Introduction

Hardware Labs Black Ice® GTX 360mm Radiator Review

The intention of this write-up is to provide information and performance data about the HWLabs GTX 360mm Radiator for inclusion in the Extreme Rigs Rad Round Up 2015. I’ll be keeping the review section short and factual, focusing on presenting the performance data compiled through the numerous tests carried out.

Newer HWLabs models are available, but the original GTX remains available through retailers.

The GTX 360 sample was pulled from a working system and thoroughly cleaned prior to any testing was commenced. The dent on the tank corner does not affect the GTX’s performance and we are confident that it is representative of a new/boxed retail unit’s cooling potential.

hwl-logoHWLabs GTX - logo

Note: You may notice changes in the data presentation in this rad review as I’m now also including even more thermal testing data for both Push/Pull and Push Only fan installations.

What’s in the Box?

My used test/review sample no longer has any of its packaging available, or any of the original accessories. I grabbed this box shot off the web.

The next pic is not of the GTX, but it is the standard HWLabs inner box layout.
I really like the HWLabs style of boxing as the compartment at one end keeps the provided accessories away from the core. I believe that the GTX is now shipped with black screws.

This picture courtesy of Martin gives an indication of what to expect, now with black M4 screws.

Extent of delivery:
Included in the GTX package is the following
1 x GTX 360 Radiator
12 x M4 x 28mm screws for mounting fans.

I’m sorry I can’t show more on the unboxing – but I just don’t have any of it to be able to show you!

Onwards to technical specifications!

Technical Specifications

 

 

 

Technical specifications as listed by HWLabs:

• Two-pass pre-cool/aftercool front-rear dual-core flow configuration with a 30% increase in internal flow rates.
• Custom MaxFin™ 25 micron Copper Splitter Fin Configuration utilizing 45% thinner fin material yielding up to 50% less pressure drop.
• Unique fin configuration eliminates intra-louver accumulation of dust particles for trouble free operation.
• Radically optimized 20 FPI (Fins Per Inch) fin density providing dramatically increased heat transfer surface area.
• Custom low-profile 2-row 19.0 x 1.2mm MaxFlow™ tubes with twice the waterside surface area of the GTS and 60% frontal area reduction for superior low air-resistance aerodynamics and lower internal flow requirements.
• Yields up to 45% more heat exchange capacity than the Black Ice® Xtreme in both Stealth or Performance modes.
• Achieves Black Ice® Xtreme level performance in stealth mode (low-noise/low-airflow conditions).
• M4 Threaded Screw Holes for easier mounting and greater adaptability.
• Now standard G 1/4″ female threaded fittings.
• Full high-temper brass structural construction for weight reduction and superior corrosion resistance.
• High performance compact radiator compatible as an upgrade to the Black Ice® Xtreme and most 120mm form factor radiators.
• Uses 100% Non-corrosive water-solluble fluxing process.
• Fully RoHS compliant.
• Full electrostatic polyurethane painting finish for uniform coating with high temperature curing for increased finish durability.
• Renowned Black Ice® quality.
• Patent Pending Design.

Tech Drawing of the GTX 360.

Dimensions Measured on the Radiator Tested:

Radiator Core Dimensions:

The core is made up of 2 rows of 12 tubes arranged in a dual pass front to back configuration which I call “Opti-Flow”. Opti-Flow needs some explanation so we’ll examine that shortly. The fin arrangement is made of split fins with a 20 FPI count. The high fin count should equate to some great high speed fan results, but might struggle with low fan speeds.

The GTX is a dual pass radiator, but not in the tradition sense where the coolant flow path is side to side such as in this reference picture below. This is known as U-Flow because the coolant travels up all the tubes on one side (left in pic) of the rad and then returns down the other side. U-Flow is most easily recognized when the port end has 2 separate tanks:

cool8

Instead the GTX flow path is front to back OR back to front depending on which port is used as the inlet. In other words the coolant travels up the tubes the full width of the radiator core then changes direction in the non-port tank and travels down the other set of tubes on the opposite face of the core. To get this flow path to work HWLabs have installed a plate and tube in the port tank ends.

HWLabs recommends that the “hot face” (coolant in) be the first pass and act as a pre-treatment for the second pass, through the “cool face” tubes (coolant out).

I have done a small amount of testing on this myself, and at higher fans speeds the Opti-Flow performed better. As the GTX certainly appears tuned for high fan speeds, that is how I will run the thermal testing – set up for Opti-Flow.

The split fins are not louvered and are spaced at 20 FPI. The fin spacing is quite uneven between the tubes, but this is to be expected considering it is such a high FPI count.

