Corsair revisits CPU cooling: introducing the Hydro Series H50

Printable View

Show 100 post(s) from this thread on one page

06-25-2009, 10:50 AM
Zaskar

Quote:

Originally Posted by SNiiPE_DoGG

you guys are idiots.

not only does running such stress tests as IBT and linX for long periods of time physically shorten the lifespan of your CPU (even prime is bad for the CPU) but it also is pointless as leeghoofd says there is no REAL PROGRAM that will stress your CPU as much as prime and therefore it is not worth clocking for stability in anything more stressful than prime. period. You all think your proving something by running these extreme torture tests but in reality your just proving you have no regard for reality or the health of your CPU.

It will not shorten the life of the CPU if your max load temp under those programs is still safe, if however your at like 75c Prime with an i7 CPU and with Linpack/LINX your 85c-90c then thats different, but that means its clocked too high anyways.

The point of stressing the CPU that bad is to drag into the light any instability in a relatively short amount of time, that way a few days, weeks or even months from now your not playing a game and it just locks up or BSOD's and reboots on you.

The flaw in peoples thoughts on this seem to be that if your system is stable for hours on Prime but not on LINX, then it will be stable forever on anything that doesn't stress it over Prime levels, thats wrong.

The less stress just means less of a chance to run into an error, but if its not truly stable then you will eventually get one.

If you don't care about that, its fine, completely up to you and what you find important.

But when people here talk in terms of "Stable" their talking about Perfectly stable under the highest stress testing programs available to us, not "doesn't crash in my games" stable.
06-25-2009, 10:57 AM
Leeghoofd

Explain me then why I am linX stable (max mem and 64bit OS) yet cannot complete a superpi 32mb run ? Really guys there's no ultimate testing tool, each one has it's advantages but sadly also it's flaws. I opted for custom prime 20k for the database to avoid getting different numbers and comparisons.

Quote:

Originally Posted by Ender17

I'm tired of waiting :p:
plus I would kinda like to do my own testing as I've said before, my P182 is pretty starved for airflow I think - seems like my case would be the optimal environment for the H50

Keep us posted how it performs sir
06-25-2009, 11:00 AM
Zaskar

Quote:

Originally Posted by Leeghoofd

Explain me then why I am linX stable (max mem and 64bit OS) yet cannot complete a superpi 32mb run ? Really guys there's no ultimate testing tool, each one has it's advantages but sadly also it's flaws. I opted for custom prime 20k for the database to avoid getting different numbers and comparisons.

Very true, on my old 780i I was like 30 passes of IntelBurnTest stable (Linpack based) but failed Memtest86 after a few hours.
06-25-2009, 11:05 AM
JPQY

That's i want to say many times,just use your most stressing daily program and tune your system to it like i do with my chess program!

I want my daily program 24/7 Stable!

The rest is nice for testing,benching!

JP.
06-25-2009, 11:10 AM
Ender17

Quote:

Originally Posted by Leeghoofd

Explain me then why I am linX stable (max mem and 64bit OS) yet cannot complete a superpi 32mb run ? Really guys there's no ultimate testing tool, each one has it's advantages but sadly also it's flaws. I opted for custom prime 20k for the database to avoid getting different numbers and comparisons.

from what I've heard, LinX isn't the best test of RAM
to test my final overclock on my HTPC I ran LinX, Intel Burn Test, Core Damage, OCCT, Prime95 blend, HyperPi and FurMark

still going to do 24 hours of Prime95 blend, 24 hours of Prime95 small FFTs, and 24 hours of Memtest

Quote:

Originally Posted by Leeghoofd

Keep us posted how it performs sir

Will do!

http://i43.tinypic.com/x5r0n9.png
06-25-2009, 11:15 AM
Manicdan

should i ask why you payed 21$ for shipping?
06-25-2009, 11:24 AM
Ender17

Quote:

Originally Posted by Manicdan

should i ask why you payed 21$ for shipping?

these were my options
an extra $7 to get it about a week faster seemed like a good deal to me

Code:

FedEx Express Saver USA $14.49 Usually takes 3 business days to ship. FedEx Priority Air USA $21.19 Usually takes 1 business days to ship. FedEx Economy Air USA $24.58 Usually takes 2 business days to ship. CanadaPost Xpress USA $25.76 Usually takes 2-3 business days to ship.
06-25-2009, 11:36 AM
Smartidiot89

Quote:

Originally Posted by Ender17

these were my options
an extra $7 to get it about a week faster seemed like a good deal to me

Code:

FedEx Express Saver USA $14.49 Usually takes 3 business days to ship. FedEx Priority Air USA $21.19 Usually takes 1 business days to ship. FedEx Economy Air USA $24.58 Usually takes 2 business days to ship. CanadaPost Xpress USA $25.76 Usually takes 2-3 business days to ship.

