesus, but we don't want you to actually die for our PC sins :)
@@TROPtastic ded LMAO
@@GamersNexus that's good but don't forget to take some out for yourself. You work hard and we appreciate it.
@@GamersNexus cognac is good medicine
I love that Steve does 20 min videos where the conclusion ist "put the dank cooler on the IHS and carry on with your day" 😂👍
Late upload Steve? It's 12:17 AM. Hope you guys have a good holiday
@FOG BLADES So... Can we go with the theory of an abdominal bazooka?
What time zone are these guys in?
@FOG BLADES Well i havent seen him completely strigth, so its a possible abdominal bazooka, that or he has a deep depression thanks to every GPU manufacturer making always the same boring graphics aesthetically speaking.
@FOG BLADES If it was something like that either we would be already conquered by aliens or destroyed by them, and i vote for the later.
I never thought about orientation of the heatpipes, but more about if I wanted the heat going up or towards the rear of the case
If you guys want to do some more cooling testing, it seems like there are a lot myths out there about what types of fans to use (i.e. static pressure vs airflow). You'll hear things like for a heatsink or radiator you want a static pressure fan, or "if you're blowing through a hard drive cage".
I couldn't find any real data on this out there, just a bunch of "conventional wisdom", so I did some quick tests myself about a year ago. The funny thing is in my own testing a lot of the conventional wisdom doesn't actually pan out. Even blowing through an AIO radiator I found airflow oriented fans to work just about as well on the PUSH side (this is important). In fact, there were very few instances where static pressure fans were clearly better than airflow fans. The one place static pressure did work better was when PULLing through a radiator, where airflow fans (thin, widely spaced blades) had trouble generating enough suction and made a horrible turbulence noise (sounded sort of like a grinding noise, the best way I can describe it).
I think the problem is a lot of static pressure fans are too static pressure oriented. There just isn't enough resistance in most cases, even blowing through an AIO radiator, for them to reach their full potential (pull orientation, once again, is the exception where it makes a big difference, at least compared to very airflow oriented fans). I think this is why the Corsair ML fans are a little less static pressure oriented than the SP series fans, since it's just a better design in general for most situations. Blowing through a really thick radiator obviously could be different, but for most radiators and I'd guess heatsinks, you don't appear to need that much static pressure.
In any case, it seems like this is one area where few people have done any concerted testing, and it could help a lot to set things straight about what situations you'd actually want a static pressure fan, vs an airflow fan. I think the results would probably surprise a lot of people, seeing as my own limited testing certainly surprised me.
Great Video! So much respect for Steve as a researcher for publishing this solid result. I'd love to see more!
I also did a post about this on techpowerup, thanks for testing!
Hey Steve, silverstone is coming out with a new case January 15th, the Seta A1. I was wondering if you plan on doing a review for it!
Speaking of coolers, mine went out yesterday. I have been wanting to try the MSI Frozr L because Steve shows it a lot in his videos so he must like it. It ended up being on a great sale at Amazon for 43.99 so I had it same day delivered. It wasn’t the funnest to install but it is running very cool and quiet. If anyone wants a good cooler for a steal of a deal this is one to look at.
Love this video
Did you AI upscale the graphic used in the first graph? Looks good
The important question here is: did you, or did you not pray to the machine spirit before doing the tests?
I would be interested in seeing the same type of testing and evaluation but with AIO water blocks.
Well Bed time can wait!
on a blow down cooler i'm guessing the fins have to be perpendicular to the back fan so that it helps blowing the air out ?
is that an msi core frozr L @6:03 ? and would you suggest that cooler for ryzen 9 3900x?
@GN, I was actually curious about this. My thinking was the direct contact heatpipe may be more effected by an A- symmetric cpu layout. I would have tested this myself but AMD won't send me free samples for testing.....
Thank you - sometimes debunking an intuitive theory is as helpful as proving it.
This video was super helpful Steve! And very timely, too. I'm working on a new 3900X/D15 build right now.
Good video.
What’s that blue print type picture you have back there on your wall?
