2 Socket 2011 water blocks to test
AC Cuplex Kryos HF £67.70
EK Supreme HF Full Nickel EN (discontinued) (£57.99)
Aquacomputer Cuplex Kryos HF for 2011
The Cuplex Kryos is a revision of the older Kryos XT and it was originally designed to improve the XT’s poor fitting compatibility, allowing it to support larger, less restrictive tubing and fittings
With that in mind the HF (High Flow) moniker is a bit of a misnomer really, as the internal design of the block isn’t actually less restrictive
Packaging
Yes, that is how it arrived, a thin and slightly worn cardboard box that looks as if it has sat on someone’s shelf for too long
The block itself is well wrapped up in bubble wrap and in a vacuum packed plastic bag
The only fittings supplied (socket 2011) fit in a small plastic bag
A small packet of Prolima Tech’s excellent PK1 thermal compound is supplied as the paste of choice and a thicker o-ring should you wish to introduce a more pronounced bow to the block’s base
The mounting studs have a hexagonal shape grip at the base of the stud with the main body of the pole being round, the top thread on these poles are quite long and require a lot of turns for the spring nuts to fully engage
The block
The top is unpolished copper compared to the base which has a mirror finish which, despite the lack of focus in the photo, is really well polished
The base has a slight bow to it with the stock o-ring
Internals
As this is socket 2011 only I don’t have to open the block to change anything over, but that didn’t stop me
The base plate is a pin design that is the favoured design of most these days with rectangular micropins machined into the base
If you take a close look at the top you can clearly see the marks left by the machining
Reassembling the block was not the easiest thing to do as the mounting wings just sit sandwiched between the base and the top
With nothing to guide or hold them in place while you desperately try to get a screw in there without disturbing the o-rings, the darn things kept slipping out, so if you do have to change mounts it it’s not fun
EK Supreme HF Easy Mount Full Nickel EN 2011
Packaging
[FONT="]A sturdy inner box with an outer cardboard sheath with the block packed in a bag in its own compartment away from the fittings
[/FONT]
A tube of EK Gelid thermal compound was the paste of choice (not in the picture)
The EN in the title stands for Electroless Nickel which is the chosen method of applying the nickel plating
Although this is labelled as the socket 2011 model it also comes with all the fittings to enable you to mount this on most of today’s system from AMD and Intel
The mounting studs have a thumb screw shape at the base with the main pole being round
These are a bit shorter than the Kryos and the threads require less turns of the thumb nuts to fully engage but they do require a bit of pressure on the spring to engage in the initial stage
The block
Both the top and the base are fully polished to a mirror finish with no apparent bow to the base
You’ll have to forgive the damage to the base in these pictures, an accident involving a ginger cat and my photo bench, it really is a mirror finish
Internals
This base plate uses a micro channel design rather than pins with the inlet feeding into a small chamber and the water passing through an injection plate before striking the channels
There is no guide or holding mechanism holding the injection plate in place rather it relies upon the plate sitting in the correct orientation to align with a small ridge in the base plate to hold it in place
Test setup
The water testing system consists of
· 1 x Koolance INS-FM17N Flow meter with Koolance ADT-FM03 Frequency Adapter
· 4 x Phobya In-line water temperature sensors (1 radiator inlet - 1 radiator outlet - 1 CPU inlet - 1 CPU outlet)
· 2 x Air temperature sensors (1 air in to radiator – 1 air out of radiator)
· 1 x Aquacomputer Aquaero5 Pro system controller
My current test bench cooling system consists of
Loop 1 CPU only
· 1 x XSPC dual bay reservoir for Laing D5
· 1 x Alphacool VPP655/D5 Vario pump (Speed set at 3)
· 1 x Alphacool NeXxos XT60 Full Copper 420mm Radiator
· 6 x Phobya G-Silent 1100rpm 140mm fans (push pull)
· 13mm (½”) barbs
· Masterkleer 13mm tubing
· Worm drive jubilee clamps
Loop 2 - Motherboard
· 1 x Koolance MB-ASR4E Motherboard kit
· 1 x XSPC dual bay reservoir for Laing DDC pump
· 1 x Laing DDC-1T plus 18 Watt pump
· 1 x Alphacool NeXxos UT60 240mm radiator
· Barbs, tubing and clamps as above
Test setup
Asus Rampage IV Extreme motherboard
Intel i7 Q3960X EE processor
16GB G-Skill RipjawsZ DDR3 2133 CL9-11-10-28-1T system memory
Kingston HyperX 240GB SSD drive
XFX HD6770 1GB DDR5 GPU
[FONT="]Arctic cooling MX-4 thermal compound[/FONT]
Performance and testing
For the first test I will use 3 mounts with the CPU clocked at 4.6GHz - 1.45 vcore
Each mount will be allowed to idle for 1 hour to allow the CPU to reach an average temperature
Then 1 hour under load with Prime95 with the temps being logged every 10 minutes, the first 10 minutes will be ignored as the warm up period and the remainder will be used to calculate an average for the cores
The second test will have a further 3 mounts each with the CPU clocked at 4.8GHz - 1.48 vcore following the same procedure
