Swiftech Apogee GT vs the D-Tek FuZion CPU blocks @ CES

[Cathar]

why not? If we could make the block cheap enough? If watercooling goes mainstream cost of production could drop to a few dollars. Then OEM's would certainly adapt each solution to their particular needs.

If true (< $25 blocks), then yeah, that'd be nice.

Still, it comes to this. What is better from a customer perspective?

1 block @ $75 that lasts for 5 CPU changes and offers within 1C of the best possible across all the CPU's.

5 blocks @ $30 for each of the 5 CPU's, each of which is 1C better than the $75 block on their own CPU, but 1-2C worse on other CPU's.

I guess I've always been a subscriber of the first approach (but instead strived to get that 1C difference as small as possible).

[Turtle 1]

why not? If we could make the block cheap enough? If watercooling goes mainstream cost of production could drop to a few dollars. Then OEM's would certainly adapt each solution to their particular needs.



That's the nature of a good debate isn't it?

I am assuming than you are talking about buying your own casting equipment. I would think D-teck is paying around $8 for the base using an outside source for casting and about $5 for the middle and top block. than about $4 for packaging and handling . Barbs depends on qaunity. lets say $2 for 2. I would imagine D-teck has around $21 dollars in these units. Without mold cost.

Now to get better prices you have to do in house = equipment cost . Or higher quanities out of house.

Now I may be way off on these prices. But I will tell you guys when we release our blocks what my exact cost is. I already know its extremely low.

Mold cost I won't figure in but will show there cost. . The guys already know that the plastic derlin mold cost was just a bit over $5200.
And were casting our copper blocks in house. So I guess its going to be rather competive business.

[WeStSiDePLaYa]

ok, i still dont think that people understand this.


a flat base will warp even if the base it is being matted up against is also flat. due to the one base being larger than the other, and the pressure being applied.

therefor, it does not matter that certain ihs's vary, because it will happen with them all.

this even happens with bare cores! just on a minute scale.

i think it is a technology that already showing its usefullness, and is already being adopted by the newest leading blocks the apogeeGT and the fuzion. I also think the ocz block will implement this feature.

so i dont understand why people are fighting this idea. why fight improvement? these are the types of designs we need to cool IHS cpu's these days, and since ALL cpus have fairly permanent IHS(soldered), these are the blocks we need.

i for one am now going to figure out a way to warp my apogee block slightly.

[EnJoY]

If true (< $25 blocks), then yeah, that'd be nice.

Still, it comes to this. What is better from a customer perspective?

1 block @ $75 that lasts for 5 CPU changes and offers within 1C of the best possible across all the CPU's.

5 blocks @ $30 for each of the 5 CPU's, each of which is 1C better than the $75 block on their own CPU, but 1-2C worse on other CPU's.

I guess I've always been a subscriber of the first approach (but instead strived to get that 1C difference as small as possible).

But is there one block that can do it all anymore Stew? The Storm used to be that block. Low flow, high flow, IHS, bare die, it conquered. Now with dual cores and quad cores, everything has changed. Can one block be made anymore that can perform "the best" on every cpu?

[Turtle 1]

For the record (phylly_boy can confirm) I announced the bow technique first and clearly presented it as a one week old discovery. No engineering flags here. We plain and simple stumbled on this, then decided to hand it over to the community. After I made that announcement, Danny remarked that his block was also bowed, and provided some explanations that do not coincide with your above assumptions, but I will let Danny answer this if such is his inclination.



From a statistical point of view, probably true. But I also know for a fact that the top 30% of our customers are tweakers and will spend uncounted amounts of time to try to improve the perf of their systems by any means. The mere scale of what we have recorded (2 to 3C for us is huge) warrants that we disclose it to our extreme crowd.



Our parts are not cast. D-Tek parts are cast. Our parts are suplied flat in stock config.



"Disappear" is wrong choice of words. "Reduce" is likely to be a more adequate description.

