Cooling DangerDen Maze4

I completely understand what you're saying and I, nor beta124, have taken any offense. In fact, your opinion has helped us to better our site for now we now how to be better than most sites. And so, now we're gonna work towards getting proper equipment, such as BillAs.

 

In that case, if you are serious about this, I would suggest joining the forums over at procooling ( forums.procooling.com )

BillA was/is a regular there and most members are getting clued up on the dangers of system testing.

There has been talk about waterblock testing, and even putting together some "standards" for testing, so that results from different sources could be accurately compared. This would be VERY good for the community, but would require a HUGE amount of work. I plan to get involved once I've got my finals out of the way.

Glad you didn't take offence

8-ball

 

Ok, the new revised, shorter, to the point, less general info, comparison review is finished.

http://www.elite-tech.org/reviews/maze4/

sorry for the wait.

 

you compared a maze4 to the volcano9?


why didn't you compare it to a maze3?

"Another interesting concept was having my CPU temps below my case temps"

that's complete and utter trash

"The Volcano 9's heat sink itself is more than 3 times the height of the Maze 4 block."

 

Thats all we had available to us. We are in contact with DangerDen right now seeing if they'll send us a MAZE3 to compare against. I was switching from the Volcano9 to watercooling and posted what I thought was relevant.

 

Hey mate, I don't usually go off at people, but I think you should read the rest of the thread. Particularly the pages I linked to earlier.

Why not, lets think about that shall we.

How much difference are we gonna get, reasonably?

Just a few degrees most likely. But what is going to cause these differences in temp. The block you say?

B@llocks I say!

Well it will, partly, but the maze 3 and the maze 4 do not have the same restriction to flow as one another.

My point, this will affect two other major components. The pump and the radiator.

Changing the pressure drop seen by a pump will affect how efficiently it runs, and consequently how much thermal energy it dissipates into the coolant stream. A change in flow rate resulting from a more or less restrictive block will also affect the thermal resistance associated with the convective heat transfer at the interface between the water and the walls of the coolant channels in the radiator. (I won't go into secondary heat paths other than to say that this adds another complication)

So, how much did the block improve your temps?

I suppose it would be ok if you had the same rad and the same pump, but otherwise, it would be useless. With one combination of pump and rad, changing a maze 3 to a maze 4 might yield better results, but on another, that might not be the case.

These are some of the finer points which need to be addressed with cooling reviews, and it seems that these guys are actually keen to learn how to improve their technique.

brwnvij & beta124,

I have a couple of suggestion/corrections, as you still seem to be missing the mark slightly.

However, I am impressed to see that you are keen to understand more. As I said before, join up at the proforums. You can learn A LOT there.

To start with, I think you took some of the character out of the review. From what I can tell, it appears that you are still not sure who you are aiming this review towards. I understand that you want to be able to provide more accurate results which will be more useful, but to how many people. Not many.

Remeber that this is one of the first maze 4 reviews on the web, so general feedback, ease of use, first impression and all that count. People want to know what this block is like.

Equally, with your current testing apparatus, you can't delve in to the comparisons of different waterblocks, so the volcano will be fine. At least you compared it to something. I've seen quite a few, (too many in fact) reviews of watercooling gear, where the components aren't compared with anything, and at the end, the reviewer tries to tell you that this is the best block/rad/heatsink on the market.

So, inject some of the character back into it, but as you have obviously realised, don't pad it out too much. It's all about getting a balance of opinion/feeling, and fact/data.

Secondly, I think you might have got a little muddled up with the thermal resistance bit.

A number of times, you quoted the thermal resistance as W/C, when it is actually C/W.

Also, what are you using for your temperature. It should be the difference between the recorded temperature and the ambient temperature, and I don't think you have done this.

Since the ambient temp and the cpu temp are recorded on two different diodes/thermometers, it is very difficult to make any comparisons.

As such, I wouldn't include the resistance figures.

My reason for bringing them up was to demonstrate that your recorded cpu temps were off.