The CPU is never removed from the socket and between each mount the TIM is thoroughly removed using 99% Isopropyl Alcohol and a tissue swab
Flow rate
Before moving on I must mention the Aquaero5 flow rate interface
Left at its default values the speed indicated is unrealistic
I can enter a calibration value manually, but without a user manual or guide other than what’s found on a few forums, I am still figuring out how to use this feature, so I used the same calibration value for both blocks
With that in mind the following values are only indicative of the flow rate of each block with the Alphacool D5 set to speed 3
The Kryos has a much lower flow rate compared to the Supreme but flow rate isn’t everything
Even though I try to keep my system running at a high flow rate a high restriction/low flow rate block can perform better than a low restriction/high flow rate block depending on the rest of the cooling system
Test run 1 at 4.6GHz - 1.45 vcore
System Temperatures
In order to gauge how well the rest of the system copes with the blocks I use a system delta temp which is derived from the ambient room air temp and the water temp at the radiator outlet
A good delta t averages below 10c and anywhere near 5c is better for a test system
Despite a 0.2c difference in room temp the Kryos is moving a bit more heat into the system as a whole but the delta t of 5.12c is still very good
Comparing the temps at CPU out and radiator out there is a notable difference in the amount of heat the systems are discarding
With the Kryos system the radiator is removing around 1.5c while the Supreme is only dropping around 0.22c
CPU Temperatures
A difference of around 2c in favour of the Kryos was consistent across all six cores on this test
The high flow rate of the Supreme doesn’t seem to remove as much heat from the CPU at these temps
Test run 2 at 4.8GHz - 1.48vcore
[FONT="]System temps[/FONT]
The Kryos continues to dump more heat into the system at these higher temps
Again at the radiator out the Kryos is removing around 1.5c while the Supreme is removing around 0.5c
CPU Temperatures
The Supreme pulls the difference back at these higher temps with a decreased difference of less than 1c across the cores, even so the Kryos still keeps the cores cooler
Removing the ambient shows a greater degree in favour of the Kryos again
Conclusion
The mounting mechanism on socket 2011 removes many of the bad points relating to mounting any cooling system
Without the need to install back plates and mounting bolts it’s a much easier task than it used to be and both blocks are equally simple to mount
That leaves only usability, performance and possibly aesthetics
Usability
Despite the HF branding on the Kryos indicating it has more space for higher flow fittings I couldn’t fit two of the in-line sensors on to the block itself so I had to add a 45 degree bend to the outlet in order to fit the second (only because this was all I had to hand)
I had no such problem with the Supreme but I also added one to this outlet as well just to keep the test systems equal
The offset design of the Supreme inlet which feeds into a chamber in the top before entering the block allows for a bit more space between fittings, the Kryos accepts compression fittings, it’s just a bit more cramped
Under normal conditions you wouldn’t disassemble the blocks but when it comes to cleaning your loop some time down the line you may want to
The individual wings of the Kryos block are a bit more work to hold in place when re-assembling while the EK system helps hold the outer o-ring in place
The addition of the striker plate in the Supreme does make re-assembling the block trickier because you have no guide other than the ridge in the base plate
Performance
Of the two the Kryos was the clear leader throughout testing
With a difference of 1.5 to 2c across all six cores overall on the first test there was no question
Core 3 behaved erratically during testing so I removed it from the equation on both tests
During test 2 the difference wasn’t quite so clear with some of the cores matching across both blocks
With the average and the delta in its favour the Kryos took the lead, but it wasn’t as clear cut as the first tests
It’s either because the Kryos is peaking or the higher flow rate of the Supreme starts to come in to its own at these temps, flow rate can make that much difference if the rest of your system can handle it
Aesthetics
Opinions, we all have our own and when comes to looks there is no bigger debate than what looks better than what
I like the copper coloured blocks more than the silvery mirror finish which just screams chrome to me and in that respect the Kryos is my favourite
On the board both blocks look equally impressive but I never did like pimping my vehicles up by adding chrome here and there so I am biased towards that warm copper effect of the Kryos and the Supreme will have to take a back seat on this
I did originally order the full copper version Supreme but it was unavailable when needed so after reading a few reviews I went for the full nickel version rather than wait a couple of months for the latest revision to arrive
Overall then I have to give it to the Kryos for presentation and performance
