Do i need TWO D5 pumps?

[EarthDog]

Obviously I struck some kind of nerve with my comments?

Any increased overclocking voltage at all is accelerating degradation, which is an acceptable side effect to any of us that overclock, it truly depends on just how far you desire to go!

Why use voltage degradation as an excuse on a CPU that has not even been on the market long enough to even know for sure what it's daily run voltage limits even are when it relates to CPU overclock longevity?

My comments regarding lower DeltaT allowing higher overclocks seem to be past common acceptable knowledge here, which I am baffled by since this is Overclockers.com.

Water cooling with radiators is not simply a cookie cutter setup, unless you have zero overclocking intentions, but if you have overclocking intentions, DeltaT is a major factor relating to ambient water cooling.

DeltaT will be a determining factor on just how far you get overclocking, and DeltaT is relevant to the radiator cooling field.

Seems like the point that was missed here in the minutia is that you reach a point where you are uncomfortable or using too high of voltage. Ambiently cooled water just isn't enough. Typically, you run out of voltage headroom. So while deltaT is a consideration, there other factors you run into to FIRST on these chips. You are always welcome to disagree, but seeing where these chips top out under "acceptable" voltages is how I came up with my original post.

That's what say I.

(Sorry for the delay in response, I was moving over the past week)

Edit: let's bring the focus back on the op, and stop double posting while we are at it (combined a few).

 

[Silver Surfer]

Seems like the point that was missed here in the minutia is that you reach a point where you are uncomfortable or using too high of voltage. Ambiently cooled water just isn't enough. Typically, you run out of voltage headroom. So while deltaT is a consideration, there other factors you run into to FIRST.

That is only relative if you have actually tested down to a zero DeltaT, being uncomfortable or using too high of voltage, is the users choice.

And you will not reach a zero DeltaT with a single 480 radiator with the 6800K and GTX 1080.

So to me all these arguments are irrelevant.

I have actually tested down to zero DeltaT with my Chilled Water Cooling and seen for myself the limitations of ambient, but even those limitations can be exceeded if the individual is willing to go there, and the DeltaT that he or she is running is low enough to support it.

Now I could have written can be exceeded to a certain extent, but that's like you saying, "deltaT is a consideration", it is not the fact that it can be exceeded to a certain extent, it is that it can be exceeded, (which is my point).

Especially when it has been proposed running the CPU and GPU on the same loop, which apparently doesn't make any difference in independent CPU or GPU overclocking head room, according to whats being proposed in this thread.

 

[ATMINSIDE]

Degraded silicon says there is a limit not within the user's choice

 

[GTXJackBauer]

You're either a troll or a person with sub-ambient ego's which are fairly rare. They don't come around speaking nonsense in watercooling, especially coming from sub-ambient since those guys usually know what they're talking about but only leaves me to believe you're a forum troll with barely any posts and experiences. That's just my "ambient opinion" I got out of you as its pretty cool where I'm from. There are many people you disrespect here that have gone far and beyond you have ever gone.

I'm waiting for you to rewrite our whole water cooling sticky any moment now.

Peace be with you.

 

[Silver Surfer]

You're either a troll or a person with sub-ambient ego's which are fairly rare. They don't come around speaking nonsense in watercooling, especially coming from sub-ambient since those guys usually know what they're talking about but only leaves me to believe you're a forum troll with barely any posts and experiences. That's just my "ambient opinion" I got out of you as its pretty cool where I'm from. There are many people you disrespect here that have gone far and beyond you have ever gone.

I'm waiting for you to rewrite our whole water cooling sticky any moment now.

Peace be with you.

Post count obviously does not translate to intelligence!

I did not realize I was disrespecting anyone, that is definitely not my goal!

These people you say I have disrespected that have gone far and beyond I have ever gone, that is truly an assumption on your part.

What I am personally taking from this encounter is partial admissions to what I said without backing any of the claims to the contrary.

Admitting I have valid points but dismissing them.

When I first stated DeltaT would make a difference with overclocking head room, the response is DeltaT would make a difference, BUT.

You cannot honestly believe that between a 15c and 0c DeltaT there is no overclocking headroom, I don't care what CPU it is.

So my opinion is that the consensus comes from he said/she said, without actually testing it yourself.

