Opinions on this case ?

[Diegoahq]

Hello, first of all i don't speak english soo sorry if this is hard to read and i don't post here soo often sorry if i'm breaking any rule or something.

I reciently bought a rtx 3080 and i'm having thermal problems because my case doesn't have good airlfow. (Aereocool SI-5200) I managed to fix it using vsync in some games and limiting fps in others and undervolting (i'm sitting around 60°C).

But i'd like to use the card at the full potential and play heavy games without having to worry about the temperatures and upgrade my monitor in the future. I was thinking of getting a new case. Where I live there isn't a lot of alternatives and besides that i need at least 165mm for the cpu cooler. I was thinking about this case: Deepcool MATREXX 55 MESH 4 fans.

What do you think of this case ? The airflow will be enought for the 3080 ? Any recomendation ?

Rest of my system: 

 

i5 9600kf (no oc)

16gb ddr4 (8x2)

Seasonic focus gold 750w

msi b360m-pro-vdh

240 ssd, 500 hhd, 1tb hhd

 

Sorry I feel like this is really poorly written. Thanks for reading.

 

[Bombastinator]

I would like to first suggest a hack for your current case: remove the front panel.  It will give vastly less obstructed airflow to your fans and might make a new case unnecessary.  The main difference between your current case and your proposed new on is the new one has a perforated front panel whereas the old one does not.  If nothing else it would at least tell you if the proposed case will do the job for you or not.

[Diegoahq]

31 minutes ago, Bombastinator said:

I would like to first suggest a hack for your current case: remove the front panel.  It will give vastly less obstructed airflow to your fans and might make a new case unnecessary.  The main difference between your current case and your proposed new on is the new one has a perforated front panel whereas the old one does not.  If nothing else it would at least tell you if the proposed case will do the job for you or not.

Okay, I did that and I tested temps and noise with Superposition Benchmark at 8K optimized. 

Here are the results.

Front panel on: https://imgur.com/a/loaUvwZ

Front panel off: https://imgur.com/a/NWd4elL

The difference that I noticed its between 72 °C and 80 °C max. with stock fan curve and undervolted at 1800 MHz 0.831. Maybe taking the front panel off didin't change much because i have bad intake fans ? and with the better ones in the new case the difference will be more notable ? Or this is it ? Also by changing the case maybe my cpu temps will get lower ? I'm currently in 65°C - 80°C playing.

[Bombastinator]

6 minutes ago, Diegoahq said:

Okay, I did that and I tested temps and noise with Superposition Benchmark at 8K optimized. 

Here are the results.

Front panel on: https://imgur.com/a/loaUvwZ

Front panel off: https://imgur.com/a/NWd4elL

The difference that I noticed its between 72 °C and 80 °C max. with stock fan curve and undervolted at 1800 MHz 0.831. Maybe taking the front panel off didin't change much because i have bad intake fans ? and with the better ones in the new case the difference will be more notable ? Or this is it ? Also by changing the case maybe my cpu temps will get lower ? I'm currently in 65°C - 80°C playing.

The thing I notice about the difference in gpu temps is with the “on” one the temperature was still rising when the test was conclude while the “off” one had reached steady state. There is no temperature scale on the graphs I can see though which makes their utility limited.  A final temp is given but only after a cool down period making it near useless.
 

You state that the difference in case temp between removed and not removed was 8°c? 

if the difference between the front panel “on” and the front panel “off” is not enough for what you want, I think more than a case change may be necessary.  The “new”proposed case isn’t all that different from the old one with no front.  It may have more width which could allow a bigger air cooler.

[Diegoahq]

49 minutes ago, Bombastinator said:

The thing I notice about the difference in gpu temps is with the “on” one the temperature was still rising when the test was conclude while the “off” one had reached steady state. There is no temperature scale on the graphs I can see though which makes their utility limited.  A final temp is given but only after a cool down period making it near useless.
 

You state that the difference in case temp between removed and not removed was 8°c? 

if the difference between the front panel “on” and the front panel “off” is not enough for what you want, I think more than a case change may be necessary.  The “new”proposed case isn’t all that different from the old one with no front.  It may have more width which could allow a bigger air cooler.

"The thing I notice about the difference in gpu temps is with the “on” one the temperature was still rising when the test was conclude" That maybe translate to higher temps in longer game sesions ? I belive that i should test that with a high demanding game (Red dead Redemption II or Doom Eternal) maybe  ? anyways i don't have either on steam ;/. I don't really know alot about this.. Acording to youtube videos my temps should be way lower. 

 

"You state that the difference in case temp between removed and not removed was 8°c? " I meant the GPU temps.

[Bombastinator]

3 minutes ago, Diegoahq said:

"The thing I notice about the difference in gpu temps is with the “on” one the temperature was still rising when the test was conclude" That maybe translate to higher temps in longer game sesions ? I belive that i should test that with a high demanding game (Red dead Redemption II or Doom Eternal) maybe  ? anyways i don't have either on steam ;/. I don't really know alot about this.. Acording to youtube videos my temps should be way lower. 

 

"You state that the difference in case temp between removed and not removed was 8°c? " I meant the GPU temps.

So resting idle case temp is 72°c?  That’s extremely high.  A case temp of 50°c is high.

