Skinnee Labs - GTX480 Block Round-Up (working thread)

[Alexandr0s]

Take what we can get. We can call Danger Den Flo ridah (which I like cause I have to fly to Miami in a couple weeks for the sole purpose of doing an insane water-cooled build).

triple pun... D'OH!

[gabe]

with Dynamic Head Pressure from the D5-S5 with EK V2 Top (Top Outlet) and the DDC3.2 with XSPC V3 top.

That looks great!

You might want to clarify for the benefit of casual readers that the DDC3.2 with XSPC V3 top is a dual pump.

Also, it would be interesting to show where a single DDC falls.

Thanks for the hard work!

[Vapor]

DDC3.2 + XPSC V3 top is single pump

http://www.performance-pcs.com/catal...ducts_id=22700

[Ketzer7]

You might want to clarify for the benefit of casual readers that the DDC3.2 with XSPC V3 top is a dual pump.

Also, it would be interesting to show where a single DDC falls.

I don't think that's the dual DDC XSPC top, looks to be the single from the dual DDC article:

[gabe]

DDC3.2 + XPSC V3 top is single pump

http://www.performance-pcs.com/catal...ducts_id=22700

ahh, I see.. that's the "unreviewed" V3 top from the dual top review that got me confused. thanks for clariyfing.

[skinnee]

ahh, I see.. that's the "unreviewed" V3 top from the dual top review that got me confused. thanks for clariyfing.

Hey now... don't make me bring up the PWM pumps.

Pump top reviews are changing on this next go round... less fancy review, more data driven which should mean faster turn around. I take too long.

[Captain H.N.]

Pump top reviews are changing on this next go round... less fancy review, more data driven which should mean faster turn around. I take too long.

Just some Qs if you don't mind answering about that:

What exact aspects will change?
Will you compare a topped DDC 3.2 to a topped DDC 3.25 using all the tops you have?
How did the last roundup not turn out consistent to European reviews that all showed EK's V2 to be better than XSPC's tops, and do you expect that to change?

And a more on topic question; when should we see the GTX480 blocks roundup get published?

[Vapor]

Based on all the testing I've seen, EK V2 top is only better for max flow, but the PQ curve isn't as good as the XSPC V3 in the important/usable portion of of the curve.

On top of that, in my own personal use (with flowmeters), the XSPC V3 is ahead of the EK V2, by a noticeable amount.

[Systemlord]

Based on all the testing I've seen, EK V2 top is only better for max flow, but the PQ curve isn't as good as the XSPC V3 in the important/usable portion of of the curve.

On top of that, in my own personal use (with flowmeters), the XSPC V3 is ahead of the EK V2, by a noticeable amount.

Are there accurate electronic flow meters that you can attach to your loops so you know exactly what flow rates you are getting? I've never heard of a XSPC V3 yet.

-Systemlord

[luke997]

Are there accurate electronic flow meters that you can attach to your loops so you know exactly what flow rates you are getting?

-Systemlord

Sure, sample one from Aquacomputer:

Click

[gabe]

Hey now... don't make me bring up the PWM pumps.

Pump top reviews are changing on this next go round... less fancy review, more data driven which should mean faster turn around. I take too long.

haha, no I wouldn't do that to you

...but I am still confused by the DDC curves because the curves in the link that Eric posted above (http://www.performance-pcs.com/catal...ducts_id=22700) do not match those of Martin's shootoot http://martin.skinneelabs.com/DDC32PumpTopTesting.html

#1 Pump Top Test comparision


#2 Performance PC


It looks like you are using the #1 set, right? and equally important, why the difference? maybe Martin could clarify?

[skinnee]

Completely different pumps, newer revision of the top and completely different test gear, there are going to be differences. I've even retested the same RD-30 as Martin, my amp draw and dynamic head were lower.

Cross comparing benches...

[NaeKuh]

when will the good bread crumbs start falling skinnee?

Your shedding your results like a trail... we want the main course!

[gabe]

Completely different pumps, newer revision of the top and completely different test gear, there are going to be differences. I've even retested the same RD-30 as Martin, my amp draw and dynamic head were lower.

Cross comparing benches...

