ati/amd 6970 v2 waterblock

[Sorcos]

I need to get a new waterblock for my ati/amd 6970 v2.

My original XFX 6970 that I had recently died on me and I was able to rma it and get a replacement from XFX. However, the one they sent back to me is a revision 2 card (since they obviously don't make the originals anymore). My problem is that the waterblock I had on my card (a Koolance AR697) does not fit the version 2 card and I must now find a replacement waterblock.

I have done quite a bit of research over the past few days and have found that there is really only 4 companies I can choose from for this new version 2 card. The following list is what I am trying to get information on:

• EK FC6970 v2 (3 varieties - all nickel plated)
• Swiftech Komodo 6900-2 (I cannot find this block. However, it is my understanding that the Komodo 6900 v1 is adjustable and I can replace the VR coldplate to match the v2 model.
• Aquacomputer Aquagratix 6970 type 2
• Danger Den 6970 v2

I have seen many peoples suggestions scattered across the internet, but I am mainly looking for some type of statistics from tests that were performed on these blocks. So far I have only come across Cartago's test results, which tests the v1 variants of the 6970 blocks (and does not include the danger den block) - http://www.techpowerup.com/forums/showthread.php?t=148860. His test results seem to indicate that the Swiftech Komodo has the best thermal performance but it has the worst flow rate.

Based on Cartago's tests, since they are the only statistics I can find, I was leaning towards getting the Komodo but I am worried that the flow rate will affect my cpu temperatures - I have a Koolance CPU-370 cooling a i7-950 and a XSPC RX360 radiator. I was also shying away from the ek block because of the nickel plating issue (not because of customer service, etc.), unless someone can prove it is worth the risk.

So is there anyone who can point me to some other type of tests that were performed on these blocks that I can compare and make an informed decision? I would also love to hear others unbiased opinions on these blocks based on actual performance and experience.

 

[jojo69]

DD closed the doors last week, so good luck finding those.

EK is on my "never again" list after their behavior around the failed plating fiasco.

 

[Skripka]

I have a Komodo with v2 coldplate your understanding of swapping plates is correct...swapping can be done without pulling tje block apart. Nice block, i was able to get it used for a pittance. System is in my sig. Switching from my old ApogeeXTv2 to the non-restrictive DT5Noz dropped my GPU temps 5C across the board.

Note that depending on the topography of your pcb you may have no choice but to use tape or pads for mosfets and vram. On my Powercolor the chokes on the vrm rise to high to allow the coldplates to sit flush on the FETs.

 

[Sorcos]

DD closed the doors last week, so good luck finding those.

EK is on my "never again" list after their behavior around the failed plating fiasco.

Thank you for your response. I did read about Danger Den closing up shop but I did happen to find a place that has one of their 6970 v2 blocks left.

As far as EK goes, I understand everyone's frustration with the company but I would be willing to try one of their blocks if it outperforms the others listed and possibly justify the risk of the nickel flaking. Maybe adding an inline filter?

I have a Komodo with v2 coldplate your understanding of swapping plates is correct...swapping can be done without pulling tje block apart. Nice block, i was able to get it used for a pittance. System is in my sig. Switching from my old ApogeeXTv2 to the non-restrictive DT5Noz dropped my GPU temps 5C across the board.

Note that depending on the topography of your pcb you may have no choice but to use tape or pads for mosfets and vram. On my Powercolor the chokes on the vrm rise to high to allow the coldplates to sit flush on the FETs.

In regards to the effect the flow rate of the Komodo has on the system, did you happen to switch from another less restrictive GPU waterblock? If so, I know you mentioned switching your CPU block but did your CPU temps change much (in either direction) after switching to the restrictive Komodo. I have read that the Komodo is designed to be a bit more restrictive and work more efficiently with multi-waterblock systems, whether that was intended or not. I am not sure if this means they are only efficient when the multi-waterblock system is all Swiftech waterblocks, considering my CPU block is a Koolance.

As a side question, do you happen to know if there is a difference internally between the different versions of the Komodo? For example, the only differences I can tell is the coldplate and the different outer design/sticker.

 

[Skripka]

Internally the Komodo 6900 blocks are the same configuration within my knowledge. I haven't had another block on this GPU to compare to. I could give you my raw temps, but unlesd you're running eyefinity my temps will seem warm...because simply idling in Eyefinity raises gpu temps 10c compared to single head.

If you're cpu and single gpu and nothing else, and areusing a qquality pump restriction shouldn't be qn issue...unless you trying to run a very restrictive cpu block with a 35x at 1400rpm (low idle throttle)...and as I recall that koolance370 is much more comparable in restriction to my 5noz 5han a high restriction block like the apogee line.