Apogee from Swiftech...

[AngryAlpaca]

This quality control question raises a question: how good/bad were previous blocks? I don't think I've seen close scrutiny of any other block's insides.

[gone_fishin]

Originally Posted by Orkan

I'm not out any money... My fame? lol

Nice productive post... troll.

[deep voice]Trollin trollin trollin, get your bodies trollin, trollin trollin trollin RAWHIDE!!!!..........[/deep voice]

[Annirak]

Originally Posted by bigben2k

The only person I know that does water-in water-out is Derek (pHaestus).

Well, if someone's already doing it, then I'm just going to butt out now. No point in tossing around ideas that people are already using.

One point to note is that I believe the AMD user accessible thermal diode is, like the intel one, located in a relatively cool section of the die. It may be misrepresenting the actual temperature of the hotter parts of the die.

[bigben2k]

Originally Posted by Annirak

...One point to note is that I believe the AMD user accessible thermal diode is, like the intel one, located in a relatively cool section of the die. It may be misrepresenting the actual temperature of the hotter parts of the die.

Don't know about AMD, but I believe that the Intel one has been demonstrated, indeed.

It isn't pleasant to know, but a heat die is not forgiving that way. It then comes down to applying a "correction" to match what a CPU reports. Either way, it should give anyone "something" to measure.

[ricecrispi]

Originally Posted by Swiftech

You might be interested to learn that we have since conducted severe stress tests on a number of housings with 100% negative results (no failures). I actually sent a response to Lee's (Robotech) concerns, including videos of the stress test procedure, featuring yours truly hammering these blocks like a madman It's actually funny, specially the part when I hit my fingers with the hammer. I hope they post it.

Seriously, and for the record every single block is pressure tested. How your block escaped QC is still under investigation. Our organization is composed of people. People sometimes make mistakes. We learn from them, and try not to repeat them.

My experience with acetal/delrin is it's good at absorbing impact (at certain angles). It doesn't responded as well as stress loading. That would be a better way to test the block. Lay it on angle so it's resting at the top on the outside edge of the top and edge of the inlet/outlets edge and press down on it. I thick the stress (weight) at that angle might form a crack over time. It will take the whack with hammer very well. Used in paintball for years now.

[Althornin]

Originally Posted by gone_fishin

[deep voice]Trollin trollin trollin, get your bodies trollin, trollin trollin trollin RAWHIDE!!!!..........[/deep voice]

Thats been you this whole time.
You pop in, post some shit, and 100% do nto respond to the arguments against what you've said.
You used to be a decent poster -- what happened?

[Albigger]

@ annirak:

guess I should've looked more before posting so much. Here: relevant threads, but old, tried to calculate wattage. Still reading the procooling one:
http://forums.overclockers.com.au/sh...threadid=60614
and
http://forums.procooling.com/vbb/showthread.php?t=8892

Have to finish reading but may be nice to also try to measure/calc. wattage from newer cpus...

Edit: Haven't read this one either but seems pertinent
http://forums.procooling.com/vbb/showthread.php?t=8433
(seems like I try to read one thread only to be linked to 20 others that I have to subsequently read in order to hope to understand...)

[bobo5195]

beware of the second law of thermodynamics as it can effect C/W. I would say more but second law stuff still makes my head hurt.

I found a thermal distribution map for an old p4
http://www.inqst.com/articles/athlon4/article.htm
Clearly not a uniform temprature distribution.

There is also one for the sun niagara (8 core cpu, low transistor count per core) on www.aceshardware.com. But im guessing that is of no interest what so ever.

[Brians256]

Originally Posted by bobo5195

beware of the second law of thermodynamics as it can effect C/W. I would say more but second law stuff still makes my head hurt.

I found a thermal distribution map for an old p4
http://www.inqst.com/articles/athlon4/article.htm
Clearly not a uniform temprature distribution.

There is also one for the sun niagara (8 core cpu, low transistor count per core) on www.aceshardware.com. But im guessing that is of no interest what so ever.

This is a better link, bobo, as it points out the positions of the thermal diode vs. the actual hotspot: SPCR Article

[Swiftech]

Originally Posted by ricecrispi

My experience with acetal/delrin is it's good at absorbing impact (at certain angles). It doesn't responded as well as stress loading. That would be a better way to test the block. Lay it on angle so it's resting at the top on the outside edge of the top and edge of the inlet/outlets edge and press down on it. I thick the stress (weight) at that angle might form a crack over time. It will take the whack with hammer very well. Used in paintball for years now.

concur. we've had several blocks in a vise under an estimated 400lbs of pressure for 4 days now. no cracks. Not surprising when one understands that the 0.030 spot is at the apex of an eliptic funnel. A cross-sectional view would show that the thickness increases as one moves away from the apex of the elipse until it reaches the maximum wall thickness of 0.045.