Have you ever even run a zero deltaT water cooling?

And we all know that the exact same identical brand hardware would each overclock differently, Sam may reach 4.5ghz, but Bob next door cannot even reach 4.1ghz, with the same settings Sam used, when John across town is running 4.8ghz with the same settings.

Overclocking is not a cookie cutter fit it nicely in a box and sell it situation.

And Sir, I am not rewriting anything because of anything you say, now if I have broken any forum rules by challenging what's been said here, what rules have I broken to be dismissed as a troll.

Actually if there is any insulting going on here it is presently coming from you.

Especially starting out with your Picard Face Palm, and then the cat picture with the Dumbrella caption, what do you expect from me when that's what I see from you.

Then you sarcastically end your post with, Peace be with you.

You really should re-evaluate yourself so I'll leave you with.

Sweep around your own backdoor, before you worry about sweeping around someone elses.

 

[EarthDog]

That is only relative if you have actually tested down to a zero DeltaT, being uncomfortable or using too high of voltage, is the users choice.

And you will not reach a zero DeltaT with a single 480 radiator with the 6800K and GTX 1080.

So to me all these arguments are irrelevant.

I have actually tested down to zero DeltaT with my Chilled Water Cooling and seen for myself the limitations of ambient, but even those limitations can be exceeded if the individual is willing to go there, and the DeltaT that he or she is running is low enough to support it.

Now I could have written can be exceeded to a certain extent, but that's like you saying, "deltaT is a consideration", it is not the fact that it can be exceeded to a certain extent, it is that it can be exceeded, (which is my point).

Especially when it has been proposed running the CPU and GPU on the same loop, which apparently doesn't make any difference in independent CPU or GPU overclocking head room, according to whats being proposed in this thread.

Sure... if someone wants to push past what most would consider 24/7 voltages (which is the antithesis of what I was saying in the post you quoted and disagreed with me), of course you are correct! Not sure anyone said otherwise? Clearly the discussion here is with 24/7 type voltages and without BIOS modding the GPU for additional voltage and bypassing power limits. THen you jumped in thinking it is balls to the wall and 0 deltaT, etc etc...

It may have been "4.5 years" since you were satisfied with 4.5 GHz and that is great. That but a 6800K isn't a 3770K either (what ATM said earlier). Where the limits of sanity and the silicon meet with Ivybridge are different than with Broadwell-E. Hence why we are saying around 4.5 GHz. Voltage really starts to climb around the 4.2-4.3 GHz range is what we are seeing so far.

I have to say though, I'm making FG's on the goal post I setup, not the one you are talking about on a different field, which you are also making.

 

[LawShadow]

"Regarding the MO-RA3 I honestly do not remember if it was actually soldered or not I was flushing many rads during that time period as I gave away all my smaller rads so whether it is soldered or not I do not know, if it is O-ring sealed tubes that may account for it's extra flow restriction it was labeled as having.

Sealed or soldered, the only thing I know for sure regarding the Watercool MO-RA3 is how well it performs, it is some of the best water cooling equipment I've spent money on yet!

I do know when I flushed it, almost nothing debri wise came out of it, of that I was most impressed.

I use D5 pumps."

(something went wrong with quoting")

Thanks for the feedback Silver Surfer,


Good news with the MO-RA3 coming out clean after flushing and from the looks of it it performs really well. Hopefully my AMS 840 will perform and come out clean just as well. Speaking of, it arrived today after 7 days of waiting. The thing weighs a metric ton but build quality is top notch also the packaging was really well done 2 boxes fully protected. Once i have my new rig which will take about a month from now. I will do a "Ghetto review" on the Aquacomputer AMS 840 copper including temps on both cpu and gpu idle and full/max load on stock as well as overclocked plus flow chart ie gpm for anyone who might contemplate getting this radiator. I had a hard time finding people using this rad.

 

[Silver Surfer]

Hopefully my AMS 840 will perform and come out clean just as well. Speaking of, it arrived today after 7 days of waiting. The thing weighs a metric ton but build quality is top notch also the packaging was really well done 2 boxes fully protected. Once i have my new rig which will take about a month from now. I will do a "Ghetto review" on the Aquacomputer AMS 840 copper including temps on both cpu and gpu idle and full/max load on stock as well as overclocked plus flow chart ie gpm for anyone who might contemplate getting this radiator. I had a hard time finding people using this rad.