Damn I NEVER EVER put up with those kinds of shippings o_O

Here in sweden it's equal to being robbed if you more then $7 for shipping, unless you're buying a Case or something then you could dubble add $3-5 :confused:
06-25-2009, 11:38 AM
WhiteFireDragon

can these tubes be taken off and shortened?
06-25-2009, 11:46 AM
Ender17

Quote:

Originally Posted by Smartidiot89

Damn I NEVER EVER put up with those kinds of shippings o_O

Here in sweden it's equal to being robbed if you more then $7 for shipping, unless you're buying a Case or something then you could dubble add $3-5 :confused:

yeah normally I would never pay $20 for shipping, but their base price was about $10-15 cheaper than anywhere else that has it listed and that's for international air so in my mind it kinda evens out
06-25-2009, 12:00 PM
Leeghoofd

Quote:

Originally Posted by WhiteFireDragon

can these tubes be taken and shortened?

I'll have to verify that with Corsair. But since the unit is prefilled I don't think you can, otherwise I would hook it up to a 240mm rad to check it's performance gain...
06-25-2009, 12:03 PM
SNiiPE_DoGG

you could hook it to an mcr220qg res and check perf gain, I think that would be a really good idea :up:

get one of these kits and do a comparitive review- you would get the rad for the corsair and a competitor in one package. PM gabe I am sure he could get you one http://www.newegg.com/Product/Produc...82E16835108105
06-25-2009, 12:13 PM
Redbeard

This unit is designed for zero maintenance so you shouldn't have to worry about evap or filling. You're more than welcome to cut the tubing, drain it, and add your own radiator, but that would void the warranty of course. We decided to go with the single 120mm radiator for maximum compatibility. There aren't many cases where you can fit a 240mm rad without modding, and certainly the minute you use the word "cut" in your instructions you lose like 75% of the people who are interested in buying it at least.

Again, we want people to think of this like a high end aircooler, not a low-end watercooler. It's designed to compete with aircooling more than anything, which is why the price point is so low comparatively.
06-25-2009, 12:24 PM
Manicdan

im sold on this, i like building compact, and i mean very compact cases, and a good high perf air cooler in a much smaller size that can be put anywhere.

whats its total weight and how many watts does the pump use?
06-25-2009, 12:31 PM
NeXaR

When could I buy this on Spain? I have an HTPC proyect in mind and this will suit well.
06-25-2009, 12:48 PM
WhiteFireDragon

Quote:

Originally Posted by Redbeard

This unit is designed for zero maintenance so you shouldn't have to worry about evap or filling. You're more than welcome to cut the tubing, drain it, and add your own radiator, but that would void the warranty of course. We decided to go with the single 120mm radiator for maximum compatibility. There aren't many cases where you can fit a 240mm rad without modding, and certainly the minute you use the word "cut" in your instructions you lose like 75% of the people who are interested in buying it at least.

this looks great for people that don't have any WC equipment yet, but for those that do and already have some spare tubing, what size are the barbs and tubing? they look smaller than 3/8" ID. and since you mentioned "cut", how would we go about doing this? from the pics, the tubings go all the way INTO the pump. what's it connected to in there? how can we patch 3/8" tubing on this thing for a custom length?
06-25-2009, 02:15 PM
Leeghoofd

It might be 10mm tubing that could work, hardly any way to go 3/8 on this thing, there just not enough room. Since my unit has some issues with the base and the gurgling noise at the first 1-2 minutes after fire up (unlucky me ), I decided to open it...

I"m awaiting if I can post these piccies yes or no... trying to refill it with my liquid fluid and see if I can get rid of that gurgling noise...