This block may have had a defect in the material in this area, but I can only guess at this point, as the customer never sent his block back for inspection.

[GlassMan]

Originally Posted by Swiftech

concur. we've had several blocks in a vise under an estimated 400lbs of pressure for 4 days now. no cracks. Not surprising when one understands that the 0.030 spot is at the apex of an eliptic funnel. A cross-sectional view would show that the thickness increases as one moves away from the apex of the elipse until it reaches the maximum wall thickness of 0.045.

This block may have had a defect in the material in this area, but I can only guess at this point, as the customer never sent his block back for inspection.

What was the load (ft-lbs) "specified", how many repetitions?
No, I didn't think so. If anyone thought about it it would have an 1/8" minimum radius, because it would have cost nothing to do it right.

I should have specified torque, I thought my post made it clear. Squeezing it in a vice is not realistic, much like hammer blows. Posting such stuff is a waste of time.
If you want to demonstrate how strong it is put a 12 1/2" rod 1/2 inch into the opening and apply a load from side to side. Say 25lbs load. (You can impress with 50lbs) How many cycles does it last?

[BillA]

what is the stupid point here ?
why not a 36" arm ? or 72" ?
and now you move to the acrylic wbs of the world ?

what is the max possible bending load imparted through 2 flexible hoses ?
shit, hang the computer off them; add for dynamic loads (dropping the computer on the barbs ?)
what saftey factor do YOU want ?
what is the relation of that to the procedure you defined ?

torque ? with an NPSM thread ?

go to ignorantcooling.com

[AngryAlpaca]

Doesn't exist.

[bobo5195]

I believe the point bill is that a screw thread can apply an awful lot of force and that alternating loads will cause failure at much lower loading values than being hit by the hammer. For plastics I would say that screw loading is more destructive than hammer loading. The combination of plastics, warm water and additives, dynamic loading, temperature loading, possible manufacturing defects and a stress concentration MAY cause something.

Given the testing that gabe has undertaken I would say that the above concerns are valid but don’t match reality (ie don’t worry). The broken wb (orkans) is probably a 1 in million chance from something or other. If more start popping up then there maybe a problem. Not specifying a nice radius is abit of school boy error, but its not catastrophic and you could make a point for its inclusion as a matter of aesthetics. Either way looks like there is nothing in this.

[AngryAlpaca]

I thought Bill's point was that almost every other block out there has NOT been tested in a thorough manner as this has, and yet no one asks for it, and it's simply not relevant 'cause we don't mount our blocks using hammers/6 foot torque wrenches at 400 pounds load.

[bobo5195]

and my point was that it doesnt matter because it passed the test.

Admitly the block is under going undue inspection but inspections that it has past. This is a good thing.

[maxSaleen]

I don't recall anyone testing the Koolance tops to the degree that the Apogee has had its top tested. Anyone here ever held a Koolance block? The whole plastic top is about as thick as the area of the Apogee top that has caused such controversey.

[GlassMan]

what is the stupid point here ?

To duplicate , within reason, the forces that will be applied to the turret(s) that the fittings screw into when 7/16' tubing is put on the 1/2" fittings.

why not a 36" arm ? or 72" ?

75 or 150 ft-lbs is a little excessive, ie not realistic.

and now you move to the acrylic wbs of the world ?

Not sure what you mean, but they are not sold with 7/16" tubing included in the kit, and any one buying acrylic accepts the risk of a brittle material.

what saftey factor do YOU want ?

Personally, I don't care because I'm not interested an Apogee. But I think being assured you could install and remove the tubing 10 times without cracking would be a minimum "saftey factor". By my test 50 reps /joint at 50 ft-lbs would be impressive, as I said 25 ft-lbs would be sufficient. Swiftech hasn't shown any safety factor, the tests refered to by swiftech in no way duplicate the loads that will be applied in use.

what is the relation of that to the procedure you defined ?

Not sure what "that" refers to, maybe I already answered it.

what is the max possible bending load imparted through 2 flexible hoses ?
shit, hang the computer off them; add for dynamic loads (dropping the computer on the barbs ?)

Not worried about either of those loads, are you?

torque ? with an NPSM thread ?

I was not refering to the force relating to sealing the O-rings. Not worried particularly about stripping out the threads or twisting the turret out.

go to ignorantcooling.com

Sorry, can't find it BillA. I guess I'll keep looking for answers here.


As you view the picture insert the bar into the threads. If you want to take the time, thread the bar and screw it into the threads. Apply the force left to right. If the delrin bends in the circle it will crack at the corner. This test will apply the same forces as rocking tubing onto the fitting that is normally screwed into the threads. 25 ft-lbs is more than I can apply (lever arm is a couple of inches, forces higher) but not as excessively showy as a vise or hammer blow.
I'm not the one that came in talking about how it met specs., and then testing with hammer blows
Frankly Bill, I'm disappointed in you. I thought you were all for appropriate testing.