Thanks for clarifying. So if I understand correctly, you are using your own data? no offense to Martin, but I prefer that, as I verified your data before, and found it to coincide with ours.

[Systemlord]

There must be a point where flow rates (example 2 GPM or more) no longer play a role in water temperatures, where is that point of diminishing returns?

-Systemlord

[Martinm210]

I suspect its either a difference in the pump motor/impeller and or a difference in testbed t fittings or the manometers. I did all my testing on the same exact pump motor/impeller to ensure good relative data, but it wouldn't surprise me to see some varience in the performance levels between pumps. I saw some of that with different D5s. I also made my own 1/2" copper to 1/8" brass fittings to connect my manomter to. Theretically and pressure errors should cancel out if measuring pressure difference, but there could be some influences we are missing. Also calibration of the manometers..probably some differences there too. I never had mine calibrated..mine could be reading highfor all I know..

Bottom line..different test beds....we can hold precision and have very good comparative data on the same test bed, but I we don't know how accurate the testing is without some sort of check. Soo, results from two different test beds shouldn't be combined or compared..you would need standardized proceedures, equipment, and calibration efforts which we don't have nor is there any interest to do so...we already donate our free time and paid for the equipment.

It would take a manufacturer to develop a standardized testing specification to get to a cross test bed level of testing accuracy along with precision. As hobby testers...were making up the specs as we go and constantly changing them...no incentive for us to do otherwise.

IMHO the same thing could be done for all of water cooling testing including cpu/gpu c/w testing, but it would take several manufacturers to come to some agreements on the testing specification.

No one has stepped up to that plate yet..

Hopefully one day..

[skinnee]

Thanks for clarifying. So if I understand correctly, you are using your own data? no offense to Martin, but I prefer that, as I verified your data before, and found it to coincide with ours.

Yeah, all my own data.


@ Naekuh... be happy with crumbs, main course takes a while due to my poor planning.

[UrbanSmooth]

Skinnee, when you publish the results, do you plan on holding an hour-long primetime press conference?

[Martinm210]

Thanks for clarifying. So if I understand correctly, you are using your own data? no offense to Martin, but I prefer that, as I verified your data before, and found it to coincide with ours.

Disregard..I'm slow.

[wesley]

IMHO the same thing could be done for all of water cooling testing including cpu/gpu c/w testing, but it would take several manufacturers to come to some agreements on the testing specification.

No one has stepped up to that plate yet..

Hopefully one day..

Hello Martin,

At least one of the biggies would have done it but will not provide the results because of impact on the bottom line.

Wes

[gabe]

BTW, did you guys ever get around to fixing the mcp655 basic vs variable curves. The basic model is stuck at setting 4 therfore not nearly as strong as a variable at 5.

I was hoping you could verify and clarify that. It misled me into buying the cheaper pump assuming they had the same top performance...bad assumption. According to some old Laing docs, the basic model is stuck at setting 4. Yet most suppliers/vendors list the same specs for both...always bugged me..

yes, a long time ago. Note the MCP655-B label stating P4.



Note: It was unclear to me that these two sets of curves were performed on different test beds, nor was I aware of the fact that Cam used his own data. What confused me was Eric's reference to the curves published at Performance PC.

[Martinm210]

yes, a long time ago. Note the MCP655-B label stating P4.



Note: It was unclear to me that these two sets of curves were performed on different test beds, nor was I aware of the fact that Cam used his own data. What confused me was Eric's reference to the curves published at Performance PC.

I see, sorry hadn't checked in a while. I still think the lines are too smooth/simplified but I'm picky that way, don't really matter in the scale of things.

Maybe adding another note to the max pressure written spec would help emphasize it. Just one that burned me when I first started water cooling and I can't let it go...Ill get over it..someday..

[Vapor]

I was referencing the product to show it was a single pump top, not so much the curves, sorry for the confusion

[zeroibis]

when will the good bread crumbs start falling skinnee?

Your shedding your results like a trail... we want the main course!

I feel like your avatar reading this thread at this point. I make sure to refresh this thread every night before I sleep and every morning when I wake up and still nothing, ahahahahhhh

I require results or there will be more closed space....

[Vapor]

FWIW, this is just the working thread, I suspect results will be in a different one