Good points, Steve. I think a majority of people who have thermal issues are either from improper thermal paste application or poor case temps overall. Case in point? I cool a 3950x on PBO with a Noctua U9S (92mm single fan) in a thermaltake Core V1. Theres a 200mm noctua fan foe case intake and 2 80mm noctua fans for case exhaust. Under full synthetic stress, the CPU levels off at 79-80*. I can shave a couple degrees by turning up the case fans. The cooler orientation has just 1-2 pipes cooling both chiplets. Boost is no issue, as I get cores above 4.7ghz.
Oh and I'm not using paste... just an IC Cooling graphite pad.
Is there a difference between running coolers in orientation ie UPSIDE DOWN vs sideways (like in a case) vs upright (like a test bed)?
Gamers Nexus. Making sure you have all the information! And I mean ALL the information. Steve leaves no stone unturned.
Does the optimal orientation direction change if i live in Australia ? Damn - think i only got reddid to help me here now....
What about direct contact heatpipe HS. Like the Arctic Freezer 34 Esports? This is what I heard to avoid with Ryzen 3000.
Heat rises from bottom to top so does that mean the heat pipes on the bottom will always be cool and the top will be hot?
So a bed type PC withy he same cooler will cool better than a normal PC case?
Example on a test beach the heat travels from the contact point to the end(top tips) of the cooler but in a normal case the heatpipe is oriented horizontally so is it bad should allow us lay down our cpu instead of standing it up for better thermal performance?
the orientation of the heatpipes would be more effected by gravity i think, you want the cold side of the heatpipe orientated so the fluid within the heatpipe can flow back towards the cpu end of the pipe.
If you mounted a heatpipe cooler on an upside down motherboard (like the 2 system in one case with itx board mounted in the roof of the case) it's efficency would suffer greatly.
This is a noticeable problem with laptops, if you tilt some laptops 90 degrees so that they are not horizontal the cpu temp wise rise rapidly.
Man, I'm very grateful there is such a channel that pleases my computer enthusiast side so thoroughly lol
How about adding the margin of error to the testing slides so if/when they get posted elsewhere people can't claim they show a difference when it's statistically insignificant?
Remember goofy mounting with CPU blocks? There was difference there on some cold plates. Be interesting to try with water blocks.
Wasn't it the same thing back with Sandy/ivyBridge?
The long, narrow die would "prefer" a certain orientation of the heatpipes.
And on Intel you can rotate the cooler as you wish because the mounting holes are in a square.
Little note: Got my (big, 6 heatpipe) cooler blowing up because at the top I can mound 2x 140mm fans and at the back only 1x 120mm
I was looking for vertical vs sideways comparison - does it affect heatpipe performance?
answering the questions no one should ask lol. seriously trial in your own environment will give the best answers if you can build a system you can optimise your own cooling I'm a one build novice it took me half hour to work out the best cooling config for my particular situation and I had fun doing it and got sense of achievement afterwards
How about this convex thing with cryorig coolers? I saw some comments on reddit claiming the design isn't good for ryzen procies.
So I have a D15, 3700x, on a MSI B450 tomahawk max inside a Fractal Design Focus G case. And the VRM heatsink and Ram placement paired with the case constraints means the only way I can orient the D15 and use 2 fans, is to have the cooler exhaust up or down. Out the back, I can only use one fan. However, running it that way (my D15 didn't come with hardware to mount vertically out the box) I haven't seen temps that warrant the second fan for my use cases yet.
Edit: very much worth noting, for anyone looking at the NH-D15. If you mount the fans as indicated in noctua specs, the D15 could be 165mm tall. However, if you have to contend with RAM sticks and a VRM heatsink, the D15 very easily ends up a cm taller than noctua spec. Pick a case accordingly, or end up with my predicament
I do find that extra bit of clearance the Assassin III has for a GPU to be useful.
I would not have expected noctua coolers to be affected since they have a base plate spreading the heat to all heat pipes, most affected coolers would be those with direct contact heatpipes, maybe.
One thing that would have been interesting to see is for you to run them completely passive with no airflow and see which one clocks the best etc
I would want to know whether there's a difference between the heat pipes being mounted vertically (like on a bench test rig) and horizontally (like most computer cases), and maybe upside down too. The theory behind there being a difference is that vertically mounted heat pipes have gravity on their side when it comes to moving the liquid inside them to the heat source.