HWL - GTX - Lr Edited with B&W (4)

Finish and Features

The GTX has a true Gloss Black finish and while not my preferred finish for a radiator it does look stunning. Even after years of service (and abuse) this sample still looks great.

HWLabs is renowned for producing quality radiators that have long been considered as the industry standard. The GTX is one of the rads that helped build this reputation. The GTX feels solid and weighty. This GTX being tested has even been dropped (accidentally) and though dented, remains functional.


The fan mount spacing is the standard 15mm.
None of the mounting holes are located directly above tubes; however they are very, very close and no protection plates are fitted. Care must be taken to ensure the correct length screws are used to suit your mounting needs.

The GTX comes equipped with just 2 ports in total, 1 inlet and 1 outlet to choose from. Depending on installation orientation and fan assembly the inlet/outlet port might be different to achieve Opti-Flow .

While we love multi-port rads, we understand that they are difficult (though possible) to incorporate into front to back (Opti-Flow) rads.

Unfortunately this also means there is no dedicated fill/bleeder port on the return end tank.

So we have somewhat of a mixed bag of features with the Black Ice GTX 360

It has a thick core with a high FPI count that appears tuned toward high speed fans. The gloss black finish is excellent as is the overall build quality but it is lacking in features.

 

Let’s see how it performs…

Flow Rate Testing

The Data

As all the testing in this round-up was performed with the exact same equipment, using the exact same methods I have decided to keep each radiators page uncluttered by posting our testing methodology, test set-ups and equipment used in a single location. To see exactly how the tests were carried out, details of the test set ups and equipment used, please head back to the RRU Test Setup page.

Restriction Test

It’s generally agreed that radiators are one of, if not the least restrictive components in the water cooling loop. There are some exceptions however, so this must still be verified through testing:

cool13

The above photo is for referencing the restriction test bench – the GTX is not loaded so please disregard the data in the picture as it does not relate to the its test results.

Here is the raw data at the tested flow rates, displaying the measured Differential Pressure across the radiator as flow rate was increased.
The table numbers indicate that the GTX is a high restriction radiator. However numbers in isolation can only tell half the story. By plotting against other components it more easily shows the whole story.

I have decided to use a HeatKiller 3.0 CPU block as the reference in these plots for two reasons. Firstly there is no chance of the plot being cluttered by curves overlapping and secondly it gives a reference point against a fairly common loop component of average restriction.

As with all the radiator restriction plots I have limited the maximum flow rate displayed to 2.0 GPM as I suspect there are very few systems that operate above 2.0 GPM. For more information on how to read a restriction plot check out our guide.

This plot confirms the GTX as a medium restriction component.

The next three plots show the GTX’s restriction level relative to other radiators in the test group. Previously I have only shown just the 1.0 gpm plot, but I have now decided to show 3 different flow rates for you to see how the rad’s restriction levels compare against each other at different flow rates.

GTX - Restriction 0.5

So when compared to other 360mm radiators we have to call the GTX a high restriction rad. Therefore some planning of pump power required for your loop components should be considered.

 

Onwards to Thermal Performance!

Thermal Testing

The Thermal Data

Moving on from the restriction test bench the HWLabs Black Ice GTX was loaded into the thermal test chamber for a series of tests – consisting of 3 flow rates, each having 3 different fan rpm rates tested. I felt the thermal test data was most important and which you as the reader would be most interested in.

Below is the final data results gathered from at least 2 data logging runs at each flow rate and fan rpm combination. The most stable 15 minute period from each logging run was used and then averaged with the other runs to obtain the data for the table below. A total of 16 temperature sensors were used in the thermal test chamber (8 air in, 2 air out, 3 water in, 3 water out) each take a reading every second and logged via a CrystalFontz unit. The data in the table below is the averaged results of the logging runs which has then been used to create all the plots and tables there-after.

The performance metric of critical importance is the delta between the warm coolant temperature in and the cold ambient air temperature in to the radiator. Given that the system is well insulated and in equilibrium and we know the heat input to the system then we can also calculate a very important number – that is the amount of power required to raise the coolant temperature 1C (or 10C which is more useful reference point).

Let’s take a look at the Delta T results from the tests, in bar chart form first:

I was not too concerned about the actual delta numbers, more so the trend pattern. Most noticeable here is the lack of variation between the flow rates at each fan speed in Push/Pull.