Sorry Gabe I was referring to the D-tek blocks. I understand perfectly what happened with your blocks. I think also gabe your going to have to go to castings to be price competitive.

[STEvil]

Arguable. since joint between IHS and silicon is now soldered, amount of deflection in the contact patch area could remain negligible. Whatever the case may be, the problem is there is nothing we can DO about it :-(

Exactly. One series may be concave, next could be convex, and later they could get their process fixed and have them nice and flat.

All the waterblock makers can do is make blocks that perform as best they can on the multitude of configurations and then let the end user do with as they please with the product.


Oh yeah, any easy way of getting that stainless steel "U" out of my MCW6002 so I can build a 6002-FuZion hybrid?

Probably just going to have to drill it out

[nikhsub1]

I'm interested to see how a bowed base will perform on a single die CPU like a 6700.

[Turtle 1]

ok, i still dont think that people understand this.


a flat base will warp even if the base it is being matted up against is also flat. due to the one base being larger than the other, and the pressure being applied.

therefor, it does not matter that certain ihs's vary, because it will happen with them all.

this even happens with bare cores! just on a minute scale.

i think it is a technology that already showing its usefullness, and is already being adopted by the newest leading blocks the apogeeGT and the fuzion. I also think the ocz block will implement this feature.

so i dont understand why people are fighting this idea. why fight improvement? these are the types of designs we need to cool IHS cpu's these days, and since ALL cpus have fairly permanent IHS(soldered), these are the blocks we need.

i for one am now going to figure out a way to warp my apogee block slightly.

Just use a larger diameter oring it will deflect.

[gabe]

[QUOTE=WeStSiDePLaYa]gabe, seems to understand, but he appears to work for a company. (swiftech? )
QUOTE]

My apologies, I assumed everyone knew that I was Swiftech's founder. Not too used to forums etiquette..

I will change my profile to reflect this fact as soon as I find out how to do it :-)

[Cathar]

But is there one block that can do it all anymore Stew? The Storm used to be that block. Low flow, high flow, IHS, bare die, it conquered. Now with dual cores and quad cores, everything has changed. Can one block be made anymore that can perform "the best" on every cpu?

Well, I did say 5 CPU's, so yes, there does need to be change.

I've never said that one block can do it all. Still, cooling CPU's doesn't involve large differences in cooling size. It's not like we're cooling things that differ in size by 100x.

Can we do it all with 1 block? No. Well, no, we probably could, but it'd likely have some pump requirements that fall outside of what hobby pumps can do. This is really the main issue then, can we do it all given typical pumps?

Could we do it all with 2 blocks. I think so. Well, at least until the CPU makers start pumping out wafer CPU's that are 2"x2" in size, with 32 CPU's on the wafer, and of which 0-6 of them don't work. At that point, we'd all be going back to the sorts of designs that work best with 50x50mm TEC's.

We don't know what the future holds, true, but we can still design products for the mid-term.

[cyberspyder]

so after all the debates, should I still purchase the FuZion over the Apogee GT (sorry Gabe )? I have an E6300, so will this make a difference over the QX6700?

Brendan

[gabe]

I'm interested to see how a bowed base will perform on a single die CPU like a 6700.

we noted 2C improvement on our 6600 test bed

[Plywood99]

Exactly. One series may be concave, next could be convex, and later they could get their process fixed and have them nice and flat.

All the waterblock makers can do is make blocks that perform as best they can on the multitude of configurations and then let the end user do with as they please with the product.


Oh yeah, any easy way of getting that stainless steel "U" out of my MCW6002 so I can build a 6002-FuZion hybrid?

Probably just going to have to drill it out

I did that. Heated the block up on a gas stove till the solder joint melted. Then used a small screwdriver to lightly knock the cover from the base. I think the U is made of brass.

Tricky part was getting it all soldered back together. Clean surfaces then use flux to help wick solder into the joint...


