Monitoring Temperatures with HWInfo64 - Best Practice?

[RevGAM]

I'm currently working on improving my testing of an AIO to corroborate MSI's AIO mounting article. I've just given them some feedback on problems with that article (e.g. intake arrows in images that should be blue, not red; the ideal position of the water block's tubes should be from the top; pump location influence on where the rad can safely be mounted and its orientation).

 

I'm revising which temperatures I'm monitoring during testing. Previously, I was using readings from the mobo for "CPU", "CPU Package", "VRM", "Mobo", and I use the "GPU Temperature" for my GC. However, after looking at the brief descriptions, it seems that some of these are not ideal, so I've switched to the CPU temps from the "DTS" and "Enhanced" sections. HWInfo suggests that some of the readings from the mobo may not be reliable because of variance from one manufacturer to another.

Would it be advisable to monitor any other temperatures, such as "GPU Memory Junction Temperature", "Core Temperatures (DTS)", "CPU IA Cores", and "CPU GT Cores"?

I'm also monitoring the RPMs of the AIO fans, but not the exhaust fan.

 

I'm considering removing the VRM temps as they do not fluctuate more than 3^o C most of the time, even during benchmarking (however, Time Spy Extreme pushes it beyond 10^oC variance).

I haven't been monitoring the pump, since it's set to 100% on the pump header, and is always above 5,000 RPM.

 

This is my current view.

 

Any thoughts? @TatamiMatt@NorKris@Agall@freeagent@thrasher_565@jaslion@TeraSeraph

Edited February 10 by RevGAM
corrected which fans I'm monitoring

[NorKris]

16 minutes ago, RevGAM said:

I'm currently working on improving my testing of an AIO to corroborate MSI's AIO mounting article. I've just given them some feedback on problems with that article (e.g. intake arrows in images that should be blue, not red; the ideal position of the water block's tubes should be from the top; pump location influence on where the rad can safely be mounted and its orientation).

i think we have been clear on this: dont buy their AIO's dont buy their numbers...(the numbers dont make any sense) 

 

 

when it comes to the use of HWinfo64 (when looking at CPU stuff)  i expand the cores to also look at the diff between the hottest cores and the "chillest" core 

[freeagent]

I am running AMD right now, so we would be looking at some different things on the CPU, but this is generally how I have my screen setup.

 

[RevGAM]

9 minutes ago, NorKris said:

i think we have been clear on this: dont buy their AIO's dont buy their numbers...(the numbers dont make any sense)

And here I thought you liked that article, since it supports your claim that the best location is front-mounted.

10 minutes ago, NorKris said:

when it comes to the use of HWinfo64 (when looking at CPU stuff)  i expand the cores to also look at the diff between the hottest cores and the "chillest" core 

7 minutes ago, freeagent said:

I am running AMD right now, so we would be looking at some different things on the CPU, but this is generally how I have my screen setup.

I have too many sensors showing up in that view - even double-column isn't enough to show them all, so I use the charts.

What suggested charts should I have open?

[NorKris]

7 minutes ago, RevGAM said:

And here I thought you liked that article, since it supports your claim that the best location is front-mounted.

well, every article / YR test would.. but this..

 

i dont even..

 

 

[RevGAM]

4 minutes ago, NorKris said:

well, every article / YR test would.. but this..

i dont even..

What? Is it because you disagree that tubes down should be cooler? Or is it because the image shows 240mm AIOs? Or what?

One of the other points I found confusing is most drawings show a 240, while a few show a 360, like this set. They didn't even properly present all the data for the 360s and 240s.

Well, at least this one only has problems with arrow colors. I hope. It's a flawed article.

 

Anyways, this is all beside the point. What about my questions?

[NorKris]

2 minutes ago, RevGAM said:

What? Is it because you disagree that tubes down should be cooler? Or is it because the image shows 240mm AIOs? Or what?

well let me put it this way, try to logical explain it  

 

3 minutes ago, RevGAM said:

 

Anyways, this is all beside the point. What about my questions?

i dont see any reviewers having charts open,  min, max and avg is a nice thing 

[RevGAM]

3 minutes ago, NorKris said:

well let me put it this way, try to logical explain it

I've pointed out some things I don't like about it. Why not just state what your concern is instead of beating around the bush?