[Budwise]

This is Gabes post ealier this evening at Xtremesystems.org:


The following tests have been conducted specifically for the purpose of verifying the allegations published by a forum member (Orkan) comparing overclocking performance between an Apogee and a Storm water-block, where it was reported that under the same circumstances (same CPU/motherboard platform etc..)the Storm yielded a 100Mhz higher overclock than the Apogee (2.8 Ghz instead of 2.7 Ghz).

The reason why we questionned this report in the first place was the scale of the alleged difference. In effect a 100 Mhz increase on a CPU operating at 2.7Mhz is a 3% increase in processor frequency. Why doubt such a difference ? An Integrated Circuit Industry rule of thumb states that an IC will gain an average of 2 to 3% in frequency for every 10°C drop in operating temperature. We have practically verified this rule many, many times. So, while the actual CPU temperatures from one testing methodology to another can be argued ad nauseam, the report published by this user suggested to us that the CPU was actually "seeing" about a 10C drop in junction temperature, in effect resulting in this 3% frequency increase.

This, we knew to be highly implausible.

Our tests were conducted in the same fashion as a typical overclocker would.

Equipment:
Motherboard: Asus A8N Sli Deluxe
CPU: AMD A64 X2 4400+ - Unmodified (with IHS)
Waterblocks: Storm and Apogee taken straight from inventory and used "as is"
Radiator: Triple 120mm CF prototype with fans operating at 5 Volts (silent mode)
Pump: MCP655
Vcore: 1.52V (motherboard max)
OS Window 2003 Server

Since CPU overclock was the main object of this test, the memory was set to a 1/2 ratio in order to make sure that no instability would result from the memory.

Processor Load was induced by using 2 tasks of CPU Burn, one per core (Set Affinity function).

The CPU temperature was read using Asus Probe. The Air temperature was measured with an Omega thermometer and its thermocouple placed 1 foot away from the fans intake.

OC procedure:
Starting windows at 2750MHz (250*11), starting CPUZ (v1.30), set 2 tasks of CPU burn (as described above).
If no crash after 20minutes, record the results, launch clockgen and increased the HTT by one increment.
Repeat until windows crashed.

Important Note: this procedure does not provide a full measure of CPU stability but we feel is perfectly adapted to indicate overclockability within a relatively short period of time.

This being said, the results are:

With Storm, we started at 2750, and subsequently passed 2755, 2764, 2771 and reached an immediate fail at a 2781Mhz setting. So for the Storm the final "stable" overclock was 2771MHz with a final air temperature of 23.1C. The CPU temperature (reported by Asus probe) occillated between 47C and 48C.

With Apogee, we passed 2750, 2755, 2764, 2771 and 2781MHz. Windows crashed finally at 2788MHz after 10minutes. So the "pass" for this test is 2781 at an air temperature of 23.2C and CPU temperature stable on the 47C mark.

To us, the above results simply indicate one thing: there does not appear to be ANY significant difference in overclocking abilities between these two water-blocks and with this processor.

[flatline]

"Our tests were conducted in the same fashion as a typical overclocker would.
Equipment:
CPU: AMD A64 X2 4400+ - Unmodified (with IHS)"


whats a "typical overclocker"

[Orkan]

Differs greatly from my own results.

[Budwise]

before you start tearing him down and nitpicking the details, at least give credit to the fact that he is following through. He is doing his best to satisfy our requests. How many other business owners do you know that would take the time to do this? We asked for more testing, and at least we see that more is being done whether its high tech testing or a simple max OC test.

[Albigger]

Originally Posted by Budwise

before you start tearing him down and nitpicking the details, at least give credit to the fact that he is following through. He is doing his best to satisfy our requests. How many other business owners do you know that would take the time to do this? We asked for more testing, and at least we see that more is being done whether its high tech testing or a simple max OC test.

I didn't know people asked for more results (from Swiftech specifically) and was thinking the exact opposite: Why is Gabe going through ALL this trouble to assure us of the abilities and soundness of construction of this block? For one user (Orkan)? I mean people post negative comments in forums all day long about products and usually those are the ones that won't believe company data anyhow - they'll test it themselves or wait for others to test it.


But I guess you're right - at least they (Swiftech) are adressing the issues to the best of their abilities.

Edit: Also - at least the procedure for testing stability was defined.
Orkan - saw you popped the top on your X2 - any chance of repeating the apogee vs storm overclock test now without the IHS? Maybe also using Gabe's method (instead of prime)?

[nikhsub1]

Originally Posted by Albigger

Orkan - saw you popped the top on your X2 - any chance of repeating the apogee vs storm overclock test now without the IHS? Maybe also using Gabe's method (instead of prime)?

Yeah this would be a good test. I would bet the storm should really shine over the apogee with no IHS... let us know orkan. HTFU already.

[GlassMan]

The on die memory controller puts its own limit on OCing.