It might be worth mentioning that for an ambient of 26 degrees and a cpu load temp, (assuming a 70W heat load) of 31 degrees, the apparent thermal resistance is simply too low. This wouls suggest that the readings from the thermal diode are off. Make a point of this, since it is something that needs to be got across to the wider community.

However, as you rightly pointed out, you have readings for the watercooling and the aircooling, taken with the same diode, so a comparison is acceptable, provided you make it clear that the temperatures might be off. Even then, it should be clear that the watercooling performs significantly better than the air cooling.

It's also good that you have listed the specs of your system, and I like the point about being able to overclock your cpu further. This reinforces the point.

For future information, if you collect data, think hard about how accurate it is, and what it proves for sure, and what you would like it to prove. They are often quite different. Realising this is the first step towards good, honest reviews, something we need many more of!

Also, I'm not sure if you mentioned it, but perhaps an appraisal of the difference in sound levels when in use, compared with your volcano 9. This is something which attracts new watercoolers, perhaps even more than the performance aspects.

As for becoming the next BillA. Understand that it is not something that can be halfbaked. It's all or nothing, otherwise people will come and pick holes in your testing procedure. If you have a good look round BillA's site, you will no doubt get a feeling for how thorough he is. His radiator roundup required 6 months of tests!!!!

I hope this has been helpful, and I'm sure folks here will be able to help you out with what they would like to see from a review, though I thinkk they will also have to be realistic about what can be provided without the equipment BillA has in his "lab".

Keep it up

8-ball

 

"I suppose it would be ok if you had the same rad and the same pump, but otherwise, it would be useless. With one combination of pump and rad, changing a maze 3 to a maze 4 might yield better results, but on another, that might not be the case."

comeon?

OF COURSE i expect all other equipment to stay the same if they compare two blocks.
Get a eheim 1250 and a bix and compare the results of the blocks, that would be better than comparing aircooling to watercooling, that gives absolutely zero indication of how the maze 4 performs. lol

it's probably the worst review i've ever read, and that's being nice.

 

I'm saying that if you don't have the same pump and rad as them, then you might get better results with a maze 3 than a maze 4.

There are so many variables tht cannot really be controlled when doing on system testing that comparing blocks with relatively similar performance becomes pointless. You don't know whether the difference in temperature is a result of the new block design or the resulting change in flow through the radiator, and this could be for better or for worse.

The equipment needed to say for SURE that the maze 4 is better than a maze 3 is horrendously expensive. Essentially, you nedd to calculate the thermal resistance as a function of flow rate, for a given heat load.

This is what BillA does, and no one else has come close to doing, yet no one else is prepared to spend nearly $20,000 on equipment purely for testing waterblocks. And this is why I always take any given temperatures with a pinch of salt. You don't know what factors are affecting the temperature.

So what if they haven't compared the maze 4 with the maze 3. At least they're not trying to convince people that they can tell the difference between the two with enough accuracy to say for sure that one is better than the other.

This is why a the web is a minefield of dodgy reviews, and why companies selling dodgy equipment still make money.

8-ball

 

Thanks for the info 8-Ball, much appreciated.

You are definitely right about the audience. With the watercooling article, I just didn't know who to cater to..

But again, thanks for the input

 

lol @ 8-Ball and his book writing....

8-Ball is right though...Soo many variables to consider when testing waterblocks and pretty much any cooling device that theres almost no noticeable difference between the temps that couldnt be from something that was different between the tests....

lol $20,000 on testing equipment? Must be deep into his hobbie thats a lil extreme for my pocket book but then again thats why I only give out advice and things Ive seen and experienced myself....

 

Yeah, I think BillA's quite unique. Though it was actually a company, and I think he did work in that area before, so I think he might have had a lot of the kit already.

It's a shame that he isn't offering his services any longer. Just as people were starting to understand what needed to be done to get a waterblock tested right, Swiftech snap him up as there engineering director (or something like that).

At least we mightsee some nice blocks from Swiftech in the coming years. He has played a part in the design of many waterblocks, including the white water, and his knowledge of the theory behind waterblock design is just about second to none. (unfortunately, he knows it).

