FAQ FAQ  Forum Search   Register Register  Login Login

LC vs Air

Post Date: 2016-08-28

 Post Reply Post Reply
Author
  Topic Search Topic Search  Topic Options Topic Options
Batman34 View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 15 Aug 2016
Online Status: Offline
Posts: 366
  Quote Batman34 Quote  Post ReplyReply bullet Topic: LC vs Air
    Posted: 28 Aug 2016 at 2:47pm
3 questions about lc vs air:

1. How long would a lc system last if properly maintained?

2. What benefits do I get with going with the lc over air?

3. If lc lasts only 5-6 years, why would someone spend $1000 on open loop lc?
"It's not who I am underneath, but what I do that defines me." - Batman Begins
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 3:07pm
1. an open loop will last as long as you maintain it. aio can last 3 to 5 years depending.

2. unless you are looking for extreme oc or need to have lower temps because for some reason air couldn't do the job, you have no benefit over air other then lower temps, but if your temps are already in acceptable range with air then you have nothing to gain really. In some cases if the loop is custom with quiet fans then LC will be more quiet then air but again that depends on the fans on the rad, at the end it all comes back to the fan whether its air or lc. Having said that, keep in mind that the main difference of lc over air is that liquid has much better heat exchange then air, in other words temps should always be lower on lc than air, as long as we are talking about an open loop LC with proper parts and properly setup, otherwise that statement will not hold true.

3. I think you are confused with LC. All LC are not the same, if you think you are getting the same performance out of say hydrolux lite or corsair as an open loop you are heavily mistaken. Real LC performance does not start till you get to an open loop, even then you are only as good as the parts used in that loop, you have to have the proper waterblock, pump, rad, fans, tubing kind and size that matches that wb and rad setup.

Most of the aio match or beat noctua by very few degrees, not even worth mentioning, an open loop lc will put both aio and noctua to shame.

Now I have a question for you:

4. why bother going with aio if you can get the same round about performance from noctua without having to worry about pump failures and leaks? While noctua will be more quiet.

It is important to not go by people's opinion due to the fact that each person has a different experience due to different variables, ambient temps being one. So sticking with facts based on creditable tests, proper open loop has the best performance, followed by air/aio.

As always do some research yourself on creditable sites and see the results for yourself.

Edited by DST4ME - 28 Aug 2016 at 3:36pm
Back to Top
EdgeCrusher View Drop Down
Newbie
Newbie

Email address used to purchase matched with forums account email.

Joined: 17 Jul 2010
Online Status: Offline
Posts: 66
  Quote EdgeCrusher Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 3:18pm
I Generally upgrade my system every 6 years or so. Maybe that's not for everybody but I also have never used open loop myself just air or all in one water coolers.

Air coolers will need dusting and possibly fan replacements over 6 years and all in one coolers will need the same on the radiators. If either fail its easy to replace.

I personally like the AIO coolers because they are much smaller profiled in the case and less weight/stress on the board as a big air cooler. The Noctua air coolers DS sell perform the same as the Corsair 100i GTX's that DS sell.

The reason why people go with open loop coolers is that technically you can get the most performance out of it but I would guess its mostly a looks thing. Air or AIO's can do just fine on CPU overclocks but that's the other thing, the coolers I'm talking about above only really do the CPU. Full system water loops do the CPU, GPU and possibly other parts in the case.

Bottom line is, all 3 types of cooling require some kind of maintenance and over 6 years that's totally normal. I'd say most of it comes down to what kind of overclocking your planning on doing and your own personal preferences to looks and budget. I'm sure a certain somebody will be by with charts and numbers saying how one method is far superior then the other but done right any of the three options are viable.
Back to Top
Batman34 View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 15 Aug 2016
Online Status: Offline
Posts: 366
  Quote Batman34 Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 4:35pm
Originally posted by DST4ME

1. an open loop will last as long as you maintain it. aio can last 3 to 5 years depending.

2. unless you are looking for extreme oc or need to have lower temps because for some reason air couldn't do the job, you have no benefit over air other then lower temps, but if your temps are already in acceptable range with air then you have nothing to gain really. In some cases if the loop is custom with quiet fans then LC will be more quiet then air but again that depends on the fans on the rad, at the end it all comes back to the fan whether its air or lc. Having said that, keep in mind that the main difference of lc over air is that liquid has much better heat exchange then air, in other words temps should always be lower on lc than air, as long as we are talking about an open loop LC with proper parts and properly setup, otherwise that statement will not hold true.

3. I think you are confused with LC. All LC are not the same, if you think you are getting the same performance out of say hydrolux lite or corsair as an open loop you are heavily mistaken. Real LC performance does not start till you get to an open loop, even then you are only as good as the parts used in that loop, you have to have the proper waterblock, pump, rad, fans, tubing kind and size that matches that wb and rad setup.

Most of the aio match or beat noctua by very few degrees, not even worth mentioning, an open loop lc will put both aio and noctua to shame.

Now I have a question for you:

4. why bother going with aio if you can get the same round about performance from noctua without having to worry about pump failures and leaks? While noctua will be more quiet.