Looking forward to seeing what that bad boy can do, that is one awesome radiator!

 

[Dlaw]

Silver Surfer, look at it this way:

Your sub-ambient system cost you a crazy amount of money, I'll wager. Like, most people wouldn't spend that kind of money on the entire computer, let alone just the cooling system, crazy amount of money. My custom loop cost more than $1500, which is more than most people consider paying for their entire computer, so I can imagine what that cost.

I'm bringing this up for one reason: to get a 0 delta, you will need to spend way more money than just what that 480 would cost. Then, consider that you very well may only get 100mhz, if anything, after spending all of that money on the additional cooling.

You need to look at this from the perspective of someone who is asking whether or not they need two D5 pumps for their system, not from someone who set up an elaborate sub-ambient water cooled beast. Most people don't need a 0 dT, so they don't need to spend extra money on getting it. Keeping things in the realm of "this will work better than any off-the-shelf cooler, look better and be worth the extra money" while staying out of "completely overkill for anyone not trying to competitively (professionally) bench or something similar" is something we try to do here, particularly for people without the extensive experience that someone like you has.

For these reasons, the single 480 recommended by some of our most senior members was indeed the correct recommendation. There's no real difference between 4.4 and 4.5GHz in day to day usage, which very well could be the only increase the OP would see in adding more rad. Therefor, there's no reason to spend more money on more rad.

 

[Silver Surfer]

Silver Surfer, look at it this way:

Your sub-ambient system cost you a crazy amount of money, I'll wager. Like, most people wouldn't spend that kind of money on the entire computer, let alone just the cooling system, crazy amount of money. My custom loop cost more than $1500, which is more than most people consider paying for their entire computer, so I can imagine what that cost.

I'm bringing this up for one reason: to get a 0 delta, you will need to spend way more money than just what that 480 would cost. Then, consider that you very well may only get 100mhz, if anything, after spending all of that money on the additional cooling.

You need to look at this from the perspective of someone who is asking whether or not they need two D5 pumps for their system, not from someone who set up an elaborate sub-ambient water cooled beast. Most people don't need a 0 dT, so they don't need to spend extra money on getting it. Keeping things in the realm of "this will work better than any off-the-shelf cooler, look better and be worth the extra money" while staying out of "completely overkill for anyone not trying to competitively (professionally) bench or something similar" is something we try to do here, particularly for people without the extensive experience that someone like you has.

For these reasons, the single 480 recommended by some of our most senior members was indeed the correct recommendation. There's no real difference between 4.4 and 4.5GHz in day to day usage, which very well could be the only increase the OP would see in adding more rad. Therefor, there's no reason to spend more money on more rad.

That's a very reasonable, calm, uninsulting, explanation. Thank You! SS

 

[EarthDog]

That makes two of them...

 

[Lochekey]

As the others have said you will hit a point on these newer chips where a lower delta t gains you nothing.

For an example my one setup originally consisted of a it 4690k cooled with an alphacool xt45 360 rad. At a 4.5ghz overclock running 1.3 volts I was able to maintain a delta t of around 10 degrees, mainly due to my fan profile.

I have recently switched this setup to a phobya 1260 rad and can now achieve a delta t of less than 2 degrees(I don't trust the accuracy of my devices to give a more solid delta t reading).

Since this change my reported core temp is about 8-9 degrees less than before. Even with this lower core temp I have not been able to raise the multiplier to 46x without pushing 1.375 volts and I still need 1.3 volts for a stable 45x multiplier stop I really did not gain anything.


Even when I am benching at 1.6volts+, I am unable to do better on this new rad than I did before.

So my conclusion is that, as the others have said, you hit a wall on ambient cooling with the new silicone that cannot be overcome just by lower temps by 5-10 degrees. So while I'm all for getting a little extra rad to make sure you leave some head room for overclocking there is a point where there is no return on your investment and chasing that delta t of less than 10 degrees is only worth it if you are benching or you want the headroom for an ultra quiet setup.

 

[Silver Surfer]

As the others have said you will hit a point on these newer chips where a lower delta t gains you nothing.

