Corsair H70 CPU Cooler - First Install @ [H]

FrgMstr

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,602
Corsair H70 CPU Cooler - First Install - We showed you the new Corsair H70 CPU cooler a couple of days ago and finally today we had the opportunity to place it inside our Quiet Fermi Build replacing the Corsair H50 we had in the system. Come along to see our install and what our preliminary temperatures look like.
 
WIll you guys be comparing it to the ECO C240 you recently reviewed? That didn't have a comparison to the H50, and I'm really curious to see how all 3 stack up.
 
FaRKle0079 said:
Hrm, so in the load tests you guys saw a 5C drop in temps. Worth it or not...
Click to expand...

At stock clock speeds the 5-6C drop in temps is accurate. However, the difference is much more noticeable when overclocked. The gap between the H50 and H70 widens as the heat produced by the CPU goes up.

In the Fortress or Raven cases, I would use the H50/H70 as an exhaust as well, but in most cases with more standard designs (psu on bottom, intake in front/bottom, exhaust in top, rear) using it as an intake can drop CPU temps significantly if your top exhaust fans are installed.
 
Redbeard said:
At stock clock speeds the 5-6C drop in temps is accurate. However, the difference is much more noticeable when overclocked. The gap between the H50 and H70 widens as the heat produced by the CPU goes up.

In the Fortress or Raven cases, I would use the H50/H70 as an exhaust as well, but in most cases with more standard designs (psu on bottom, intake in front/bottom, exhaust in top, rear) using it as an intake can drop CPU temps significantly if your top exhaust fans are installed.
Click to expand...


We were running a 980X at 4GHz at 1.32 vCore.
 
Ooooohhh... I want a new case... and that cooler. My i7 920 runs 4.2ghz at 1.4V. That's hot.

My Antec Twelve Hundred has great airflow, but crappy cable management. The LZG-1000's (PSU) cables are just too big for this case. For four pci-e (2x6, 2x8) cables, seven SATA devices, a sound card with a floppy disk power connector, and standard motherboard connectors, there just aren't enough slots to route the cables behind the motherboard tray. And even with everything one can fit behind the motherboard tray, the lid that goes on top of that tray leaves much too little clearance.

I only mention this because Kyle keeps pimping that Silverstone Raven RV02... and it looks so much more adequate than my Twelve Hundred. The only things stopping me from picking one up are:
1) I'd have to buy 5.25" to 3.5" adapters for all my hard drives
2) It doesn't look like there is nearly enough space for a third dual slot video card without interfering with the plugs on the hard drives. Every week, I reconsider grabbing a GTS 250 for PhysX for Mafia II.
3) I don't think I'd be able to mount the H70 in that case with my E758. The very large MOSFET heat sink looks like it would interfere.

*sigh*

I don't know if I can settle for the 800D's sub-par (compared to Raven 02 and Twelve Hundred) airflow, but it seems to have everything else I need.
 
IRSmurf said:
Ooooohhh... I want a new case... and that cooler. My i7 920 runs 4.2ghz at 1.4V. That's hot.

.....

I only mention this because Kyle keeps pimping that Silverstone Raven RV02... and it looks so much more adequate than my Twelve Hundred. The only things stopping me from picking one up are:
1) I'd have to buy 5.25" to 3.5" adapters for all my hard drives
2) It doesn't look like there is nearly enough space for a third dual slot video card without interfering with the plugs on the hard drives. Every week, I reconsider grabbing a GTS 250 for PhysX for Mafia II.
3) I don't think I'd be able to mount the H70 in that case with my E758. The very large MOSFET heat sink looks like it would interfere.

*sigh*

I don't know if I can settle for the 800D's sub-par (compared to Raven 02 and Twelve Hundred) airflow, but it seems to have everything else I need.
Click to expand...

Both the Raven2 and the Fortress2 have 8 I/O slots, so there should be space for a tri-card setup.
The big caution would be making sure that your optical drive fits with the 3rd card installed.
Space is pretty tight between the drive bays and the edge of the motherboard area.
 
gathagan said:
Both the Raven2 and the Fortress2 have 8 I/O slots, so there should be space for a tri-card setup.
The big caution would be making sure that your optical drive fits with the 3rd card installed.
Space is pretty tight between the drive bays and the edge of the motherboard area.
Click to expand...