It is important to not go by people's opinion due to the fact that each person has a different experience due to different variables, ambient temps being one. So sticking with facts based on creditable tests, proper open loop has the best performance, followed by air/aio.

As always do some research yourself on creditable sites and see the results for yourself.

Thanks for the response. I always feel like I'm in a classroom when you respond to my questions. Anyways, what kind of oc would you consider to be extreme for open loop lc? 4.8 ghz? Also, what monitor would you recommend for 1080p?
"It's not who I am underneath, but what I do that defines me." - Batman Begins
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 4:52pm
My pleasure and thank you, you are very kind.

Extreme OC depends on the cpu, most sites/benches will consider 5.0Ghz an extreme oc, but that really depends on the cpu, for example I would say 4.6GHz would be an extreme OC for the 6950x and 4.8GHz to 5.0GHz would be an extreme oc for the 6700Km but those are my opinions based on how far one has to push clock/vcore on those cpus to get there.

Keep in mind extreme oc is not meant for long term use, but rather short term benches.

Now when you say open loop you have to remember all open loops are not the same either, one loop can have multiple triple or quad rads on one loop and that system will be far more superior than a loop with say just one trip rad.

As a base when it comes to LC, for a minimum consider one 120mm/140mm rad for each part, so for example for cpu at a minimum we need one 120mm/140mm rad, now if we decide to use double rads instead of a single, we are going to have lower temps, triple rad lower temps, and so on and so forth, of course at some point the performance gain diminishes depending on what you are doing.

For monitors first you want to look at the panel, IPS is the one with best image quality, then look at the refresh rate, for gaming you want a 120/144Hz refresh rate.

For monitor guide have a look here, everything you need to know is here

Edited by DST4ME - 28 Aug 2016 at 4:53pm
Back to Top
Batman34 View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 15 Aug 2016
Online Status: Offline
Posts: 366
  Quote Batman34 Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 5:31pm
Originally posted by DST4ME

My pleasure and thank you, you are very kind.

Extreme OC depends on the cpu, most sites/benches will consider 5.0Ghz an extreme oc, but that really depends on the cpu, for example I would say 4.6GHz would be an extreme OC for the 6950x and 4.8GHz to 5.0GHz would be an extreme oc for the 6700Km but those are my opinions based on how far one has to push clock/vcore on those cpus to get there.

Keep in mind extreme oc is not meant for long term use, but rather short term benches.

Now when you say open loop you have to remember all open loops are not the same either, one loop can have multiple triple or quad rads on one loop and that system will be far more superior than a loop with say just one trip rad.

As a base when it comes to LC, for a minimum consider one 120mm/140mm rad for each part, so for example for cpu at a minimum we need one 120mm/140mm rad, now if we decide to use double rads instead of a single, we are going to have lower temps, triple rad lower temps, and so on and so forth, of course at some point the performance gain diminishes depending on what you are doing.

For monitors first you want to look at the panel, IPS is the one with best image quality, then look at the refresh rate, for gaming you want a 120/144Hz refresh rate.

For monitor guide have a look here, everything you need to know is here

How many rads does the h2o hydrolux pro have? Also, is the hardline or copper tubing worth it?
"It's not who I am underneath, but what I do that defines me." - Batman Begins
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 5:38pm
I want to say one triple rad which should be enough but they may do different configs for different systems so check with them to be sure.

Hardline and copper look amazing but they are very difficult to work on.
Back to Top
Batman34 View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 15 Aug 2016
Online Status: Offline
Posts: 366
  Quote Batman34 Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 5:53pm
Originally posted by DST4ME

My pleasure and thank you, you are very kind.

Extreme OC depends on the cpu, most sites/benches will consider 5.0Ghz an extreme oc, but that really depends on the cpu, for example I would say 4.6GHz would be an extreme OC for the 6950x and 4.8GHz to 5.0GHz would be an extreme oc for the 6700Km but those are my opinions based on how far one has to push clock/vcore on those cpus to get there.

Keep in mind extreme oc is not meant for long term use, but rather short term benches.

Now when you say open loop you have to remember all open loops are not the same either, one loop can have multiple triple or quad rads on one loop and that system will be far more superior than a loop with say just one trip rad.

As a base when it comes to LC, for a minimum consider one 120mm/140mm rad for each part, so for example for cpu at a minimum we need one 120mm/140mm rad, now if we decide to use double rads instead of a single, we are going to have lower temps, triple rad lower temps, and so on and so forth, of course at some point the performance gain diminishes depending on what you are doing.

For monitors first you want to look at the panel, IPS is the one with best image quality, then look at the refresh rate, for gaming you want a 120/144Hz refresh rate.

For monitor guide have a look here, everything you need to know is here

What do you mean by "extreme oc is not meant for long term use, but rather short term benches." Also, from your experience, is open loop cooling worth it if you have the money?
"It's not who I am underneath, but what I do that defines me." - Batman Begins
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 5:59pm
I would only get open loop if you need extreme oc or its just preference.