AC Cuplex Kryos HF £67.70
EK Supreme HF Full Nickel EN (discontinued) (£57.99)
Aquacomputer Cuplex Kryos HF for 2011
The Cuplex Kryos is a revision of the older Kryos XT and it was originally designed to improve the XT’s poor fitting compatibility, allowing it to support larger, less restrictive tubing and fittings
With that in mind the HF (High Flow) moniker is a bit of a misnomer really, as the internal design of the block isn’t actually less restrictive
Packaging
Yes, that is how it arrived, a thin and slightly worn cardboard box that looks as if it has sat on someone’s shelf for too long
The block itself is well wrapped up in bubble wrap and in a vacuum packed plastic bag
The only fittings supplied (socket 2011) fit in a small plastic bag
A small packet of Prolima Tech’s excellent PK1 thermal compound is supplied as the paste of choice and a thicker o-ring should you wish to introduce a more pronounced bow to the block’s base
The mounting studs have a hexagonal shape grip at the base of the stud with the main body of the pole being round, the top thread on these poles are quite long and require a lot of turns for the spring nuts to fully engage
The block
The top is unpolished copper compared to the base which has a mirror finish which, despite the lack of focus in the photo, is really well polished
The base has a slight bow to it with the stock o-ring
Internals
As this is socket 2011 only I don’t have to open the block to change anything over, but that didn’t stop me
The base plate is a pin design that is the favoured design of most these days with rectangular micropins machined into the base
If you take a close look at the top you can clearly see the marks left by the machining
Reassembling the block was not the easiest thing to do as the mounting wings just sit sandwiched between the base and the top
With nothing to guide or hold them in place while you desperately try to get a screw in there without disturbing the o-rings, the darn things kept slipping out, so if you do have to change mounts it it’s not fun
EK Supreme HF Easy Mount Full Nickel EN 2011
Packaging
[FONT="]A sturdy inner box with an outer cardboard sheath with the block packed in a bag in its own compartment away from the fittings
[/FONT]
A tube of EK Gelid thermal compound was the paste of choice (not in the picture)
The EN in the title stands for Electroless Nickel which is the chosen method of applying the nickel plating
Although this is labelled as the socket 2011 model it also comes with all the fittings to enable you to mount this on most of today’s system from AMD and Intel
The mounting studs have a thumb screw shape at the base with the main pole being round
These are a bit shorter than the Kryos and the threads require less turns of the thumb nuts to fully engage but they do require a bit of pressure on the spring to engage in the initial stage
The block
Both the top and the base are fully polished to a mirror finish with no apparent bow to the base
You’ll have to forgive the damage to the base in these pictures, an accident involving a ginger cat and my photo bench, it really is a mirror finish
Internals
This base plate uses a micro channel design rather than pins with the inlet feeding into a small chamber and the water passing through an injection plate before striking the channels
There is no guide or holding mechanism holding the injection plate in place rather it relies upon the plate sitting in the correct orientation to align with a small ridge in the base plate to hold it in place
Test setup
The water testing system consists of
· 1 x Koolance INS-FM17N Flow meter with Koolance ADT-FM03 Frequency Adapter
· 4 x Phobya In-line water temperature sensors (1 radiator inlet - 1 radiator outlet - 1 CPU inlet - 1 CPU outlet)
· 2 x Air temperature sensors (1 air in to radiator – 1 air out of radiator)
· 1 x Aquacomputer Aquaero5 Pro system controller
My current test bench cooling system consists of
Loop 1 CPU only
· 1 x XSPC dual bay reservoir for Laing D5
· 1 x Alphacool VPP655/D5 Vario pump (Speed set at 3)
· 1 x Alphacool NeXxos XT60 Full Copper 420mm Radiator
· 6 x Phobya G-Silent 1100rpm 140mm fans (push pull)
· 13mm (½”) barbs
· Masterkleer 13mm tubing
· Worm drive jubilee clamps
Loop 2 - Motherboard
· 1 x Koolance MB-ASR4E Motherboard kit
· 1 x XSPC dual bay reservoir for Laing DDC pump
· 1 x Laing DDC-1T plus 18 Watt pump
· 1 x Alphacool NeXxos UT60 240mm radiator
· Barbs, tubing and clamps as above
Test setup
Asus Rampage IV Extreme motherboard
Intel i7 Q3960X EE processor
16GB G-Skill RipjawsZ DDR3 2133 CL9-11-10-28-1T system memory
Kingston HyperX 240GB SSD drive
XFX HD6770 1GB DDR5 GPU
[FONT="]Arctic cooling MX-4 thermal compound[/FONT]
Performance and testing
For the first test I will use 3 mounts with the CPU clocked at 4.6GHz - 1.45 vcore
Each mount will be allowed to idle for 1 hour to allow the CPU to reach an average temperature
Then 1 hour under load with Prime95 with the temps being logged every 10 minutes, the first 10 minutes will be ignored as the warm up period and the remainder will be used to calculate an average for the cores
The second test will have a further 3 mounts each with the CPU clocked at 4.8GHz - 1.48 vcore following the same procedure