Yes there is actually a pretty significant difference. The best orentation for most all heatpipes is as follows : heat source at the bottom , vertical heatpipes condenser at the top. In this case vapor is allowed to rise and gravity assists with returning the fluid back to the hotter base. Any deviation from this lowers the Qmax of the heatpipes. By how much is dependent on a lot of things. They (as in cooler companies) actually change the wick type , wick thickness , and wick porosity to design around and mitigate orentation. Porosity and wick thickness for example have a pretty good effect on the ability of the heatpipe to work against gravity or in different orentations. Some wick types also work better in some orentations vs others. If your interested I attached a reply i made to another comment on this video to the bottom of this that contains some good info on heatpipes
there is a lot that goes into it. Starting off diameter plays a huge role. A 3mm heatpipe might have a qmax of 15w and a 8mm heatpipe qmax of 63w.... Thats a big difference for less then a 3x boost in diameter thats nearly 4x the orgional energy for sub 3x larger dia. .... Now lets say you have a low profile heatsinks (laptop for example) and you need to flatten a heatpipe ... If you flatten a 3mm heatpipe by 2mm height thats 3.57mm wide and now has a qmax of 10w instead of 15w .... If you take a 8mm heatpipe and flatten it so the width now becomes 11.14mm @ widest point the qmax is 52w rather then 63w of the round 8mm heatpipe....now some of the factors here are the wick limit (ability of the wick to transfer condensed fluid) and vapor limit (ability and space for the vapor to form and travel.) Thickness and porosity of the wick can be changed to tune it for a application. Which would allow for Qmax changes and / or the ability to work against gravity. .... A typical water filled sintered wick heatpipe (very common for pc cooling ) can be flattened up to around 60% though 30% is much more reasonable figure. As you flatten a heatpipe you will lower the vapor limit. A 3mm heatpipe flattened 33% now has Qmax limited by vapor limit rather then wick. Where as a 8mm heatpipe can be flattened by about 66% before the Qmax is now limited by vapor limit. ... Now if we start making bends we further reduce efficency. If we take the 8mm flattened heatpipe talked about above (52w Qmax) and we bend it 90 degrees we lose about 5% of our 52w Qmax. Which would be about what 49w after that off the top of my head ...then you have orentation. The most ideal situation for most heatpipes is vertical orentation as opposed to the heatsource with the condenser above the source of heat so vapor can rise and gravity can assist with pulling the condensed fluid back down to the source of the heat. Now to makes things even more complex again wick thickness and porosity effect this. As we as thsre is quite a few differnt forms of wick types that each work better for different things. Then if we want to complicate this more the length matters... shorter = more efficient (assuming you can still matain the same ability to cool the vapor in the condenser) now it varies for many reasons (ex. Working fluid , orentation , wick thickness , wick porosity , wick type , ect) however many fairly traditional heatpipes start to fall off around 150mm in length. However there is factors you can adjust to mitigate this. .... This info is still condensed a lot. And there is a lot of types of heatpipes. The ones discussed above are constant conductance heat pipes. There are also variable conductance heat pipes , vapor Chambers, pressure controlled heat pipes , rotating heatpipes , thermal syphons, loop heat pipes , oscillating or pulsing heatpipes, and diode heatpipes. Even within diode heatpipes there are a few sub types like vapor trap and fluid / liquid trap types.
Isnt that the point of the IHS to spread the heat so that aslong as the heatpipes or contact plates touch the ihs it should be equally distributing the heat?
I have a request. Would you consider doing a test with venting the heatpipes? I am interested in seeing how much effect the liquid actually has. We know heat travels by 3 methods, Conduction, Convection, and Radiation. The copper is always there and conduction is a constant. The fluid introduces both phase change and convection. You guys seem to be able to afford to do destructive testing so how about it?
If you ever made videos on the methodology I would watch the shit outta that
What happened to the Review for XFX 5700XT THICC 3 ULTRA ?
you could have also talked about thermal paste application and that the typical pea sized or line method is not good for ryzen given the die layout. but i know it dosen't affect your testing since of the general thermal paste application video you guys did.
How about water block channel direction?
Great work! This is as close to scientific method as you can easily do with coolers. The only thing possibly missing is testing for the gravity effects (e.g. you could run the whole test bench upside down and see if it affects the performance of heat pipes - logically cooled liquid should fall downwards and if you have the system upside down, the liquid inside the heat pipes might never contact the CPU end of the heat pipe).