It seems a good a spot as any to mention that the OptiFlow testing on all the rads that could benefit from it produced some results that were not as expected. The Black ice GTX was no exception and despite numerous retesting the results left us bewildered at times at some of the data which was produced. Some of the discrepancies we can attribute to test error margins – that being uncontrollable fluctuations of the test rig such as power fluctuating which can effect fan speeds, flow rate and heater power. This despite having a 1000watt power conditioner between the wall and the PSU.

What I am saying is that despite our best efforts some of OptiFlow Vs. Reverse Flow comparisons never made sense to us. I wish I had a proper explanation for some of these results, but at this stage unfortunately I do not. So I am going to call it as the data presents and try to ignore the abnormalities that may exist.

Delta T results (as above) are not always helpful when thinking about how many radiators you would need to cool your system. Instead it’s more useful to know the delta/W, or more usefully, the inverse metric of W/delta C. The metric plotted below tells us how many watts are dissipated by the radiator when the coolant rises 10C above ambient temperatures. (W/10 Delta T):

As expected increasing airflow through the radiator is the primary determinant in changing the radiators performance.

This data can now be plotted as a pretty curve so that an end user can interpolate their own fan speed. Note again that the extrapolation of the curve is much more sensitive to error than in between the tested range. Let’s exclude the push data for now and come back to it:

Evidence again that increasing flow rate is yielding very little benefit in performance as the fan speed is increased. These straight lines indicate some very good Push/Pull comparison results coming up for the GTX. Due to the lack of curvature (drop off) at higher fan speeds it also indicates that the GTX’s performance would benefit from even higher fan speeds (if noise was not a concern)

If we now come back to the Push Only data it’s good to compare only the push to push/pull in an “apples to apples” fashion by only looking at the 1GPM data:

It’s an unusual looking Push Only curve due to the GTX having such poor performance at low fan speeds. Push only gives around 72.5% (averaged) of the performance of the Push/Pull 1.0 gpm test results. This was by far the lowest average of the test group and such big variance shows that Push Only fan assemblies do not offer great performance on the GTX.

In this next plot all three Push/Pull results at each flow rate were averaged to produce one curve and we compared the results against one of the best and worst performers from the first test group. Averaging reduces test error of course so this helps further to be sure of our data and is more useful therefore for comparing to other radiators.

As you can see the GTX is ahead of the Koolance, which has a similar core design and is also targeted for high speed fan performance.

Now let’s analyze that data some more…

Data Analysis

This first table shows the GTX’s Watts/10 Delta Temp numbers in a quick glance chart format.

Using the data compiled from the GTX thermal testing I have compiled the following tables in an attempt to show other ways of how its performance varies against itself at the flow rates and fan speeds tested. Effectively these show percentage gains relative to a reference point. It’s an interesting way to show gains/losses while changing a variable.

First let’s focus on 1300 RPM as our reference and see how much gain or loss in performance we get by changing fan speed.

The high percentage changes between 1300 rpm to 750 rpm indicate a significant performance drop when the fan speed is reduced. Also we notice huge changes in the Push Only results on either side of the 1300 rpm reference.

In the next plot we focus on changing the flow rate with 1.0 GPM as the base line.

With only slight percentage changes with adjustments to the flow rate could mean that flow rate is not a big determining factor for performance on the GTX.

How about we combine both flow rate and fan speed as reference points and have a look at 0.5 GPM & 750 RPM as the reference point.

Massive performance boost is seen with each increase of the fan speed.

Lastly, we change the reference and choose our center 1300 RPM and 1.0 GPM as our reference point to show both effects concurrently:

So from the data above we can get a very good idea of how the GTX radiator performs relative to itself. But there is a large selection of 360mm radiator models to choose from, released from numerous manufacturers. So, we need to start comparing performance between them. To see how the GTX performed against the other radiators tested, I have included the averaged flow rate comparison charts from the Round Up. We know that the flow rate has little impact on thermal performance so averaging of the 3 flow rate results gives us a good look at head to head performance at the rpm speeds tested at with even less error.

Push Only Data vs Competition

Let’s focus on the Push Only results for now and come back to the Push/Pull data later. While for Push/Pull we tested three flow rates, the Push Only testing was conducted only at 1GPM in order to save time.

Let’s start with 750 RPM and see how it compares to the competition:

Dead last – with the lowest score of any data point of the entire group. There is no surprise here though. It’s thick dense core is not designed for this fan set-up, it fact being so bad could be a promising sign that it will perform really well at the other end of the scale (high speed Push/Pull)

How does the GTX do at Push Only 1300 rpm?

Some improvement, moving up the ranking, but is still 15% behind the winner.

OK then, with the high fan speed the GTX is starting to breath, after having been choked in the previous results. In 6th place it is now just 1.5% behind the winner.