For an example my one setup originally consisted of a it 4690k cooled with an alphacool xt45 360 rad. At a 4.5ghz overclock running 1.3 volts I was able to maintain a delta t of around 10 degrees, mainly due to my fan profile.

I have recently switched this setup to a phobya 1260 rad and can now achieve a delta t of less than 2 degrees(I don't trust the accuracy of my devices to give a more solid delta t reading).

Since this change my reported core temp is about 8-9 degrees less than before. Even with this lower core temp I have not been able to raise the multiplier to 46x without pushing 1.375 volts and I still need 1.3 volts for a stable 45x multiplier stop I really did not gain anything.


Even when I am benching at 1.6volts+, I am unable to do better on this new rad than I did before.

So my conclusion is that, as the others have said, you hit a wall on ambient cooling with the new silicone that cannot be overcome just by lower temps by 5-10 degrees. So while I'm all for getting a little extra rad to make sure you leave some head room for overclocking there is a point where there is no return on your investment and chasing that delta t of less than 10 degrees is only worth it if you are benching or you want the headroom for an ultra quiet setup.

That Phobya 1260 is a nice rad, I almost bought that instead of my Watercool MO-RA3, but at the time the MO-RA3 allowed running 180mm cooling fans and the Phobya did not.

You say with your 4690K you were at 4.5ghz @ 1.3v, 4.6ghz at 1.375v, then you mention benching at 1.6 volts +, what clock were you running then?

I know there is a limit to what ambient cooling can do, running chilled water cooling my 3770K easily ran 5ghz at 1.375v but that was running coolant temperature 14c below ambient.

Running ambient cooling alone the 3770K can run at 4.9ghz but cannot reach 5ghz, with temperatures I am comfortable with running everyday.

The entire past points I have been attempting to make is that higher is possible, but it is up to the individual as to whether they want to run like that or not.

What is caution to you or I may not be to someone else, and apparently your caution is way less than mine as I would not throw 1.6v+ to my 3770K?

 

[EarthDog]

Higher is possible...not a soul denied that.

There is a difference between benching and 24/7 speeds/voltages. As Lochey eloquently explained, for an average user, getting down to a 0dT really doesn't help out much for 24/7. I mean, what do you have on a 3770k at 5ghz that you don't have at 4.9ghz? The answer, 100 mhz. This isn't the difference between playable and unplayable or upping settings in a game. So why spend the extra cash on a goal that truly helps only those in benchmarking? I'd rather take the difference and buy a better gpu or m.2 ssd or something that is a lot more tangible than lower temps and a potential 100 mhz increase (assuming voltages allow).

To each their own.

 

[EarthDog]

Thread cleaned up... a bit tired of the needling on both sides.

 

[LawShadow]

Hello all,

I finally finished my new rig and i wanted to give an update.

Me being overly worried about the flow rate with my setup:

Aquacomputer AMS 840 full copper rad and 2 water blocks.

I tested the rig:



I got 1.26 GPM with the D5 which was higher then expected with this beast of a radiator.

Here is a pic of the finished build:



The rad is heavy so it is also supported underneath the rad the case does not carry that monster by it self if some might wonder just in case i reinforced the backside of the case with a steel rad grill.

Now onto the temps and i am very pleased

Today i checked my temps again with HWIinfo64

Intel I7 6800K

Idle cpu 22c to 23c
Idle gpu 25c

Gaming session 2 hours of Crysis 3

cpu 25c to 28c
gpu 28c to 31c

Its Autumn now so maybe that helps also with these temps. I live in a old house. The first time i started my new rig temps where even lower but that has to do with my window open and no heater lol had 17c idle on cpu it was a bit chilly inside after cooking i always open my windows for fresh air and forgot to close them. But the above temps are my normal room temp with the gas heater opened for approx 30 min to warm my room up.
Before this setup was using a Cooler Master Seidon 120V and on my EVGA GTX 1070 sc just stock cooler

temps where while gaming Crysis 3
cpu 45c to 49c
gpu 67c to 71c

So massive improvement in my book. This radiator has top notch build quality and lots of heat dissipation even at 600rpm fans on push/ pull config!