Heat pipes are built with wicking inside.. Sintered metal powder based heat pipes should not in anyway be disturbed by gravity! Liquid has no issue traveling against gravity in a wick
@@DarkAttack14 I totally agree that a *correctly* implemented heat pipe should depend on wicking only. My trust in average hardware manufacturer is low enough that I think the implementation quality should be validated and that's why I think at least one test done upside down would be warranted. If that doesn't show a difference, then you can be pretty sure that the wick implementation was actually correctly done.
@@MikkoRantalainen yeah I'll give you that one, I'm operating from the assumption they produced it properly! You're not wrong that it is not a guarantee. One of the first uses for heatpipes was inside satellites and spacecraft so a properly built one can handle all types of gravitational situations but one would assume NASA was building proper heatpipes XD
I am guessing that the CPU heatspreader and Noctua base plate is doing its job well enough to spread the heat out to all heatpipes.
I am running a Noctua D15se AM4 with horizontal fans, to suck heat away from the GPU and send it up and out through mesh top.
A Ryzen 1700 overclocked to around 110W does not mind sucking in a little hot air from the GPU.
I have both Be Quiet and Noctua big air coolers, for case builds you are right the GPU being jammed against the cooling tower is the worst issue.
Is there any difference in heatpipe performance when placed vertical(ala test bench) vs. horizontal(ala most cases)?
Yes there is - i attached a reply i made to another earlier comment that goes over this as well as some other info on heatpipes mostly being constant conductance heatpipes.
there is a lot that goes into it. Starting off diameter plays a huge role. A 3mm heatpipe might have a qmax of 15w and a 8mm heatpipe qmax of 63w.... Thats a big difference for less then a 3x boost in diameter thats nearly 4x the orgional energy for sub 3x larger dia. .... Now lets say you have a low profile heatsinks (laptop for example) and you need to flatten a heatpipe ... If you flatten a 3mm heatpipe by 2mm height thats 3.57mm wide and now has a qmax of 10w instead of 15w .... If you take a 8mm heatpipe and flatten it so the width now becomes 11.14mm @ widest point the qmax is 52w rather then 63w of the round 8mm heatpipe....now some of the factors here are the wick limit (ability of the wick to transfer condensed fluid) and vapor limit (ability and space for the vapor to form and travel.) Thickness and porosity of the wick can be changed to tune it for a application (like differnt orentations) Which would allow for Qmax changes and / or the ability to work against gravity. .... A typical water filled sintered wick heatpipe (very common for pc cooling ) can be flattened up to around 60% though 30% is much more reasonable figure. As you flatten a heatpipe you will lower the vapor limit. A 3mm heatpipe flattened 33% now has Qmax limited by vapor limit rather then wick. Where as a 8mm heatpipe can be flattened by about 66% before the Qmax is now limited by vapor limit. ... Now if we start making bends we further reduce efficency. If we take the 8mm flattened heatpipe talked about above (52w Qmax) and we bend it 90 degrees we lose about 5% of our 52w Qmax. Which would be about what 49w after that off the top of my head ...then you have orentation. The most ideal situation for most heatpipes is vertical orentation as opposed to the heatsource with the condenser above the source of heat so vapor can rise and gravity can assist with pulling the condensed fluid back down to the source of the heat. Now to makes things even more complex again wick thickness and porosity effect this. As we as thsre is quite a few differnt forms of wick types that each work better for different things. Then if we want to complicate this more the length matters... shorter = more efficient (assuming you can still matain the same ability to cool the vapor in the condenser) now it varies for many reasons (ex. Working fluid , orentation , wick thickness , wick porosity , wick type , ect) however many fairly traditional heatpipes start to fall off around 150mm in length. However there is factors you can adjust to mitigate this. .... This info is still condensed a lot. And there is a lot of types of heatpipes. The ones discussed above are constant conductance heat pipes. There are also variable conductance heat pipes , vapor Chambers, pressure controlled heat pipes , rotating heatpipes , thermal syphons, loop heat pipes , oscillating or pulsing heatpipes, and diode heatpipes. Even within diode heatpipes there are a few sub types like vapor trap and fluid / liquid trap types.