Very high oc will have very high vcore, very high vcore will kill your cpu faster. For an OC that you want to use on consistent basis you are looking for a vcore of preferably lower than 1.3v, 1.4 is the max I would recommend and not for constant use either. There are other factors that go into it but mainly vcore is a big part of the equation.

Edited by DST4ME - 28 Aug 2016 at 6:00pm
Back to Top
Batman34 View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 15 Aug 2016
Online Status: Offline
Posts: 366
  Quote Batman34 Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 6:06pm
Originally posted by DST4ME

I would only get open loop if you need extreme oc or its just preference.

Very high oc will have very high vcore, very high vcore will kill your cpu faster. For an OC that you want to use on consistent basis you are looking for a vcore of preferably lower than 1.3v, 1.4 is the max I would recommend and not for constant use either. There are other factors that go into it but mainly vcore is a big part of the equation.

What is the ideal vcore. What should it be lower than?
"It's not who I am underneath, but what I do that defines me." - Batman Begins
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 28 Aug 2016 at 6:43pm
Depends, below 1.3 is best. Max 1.4 but the further below 1.4 the better but those would be on the high side.
Back to Top
SnowandSand View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 21 Jan 2013
Online Status: Offline
Posts: 278
  Quote SnowandSand Quote  Post ReplyReply bullet Posted: 29 Aug 2016 at 3:44am
I'm currently running an OC of 4.6 on my 6700k in my Hailstorm II w/ 2x 360mm Rads and after using Aida x264 to stability test it (I know DST4ME you prefer others Hahaha) over night I was only seeing max temps of 62c.  My 2x 980Tis never really see above 40c even when OC'd and gaming for long periods.  Needless to say this is my first computer with LC and my first adventure in to OC'ing.


Edited by SnowandSand - 29 Aug 2016 at 3:47am
Hailstorm II(Retired)
R7 2700x
16GB Team Dark Pro @ 3200 CL14
Asus x470 CH VII Hero
Corsair Hx1000i 1000W PSU
250GB Samsung 850 EVO
1TB & 6TB WD - Black Edition
2x GTX 980TI Strix
Hydrolux Stage 3
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 29 Aug 2016 at 4:03am
Well the reason I like others is because that one will never get your temp high to begin with. Its like testing our top speed on a very curvy road, its not gonna happen.

If one wants to test to see what their max temps are they have to put the part they want to test under max stress.

with your setup for example if you can't run the test I would do, then there is something wrong, which is what those test are there to reveal. They can test stability (by crunching numbers) and heat at the same time.

Did you request the 2 x 360mm rad?

Edited by DST4ME - 29 Aug 2016 at 4:05am
Back to Top
SnowandSand View Drop Down
Groupie
Groupie

Email address used to purchase matched with forums account email.

Joined: 21 Jan 2013
Online Status: Offline
Posts: 278
  Quote SnowandSand Quote  Post ReplyReply bullet Posted: 29 Aug 2016 at 5:24am
Originally posted by DST4ME

Well the reason I like others is because that one will never get your temp high to begin with. Its like testing our top speed on a very curvy road, its not gonna happen.

If one wants to test to see what their max temps are they have to put the part they want to test under max stress.

with your setup for example if you can't run the test I would do, then there is something wrong, which is what those test are there to reveal. They can test stability (by crunching numbers) and heat at the same time.

Did you request the 2 x 360mm rad?


I haven't run them simply because I haven't got around to downloading them lol. I do plan on doing it in the near future.  No, I didn't request the 2x 360mm rads it is what they put in with my hyrolux pro lvl 3.  I could fit a bigger one on the bottom if I needed to, but for the top I would have to move/re-move the dvd/rw.  For now it is doing the trick, my near term upgrades will be to add a 950 pro and a larger SSD for primary storage.  Do my temps seem high to you for that OC?  I think i set the V-core around 1.3/33 forget off the top of my head.
Hailstorm II(Retired)
R7 2700x
16GB Team Dark Pro @ 3200 CL14
Asus x470 CH VII Hero
Corsair Hx1000i 1000W PSU
250GB Samsung 850 EVO
1TB & 6TB WD - Black Edition
2x GTX 980TI Strix
Hydrolux Stage 3
Back to Top
DST4ME View Drop Down
DS ELITE
DS ELITE

Email address used to purchase matched with forums account email.

Joined: 14 Apr 2008
Online Status: Offline
Posts: 36758
  Quote DST4ME Quote  Post ReplyReply bullet Posted: 29 Aug 2016 at 5:35am
You want to be below 1.3v but personally I would be ok with 1.33v but as mentioned below 1.3 is what is preferred.

In order for to judge temps I have to use my methods, whatever your max temps are via my methods then I could say your temps are fine or not.

If you are going to run those apps make sure to disable adaptive voltage in th bios if its enabled, you don't want to run avx tests with the adaptive voltage on.

Max temps at or below 80c are in the acceptable range. aka no harm.

Edited by DST4ME - 29 Aug 2016 at 5:37am
Back to Top
 Post Reply Post Reply

Forum Jump Forum Permissions View Drop Down



This page was generated in 0.0703125 seconds.