Does Sapphire's Vapor-X Technology work? You betcha!!

[SKYMTL]

I know that we have all seen a million Sapphire HD3870 Toxic reviews but I am going to be touching on something a bit different here.

The reason I am posting this here is because I think all of you Air Cooling fanatics will get a kick out of some of the results this little heatsink was able to pull off. The only problem is that Sapphire has the default fan speed set FAR too high and we set out to prove this.

Anyways, I think the results speak for themselves. Below I link to a few pages in the review since I am pretty sure you don't want to read through all 20 pages. ;)

First of all, there is the look at the HEATSINK ITSELF (http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/7245-sapphire-hd3870-512mb-toxic-edition-review-5.html) and it ain't pretty.

Here we have finally come to some very lurid and disappointing shots of this heatsink and let me tell you: it ain’t pretty. Copper corrosion is the name of the game here where it looks like the glue which holds the metal support plate in place has begun eating away at the copper itself.


Then its on to the INNER WORKINGS OF VAPOR-X (http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/7245-sapphire-hd3870-512mb-toxic-edition-review-6.html) which is basically a technology overview.

Vapor-X technology is basically a patented Sapphire technology which is used to cool off the GPU core using a vapor chamber. In a nutshell, a vapor chamber cooler will hold water which is vaporized by the heat generated by the core. This vapor will carry the heat to a condensation wick which will then be dispersed through the top plate and transported back to the base plate where the process repeats itself.


Finally, it is the TESTING PORTION (http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/7245-sapphire-hd3870-512mb-toxic-edition-review-17.html) where all of this comes together where the heatsink is tested at three different fan speeds.

http://images.hardwarecanucks.com/image/skymtl/GPU/SAPPHIRE-TOXIC/TOXIC-44.jpg

While it isn't 100% scientific, it shows that this really does work. Actually, better then even Sapphire thought it did.... :up:

[Spawne32]

wow, very interesting indeed, i wonder how well it could have worked had they actually put some effort into manufacturing

[STEvil]

The corrosion on the bottom of the heatsink is flux from the rosencore solder used to solder the hold-down plate to the base I think.

From the pictures I cant see any corrosion where the top and bottom plates are held together.

Also we'll need to have better testing done, that just isnt enough heat to really test this.

[SKYMTL]

Also we'll need to have better testing done, that just isnt enough heat to really test this.

Actually, it is being testing in the environment it was mean to be used in. It wasn't designed to take ridiculous amounts of heat, it was designed to work better than the stock heatsink which it does even though it is more compact than the heatsink it replaces.

If and when Sapphire releases a card with a heatsink designed to take higher heat loads then it will be tested in the same manner. Until then, it is more than obvious that the Vapor Chamber improves cooling efficiency.

[tiro_uspsss]

soo.. its like a giant flat heatpipe?

[SKYMTL]

soo.. its like a giant flat heatpipe?

In layman's terms, yes.

[it_burns_when_i_pee]

looks good

i think we could see more variations of heatpipe technology. they are, in effect, a self containted water cooling system with radiator

next way forward i believe is to have core's exposed both sides. effectively allowing a front and rear heatsink.

can you imagine the cooling potential if CPU's went back to the card state like the P2 / P3 era and heatsinks mated either side?

[Boogerlad]

where can we buy flat heatpipes?

[Retro]

Nice revew, SKYMTL:up:
It's interesting to see Sapphire's latest "Toxic" heatsink technology. Quite a leap from their earlier ATI cards which simply used an Arctic Cooling heatsink.
Very good temp. reductions, but it's a shame that a quieter, more efficient fan wasn't used. I suspect that with a better fan running at 100% a few more degrees could be shaved off:cool:

[STEvil]

Actually, it is being testing in the environment it was mean to be used in. It wasn't designed to take ridiculous amounts of heat, it was designed to work better than the stock heatsink which it does even though it is more compact than the heatsink it replaces.

If and when Sapphire releases a card with a heatsink designed to take higher heat loads then it will be tested in the same manner. Until then, it is more than obvious that the Vapor Chamber improves cooling efficiency.

I agree mostly except that the fin density may be what is "improving efficiency" moreso than the vapor chamber.

What would be a good test of the vapor chamber would be to test it on an HD3850 that is volt modded vs both the stock cooler (non vapor chamber) and a stock HD3870 cooler (non vapor chamber)

[SKYMTL]

I agree mostly except that the fin density may be what is "improving efficiency" moreso than the vapor chamber.

What would be a good test of the vapor chamber would be to test it on an HD3850 that is volt modded vs both the stock cooler (non vapor chamber) and a stock HD3870 cooler (non vapor chamber)

Now THAT I agree with 100%. The stock HD3870 heatsink is a joke and does do well to illustrate my point here that there is just no way Sapphire needed the fan speed running at 50%. It was this point moreso than the actual cooling potential I was trying to touch upon when I set out to do these few tests. Fortunately, it sort of morphed into something else when I saw what the temperatures looked like.

I should point you to the actual heat tests I did across the cards tested which you can see HERE (http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/7245-sapphire-hd3870-512mb-toxic-edition-review-16.html). The stock HD3850 heatsink at the original reference fan speed runs between 25% to 30%. Compare that to the Sapphire number posted above and I have to say one thing for sure: I don't want to be running a volt modded HD3850 on the stock single-slot cooler. ;)

[STEvil]

If the fan is only running at 25-30% though you have no idea of what the maximum performance of the cooler is.

Besides that the different fan designs (size, depth, blade pitch and angle, etc) could mean that fan at 25-30% works as well as the fan in the vapor cooler at 50%

So to really test you'd need some tests at a few of the speed settings (25,50,75,100?).... yay work! :rolleyes: ;)

[Spawne32]

If the fan is only running at 25-30% though you have no idea of what the maximum performance of the cooler is.

Besides that the different fan designs (size, depth, blade pitch and angle, etc) could mean that fan at 25-30% works as well as the fan in the vapor cooler at 50%

So to really test you'd need some tests at a few of the speed settings (25,50,75,100?).... yay work! :rolleyes: ;)

Yah i definitely liked your tests on the vaporx better, had alot more information...oh wait...you never tested it before :rofl:

[SKYMTL]

So to really test you'd need some tests at a few of the speed settings (25,50,75,100?).... yay work! :rolleyes: ;)

Yup. I know. 25% vs 25%, 35% vs 35% and so on and so forth. I just can't see the stock cooler of the HD3850 keeping up that well one way or another but it is definately worth a try. If I ever have the time. ;)

[cegras]

looks good

i think we could see more variations of heatpipe technology. they are, in effect, a self containted water cooling system with radiator

next way forward i believe is to have core's exposed both sides. effectively allowing a front and rear heatsink.

can you imagine the cooling potential if CPU's went back to the card state like the P2 / P3 era and heatsinks mated either side?

Well, more like phase change, actually.

[railmeat]

skymtl,
your site is VERY nice and informative.i like the 8800gts 512 totally flattening the ati cards into the ground,rofl sorry of topic but true.ati needs to pack it in already :p
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/7245-sapphire-hd3870-512mb-toxic-edition-review-8.html

back on subjet,that cooler owns!!!!

[STEvil]

Yah i definitely liked your tests on the vaporx better, had alot more information...oh wait...you never tested it before :rofl:

:welcome::clap:

[SKYMTL]

skymtl,
your site is VERY nice and informative.i like the 8800gts 512 totally flattening the ati cards into the ground,rofl sorry of topic but true.ati needs to pack it in already :p
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/7245-sapphire-hd3870-512mb-toxic-edition-review-8.html

back on subjet,that cooler owns!!!!

Thanks for the compliments.

I think I might have time this coming weekend to at least overclock the HD3850 and test its stock heatsink...but first I'll see how it does at stock clocks. :up:

[RADCOM]

I really thought this technology would be going somewhere but it hasn't been used on the 4800 series has it? I first read about it in Custom PC here's the article.http://www.custompc.co.uk/features/602193/vapour-chamber-cooling-explained.html

[LuckyNV]

Its here
http://www.hexus.net/content/item.php?item=15457

[BreeSpree]

Nice cooler and good temps for air. Maybe they could use flat heatsinks for other things as well. Like chipsets, or HDD's.....

[cegras]

The cooler uses Vapour Chamber Technology (VCT) and features three heatpipes and a thermally-controlled fan

What D:

[3oh6]

Nice cooler and good temps for air. Maybe they could use flat heatsinks for other things as well. Like chipsets, or HDD's.....
the Mushkin Ascent series heat sink is essentially just that, large flat heat sinks that act as a vapor chamber.

[Nanometer]

As Jermey Clarkson would say, it's like making an exquisite souffle and garnishing it with lard.

[philbrown23]

nice that sir is one kickass product!

[happychappy]

As Jermey Clarkson would say, it's like making an exquisite souffle and garnishing it with lard.
Top Gear ftw!:D