Reviewed: Deep Cool Gabriel CPU Cooler

Installation Notes

Apply the thermal paste, screw it on from the back, connect the fan header and you’re basically done. This is on par with the Noctua NH-L9i in terms of ease to install – well the NH-L9i has a slight edge because the Gabriel fan clips are fiddly. It took less than 5 minutes from the time I applied the thermal paste to have the installation complete, probably about 3 minutes.


Clearance around the cooler was pretty good for a horizontal cooler on our Micro-ATX ASUS Maximus VI Gene test bench. Unlike tower coolers, a 120mm footprint in a flat orientation is always going to cover the RAM. There were no issues at all accessing the first PCI-E slot so graphics card compatibility is fine. The fan cable is 44cm long which is way too long for a cooler like this. I managed to route the excess away on the test bench but it was more challenging in the concept builds – I think the fan cable should be braided and shortened by about 30-32cm to avoid ‘spaghetti’.

When doing the MSI Mini ITX review I tested the overclocking with the 120mm Noctua NH-U12S because it handled the temperatures better but the NH-U12S blocked the first PCIE slot on the MSI Z97I Gaming AC so I couldn’t use it in the practical build. I also couldn’t use the Noctua NH-L9i because it was too low and wide in that the base hit the chokes that are closer to the socket than normal. Then the H100 was out of the question because the Node 304 case can’t fit a 240mm radiator… The Gabriel was the only cooler I had (other than the stock Intel Cooler) that would fit that board and allow access to the PCIE Slot because of the mounting system and profile.


No backplate and only the 4 screws that go in from the back make this dead simple but it does limit the firmness of the fit.

Fan Clips

The included fan clips allow you to fit the included 120x120x20mm fan but nothing thicker like a standard 25mm wide fan. This is a shame because it makes life harder for enthusiasts who want to use a more powerful, colourful or simply different fan for whatever reason – It limits the versatility of the cooler in a custom build. I installed the Gabriel no less than 5 times and didn’t make friends with the clips.

Contact with the CPU Heat Spreader

After testing the Gabriel in our load tests, I was surprised to see the results – the temps were higher than I was expecting. I thought it might be due to the thinner fan so I rigged a thicker fan on it and got basically the same results. The heat pipes didn’t feel too hot so I wondered about the thermal paste application. After several runs with different methods, it looked like the Gabriel base didn’t sit as close to the CPU heat spreader as other coolers like its bigger brother the Lucifer. This was confirmed when I applied a very thin layer of thermal paste across the CPU, fitted and removed the Gabriel to see the underside – there were a few spots that showed transfer of paste to the bottom of the cooler. By comparison, the Lucifer made much better contact due to the mounting clamp. The best results were achieved using a larger than normal ‘rice grain’ application.

IMG 5971  IMG 5964

Room for Improvement

For starters, it’s not that I don’t like the cooler – I hope Deep Cool take the following on board and implement some of these changes in version 2. Affordable coolers are great for gamers because it means that they can better decide how to balance their costs to get the best bang for their buck.

  • The mount needs to be firmer on the socket to get better heat transfer – better heat transfer means that the fan can operate at lower RPM and will have a better acoustic profile.
  • The fan mounting clips could be made out of firmer wire and maybe have longer hooks where they attach onto the corner holes in the fan
  • Shorten and braid the fan cable
  • As with the Lucifer the included thermal compound works as expected and there is enough in the small tube for several applications. Although the texture is stickier and harder to clean up than Noctua NT-H1, as with the previous Lucifer review I couldn’t see any consistent difference in the temperatures between the paste. The NT-H1 is also non-conductive – I’m not sure if the compound included with the Gabriel is conductive or not.


Please enter your comment!
Please enter your name here

Captcha loading...

This site uses Akismet to reduce spam. Learn how your comment data is processed.