I was just going to suggest the FT02, which wouldn't interfere with the HDD sata connectors (mentioned above about the RV02), but the FT02 only has 7 expansion slots, not 8. The newer RV02-E will also have the same HDD setup and 7 expansion slots the FT02 currently has. All you have to with the optical drive is make sure it fits in the top slot, in which case there's no interference with 3 way.

Anyway, I guess I'm not taking my $60 H50 back to best buy.

Edit: Plus you'll get AP181's with the RV02-E :)
 
Last edited:
MissJ84 said:
I was just going to suggest the FT02, which wouldn't interfere with the HDD sata connectors (mentioned above about the RV02), but the FT02 only has 7 expansion slots, not 8. The newer RV02-E will also have the same HDD setup and 7 expansion slots the FT02 currently has. All you have to with the optical drive is make sure it fits in the top slot, in which case there's no interference with 3 way.

Anyway, I guess I'm not taking my $60 H50 back to best buy.
Click to expand...

Magnificent! Thank you for referring me to the RV02-E:
http://www.bit-tech.net/hardware/cases/2010/05/04/first-look-silverstone-raven-rv02-e/1

That's my next case, for sure!
 
kllrnohj said:
WIll you guys be comparing it to the ECO C240 you recently reviewed? That didn't have a comparison to the H50, and I'm really curious to see how all 3 stack up.
Click to expand...
+1
Although, this cooler would do better if the price is under $100.
 
It's better but nothing to sell my H50 over. Only a custom loop will make a significant difference.
 
All this showed is that if you own an H50 its really not worth selling for the H70.For the price its not worth a couple degrees less
 
Wonder if lapping would help this thing out any?
 
Last edited by a moderator:
Wonder if lapping would help this thing out any?
Click to expand...

Only if you did the IHS on the CPU as well, and then the results would depend on how "unflat" the two surfaces where to start with. A simple test with a razor blade can determine the approximate "flatness" of each surface and help one decide if lapping would be worth the trouble. On the other hand, I for instance, lapp everything applicable as a matter of course regardless as I can have it half done in the time it would take me to actually find a razor blade and get a lamp set up. Its kinda like going to the drag strip with street tires on the vehicle and then leaving them on for the timed run, makes no sense to me and the slicks go on first thing without thought. In general lapping is always worth a deg C or two. Mainly because of the IHS of the CPU is so irregular.


@ article

Yaaaaa ! Coretemp temps !!!!

Every different case presents different issues regarding WC radiator installation. It also appears that if you had mounted the fans as intake as far as I can tell from that case design you likely would have needed to flip all the fans in the bottom and even then the very real possibility of the H70 fans sucking in the hot exhaust from the power supply would have killed performance. But whatever, the point is, you accessed the airflow and aesthetics and made modifications/adjustments to suit the unique individual circumstances. This is how it is done. While the manuf instructions are not to be lightly disregarded WCing is still somewhat of an art and if end users of this system, especially people new to WCing, take away only that lesson, that experimentation and adaptation to their unique circumstances/equipment is not a bad thing, your article is a huge success.

It is better to give the rad as cool of air as possible but in many instances it may not be practical and the temp rise is very manageable. Frankly if the case has insufficient airflow to handle either situation one needs to be looking for a better case than worrying about which way to point the H50/70.

Now a "ding" should have fired up something to kick the video into 3D so one could see what the heat generated inside the case did do to cpu temp numbers. Most of it gets exhausted out the back but not all.
 
Oh my goodness. Thought I'd do some IHS checking of my own.

I just slid a metal straight-edge across the top of a couple of CPUs I had sitting on my desk with IHS.

Gaps you could drive a very small truck through. Not very flat at all.
 
I bet that says when Kyle finally switches the system around that those idle and load temps will drop a fair amount.. Here he is worried about dumping 130 watts into a case that already has 2 480 Fermi's which drop over 300 watts each into the case. He is not getting max performance for cooling.