Specs pc:

Intel I7 6800K cpu
Evga X99 FTW-K mb
Evga 1070 SC gpu
Evga Supernova G2 850 psu
Intel SSD 250 GB
Case: Lian Li Tyr x500

Specs watercooling parts:

Koolance CPU-380I v2 waterblock
Koolance koolance vid-nx1080 acrylic version waterblock
Koolance INS-FM18D Coolant Flow Meter
Koolance Black compression fittings (1/2 - 3/4)
Koolance Liq 702 (clear) coolant
Alphacool d5-vpp655 pump
EK-D5 X-RES CSQ + 200mm reservoir
Aquacomputer Airplex Modularity System 840 (6x140) full copper radiator
Some Primochill Advanced LTR tubing (1/2 - 3/4)


I want to test the cpu 100% but i heard that prime95 can damage the cpu specially broadwell do you guys know a good test/benchmark for my broadwell cpu without the risk of damaging it i just want to see the temp at 100% load?

I want to thank all people who contributed and helping me.

 

[gkline]

Hello all,

I finally finished my new rig and i wanted to give an update.

Me being overly worried about the flow rate with my setup:

Aquacomputer AMS 840 full copper rad and 2 water blocks.

I tested the rig:

View attachment 184182

I got 1.26 GPM with the D5 which was higher then expected with this beast of a radiator.

Here is a pic of the finished build:

View attachment 184183

The rad is heavy so it is also supported underneath the rad the case does not carry that monster by it self if some might wonder just in case i reinforced the backside of the case with a steel rad grill.

Now onto the temps and i am very pleased

Today i checked my temps again with HWIinfo64

Intel I7 6800K

Idle cpu 22c to 23c
Idle gpu 25c

Gaming session 2 hours of Crysis 3

cpu 25c to 28c
gpu 28c to 31c

Its Autumn now so maybe that helps also with these temps. I live in a old house. The first time i started my new rig temps where even lower but that has to do with my window open and no heater lol had 17c idle on cpu it was a bit chilly inside after cooking i always open my windows for fresh air and forgot to close them. But the above temps are my normal room temp with the gas heater opened for approx 30 min to warm my room up.
Before this setup was using a Cooler Master Seidon 120V and on my EVGA GTX 1070 sc just stock cooler

temps where while gaming Crysis 3
cpu 45c to 49c
gpu 67c to 71c

So massive improvement in my book. This radiator has top notch build quality and lots of heat dissipation even at 600rpm fans on push/ pull config!

Specs pc:

Intel I7 6800K cpu
Evga X99 FTW-K mb
Evga 1070 SC gpu
Evga Supernova G2 850 psu
Intel SSD 250 GB
Case: Lian Li Tyr x500

Specs watercooling parts:

Koolance CPU-380I v2 waterblock
Koolance koolance vid-nx1080 acrylic version waterblock
Koolance INS-FM18D Coolant Flow Meter
Koolance Black compression fittings (1/2 - 3/4)
Koolance Liq 702 (clear) coolant
Alphacool d5-vpp655 pump
EK-D5 X-RES CSQ + 200mm reservoir
Aquacomputer Airplex Modularity System 840 (6x140) full copper radiator
Some Primochill Advanced LTR tubing (1/2 - 3/4)


I want to test the cpu 100% but i heard that prime95 can damage the cpu specially broadwell do you guys know a good test/benchmark for my broadwell cpu without the risk of damaging it i just want to see the temp at 100% load?

I want to thank all people who contributed and helping me.

Beautiful rig, LawShadow. The D5 IS a strong pump. I am running a single D5 for my MO-RA3-420 pro and it works fine. Great temps by the way.

 

[LawShadow]

@ gkline,

Thanks for the headsup it was my hardest water cooling rig ive done and ive done 4 previous custom water cooling builds.
Indeed the D5 is a great pump and very silent even at setting 5! Coming from a Aquastream XT pump which i love btw they are pretty silent but i have to be honest the D5 is a bit quieter. Temps are great indeed was everything i hoped for.

That MO-RA3 420 pro heard allot of great things about cools like a champ! And pretty much the same internal design as the AMS rads, so they are flux less.
BTW Watercool are planning for a major overhaul for their radiators curious how this would pan out.