4 minutes ago, NorKris said:

well let me put it this way, try to logical explain it  

 

i dont see any reviewers having charts open,  min, max and avg is a nice thing 

I've seen Gamers Nexus use charts. I know others have although I can't remember who right now. In any event, as I already stated I have to use the charts because there's no way to have the sensors status window open and show everything (that I know of), plus I have both the charts AND that window open, and I screenshot them immediately after every test.

[NorKris]

1 minute ago, RevGAM said:

I've pointed out some things I don't like about it. Why not just state what your concern is instead of beating around the bush?

I've seen Gamers Nexus use charts. I know others have although I can't remember who right now. In any event, as I already stated I have to use the charts because there's no way to have the sensors status window open and show everything (that I know of), plus I have both the charts AND that window open, and I screenshot them immediately after every test.

oh charts like that (like a good visual for YT) thats great, but not a must to have it up in hwinfo   thats my take  

[RevGAM]

3 minutes ago, NorKris said:

oh charts like that (like a good visual for YT) thats great, but not a must to have it up in hwinfo   thats my take  

You haven't answered my questions. Are you okay?

To recap:
What are your logical concerns with the MSI article?

Which sensors should I monitor, with the understanding that I cannot see them all in the sensors status window so I must open charts?

[RevGAM]

50 minutes ago, freeagent said:

I am running AMD right now, so we would be looking at some different things on the CPU, but this is generally how I have my screen setup.

 

That's a lot of stuff. Since I'm only focusing on temps, not clock speed, electricity, etc., which temps do you suggest?

 

I've got 4 SSDs, which is pretty atypical, so I'm not watching them. Should I be?

[NorKris]

2 minutes ago, RevGAM said:

You haven't answered my questions. Are you okay?

 

wall of text is not my strong side, i get bored and get back to wow xd

6 minutes ago, RevGAM said:

What are your logical concerns with the MSI article?

 

when the numbers are off / strange.. thats it.. i made that clear yes?

12 minutes ago, RevGAM said:

the sensors status window so I must open charts?

when looking at cpu only stuff i use  all the core's: Max, min, Avg... i told u this too 

[RevGAM]

8 minutes ago, NorKris said:

when the numbers are off / strange.. thats it.. i made that clear yes?

As clear as mud, since you didn't explain what is off about the numbers and why you think that, or even that it was the numbers.

9 minutes ago, NorKris said:

when looking at cpu only stuff i use  all the core's: Max, min, Avg... i told u this too 

I know what you told me, but I can't just look at cores. Still, I can have the sensors window focused on the cores since opening up 12 more charts probably isn't going to be possible...Plus Core Max, since that plus Core Temps will save me a couple of calculations.

[NorKris]

1 minute ago, RevGAM said:

As clear as mud, since you didn't explain what is off about the numbers and why you think that, or even that it was the numbers.

 

CPU temps cannot go up 3-4c if tubes are up or down. 

Lets say you have tubes up right ( this is the way most ppl have tested AIOs) the water is not going to slow any slower than if the tubes are down. 

You cant say heat rises cuz the temp diff in the bottom part of the rad is MAYBE 0,1 less than the top 

 

cant only look at the cpu temps?  depends what ur after..   CPU is the hardest part in a pc to cool,, so ye some times 

5 minutes ago, RevGAM said:

Still, I can have the sensors window focused on the cores since opening up 12 more charts probably isn't going to be possible...Plus Core Max, since that plus Core Temps will save me a couple of calculations.

screen in portrait is OP   

[RevGAM]

9 minutes ago, NorKris said:

CPU temps cannot go up 3-4c if tubes are up or down. 

Lets say you have tubes up right ( this is the way most ppl have tested AIOs) the water is not going to slow any slower than if the tubes are down. 