Remember that without some other outside source to cool the water (TEC) setup of a Refrig setup, the lowest temps that can be achieved are those of Ambient Air Temps.

If he switches the fan's in the correct order, I'm sure that he will see a difference of about 5-7 degrees C or more and that running under load the Fermi's really won't run much hotter than they are now..... If you take your time and route your airflow correctly then you can have the best of both worlds, but this is far from optimal.

An example is my friends Alienware. It came with a rebranded H50. The case was tightly packed and it was set to exhaust air. His system would idle at 50 and run Prime95 up to over 80 degree's C. Having already having this system, we went out got a slightly higher flow fan and turned it around to pull air in.

Idle dropped to 34 degrees C, prime 95 with small FFT's was 58 degrees.

This was a Core i7 920 at stock speeds....

Those that worry about the fans sucking in the hot air of the Power supply need not worry much as it is not a direct route ie channeled into the intake.... it pulls in more cool air than you think, but it does require changing the fan's which is not a big deal and can gain you a huge margin.

Thats why WC isn't as big as it should be... people put systems together as if they were putting together an air cooled system and wonder why WC cooling doesn't seem to work that good.....

They are two different beasts.... Those idle temps tell me that Kyles case is already warm inside and that at idle that rad is actually acting like a heater for the CPU.

My WC rig will run at 4.4 GHZ all day long (stock 2.8) and never come close to 75 degree's C. But I do have an external Rad...

As my previous post said I had one of these on a Q6600 in both positions and as a rear intake, it made all the difference in the world.... why does everyone feel that the front fan needs to pull air in??? Why can't it be the exhaust, or is everyone stuck in their rut, not willing to try things.... I have heard an argument from a close friend about needing to cools his HD's.... only problem was he had replaced his HD's with SSD's.... and unless you are running 10,000 RPM HD's, they really are not going to suffer from the heat....

I'm still in shock about Kyle worrying about dumping 130 into his system when he is dumping nearly 5x that or more with the GTX480's....

It is just sad to see people pay money and not get what they paid for in full. Whether it be a 100 cooler or a 4 billion dollar sub.... just because things have been run a certain way for all these years does't mean that is the correct way to do everything with new technology. I know you have more than one fan in that case Kyle... given your ambient, and the specs on your system, you should see a decent drop in idle temps and a much more radical drop in loaded temps.... As long as the CFM of the fans are running fast enough, the thermal dissipation into his case can be minimized....

Don't let Kyle's ineptness in this matter stop you from doing it right. Get a cheap thermometer and measure the air right outside the case where corsair says to have it intake... make sure no fan is running in that slot and have it just below the opening. (Hot air from the power supply rises BTW, actually causing cooler air from below to fill the void, even if you have Fermi's they blow far enough out to mix with the cooler air that you will find that the air is cooler outside the case...

Open your case and put the thermometer where the case fan would be.... it should be warmer in this closed case.... the temp that you see is going to be the absolute minimum you will see from your cooling setup in each scenario.

Thats why Corsair says use the rear fan as an intake because that is so it can get the coolest air to work with.

Even the editors of sites you admire and follow like sheep can be wrong.... in this case from experience with the H50 and the intake/outflow test on 2 different machines, cooling gains will be had from following Corsairs directions, they didn't choose that setup at random.

As for heating the cards, I'm sure the system probably has fans blowing air directly onto the cards, so that would probably pretty much negate any Fermi temp increase, and turn the front intake fan into an exhaust fan.... it will work.... its not 100% optimal, but its better than turning your WC cooler into a WC heater loop....

Some of you may question my experience... well 9 years running your tax payers billion dollar subs reactor units and having to deal with heat exchange issues has taught me over the years how to deal with running a reactor, getting maximum power without leaving enough heat behind to be tracked by subs....

If this were a Fermi like 400+ watt draw at max power, then we would be talking much different....
 
IceEmQuick said:
I'm still in shock about Kyle worrying about dumping 130 into his system when he is dumping nearly 5x that or more with the GTX480's....
Click to expand...

You realize those video cards have external exhaust coolers, yes? The vast majority of the heat those put out gets blown out the back of the case. It's not a hard seal obviously, but move your hand around the inside portion of one sometime, then feel whats coming out of the exhaust port. You'll realize pretty quickly where the heat is going.