Why didn't you use a pcie extension cable for the GPU?
As for front-back vs side the important part is if you have the top exhaust. The "warm" air off the GPU is irrelevant. Unless your GPU has a backplate cooling problem or you're fanning an M.2 card with the CPU fan, then there might be a small difference.
"Just put the damn cooler on"
Hell yeah GN! Put Reddit in its place. Good lord they have too much time to think
They don't really think, though. They just take some bizarre idea and repeat it ad infinitum...until they believe the idea itself and accept it as true, fighting off anyone that says otherwise or points out the facts that prove it otherwise.
This whole discussion can be settled with a side-by-side thermal image of the IHSs. No problem if the CPUs throttle because you're comparing the pattern on the IHS from CPUs in the same conditions.
And under load the IHS would perform much better spreading out the heat so only if the IHS in a throttled state would be showing problematic heat spreading, would it become necessary to test if it still does under load.
What does IHS stand for again?
Ι rotated 90 degrees the am4 bykski block i have (ek supremacy evo copy), so the chiplet is fully under fins and its a little better. I believe the problem with those chips is 7nm transistors and heat dissipation, also high voltage (especially with pbo on some mobos), not offcentered heat.
Noctober!! Instant classic 🤓
Be interesting to see the U14s run with dual fans vs d15.
Hi Gamers Nexus,
Igor from IgorsLab has filmed thermal images with his industry thermal cam. Also Igor experimented with a set of modified jet plates for water cooling blocks and water flow amounts. Also Igors shows thate the IHS makes a good job spreading the heat.
His verdict was that it makes a little difference what jet plate is used and how many water flow hits the compute dies or the I/O die. He saw around 2-3°K difference in cooling performance in lowest water flow scenarios. If a higher water flow is used the differences are approaching measurement error: 0-1°K. I think that confirms your findings but for water cooling.
How ever Igor referenced AiOs and that they could be have a little bit of trouble. Cause is mostly the amount of microfins and if they positioned on the hot spot of the compute die or how many fins placed above the die or not. But this is another discussion and off topic to your video here.
Here are Igors videos, I recommend to take a look, very good content:
ua-cam.com/video/C5gTffl98Xc/v-deo.html
ua-cam.com/video/3WgtYXifHc0/v-deo.html
Minor nitpick: When referring to temperature measurements in the absolute scale, it's just Kelvin, not DEGREES Kelvin (i.e. please say 2-3K instead of 2-3°K). When referring to temperature measurements in Fahrenheit or Celcius, feel free to include the degree symbol.
Anyway thanks for the links, they're quite interesting supplemental material for this video.
Hi there,
I was wondering if y'all would be willing to do a video testing different GPUs with different types of cooling(blower style, heat pipes and fins, just fins, etc.) and their ability to cool themselves in different orientations.
This problem stems from the SFF community where there are many cases with an upright GPU, which is now being introduced into Lian Li's cases as well. The SFF community's general consensus has been an upright GPU does not cool as well because of heat pipes pointing down, vapor chamber liquid being further from the processor, and blowers having more resistance blowing up than sideways.
I would think case designers would not make upright mounts if this were the case, however, Thermaltake has just pumped out a new set of 90 degree mounted cases and a new Tower case with the GPU in an upright orientation.
I would appreciate it if there was a test of some sort done specifically for GPU orientation in open air, not taking into account things like distance to side panels and such. The only tests I can find for heat pipes are industrial or zero gravity applications.
If you're going to do cool mythbusting videos like this, then can you please test the effect of vertical/horizontal GPU cooler fins on internal case/motherboard/M.2 SSD temperature please.
The orientation of the grills on GPU coolers is almost never mention in the reviews. Generally the lower end board partner designs have horizontal grills and this may be to the detriment of cooling performance but does allow /some/ hot air to be vented out through the PCI-e bracket whereas vertical grills throw hot air into the motherboard and case-side possibly heating up a M.2 SSD mounted between PCI-e slots. Thanks!
You could eliminate GPU thermal influence by either using a riser cable to move it away from the CPU, even blocking it with cardboard or something, or using something like an R7 240 for CPU-only tests... But I think you proved well enough that there isn't a practical difference, even if there might be a technical difference, so it wouldn't really be a useful test at this point.
we love you steve