You cant say heat rises cuz the temp diff in the bottom part of the rad is MAYBE 0,1 less than the top 

 

cant only look at the cpu temps?  depends what ur after..   CPU is the hardest part in a pc to cool,, so ye some times 

Well, I'm not an expert in thermodynamics but I would suspect that the air building up where the tubes are (tubes at the top) is going to have some sort of impact, as opposed to tubes down. Since I've not been satisfied with my previous tests, I've thrown out the data, so I cannot say what kind of differential I'm experiencing.

 

No, I didn't throw out the data because I disagreed with it. It's because I realized that I did something wrong or sub-optimal....Like, say, not heating up the loop enough prior to the first benchmark. Or my mysterious blood sugar problem hit me (my Hb A1C is always good, and so is my blood sugar according to my glucose monitor testing, but I still crash enough to have to take a nap) while I was testing, causing me to make some sort of mistake. Or I forgot to record the ambient temp. Or...or...or. Impending senility, too.

[NorKris]

9 minutes ago, RevGAM said:

Well, I'm not an expert in thermodynamics but I would suspect that the air building up where the tubes are (tubes at the top) is going to have some sort of impact, as opposed to tubes down. 

here you might be onto somthing, but look how big that "airgap" / "air bobbles" has to be from GN's testing

its massive. 

 

that would take years to "develop"..?  happen?  happen might be the word

 but it might happen after 2weeks on a MSI AIO  KEKW 

[RevGAM]

2 minutes ago, NorKris said:

here you might be onto somthing, but look how big that "airgap" / "air bobbles" has to be from GN's testing

its massive. 

 

that would take years to "develop"..?  happen?  happen might be the word

 but it might happen after 2weeks on a MSI AIO  KEKW 

Thank you for clarifying that for others! My grammar implied it ("building up") but that's not enough for many users here.

 

I wonder how fast permeation can happen when an AIO is just sitting in its box...A topic for another time.

 

So, at this point, I've added the Core Temps (DTS), the 11 cores, and the Core Max via the sensors status window, and the rest is shown via the graphs. I'm going to add in the exhaust fan.

[RevGAM]

Does anyone else have any suggestions?

I mean, I could certainly add in clocks and electricity for other tests, but I'm not sure how useful (and which) for this test. Any feedback from anyone?

[thrasher_565]

aios vary in  quality and how full they fill em. for the less full ones tubes down should be used to keep bubbles out of the pump and it should be the hiest part over the pump. a good aio filled right dose not matter thow over time it will lose liquid so like i said tubes down is best imo.

[freeagent]

The pump on mine is pretty strong, tubes are nice and stiff.. might get good life out of it.. tubes down is nowhere near a concern for me, and seems almost comical.

[RevGAM]

3 minutes ago, freeagent said:

The pump on mine is pretty strong, tubes are nice and stiff.. might get good life out of it.. tubes down is nowhere near a concern for me, and seems almost comical.

Why does it seem comical?

[thrasher_565]

4 minutes ago, freeagent said:

The pump on mine is pretty strong, tubes are nice and stiff.. might get good life out of it.. tubes down is nowhere near a concern for me, and seems almost comical.

im old school and thats how it was for custom loops so make sense to me. thow aios back in the day were vary good quality

people say getting like 7+ years from them. but the quality not as good these days from alot of useser.

[freeagent]

4 minutes ago, RevGAM said:

Why does it seem comical?

For one, it looks absolutely ridiculous.. and on my H100 tubes down didn't do anything meaningful. I originally attempted it to quiet the pump, sounded like a midget with a rattle can in my ginormous case at the time.

[IkeaGnome]

27 minutes ago, freeagent said:

it looks absolutely ridiculous.

And length. A lot of AIOs don't have long enough tubes to reach the front of the case and the bottom. Especially once you add in a GPU. 

[freeagent]

1 minute ago, IkeaGnome said:

And length. A lot of AIOs don't have long enough tubes to reach the front of the case and the bottom. Especially once you add in a GPU. 

I cant do it with my current one and have a GPU at the same time, it was how I was going to initially run it.. because I saw the videos too

 

But the hose length is too short, but girth is fine