Ply

[Turtle 1]

so after all the debates, should I still purchase the FuZion over the Apogee GT (sorry Gabe )? I have an E6300, so will this make a difference over the QX6700?

Brendan

Buy the cheapest and the 1 that appeals to you the most. Looks count.not much but does count.

[EnJoY]

My apologies, I assumed everyone knew that I was Swiftech's founder. Not too used to forums etiquette..

I will change my profile to reflect this fact as soon as I find out how to do it :-)

Send Charles a pm, he'll place you in the vendor user group and give you a custom title so that there's no confusion.

[Fr3ak]

Send Charles a pm, he'll place you in the vendor user group and give you a custom title so that there's no confusion.

He doesnt need to, because it is already taken care of

[Barb]

Are there temperatures for the same cpu with the original apogee and/or storm?
I'm curious how they compare to the apogee gt and the fuzion.

[gabe]

The Apogee GT specifications are now posted on our web site.

[Praxis1452]

it seems that the Fuzion is 55.25mm^2 so a 50x50mm TEC will fit but barely. How large is the apogee GT? same size as before so it'd be a bit smaller yes?

[cyberspyder]

The Apogee GT specifications are now posted on our web site.

Thanks alot BTW, what are you guys setting the price to?

it seems that the Fuzion is 55.25mm^2 so a 50x50mm TEC will fit but barely. How large is the apogee GT? same size as before so it'd be a bit smaller yes?

Is this including the screws?

Brendan

EDIT: NVM, saw the MSRP on the site.

[nikhsub1]

I'm curious how they compare to the apogee gt and the fuzion.

I will have those numbers tomorrow... on the last test with the GT the machine kept BSOD'ing. Need to remount.

[Marlowe]

I think it's interesting how one block makes a computer bsod and one don't..

I don't remember what it was about, but I read a review a good time ago where a certain heatsink gave them higher overclock than another, even tho temps were not that different. They couldn't explain it and neither can I :P

Funny to see the same thing here

What about trying max clock with both blocks?

[nikhsub1]

I think it's interesting how one block makes a computer bsod and one don't..

I don't remember what it was about, but I read a review a good time ago where a certain heatsink gave them higher overclock than another, even tho temps were not that different. They couldn't explain it and neither can I :P

Funny to see the same thing here

What about trying max clock with both blocks?

I wouldn't read into it too much... probably just a bunk mount, it does happen, even to the best of us. Tomorrow first thing I will remount and run the LAST test for the GT. I have 3 tests of each done for the G5, Fuzion, Fuzion with nozzle and the Apogee GT. I await the Swiftech G4 (from Bill) and a thicker gasket (from Gabe). I plan to use this new gasket not only in the GT either

[gabe]

After paying more than $1200 for it, do you blame me for being hesitant ?

Lapping properly is actually fairly straight forward. Key factor is the work surface. You absolutely need a glass table as this is the flatest surface available in a house.

Tape a sheet of automotive 600 grid wet/dry sandpaper. Use masking tape all around so your sheet will not move around.

place the protection plastic piece that came with your CPU on the back of it, to avoid damaging anything,

place index and thumb of each hand at each corner, and slide the CPU back and forth making sure that you apply equal pressure at each corner.

finish off by repeating procedure with 1200 grit or better.

clean off surface and joint around CPU with alcohol.

verify flatness with rasor blade. looking against a bright light.

train with an older CPU if you don't want to take chances with the Kent..

[SiGfever]

The passion and desire to get that .1c advantage on our systems is why we spend so much time on forums such as this, I for one enjoy the journey with its up and downs, the system gets very boring when it is left alone.

There are so many quality blocks on the market now it makes it very tough to choose between them. Testing from independents such as nikhsub1, Free Cable Guy, and H20 frag-monger help us get an unbiased look at the possibilities but their test are unique and not the "ultimate truth". The day that all CPUs are made to the same exact specifications then the "one testbed" will provide more definite answers. But until that time every test and piece of data helps put the puzzle together.