So it appears (as expected given the high FPI count) that the GTX is well tuned for high fan speeds. Will the GTX repeat the rising performance trend as fan speeds/air pressure are increased in Push/Pull?

Let’s find out…

Push/Pull Data vs. Competition

Let’s now look at the Push/Pull results and see how the GTX compares. For Push/Pull we have averaged the results from the 3 flow rates at which we tested.

The GTX finished in 17th position in the Push/Pull low fan speed category, ~9% behind the winning rad. Certainly not as bad as the Push Only so this is a promising start.

Let’s move to the 1300 rpm then:

In the 1300 rpm category the GTX jumped up the leader table to finish in 2nd, just ~0.5% behind the category winner. The 0.5% is within our test error margin so effectively placing the GTX equal 1st. The GTX placed second in each of the 3 flow rates making up the averaged 1300 rpm performance results.

The averaged 1850 rpm results…

At 1850 Push/Pull the GTX jumps up for the category win, clear of it’s nearest competitor by ~3%. The GTX took out 1st place in each of the 3 flow rates contribute to the averaged 1850 rpm result.  Clearly this is a radiator that still delivers when airflow is high enough!

Let’s also combine the Push and Push/Pull results at the 1.0 GPM flow rate into one plot for each fan speed tested. This can be interesting to help decide if if more fans or a different radiator is the better option if looking to improve performance.

Again the 750 rpm first:

Nothing looks out of place here given the data we have reviewed already. Interestingly the GTX’s Push/Pull result is beaten by the top performing Push Only rad – the HWLabs SR2 which is optimized for lower fan speeds.

At 1300 rpm next:

At 1300 rpm everything looks in order given the data we’ve already reviewed.

From all of test results we created “Average Performance Factor” charts for both Push and Push/Pull and then a combined plot called the “Master Performance Factor”. The radiator with the best cooling ability (W/10ΔT) at each gpm/rpm combo was awarded a score of 100, and each other radiators W/10ΔT result was scored as percentage of the top performer.

This way of looking at the comparison takes away any advantages that a radiator may have at higher or lower fan speeds and looks at an overall average. While this appears fair it does tend to favor those radiators that are all-rounders and those radiators which do very well at high RPM. Most users should be more focused on their specific use case. Check in the Round Up for performance comparisons at every gpm/rpm combo for even more details and cross comparison results.

Here is the Black Ice GTX’s percentage scores at each data point that thermal tests were conducted at with the addition of the GTX’s Push Only results relative to the Best performing Push/Pull results

The Push/Pull medium and high fan speed results are impressive to say the least. It shows just how dominant the GTX was in then high speed fan tests.

Then we started averaging the percentage scores at each flow rate giving us the Averaged Performance Factors of each radiator.

Firstly the Push Only APF:

GTX - APF P
The GTX’s Push Only APF score of 80.3 is 3rd last of the test group. The High Fin count is obviously hindering performance with the low pressure provided by the Push Only fan assembly.

Now the Push/Pull APF:

The GTX’s Push/Pull APF of 96.5 places it in 2nd position. The weaker 750 rpm results held back the GTX’s result as in the medium and high speed tests the GTX was dominant with 3 2nd places and three 1st places.

Finally we created the Master Performance Factor which is calculated from the averaged results of all the thermal tests, Push and Push/Pull at all flow rates and all fan speeds.

GTX - APF All b

In the Master Performance Factor comparisons the GTX finishes in equal 16th place. The poor results with low fan speeds held back it’s overall score though dominating with medium and high Push/Pull fans.

These next couple of plots show quite nicely just how biased the GTX is towards high fan speeds or Push/Pull set ups with more air flow through the rad.

First the GTX’s Bias for Push Only…. it is almost laughable at the result that was generated.  This is not a radiator designed to be run with push only fans!

HWLabs GTX - Bias PPI really didn’t expect the GTX to perform well across the board. I suspected the dense core structure would equate weak performance in the Push Only tests, which the test data confirmed. Likewise The Push/Pull results didn’t surprise either, with a clean sweep of the high speed fan Push/Pull tests.

Next up – Summary!

Summary

Thermal Performance

Thermal Performance scores are derived from the relevant Performance Factor scores. We set this scale with 75% and below as the 0 mark, with each 2.5% increase from 75% in relative performance adding 0.5 to the awarded performance score.

Push Only Thermal Performance – 1.5/5

The GTX’s Push Only APF result of 80.3 placed it in 3rd last. It placed 5th in the 1850 rpm test but the medium and low fan speed results which were very weak drastically impacted it’s Push Only ranking and APF score.