Please get a fucking clue before you call Kyle "inept"
 
NKDietrich said:
You realize those video cards have external exhaust coolers, yes? The vast majority of the heat those put out gets blown out the back of the case. It's not a hard seal obviously, but move your hand around the inside portion of one sometime, then feel whats coming out of the exhaust port. You'll realize pretty quickly where the heat is going.

Please get a fucking clue before you call Kyle "inept"
Click to expand...


On the other hand please can we credit the Corsair engineers that they too might know what they are doing when they recommend their fan setup?

I know some companies dont think that testing a product is worthwhile these days (ahem I wont mention any names) but Corsair have never struck me as idiots. They have played quite a role in the PC enthusiast market and I think they know what they are talking about.
 
daglesj said:
On the other hand please can we credit the Corsair engineers that they too might know what they are doing when they recommend their fan setup?

I know some companies dont think that testing a product is worthwhile these days (ahem I wont mention any names) but Corsair have never struck me as idiots. They have played quite a role in the PC enthusiast market and I think they know what they are talking about.
Click to expand...

Of course Corsair knows what they are talking about, and of course the setup they recommend is going to offer the lowest CPU temperatures. No one is disputing that. They are marketing a CPU cooler, not a case fan.

What do sites measure when they test a CPU cooler? CPU temperatures. It's not surprising that Corsair wants people to use the configuration that offers the best CPU temperatures. But the few C you gain from such a configuration probably isn't going to outweigh the overall impact to case temperatures.

Corsair wants the lowest CPU temperatures possible in a given test. They don't care about the temperature of your RAM or your VRegs. That's not what they are being tested on.

They have a vested interest in making cooler look as good as possible, and there's nothing wrong with that, but Kyle and other reviewers have to recommend what they feel is in the best interest of their reader, not in the best interest of the product's sales.
 
Hello there Kyle.:p

After you set up the H70 as shown in the snappys.....can you place the top back on the Raven case without a problem?

I ask because I just got my son a new Fortress2 and that H70 looks like the shizzle!:cool:
 
magoo said:
Hello there Kyle.:p

After you set up the H70 as shown in the snappys.....can you place the top back on the Raven case without a problem?

I ask because I just got my son a new Fortress2 and that H70 looks like the shizzle!:cool:
Click to expand...

You definitely can't on the FT02 sans some kind of top panel modification, though I'm curious in hearing whether or not the RV02 can. I'm guessing no, based on some photos I just looked at.
 
Kyle,

For your temps, did you just run Prime95 or did you load up your Fermi cards with Furmark... I want to see you REAL temps... I want to see how all that hot air being dumped into your case affects your temp readings...

You think that the exit is the best way.... load up that system.... Prime95 AND Furmark and tell me the temps then.... if your system doesn't reboot or crash....

Then maybe you will reconsider turning them around.... after all, your first test of the cooler is missing 800 watts of heat being dumped into your case.... so load those cards up to max temp, then run Prime 95 and tell me that running it as an exhaust port is best....

Then swap the fan's and then swap your front intake to an exhaust and lets see the difference... how cool do you think it will be able to cool when the air in the case is roasty toasty warm then start after running prime 95 and let us know the temps....
 
Suggestions as to what would be nice to see in the actual review would be nice. Almost anything else is farts in the wind. The man just installed it and got some quick results up for us to look over.

Kyle explained why he did what he did, I commented on it, this issue has been beat to death leaving only a greasy smear smelling faintly of horse in the road in several of the H50 posts including but not limited to the stickies.

All this tells me that a lot of youse guys just jumped on this post and have not been keeping up. Dont expect us to go through it again for laggers.

Bottom line,

Its got a slighty bigger rad, different pump and two fans included. Its not going to rip a hole in the fabric of space time.
 
Not taking sides but I thought IceEmQuick's comments won the war along time ago in other Wc'ng forums. As long as your case is half decent the heat dropped back in was negligible compared with the benefits of lower cpu temps.

Obviously 2 Fermi's and a boat load of drives pound the case with heat but by all accounts don't most modern motherboards take it easily?

