Some New Blood On The Testbed :)

[Jupiler]

Yep,
that's him.

[The_Beast]

I aways loved the look of the AC Double Impact

[AAjax]

very nice looking block!

[nikhsub1]

If its the Radical_53 on this board then I just sent him a PM asking if he could drop by. Hopefully he can.

Yeah I've seen his rad testing and it is good stuff. His block testing however, I would just love to know more about it all as it just doesn't make sense to me yet.

[Vampire_Hunter]

Yeah I've seen his rad testing and it is good stuff. His block testing however, I would just love to know more about it all as it just doesn't make sense to me yet.

What I'm thinking is that he threw the TDX in there just for comparison purposes and ended up with an anomaly. No one else has tested the TDX under similar testing conditions so before we howl at the moon we should ask politely what's up. Perhaps we will learn something, yes?

[headala]

Any new numbers yet?

[Radical_53]

@nikhsub: Well, I'd have to lie if I said I wasn't surprised by the TDX performance too.
My charts on a quad core platform do look different, but that's another story.

So, the platform I used for the test that's shown was a C2D 6600 with machined top, a Thermalright backplate and Arctic Ceramique as TIM (normally I use Silmore, but I ran out of stock on it ). The CPU was running at 3.833GHz with 1.46V, tested with Windows XP 64-Bit and the according version of Prime95.
The loop I used had a PA 120.2, Laing DDC+ with Petra's top, Swissflow SF-800 flow meter, a small water filter made by Mips, t-line and 7/16" Masterkleer with EK 1/2" barbs. I also used the Aqua Computer poweradjust to run the pump at 100% and 60% to simulate a lower flow loop (for people that like to use smaller tubing or a GPU block in their loop for example). Water temperature and room temperature was recorded of course, throughout the testing the room temperature stayed within a range of ~2°C (I'd have to look that up, but the variation was small as all the testing was done in a row).

@vampire: That's right, yes. The primary goal for this test was to see what the "new" blocks could do on the "typical" modern platform. But I was interested to see what some other blocks could do, blocks that I had on my older testbed last year that used an Athlon 64 CPU (I used a naked single core FX and a capped dual core X2 back then).

[hobo man]

I don't test in my house, in my office (lab )



I've tested both o-rings of the GT and GTX, the performance is NOT very good compared to the fat o-ring... the storm I have is a rev 1 but there is no real performance difference in rev 1 and 2.

All my threads that i started, lots of info: http://www.xtremesystems.org/forums/...archid=4925317

lol@towel :P

[Vampire_Hunter]

@vampire: That's right, yes. The primary goal for this test was to see what the "new" blocks could do on the "typical" modern platform. But I was interested to see what some other blocks could do, blocks that I had on my older testbed last year that used an Athlon 64 CPU (I used a naked single core FX and a capped dual core X2 back then).

Like I said earlier I think a whole lot of people might owe you an apology about the TDX results. You can't compare results of one test method with another. Could you perhaps upload your Excel file or raw data here?

Nice nice on the use of the Aqua Computer Poweradjust for the DDC. I was trying to figure out if you meant % of CPU usage, but it didn't make any sense to me. Now it is crystal clear.

[Radical_53]

I'll see what I can do about an excel sheet I do have some flow results and the performance results at 60% pump power. Would that do?

And yes, the power adjust was the easiest way to simulate different flow rates. I did it the "hard" way on earlier tests, using two different pumps, but it works out quite nicely with just one DDC+ and this little device.

[SiGfever]

OT
@Radical_53,
How have you been? Haven't heard from you in a while, hope everyone is well. Look forward to seeing your test data.

[Vampire_Hunter]

I'll see what I can do about an excel sheet I do have some flow results and the performance results at 60% pump power. Would that do?

And yes, the power adjust was the easiest way to simulate different flow rates. I did it the "hard" way on earlier tests, using two different pumps, but it works out quite nicely with just one DDC+ and this little device.

The 60% files would be nice, thank you!

How has your use of the Poweradjust been? Does it "fix" any of the DDC starting problems or is it just a flow adjusting tool?

[RickCain]

Any updates? Things still looking good?

[headala]

Yeah, any updates?

[nikhsub1]

No guys sorry, been too cold in LA...

[Vampire_Hunter]

No guys sorry, been too cold in LA...

From the satellite look it might be a while before you guys get some warm weather.

So why not turn on the heater?

[Radical_53]

@sig: Thanks, I'm fine so far! I've been quite busy here building a house, so I wasn't around too much.

@vampire: Well, I never had any starting problems, but I do have two dead DDCs out of four here too. Lucky me, my DDC+ with red rotor still works

So, I wanted to give some numbers right? I'll limit this to the top 10 blocks, should be the most interesting anyhow.