[Diegoahq]

Just now, Bombastinator said:

So resting idle case temp is 72°c?  That’s extremely high.  A case temp of 50°c is high.

No no, GPU temps maxed at 72 °C with the front panel OFF. I don't know how to measure case temps and what exactly they are.

[Diegoahq]

5 minutes ago, Bombastinator said:

So resting idle case temp is 72°c?  That’s extremely high.  A case temp of 50°c is high.

Look at this video even the stock temps are way better than my temps, maxing at 72 ° C. And when he undervolted, at 1800MHz he has way lower temps than me. And i can't find an answer to that other than my case airflow is shitty and my fans are bad.

 

[Bombastinator]

Just now, Diegoahq said:

No no, GPU temps maxed at 72 °C with the front panel OFF. I don't know how to measure case temps and what exactly they are.

The case air temp could possibly be acquired from a sensor perhaps somewhere on the motherboard when the machine is at rest.

 

A load temp of 72°c does not strike me as extremely high for a modern system.  One complication can be ambient (or room) temperature.  Temperatures are cooling + ambient, so if the room you put a computer in is warm, the temps will be too.  Looking about I am seeing load temperatures for 3080s consistently above 72°c.  I’m seeing 78°c quite commonly.  I don’t know what throttle temp is for a 3080.  I’m having trouble finding it.  The stuff I see for various random 3080s I see online shows idle to be around 50some and load to be in the high 70’s. These will have been measured with a different ambient than you have though.

[Bombastinator]

17 minutes ago, Diegoahq said:

Look at this video even the stock temps are way better than my temps, maxing at 72 ° C. And when he undervolted, at 1800MHz he has way lower temps than me. And i can't find an answer to that other than my case airflow is shitty and my fans are bad.

 

I’m seeing a variance between 68-79°c in the top left corner.  Perhaps I’m reading it wrong. You’re running a different bench than him with different hardware though so comparison isn’t going to be very exact
 

case airflow is perhaps a bit shitty with the front cover on.  But with front cover off it more or less can’t be much worse than average.  The question was will you gain much by changing cases?  I do not think a whole lot of difference between your current case with the front off and the new case, though the added width would allow for a larger more expensive cpu air cooler.  Wider case wouldn’t make a whit of difference for a water cooler 

Edited November 25, 2020 by Bombastinator

[Diegoahq]

8 minutes ago, Bombastinator said:

The case air temp could possibly be acquired from a sensor perhaps somewhere on the motherboard when the machine is at rest.

 

A load temp of 72°c does not strike me as extremely high for a modern system.  One complication can be ambient (or room) temperature.  Temperatures are cooling + ambient, so if the room you put a computer in is warm, the temps will be too.  Looking about I am seeing load temperatures for 3080s consistently above 72°c.  I’m seeing 78°c quite commonly.  I don’t know what throttle temp is for a 3080.  I’m having trouble finding it.  The stuff I see for various random 3080s I see online shows idle to be around 50some and load to be in the high 70’s. These will have been measured with a different ambient than you have though.

Sorry I accidentaly marked the answer as solution.

"The case air temp could possibly be acquired from a sensor perhaps somewhere on the motherboard when the machine is at rest." With what program I can see that ?

I will try doom (2016) without a fps limit and with the front panel on vs off, and i will reply to you tomorrow with the results. Should I test or do something else ?

I belive that temps playing will be aloot higher than temps benchamrking, i don't have anything to support that theory but i will try it. 

[Diegoahq]

3 minutes ago, Bombastinator said:

I’m seeing a variance between 68-79°c in the top left corner.  Perhaps I’m reading it wrong. 
 

case airflow is perhaps a bit shitty with the front cover on.  But with front cover off it more or less can’t be much worse than average.  

I readed 72 as the max number at the beggining anyways it's really blurry soo let's forget about this video. 

[Bombastinator]

7 minutes ago, Diegoahq said:

Sorry I accidentaly marked the answer as solution.

"The case air temp could possibly be acquired from a sensor perhaps somewhere on the motherboard when the machine is at rest." With what program I can see that ?

I will try doom (2016) without a fps limit and with the front panel on vs off, and i will reply to you tomorrow with the results. Should I test or do something else ?

I belive that temps playing will be aloot higher than temps benchamrking, i don't have anything to support that theory but i will try it. 

Re: programs capable of logging various temp sensor data

im a bit out of date on this one.  I used to use speedfan for that but there are probably better things now.  I vaguely suggest HWmonitor as I’ve heard good things about it.  Never used it myself. Many motherboards have various temp sensors salted along the board and amongst other parts.  One far from a heat source could tell you how hot the case air is actually getting inside.  Some fans also have sensors in them though it’s more rare.  Perhaps unheard of these days.  I haven’t heard of a fan with an air temp sensor in a while.
 

I think you may be worrying too much about temperature.  As long as your hardware is running at a safe temperature level it won’t get hurt.

[Diegoahq]

12 minutes ago, Bombastinator said:

I think you may be worrying too much about temperature.  As long as your hardware is running at a safe temperature level it won’t get hurt.

Yeah, maybe, I upgraded from a asus strix 1050 TI witch in high loads had 60°C only. I'm not used to high end GPUS. Maybe it's normal. Thanks for the help.