Push/Pull Thermal Performance – 4.5/5

As we have seen in the data analysis the GTX performed very well in medium and high speed Push/Pull tests, while not so good at the low fan speed. The excellent 1300 rpm (3 x 2nd places) and 1850 rpm (3 x 1st places) kept the averages high and the GTX finished in 2nd place for Push/Pull. The GTX’s P/P APF score of 96.5 resulted in a P/P performance score of 4.5/5.

Overall Thermal Performance – 3/5

The Master Performance score of 88.4 gives a score of 3.0/5 and places the GTX in equal 16th for the overall APF ranking. The more air flow through the GTX, the better it performed, It’s 3.0 score seems appropriate given it was so strong with lots of air flow but so weak with low air flow. If nothing else it shows that our auto scoring system is working well when looking at the entire picture.

GTX - APF All b

Features & Quality – 3.5/5

The GTX 360mm has HWLabs renowned build quality built into its DNA. It feels solid and weighty, is well constructed and the glossy black finish is excellent.

The GTX is perhaps HWLabs oldest model still available and it kind of shows in the features (or lack of) that it has. Just single port options, no fill/bleeder port, and no screw guards. Maybe an updated revision of the GTX is in order with features which end users now come to expect.

The core of the GTX is heavily biased towards high speed fans, due to it’s dense core which has 20 FPI. Due to this dense core dedicated cleaning will be require more often that rads with a lower FPI count. Such is the price you pay for maximum thermal performance.

The GTX’s high restriction level mean you should consider the required pump power when planning out your loop, particularly if want to run more than one of them in your loop.

 

Summary – 3.5/5

The Hardware Labs Black Ice GTX is an excellent thermal performer with high fan speeds – particularly when run with a Push/Pull fan assembly. This is what it is designed for. Low pressure/speed fans just won’t get the performance out of the Black Ice GTX radiator.  If you want extreme performance and can deal with the associated fan noise and regular cleaning then the GTX is for you, there is nothing better!

If you desire a quite rig with low fan speeds, then it’s best to look elsewhere, as this is not the GTX’s strong point.

The design, build quality and finish are excellent. The GTX does have it’s drawbacks though with a lack of features such as any extra ports.

I’ve had 2 x 560mm GTX rads in my personal test bench for years, and the results of the 360mm GTX thermal tests were very satisfying. A test bench scenario where we’re attempting to wring every ounce performance out of our hardware is probably where the GTX is best suited. In a normal case enclosure, excellent airflow is essential to get the best from the GTX.

The GTX is not for every one or every build, but for the niche corner of the market the GTX is still at the very top of the thermal performance pack!

 

8 COMMENTS

  1. Thanks for the review again.

    I think that the issue of screw protectors again, seems to be a design flaw. This is a bigger problem with the GTX 360 than other radiators simply because it is a radiator designed for high rpm fans. By nature, you are far more likely to use 38mm fans with such a radiator in push pull for the added static pressure.

    Otherwise, it’s a solid radiator.

    I would love to see a “Monster” radiator with 20 fpi and a similar design, but I suspect it would be a niche product.

    • Keep in mind that the GTX is an older model and that what we now consider important features were not always incorporated a few years ago. The current generation of HWLabs offerings all feature protection plates.

      • If I am not mistaken though the Black Ice GTX Gen 2 though (the variant for high rpm fans in push pull) still does not have screw protectors.

        The HW Labs site is currently down so I cannot look it up.

    • Any 120mm fans would work – but if you’re buying this rad we would assume it’s to make use of it’s higher performance with highspeed fans. So something in the 2000+RPM is probably the target.

        • There is a definite intangible when it comes to fan noise. That intangible is the “ear of the beholder.” The quality of sound may not be the same to one listener as to another. This becomes more true when you start moving up to higher speed fans. Another thing to think about, if a fan is tested to emit 29 decibels of sound at 1500 RPM and you need 6 of these fans, each fan adds to the overall noise total. It is not linear, so not 29 dba * 6, it is more of a percentage. Throwing a number out there for sake of example, 6 fans together may emit a constant 38 dba at 1500 RPM. Mounting the fans can cause the noise quality to change. To answer your question with F_F target RPM (not in a specific order): Dark Side Gentle Typhoon 2150 RPM, Noctua NF-F12 Industrial 2000/3000 RPM, Noiseblocker NBe-Loop B12-P / B12-3 & B12-4, EK Vardar F4-120/120ER & FF5-120. A starting place at least.

Comments are closed.