Danger Den MC-TDX versus Swiftech Apogee GTZ.

[muddocktor]

I have been using a DD MC-TDX for the last month or 2 on one of my watercooled systems and frankly, thought it was leaving some cooling performance on the table compared to The Apogee GTZ, which I am using on another system. So, I decided to test my theory out and purchased another GTZ for this test. The system this will be tested on is a Seti cruncher (which means no decent graphics card, just something to put a picture on the monitor) and has been a good stable machine. The components of the machine are as follows:

• Silverstone TJ06-B case

• BFG ES800 psu

• Asus P5Q Pro motherboard, bios version 1406

• 2 X 1 GB Crucial DDR2-800 Ballistix (single sided), running at 400 MHz with 4-4-4-12-2T timings

• G0 stepping Q6600, 1.25v VID. Running at 400 X 9 with 1.42-1.43v vcore. This processor has not been lapped, but has shown remarkable consistancy between all the cores with whatever I have mounted on it

• ATI Radeon X1600 PCI-e video card

• 2.5" Seagate 100 gig 7200 rpm hard drive salvaged from my laptop

• 12X SATA DVD-RW

The cooling system consists of the following components:

• Swiftech MCR320-QP Res (built in reservoir) radiator

• Swiftech Radbox Revision 1 (the metal one)

• 3 X Petra's Yate Loon D12SM fans

• Swiftech MCP355 (Laing DDC 3.2) with XSPC top, 1/2" barbs

• 1/2" Tygon Silver Antimicrobial Tubing throughout the loop

• Barbs used are the 1/2" Swiftech High Flow metal and the 1/2" EK Fatboy barbs in the loop

• Arctic Cooling MX-2 used for TIM material

The 2 waterblocks being tested are:

• Swiftech Apogee GTZ, totally stock with 1/2" Swiftech High Flow metal barbs (included with the GTZ)

• Danger Den MC-TDX. equipped with the brass top and stainless steel universal mounting bracket with 1/2" Fatboy barbs (included with the MC-TDX). The mounting kit has been modified with a TR LGA775 backplate to spread mounting pressure across a wider area than the 4 individual stud approach that DD sells with the waterblock. Other than using a backplate to mount the studs instead of individually mounting each stud to the corresponding hole in the mobo, the mounting system is totally stock, using the included mounting studs, washers, springs and thumb nuts to mount the block.

I am not going to get into which mounting system is best because that is personal preference and both require you to remove the mobo from the case for the initial mount and both do secure the processor well. But I did remount the MC-TDX 4-5 times looking for a better mount before deciding to do this comparison. The mount I am testing with was the best one I had while working with the MC-TDX. The GTZ was tested with the first mount I made on this motherboard as I was sure I got an excellent mount when I fired it up.

For stress and heat testing I am using OCCT version 2.0.1 and running the default 1 hour test. Room temperature was controlled at 20-20.5 C while testing both blocks.

First up was the MC-TDX, since it was already mounted and properly seated. OCCT ran successfully with the MC-TDX and here are the temperature results of all 4 cores:

 

[muddocktor]

Now for the GTZ

Next was the Swiftech Apogee GTZ's turn to be tortured. I went ahead and changed out the back plate for the Swiftech one and mounted the block up and bled all the air out of the system. I ran it for literally 30 minutes to make sure I had no leaks and that I had a good mount and then started testing with OCCT. The test completed successfully and here are the results of the GTZ run:

I also included the graphs of the vcore on each test to show that nothing had been chaged between runs. As you can see, all core temps were lower with the GTZ compared to the MC-TDX. The breakdown is as follows:

Core #1 - 3° C cooler for the GTZ
Core #2 - 3° C cooler for the GTZ
Core #3 - 5° C cooler for the GTZ
Core #4 - 5° C cooler for the GTZ

Conclusion:

While the Danger Den MC-TDX block is a decent performer and it has very low restriction (about the lowest of all modern waterblocks), it is a step behind in performance compared to the Swiftech Apogee GTZ.

 

[grimm003]

Very nice! I just bought the GTZ and now I know I made the right decision. 3 - 5 degrees is a very nice improvement.

 

[soul23]

nice test there m8.ive got GTZ in my system too.im running Q6600 GO at 400fsb x9 1.45 vcore and temps idle are 22c and 38c on full load.

 

[axis01]

That goes right along with the notion that lower restriction blocks typically don't cool as well as higher restriction ones. Nicely done.

Axis

 

[corruption]

I would be interested to see this test done on an open die chip. The TDX, and the RBX, were designed to operate on open dies where as the Apogee is optimized for IHS's. I'm actually surprised that the TDX could keep the temps down to a "stone's throw" from what the Apogee performs at.

Thanks for the results, I love seeing comparisons like this!

 

[muddocktor]

Actually, this is the MC-TDX and not the older TDX. Danger Den states that this wb is "designed to provide optimum performance to the Multi-Core and Multi-Die processors currently on the market and soon to be released products" (Direct quote from the DD page on the block). So this is what DD is selling as the top of the line block for multi core processors. BTW, the block is only about 2-3 months old.

 

[corruption]

Ah, very nice. I haven't seen the MC-TDX, I saw TDX and automatically assumed it was the same block I have used in the past. lol Thanks for the link and the correction in my misunderstanding.

 

[phil178821]

Actually, this is the MC-TDX and not the older TDX. Danger Den states that this wb is "designed to provide optimum performance to the Multi-Core and Multi-Die processors currently on the market and soon to be released products" (Direct quote from the DD page on the block). So this is what DD is selling as the top of the line block for multi core processors. BTW, the block is only about 2-3 months old.

i have been using an mc-tdx for about a year now, and while it does perform well, I do experience slightly higher temps than others with a comparable system but different cpu block. also, at the time, i made the mistake of buying a lucite top, which appears to be getting some cracks in it.

 

[muddocktor]

Yeah phil, I had reservations about the lucite top myself, which is why I went with the brass top and the universal stainless steel mounting plate with mine. I work offshore on a 14 day on and 14 day off schedule and my machines run even when I'm at work, so leak protection is a must for me. I just don't trust the lucite for long term use either as the top material or as the mounting plate material.

 

[atomic ferret]

Thanks for the comparison. The results were about what would be expected when comparing a high end low restriction block and a high end high restriction block.

 

[outhouse]

lucite is a bad deal. my RBX came with a crack in it from the factory NIB so theres no way id trust one on my rig. i ordered the brass top to a long time ago lol.

 

[exe163]

GTX, quality stuff. TDX not bad for a non restrictive block.