100% pump power:

1. Aqua Computer Cuplex XT di, 30.0K, 3.76 l/min
(2.) Swiftech Apogee GTX bowed, 31.8K, 4,87 l/min
2. Danger Den TDX, 34.6K, 5,26 l/min
3. Swiftech Storm, 35.2K, 4,50 l/min
4. D-Tek Fuzion, 35.7K, 5,82 l/min
5. EK Wave, 35.8K, 5,56 l/min
6. Swiftech GTX small o-ring, 37K
7. Aqua Xtreme MP-05 SP LE, 37.2K, 4,27 l/min
8. Zern PQ+, 37.8K, 3,34 l/min
9. Swiftech Apogee GT, 38.1K, 5,70 l/min
10. Sub Zero ExodusFlow Altauna, 38.2K, 3,97 l/min

Short explanation: I used the result with the "normal" o-ring for both Apogees. The test with the big o-ring showed a really nice improvement on the GTX, but the bow was really extreme to me. Also, it killed the block due to corrosion shortly after. I got a new GTX for the quad core results I took later and the newer version even showed different (better) flow rates (5,53 l/min at 100%). Maybe the first GTX was crippled somehow from the start, but normally I do only get 1 piece of a standard production block to test.

Then, here's the reduced pump power chart at 60% (~3200 rpm on a DDC+).

1. Aqua Computer Cuplex XT di, 31.5K, 2,52 l/min
(2.) Swiftech Apogee GTX bowed, 32.2K, 3,27 l/min
2. D-Tek Fuzion, 36.4K, 4,02 l/min
3. Swiftech Apogee GT, 36.4K, 4,01 l/min
4. Swiftech Storm Rev.2, 36.6K, 2,98 l/min
5. Swiftech Apogee GTX small o-ring, 37.3K
6. Danger Den TDX, 37.5K, 3,70 l/min
7. Zern PQ+, 38.3K, 2,13 l/min
8. EK Wave, 38.4K, 3,88 l/min
9. Aqua Xtreme MP-05 SP LE, 38.5K, 2,88 l/min
10. Sub Zero ExodusFlow Altauna, 39.0K, 2,70 l/min

So now you've got some flow numbers with the somewhat puzzling results
All the blocks were mounted in the same way, both fittings horizontally alined, parallel to the socket clamp. I did try various orientations with some of the blocks but the results didn't really show much difference, the IHS seemed to be pretty flat. The difference between the cores was very small too for most of the blocks. Usually I saw ~1K difference between core 0 and 1, sometimes even both cores showing the same temperature.

PS: For the record, the room temperature recorded during these tests were between 17.4°C as the lowest and 18.9°C as the highest recorded temperature.

[headala]

@sig: Thanks, I'm fine so far! I've been quite busy here building a house, so I wasn't around too much.

@vampire: Well, I never had any starting problems, but I do have two dead DDCs out of four here too. Lucky me, my DDC+ with red rotor still works

So, I wanted to give some numbers right? I'll limit this to the top 10 blocks, should be the most interesting anyhow.

100% pump power:

1. Aqua Computer Cuplex XT di, 30.0K, 3.76 l/min
(2.) Swiftech Apogee GTX bowed, 31.8K, 4,87 l/min
2. Danger Den TDX, 34.6K, 5,26 l/min
3. Swiftech Storm, 35.2K, 4,50 l/min
4. D-Tek Fuzion, 35.7K, 5,82 l/min
5. EK Wave, 35.8K, 5,56 l/min
6. Swiftech GTX small o-ring, 37K
7. Aqua Xtreme MP-05 SP LE, 37.2K, 4,27 l/min
8. Zern PQ+, 37.8K, 3,34 l/min
9. Swiftech Apogee GT, 38.1K, 5,70 l/min
10. Sub Zero ExodusFlow Altauna, 38.2K, 3,97 l/min

Short explanation: I used the result with the "normal" o-ring for both Apogees. The test with the big o-ring showed a really nice improvement on the GTX, but the bow was really extreme to me. Also, it killed the block due to corrosion shortly after. I got a new GTX for the quad core results I took later and the newer version even showed different (better) flow rates (5,53 l/min at 100%). Maybe the first GTX was crippled somehow from the start, but normally I do only get 1 piece of a standard production block to test.

Then, here's the reduced pump power chart at 60% (~3200 rpm on a DDC+).