The CPU is never removed from the socket and between each mount the TIM is thoroughly removed using 99% Isopropyl Alcohol and a tissue swab
Flow rate
Before moving on I must mention the Aquaero5 flow rate interface
Left at its default values the speed indicated is unrealistic
I can enter a calibration value manually, but without a user manual or guide other than what’s found on a few forums, I am still figuring out how to use this feature, so I used the same calibration value for both blocks
With that in mind the following values are only indicative of the flow rate of each block with the Alphacool D5 set to speed 3
The Kryos has a much lower flow rate compared to the Supreme but flow rate isn’t everything
Even though I try to keep my system running at a high flow rate a high restriction/low flow rate block can perform better than a low restriction/high flow rate block depending on the rest of the cooling system
Test run 1 at 4.6GHz - 1.45 vcore
System Temperatures
In order to gauge how well the rest of the system copes with the blocks I use a system delta temp which is derived from the ambient room air temp and the water temp at the radiator outlet
A good delta t averages below 10c and anywhere near 5c is better for a test system
Despite a 0.2c difference in room temp the Kryos is moving a bit more heat into the system as a whole but the delta t of 5.12c is still very good
Comparing the temps at CPU out and radiator out there is a notable difference in the amount of heat the systems are discarding
With the Kryos system the radiator is removing around 1.5c while the Supreme is only dropping around 0.22c
CPU Temperatures
A difference of around 2c in favour of the Kryos was consistent across all six cores on this test
The high flow rate of the Supreme doesn’t seem to remove as much heat from the CPU at these temps
Test run 2 at 4.8GHz - 1.48vcore
[FONT="]System temps[/FONT]
The Kryos continues to dump more heat into the system at these higher temps
Again at the radiator out the Kryos is removing around 1.5c while the Supreme is removing around 0.5c
CPU Temperatures
The Supreme pulls the difference back at these higher temps with a decreased difference of less than 1c across the cores, even so the Kryos still keeps the cores cooler
Removing the ambient shows a greater degree in favour of the Kryos again
Conclusion
The mounting mechanism on socket 2011 removes many of the bad points relating to mounting any cooling system
Without the need to install back plates and mounting bolts it’s a much easier task than it used to be and both blocks are equally simple to mount
That leaves only usability, performance and possibly aesthetics
Usability
Despite the HF branding on the Kryos indicating it has more space for higher flow fittings I couldn’t fit two of the in-line sensors on to the block itself so I had to add a 45 degree bend to the outlet in order to fit the second (only because this was all I had to hand)
I had no such problem with the Supreme but I also added one to this outlet as well just to keep the test systems equal
The offset design of the Supreme inlet which feeds into a chamber in the top before entering the block allows for a bit more space between fittings, the Kryos accepts compression fittings, it’s just a bit more cramped
Under normal conditions you wouldn’t disassemble the blocks but when it comes to cleaning your loop some time down the line you may want to
The individual wings of the Kryos block are a bit more work to hold in place when re-assembling while the EK system helps hold the outer o-ring in place
The addition of the striker plate in the Supreme does make re-assembling the block trickier because you have no guide other than the ridge in the base plate
Performance
Of the two the Kryos was the clear leader throughout testing
With a difference of 1.5 to 2c across all six cores overall on the first test there was no question
Core 3 behaved erratically during testing so I removed it from the equation on both tests
During test 2 the difference wasn’t quite so clear with some of the cores matching across both blocks
With the average and the delta in its favour the Kryos took the lead, but it wasn’t as clear cut as the first tests
It’s either because the Kryos is peaking or the higher flow rate of the Supreme starts to come in to its own at these temps, flow rate can make that much difference if the rest of your system can handle it
Aesthetics
Opinions, we all have our own and when comes to looks there is no bigger debate than what looks better than what
I like the copper coloured blocks more than the silvery mirror finish which just screams chrome to me and in that respect the Kryos is my favourite
On the board both blocks look equally impressive but I never did like pimping my vehicles up by adding chrome here and there so I am biased towards that warm copper effect of the Kryos and the Supreme will have to take a back seat on this
I did originally order the full copper version Supreme but it was unavailable when needed so after reading a few reviews I went for the full nickel version rather than wait a couple of months for the latest revision to arrive
Overall then I have to give it to the Kryos for presentation and performance
Last edited:
