[Review] PrimoChill Typhoon III

[millertime359]

Technically, that picture does show something relevant, if you take the time and have the capacity to understand the data. I have already seen people read the initial review, have the opportunity to view that photoshopped graph, and still ask if the T3 will deliver 3+ gpm to a serial loop configuration.

The issue isn't with the T3. It's with the review. The data points selected for display appear to have been chosen to deliberately paint the T3 in the most positive possible light.

Does the T3 appear to deliver superior performance when set up in a parallel configuration and then put against its competition in a serial config? Yes.

Is the T3 the only "off the shelf" single product purchase available on the current market capable of setting up an effective parallel loop? Yes.

Does the T3 sprinkle magic dust on the D5 to make it perform more than TWICE as well as any of the other aftermarket pump top offerings? No data has been presented, but all extrapolation of available information would tend to indicate NO.

An objective product review shouldn't force this sort of extrapolation on the casual viewer.

Yes, the review could have been done better. Skinnee did us all a favor by pushing this reveiw ahead of every other one he had and pushed to get us the data as fast as he could. He even admitted that he had issues figure out the best way to test it

You have to remember something though, this is something new. There is only one other product on the market atm that is comparable to this that I have seen. The XSPC dual bay res for the DDCs.

Vapor has also confirmed Skinnee's findings with math too. Norris also has confirmed that temps are similar to a true dual loop.

I do agree that there was alot of hype around this product, and there is a need to do some further testing. I also agree that you more than likely not get 10 degrees from your loop by installing it.

So lets stop with the bashing of it. If you get one, post your results. If your not going to get one, then lets leave it at that.

[Warfarin88]

Yes, the review could have been done better. Skinnee did us all a favor by pushing this reveiw ahead of every other one he had and pushed to get us the data as fast as he could. He even admitted that he had issues figure out the best way to test it

You have to remember something though, this is something new. There is only one other product on the market atm that is comparable to this that I have seen. The XSPC dual bay res for the DDCs.

Vapor has also confirmed Skinnee's findings with math too. Norris also has confirmed that temps are similar to a true dual loop.

I do agree that there was alot of hype around this product, and there is a need to do some further testing. I also agree that you more than likely not get 10 degrees from your loop by installing it.

So lets stop with the bashing of it. If you get one, post your results. If your not going to get one, then lets leave it at that.

I never have bashed the product itself. It appears to do what it was designed to do. I've also never questioned skinnee's numbers, just the way in which they were presented and/or omitted.

I also can sympathize with skinnee to a small extent. While bastardizing his graphics, I saw first hand how difficult it is to appropriately compare the T3 to other comparable offerings, since, honestly there aren't any.

The test loop bar graphs are on the right track, but as I've stated before, I still feel they are misleading.

I'm curious if a stock pump was run on any of the test loops. Using that as a baseline, one could then chart the % of flow improvement of the D5, Detroit, and T3 in a serial configuration, and then also include the % flow improvement of separate loop sections when the T3 was set up in a parallel config.

I think this would give a much more "real world" impression of what the T3 actually does.

[millertime359]

I never have bashed the product itself. It appears to do what it was designed to do. I've also never questioned skinnee's numbers, just the way in which they were presented and/or omitted.

I also can sympathize with skinnee to a small extent. While bastardizing his graphics, I saw first hand how difficult it is to appropriately compare the T3 to other comparable offerings, since, honestly there aren't any.

The test loop bar graphs are on the right track, but as I've stated before, I still feel they are misleading.

I'm curious if a stock pump was run on any of the test loops. Using that as a baseline, one could then chart the % of flow improvement of the D5, Detroit, and T3 in a serial configuration, and then also include the % flow improvement of separate loop sections when the T3 was set up in a parallel config.

I think this would give a much more "real world" impression of what the T3 actually does.

My apologies War, I guess I may have read a little in your post. Some other posters got me a little wound up with the bashing is all.

[Darth_Penguin]

Technically, that picture does show something relevant, if you take the time and have the capacity to understand the data. I have already seen people read the initial review, have the opportunity to view that photoshopped graph, and still ask if the T3 will deliver 3+ gpm to a serial loop configuration.

The issue isn't with the T3. It's with the review. The data points selected for display appear to have been chosen to deliberately paint the T3 in the most positive possible light.

Does the T3 appear to deliver superior performance when set up in a parallel configuration and then put against its competition in a serial config? Yes.

Is the T3 the only "off the shelf" single product purchase available on the current market capable of setting up an effective parallel loop? Yes.

Does the T3 sprinkle magic dust on the D5 to make it perform more than TWICE as well as any of the other aftermarket pump top offerings? No data has been presented, but all extrapolation of available information would tend to indicate NO.

An objective product review shouldn't force this sort of extrapolation on the casual viewer.

I have taken the time to look at the graphs, read both reviews, view the photoshopped image, and read EVERYONE'S opinions. I certainly have the capacity to understand what this Res is capable of.

That graph shows the T3 only set up with one serial loop. Yes? The T3 was DESIGNED to use 2 loops in parallel. Yes? When set up correctly, does the T3 flow more to the individual components of each loop? You yourself have said Yes. I have never said anywhere that the T3 flows twice as much as other tops. I have simply stated that when set up correctly the T3 flows MORE. More flow = Better performance. How anyone can argue otherwise is beyond me.

If product A, B, C only allow 1 inlet and 1 outlet to the pump but product D offers a solution that gives you 2 allowing for multiple loops and allowing for better performance on the same pump why cant it be reviewed just the way the products were intended. There is no forced extrapolation Or deliberate attempt to fudge the numbers. He reviewed each product as it was intended to do and I'm sorry you cant swallow that. Maybe he shouldn't have added the flows together but even when not added up they still flow more to said components then the other pump tops.

[Warfarin88]

I think I may have implied in my post that I was directing my rhetoric at you personally Penguin.

That wasn't the intention.

[NaeKuh]

Once the again.. the bible of our hobby.

You get what applies to your setup and your setup only. Not what applies to another person's setup who ur seeing on a chatbox, or internet window.

[Boogerlad]

Yes I was running a true dual loop. CPU only loop with a MCP355 and XSPC top, Swiftech Micro Res and a PA120.3. The GPU loop was 3 x EVGA Hydro Copper blocks, MCP355 with XSPC top, Swiftech Micro Res and a PA120.3.

Were ambients the same and not remounted blocks? Really informative post above.

[Utnorris]

Ambients were the same and no, I did not remount the blocks. This is one reason I use QD's, makes it easier.

[Darth_Penguin]

I think I may have implied in my post that I was directing my rhetoric at you personally Penguin.

That wasn't the intention.

No Problem.

[Boogerlad]

Ambients were the same and no, I did not remount the blocks. This is one reason I use QD's, makes it easier.

Thanks. T3 is looking attractive now. t3>true dual loop>single loop, single pump?

This might be a very stupid question, but would performance be nice if you combined both inlets and outlets, or would it then become just a regular top?

[shazza]

Thanks. T3 is looking attractive now. t3>true dual loop>single loop, single pump?

I don't think you can make that generalization. And, from what I understand of what BoxGods has posted, he's not saying that either. Every system is going to behave somewhat differently.

For example, I have the i7 920 running in it's own loop with a PA120.3, the MCP355 pump and XSPC res top. I also have 3 x 285s running in their own loop - again, with the PA120.3, MCP355, and XSPC res top. I can't assume from the data presented so far that I will get better temps if I switch to one D5 and a Typhoon III. But then again, I might get comparable performance - won't know for sure til I try it. I may actually try this at some point just for kicks, but probably not in the near future.

[iandh]

I don't think you can make that generalization. And, from what I understand of what BoxGods has posted, he's not saying that either. Every system is going to behave somewhat differently.

For example, I have the i7 920 running in it's own loop with a PA120.3, the MCP355 pump and XSPC res top. I also have 3 x 285s running in their own loop - again, with the PA120.3, MCP355, and XSPC res top. I can't assume from the data presented so far that I will get better temps if I switch to one D5 and a Typhoon III. But then again, I might get comparable performance - won't know for sure til I try it. I may actually try this at some point just for kicks, but probably not in the near future.

This is pretty much how I feel about it. I really respect this product for what it is, but I hate to see people buying it expecting a major change in their loops temp performance, it may or may not happen depending on many factors. For those with pre-existing loops, chances are that the $60-70 would be much better spent on an additional radiator.

For those building a new loop, this thing is THE choice no doubt, assuming it works for their purposes. I've never been a huge fan of bay mounted reservoirs myself, so I may never end up using one of these.

[millertime359]

Thanks. T3 is looking attractive now. t3>true dual loop>single loop, single pump?

If you want to make a generalization, I would more say a T3 (set up in parallel loops) can perform as well as a true dual loop setup. Norris just saw a couple of degrees from the GPUs. I would consider that to be an improvement with just "a seat of the pants" test like Norris did. I does show it performed as well as a dual loop did in his configuration.

This is still a great thing, seeing you don't have to buy 2 res's and 2 pumps. Therefore it is a little cleaner of a setup and cheaper.

[Manicdan]

i think the definition of a true duel loop needs to be defined better

the T3 is 2 pumps sharing a res, not 2 independent loops. if you have a CPU and GPU on separate loops, using separate radiators, then switching to a T3 can give you the option to balance things out, or give more flow to certain blocks.

so for example you have a CPU, GPU and 2 radiators, and currently have CPU/Rad1 on loop A, and GPU/Rad2 on loop B, the flow of loop A may be very restricted because of the block you purchased, if you move Rad1 to loop B, the CPU block gets more flow and temps get better, with the T3, two independent loops wouldnt allow for that since your cpu would be all alone, but 2 pumps using one res would.

[millertime359]

i think the definition of a true duel loop needs to be defined better

the T3 is 2 pumps sharing a res, not 2 independent loops. if you have a CPU and GPU on separate loops, using separate radiators, then switching to a T3 can give you the option to balance things out, or give more flow to certain blocks.

so for example you have a CPU, GPU and 2 radiators, and currently have CPU/Rad1 on loop A, and GPU/Rad2 on loop B, the flow of loop A may be very restricted because of the block you purchased, if you move Rad1 to loop B, the CPU block gets more flow and temps get better, with the T3, two independent loops wouldnt allow for that since your cpu would be all alone, but 2 pumps using one res would.

I think you have yourself a little confused there. The T3 only has 1 pump, not 2.

[BoxGods]

One of the things I find kind of funny is the double standards the detractors are using. Testing proves that in a serial loop configuration the T3 moves a significantly larger amount of water, so they said well the other tops would do just as well if you bought a reservoir and Y's. Even doing that the T3 puts out more flow by a significant amount. So then they switch to the argument that it is not a TRUE dual loop but rather a parallel loop and that a true dual loop is a lot better. When users who had true dual loops (2 pumps) posted that they swapped in a T3 (with only one pump) their temps held.

Then the argument was that adding the flow of the two loops together is not accurate, even though it was listed as total system flow. If you average the loops T3 STILL moves a lot more water. Now they are saying the single loop performance is not as good as other tops. It IS in the hunt though and earlier they said that the HUGE gains in flow were irrelevant as flow doesn't matter that much...yet some how the relatively small difference in flow in the single loop DOES matter all of a sudden?

After all that I feel pretty good about the design because the T3 can obviously take a punch (and even the cheap shots) and is getting people excited and looking for different ways to do things, which is what water cooling is really all about =)

No word from Rocky on the face plate renders yet. Will post em as soon as they are done.

[Waterlogged]

i think the definition of a true duel loop needs to be defined better

the T3 is 2 pumps sharing a res, not 2 independent loops. if you have a CPU and GPU on separate loops, using separate radiators, then switching to a T3 can give you the option to balance things out, or give more flow to certain blocks.

so for example you have a CPU, GPU and 2 radiators, and currently have CPU/Rad1 on loop A, and GPU/Rad2 on loop B, the flow of loop A may be very restricted because of the block you purchased, if you move Rad1 to loop B, the CPU block gets more flow and temps get better, with the T3, two independent loops wouldnt allow for that since your cpu would be all alone, but 2 pumps using one res would.

Actually, we need some new terminology to describe the loops of a T3. Something like "symbiotic serial" might be a better phrase.

[millertime359]

One of the things I find kind of funny is the double standards the detractors are using. Testing proves that in a serial loop configuration the T3 moves a significantly larger amount of water, so they said well the other tops would do just as well if you bought a reservoir and Y's. Even doing that the T3 puts out more flow by a significant amount. So then they switch to the argument that it is not a TRUE dual loop but rather a parallel loop and that a true dual loop is a lot better. When users who had true dual loops (2 pumps) posted that they swapped in a T3 (with only one pump) their temps held.

Then the argument was that adding the flow of the two loops together is not accurate, even though it was listed as total system flow. If you average the loops T3 STILL moves a lot more water. Now they are saying the single loop performance is not as good as other tops. It IS in the hunt though and earlier they said that the HUGE gains in flow were irrelevant as flow doesn't matter that much...yet some how the relatively small difference in flow in the single loop DOES matter all of a sudden?

After all that I feel pretty good about the design because the T3 can obviously take a punch (and even the cheap shots) and is getting people excited and looking for different ways to do things, which is what water cooling is really all about =)

No word from Rocky on the face plate renders yet. Will post em as soon as they are done.

It did get hyped up a little bit too. Even though the hype was just regarding flow, people read into it as it would distractly improve temps. Therefore, it seems they are looking for anyway to shoot it down.

It is a great option to have out there though, plan and simple. Is it for everyone, no.

Once again, great work BoxGods. I am looking forward to seeing those renderings when you get them

[BoxGods]

Actually, we need some new terminology to describe the loops of a T3. Something like "symbiotic serial" might be a better phrase.

Or Trill Parallel LoL. Course if your not a Deep Space Nine fan the DAX reference is lost.

Differential Parallel Loops.

[Waterlogged]

Or Trill Parallel LoL. Course if your not a Deep Space Nine fan the DAX reference is lost.

Differential Parallel Loops.

Wasn't into Dax, now Jadzia, I'd like to break me off a little piece of that!

[JOCKTHEGLIDE]

Once the again.. the bible of our hobby.

You get what applies to your setup and your setup only. Not what applies to another person's setup who ur seeing on a chatbox, or internet window.

THATS NO FUN..that means no bashing on the boards then darn it

[Manicdan]

I think you have yourself a little confused there. The T3 only has 1 pump, not 2.

if i add the word "like" to that first sentence, it would have made more sense, i was referring to what the T3 should be compared to, not what it is.

[brinox]

i think the best thing about the T3 (purportedly) would be its pumping efficiency. i know thats more-or-less already been said numerous times, but my guess is the design of the pump interface allows it to be a majorly efficient compromise between two completely separate loops and a typical xtreme serial loop, with multiple blocks and radiators. and by efficient compromise, i mean in terms of space, power consumption and pumping efficiency.

When i first saw it, which was earlier this week in fact, i was hellbent on getting one. and i did. ordered it from jab-tech on thursday, got it friday. regardless of what it cannot achieve, that say two separate loops can, this design falls right into exactly what i needed to separate up my CPU and GPU cooling systems. i dont know if there would be any less strain on a pump thats pushing through two water blocks, two radiators and a reservoir, but thats not the point. the point is this design maximizes the aformentioned efficiency.

I applaud skinnee for taking the extra time to push out his review of the product, and i thank him for adding to my decision-making. when i first saw the results graphs, i understood everything, pulled out my calculator and did not find anything out of the ordinary. i was not double checking skinnee's math, i was double checking my understanding of how skinnee presented the data. and in my understanding, adding the two flow rates together to report the pushing power of the pump in that fashion is exactly the same as how transfer/transmit bandwidth of a network is calculated.

Of course, im not 100% sure on that parallel, and i would be most appreciative if anyone would care to correct me if i am wrong.

tl;dr:

good review skinnee. i liked it. boxgods, i really like your design, thanks for taking the time and effort to take it into mass market. i bought one already and am awaiting my 2nd radiator to really put it to the test. the T3 of my ownership sits on my kitchen table.

[millertime359]

if i add the word "like" to that first sentence, it would have made more sense, i was referring to what the T3 should be compared to, not what it is.

Got ya

[BoxGods]

Wasn't into Dax, now Jadzia, I'd like to break me off a little piece of that!

Wasn't it Jadzia DAX..jadzia is the host and DAX was the symbiot or the worm thing?

i think the best thing about the T3 (purportedly) would be its pumping efficiency. i know thats more-or-less already been said numerous times, but my guess is the design of the pump interface allows it to be a majorly efficient compromise between two completely separate loops and a typical xtreme serial loop, with multiple blocks and radiators. and by efficient compromise, i mean in terms of space, power consumption and pumping efficiency.

When i first saw it, which was earlier this week in fact, i was hellbent on getting one. and i did. ordered it from jab-tech on thursday, got it friday. regardless of what it cannot achieve, that say two separate loops can, this design falls right into exactly what i needed to separate up my CPU and GPU cooling systems. i dont know if there would be any less strain on a pump thats pushing through two water blocks, two radiators and a reservoir, but thats not the point. the point is this design maximizes the aformentioned efficiency.

I applaud skinnee for taking the extra time to push out his review of the product, and i thank him for adding to my decision-making. when i first saw the results graphs, i understood everything, pulled out my calculator and did not find anything out of the ordinary. i was not double checking skinnee's math, i was double checking my understanding of how skinnee presented the data. and in my understanding, adding the two flow rates together to report the pushing power of the pump in that fashion is exactly the same as how transfer/transmit bandwidth of a network is calculated.

Of course, im not 100% sure on that parallel, and i would be most appreciative if anyone would care to correct me if i am wrong.

tl;dr:

good review skinnee. i liked it. boxgods, i really like your design, thanks for taking the time and effort to take it into mass market. i bought one already and am awaiting my 2nd radiator to really put it to the test. the T3 of my ownership sits on my kitchen table.

Thanks. I hope to get out at least 2 more major products this year. Stay tuned!