1. Aqua Computer Cuplex XT di, 31.5K, 2,52 l/min
(2.) Swiftech Apogee GTX bowed, 32.2K, 3,27 l/min
2. D-Tek Fuzion, 36.4K, 4,02 l/min
3. Swiftech Apogee GT, 36.4K, 4,01 l/min
4. Swiftech Storm Rev.2, 36.6K, 2,98 l/min
5. Swiftech Apogee GTX small o-ring, 37.3K
6. Danger Den TDX, 37.5K, 3,70 l/min
7. Zern PQ+, 38.3K, 2,13 l/min
8. EK Wave, 38.4K, 3,88 l/min
9. Aqua Xtreme MP-05 SP LE, 38.5K, 2,88 l/min
10. Sub Zero ExodusFlow Altauna, 39.0K, 2,70 l/min

So now you've got some flow numbers with the somewhat puzzling results
All the blocks were mounted in the same way, both fittings horizontally alined, parallel to the socket clamp. I did try various orientations with some of the blocks but the results didn't really show much difference, the IHS seemed to be pretty flat. The difference between the cores was very small too for most of the blocks. Usually I saw ~1K difference between core 0 and 1, sometimes even both cores showing the same temperature.

PS: For the record, the room temperature recorded during these tests were between 17.4°C as the lowest and 18.9°C as the highest recorded temperature.

I think I hear the sound of people picking up mud getting ready to fling it your way.

Thank you for getting us your results, whether they are eclectic or not.

[headala]

Oh, and nikhsub1: TURN ON YOUR HEATER and get us some results!

[Radical_53]

Well, go ahead and test it yourself With the same basics it should roughly give the same result.
The only thing I'm really annoyed about is that after the tests I saw that the loop is "capped" somehow. The blocks with low restriction are pretty close together, closer than I guess they might be with a more open, less restrictive setup.

[headala]

Well, go ahead and test it yourself With the same basics it should roughly give the same result.
The only thing I'm really annoyed about is that after the tests I saw that the loop is "capped" somehow. The blocks with low restriction are pretty close together, closer than I guess they might be with a more open, less restrictive setup.

I'm sorry, my post wasn't meant to offend. I don't think you did anything wrong; I just think that your results go against accepted norms of performance of some of the other blocks in your test. I am most thankful that you took the time to do the test!

I live in SW China, so the availability of WC parts here is only what I can make myself or cobble together from aquarium/machine parts. I have to rely on guys like you and nikhsub1 for test data.

And, there's no practical reason that I need those results ASAP. I'm just addicted to wc'ing and have been for the last 5 years.

[mouawad]

Actually I like it, but when I first got an XT I bought a clear top and copper top as well. Sold the copper top immediately as it looked terrible to me. Kept the clear one in case I wanted to do something with different color coolant. The aqua blue top is still in my box. From what I understand, AC said the alternate tops were a commercial bust.

Aesthetics always matter. Would you buy a BMW hot rod if it looked like a soap box derby? If you have a window in your case you are probably interested in aesthetics in some manner or another.

i've got a mountain mod u2-ufo so i have a huge window but i would place a block that looked like a rancid turd in my rig without hesitation if it got better temps than the competition.

i don't lure too many women into my castle through the looks of my water block

[Radical_53]

@headala: Mine neither Just to "see" how that is. What I've seen is that the ranking is heavily related to the form and quality of the IHS and the used loop.
Thing about accepted norms, well. If you "know" something about a block, you can use that as a guideline to prevent errors on your own side. But if you double check, still getting the same result, you might want to think that it really "is that way".
For me it was quite interesting, after the dual core tests, to see how the same blocks (and some more) performed on a quad core platform. Some blocks were able to keep their good result, good ranking, others fell far behind.
I also got some feedback of the readers of the orignal magazine. Some came to the same results, others had totally different results when they switched from one block to the other.
As nikhsub quotet my radiator testing before, that was much easier to do. Radiator testing has much fewer variables you can't control, the "only" thing you need are some so far good sensors and a lot of heat to distribute. It's time consuming, but you can't mount it wrong other than not bleeding completely.
With water blocks, it starts with the big problem of the IHS. Different shapes deliver different advantages for one block over the other. Then the force you mount the block with, even when it's equal throughout the testing. It may be an advantage for some blocks if the mounting force is low, others can improve their performance with higher mounting force. What adds to that is if a backplate is used for the testing or if the test is performed "naked". Soft base plates against strong and hard base plates of the blocks is what I think makes the difference.
So, what I'm after with all this -> no matter how exact, how ideal a test setup may be, due to the vast amount of variables it can only be a rough guideline for a consumer to prefer and choose one block over the other.

[Vampire_Hunter]

i've got a mountain mod u2-ufo so i have a huge window but i would place a block that looked like a rancid turd in my rig without hesitation if it got better temps than the competition.

i don't lure too many women into my castle through the looks of my water block

So how about a pic of your box?

[Kayin]

Radical, that's the best view of rating waterblocks I've ever seen. My hat is off to you.

Is there any news to report yet?