Here were my initial results ( so to be redone once I get a new unit )

http://users.telenet.be/OAP2/Corsair...%a9S775pre.jpg

http://users.telenet.be/OAP2/Corsair...a9S1366pre.jpg
06-25-2009, 02:26 PM
WhiteFireDragon

are you allowed to post pics of the internals? please do :D
06-25-2009, 03:45 PM
BababooeyHTJ

Quote:

Originally Posted by Zaskar

It will not shorten the life of the CPU if your max load temp under those programs is still safe, if however your at like 75c Prime with an i7 CPU and with Linpack/LINX your 85c-90c then thats different, but that means its clocked too high anyways.

The point of stressing the CPU that bad is to drag into the light any instability in a relatively short amount of time, that way a few days, weeks or even months from now your not playing a game and it just locks up or BSOD's and reboots on you.

The flaw in peoples thoughts on this seem to be that if your system is stable for hours on Prime but not on LINX, then it will be stable forever on anything that doesn't stress it over Prime levels, thats wrong.

The less stress just means less of a chance to run into an error, but if its not truly stable then you will eventually get one.

If you don't care about that, its fine, completely up to you and what you find important.

But when people here talk in terms of "Stable" their talking about Perfectly stable under the highest stress testing programs available to us, not "doesn't crash in my games" stable.

I don't know of any programs that will cause as much vdroop as Linpack, and that lower vcore that you will never, ever see otherwise will be "unstable" in Linpack. I've passed 40 runs of IBT, Prime small fft, blend, Memtest 3.8, 86+ and was still having odd freezeing, then failed a custom prime run in 2 minutes. Once I got everything so I didn't fail that custom prime run I haven't seen a BSOD or freeze up since and it's been months, I don't even know that it passes IBT and I don't care.
06-25-2009, 05:23 PM
Ender17

Quote:

Originally Posted by BababooeyHTJ

I don't know of any programs that will cause as much vdroop as Linpack, and that lower vcore that you will never, ever see otherwise will be "unstable" in Linpack. I've passed 40 runs of IBT, Prime small fft, blend, Memtest 3.8, 86+ and was still having odd freezeing, then failed a custom prime run in 2 minutes. Once I got everything so I didn't fail that custom prime run I haven't seen a BSOD or freeze up since and it's been months, I don't even know that it passes IBT and I don't care.

details of custom prime blend?
06-25-2009, 06:08 PM
BababooeyHTJ

Quote:

Originally Posted by Ender17

details of custom prime blend?

Quote:

Originally Posted by cstkl1

intel and memtest doesnt stress enough on vdimm.
i can run lower vdimm and vtt with it.

try this so far the highest stress and will show errors

run prime95 25.9 64bit set large fft and then set custom, change run fft to to max mem around 5.4gb in diagnostic mode
if hang ( could be either ioh/ich voltage or inssufficient vtt)
if more than one thread error.. insufficient vdimm
if one thread error vtt insufficient or vtt/vdimm ratio wrong ( one of them is overvolted or undervolted)

:up:

It's just a prime large fft run with run fft in place unchecked and all of your available memory enabled but it stresses your mch harder than anything that I know of. If you can pass a couple hours of that and a long small fft run and your stable as far as I'm concerned.

I should also mention that on my old e8400 I have passed 40 runs of IBT and then right after failed prime small fft in around 45 minutes on a couple of different boards, nothing a little GTL tweeking didn't fix.
06-25-2009, 07:47 PM
Ender17

Quote:

Originally Posted by BababooeyHTJ

:up:

It's just a prime large fft run with run fft in place unchecked and all of your available memory enabled but it stresses your mch harder than anything that I know of. If you can pass a couple hours of that and a long small fft run and your stable as far as I'm concerned.

I should also mention that on my old e8400 I have passed 40 runs of IBT and then right after failed prime small fft in around 45 minutes on a couple of different boards, nothing a little GTL tweeking didn't fix.

thanks, I will add that to the battery of tests I do for final overclock
06-25-2009, 07:55 PM
Ender17

looks like I may have the first H50 retail sample? NCIXUS just emailed me a tracking number and it will be here in under 12 hours :D
06-25-2009, 08:38 PM
nsegative

Quote:

Originally Posted by Ender17

looks like I may have the first H50 retail sample? NCIXUS just emailed me a tracking number and it will be here in under 12 hours :D

What are you currently using as a cpu cooler? Would like a short comparison preview from you :D
06-25-2009, 08:41 PM
Ender17

Quote:

Originally Posted by nsegative

What are you currently using as a cpu cooler? Would like a short comparison preview from you :D

HR-01 Plus with S Flex E

Show 100 post(s) from this thread on one page