The heat stress point is usually the cpu directly isn't it?

I'd be following Corsairs instructions with my HAF 932 for sure but the sentence or 2 about fan noise have me a little concerned...
 
BillParrish said:
The man just installed it and got some quick results up for us to look over.
Click to expand...

Yeah and it looks like a step in the right direction. Looking forward to the full spiel in due course.

Looking forward to the H90 already.
 
uh, by all means contiure to discuss everything as you feel the need. Just keep it civil. We (well me anyway) want people to both learn and educate here. Not everything is simple or just black and white. If it is productive its good.

I just dont want this to end up as another push vs pull argument. Its been done.

Carry on.
 
Kyle,

when do the final review could you include some "apples to apples" between the two cooler? by that i mean use the same fan(s) on both coolers. I would like to see how the h50 push/pull stacks up to the h70 push/pull with the same fans.

as to the mounting. i agree with you 100%. yes you will get some better temps if you set it up as intake, but who wants that extra hot air in the case?
 
magoo said:
Hello there Kyle.:p

After you set up the H70 as shown in the snappys.....can you place the top back on the Raven case without a problem?

I ask because I just got my son a new Fortress2 and that H70 looks like the shizzle!:cool:
Click to expand...

No, it required some modding.
Posted via [H] Mobile Device
 
You think that the exit is the best way.... load up that system.... Prime95 AND Furmark and tell me the temps then.... if your system doesn't reboot or crash....
Click to expand...

That's not a very good way to test as running the CPU at 100% will inhibit Furmark's burning abilities. I've tried this before and you will get nowhere near as high temps on the GPU as you would just running furmark alone.
 
@IceEmQuick

Nonsense. The FT02 has three 180mm fans throwing 100CFM into that chassis - EACH. And some upgrade to the 150CFM FM181 for extra GPU cooling. The heat stack properties on this chassis, combined with the extremely positive pressure generated by the three 180mm fans means that the H70 should indeed be used as an exhaust.

I would turn your recommendation back on you - measure the air intake from the outside with the top panel on and the GPUs loaded. I think you'd see you're pulling in more hot air than if you were exhausting it.

With other chassis an intake setup would work. Not with a chassis with this design.
 
The only thing I hate about this review is that I just bought a H50 2 weeks ago. Now I will have to agonize over why didn't I check Corsair's website to see that a new cooler was coming out. The price of sitting on a fence for too long I suppose...
 
Good on Corsair for putting those swivels in there. I can see this being a great option for the SFF guys whom have already adopted the H50 in a lot of builds due to the small space thay can get that block/pump into.
 
Yes, an apples-to-apples comparison of the H50, H70 and Eco each in intake and exhaust tests would be great. Can't wait!
 
cageymaru said:
The only thing I hate about this review is that I just bought a H50 2 weeks ago. Now I will have to agonize over why didn't I check Corsair's website to see that a new cooler was coming out. The price of sitting on a fence for too long I suppose...
Click to expand...

If it makes you feel any better, we didn't announce it officially until yesterday so there's no way you could have known.
 
I just happen to have this case and was shopping for one of these coolers so this is great for my particular situation.

I'm curious if the full review will include the temperature difference of the video cards as well, since the configuration enables dumping the heat directly out of the case versus a normal HSF setup.
 
I'm not disagreeing with you about your setup for the raven case, in fact for that case it should work better since the motherboard is turned and essentially every part has it's own cooling zone. For other cases, take mine for example, the cm690II advanced. I get better temps with my h50 in the rear slot as intake with 2x140mm fans for exhaust.

I don't know if you can but maybe try out the h70 as an intake in a case like the 690 II advanced with more exhaust fans that just the one. I just don't think the raven series cases represents many of the users here or on other sites.

just my 2cents
 
Redbeard said:
If it makes you feel any better, we didn't announce it officially until yesterday so there's no way you could have known.
Click to expand...

But I do seem to remember someone dropping a hint about an H50 revision a while ago ;)

Someone else asked this on page 1 but it didn't get answered. How's the pump noise vs the H50?
 
You must log in or register to reply here.
Back
Top