I think there are a couple of lads over at procooling who are gonna try and fill the void, by putting together a decent testing strategy.

8-ball

 

as soon as I get back to school, I'm going to try and do some research with waterblocks through the school. hopefully, i'll be able to do that b/c they have a large budget and a nice staff. so who knows??

 

hmmm don't really see the point of upgrading to the maze 4 from a maze 3 as my load temps are on my mobo through mobo moniter 34 and on my actual thermometer 32 so it does not seem worth it I will wait for the maze 5 or 6!!!!

 

Hi, I've just been reading this thread, and had a thought about measuring the CPU temperatures. Now, even if the readings from the motherboard sensor are inaccurate, this doesn't really make a difference. Reviewers can simply test block x find out the motherboard reading, do the same for block y then find out what one cools better.

Even with completely wrong readings, the best performing block on the market can be found. You could even go as far to say block x is a % better than block y

edit: just finished reading thread...realised that this comparison method has been pointed out.

 

No you couldn't without knowing the exact room temperature and the cpu temperature measured with diodes calibrated with one another.

The performance of a block is determined by the temperature difference between the water and the cpu, or the cpu and the air, if you're testing a whole kit.

If you had two cpu temperatures measured with different blocks, then you would need to compare the two temperature deltas. ie the temp between the cpu and ambient for each block. But changing the ambient reading, ie if it were measured on a different scale, or weren't calibrated to the cpu diode, then the percentage figures would be off.

8-ball

 

Room temperature can be measured easy enough fairly accurately.

Call room temperature 100%.
Say room temperature is 25C.
Motherboard reading for block x is 30C
Motherboard reading for block y is 35C

Hmm, tries to think how a percentage could be worked out. Urm, I'm finished school now, I don't do maths. lol.

You're probably right actually. I wasn't really thinking.


edit:
could you not say block x is 120% of room temp. block y 140%. No actually you couldn't. Because it goes back to the whole problem with mobo accurancy. Still...saying one block is better than the other is good enough really?

 

I suppose, but when you are comparing blocks that perform to a similar level, you have to be careful, since the errors in the measurements can be more than the difference between the two blocks. So how can you really say which is better.

Also, changing a block will affect how other components in the loop work, since it will probably have different flow characterisitics.

8-ball

 

Well, I just talked to DangerDen and they are going to send us a MAZE3. So, in about a week or so, we'll have an article comparing the MAZE3 to the MAZE4. Its not going to be any BillA tests however we are going to use the same test bed and thus, the same testing environment. We are going to control the water by using set amounts of additives and distilled water and use the same rad, pump, and res. So all-in-all, these results should be pretty decent.

 

That's great. A lot of people are looking for comparative tests between the two.

8-ball

 

I'd like to add some info I would have liked to see added to your reveiw along with those mentioned already.

1)What pump was used, & what is it's rated flow and max head rate?

2)What rad was used?

3)What fan and what is it's rated cfm & pressure?

4)What size of water lines were used? Total length of lines in the system?

5)Was outside the case air being pulled through the rad, or was case air being pushed out through the rad. If rad was not inside the case that's fine to, but please note this.

6) How long was the system run to allow temps to stablize before readings were taken?

7) What soft ware program(s) were used to bring the system to at,or near, full load?

These are things that will add to a better understanding of the conditions of the testing for those reading your article. And they are all things that will affect the performance of the block(s) tested.

If you have a digidoc, comp-u-nurse (insert brand name here) you could test the various temp probes it comes with to find several that read alike. You can then use these for more consistent comparisons (these also are not accurate but would help some). You could also use two like reading probes to read water temps just before the block and after the block. Would be one more referance point, (and one that doesn't cost a fortune up front, and one you may already have the means to do.)

By all means continue to try to improve. 8-ball offered some good advice and tips. I hope I may have offered some helpfull ideas.

We can NEVER have to many accurate testing reports or reporters! ! KEEP GOING FOR IT ! !