Contemplating switching from SB i5 to IB i3

[Willzzz]

Use the stock cooler, it's virtually silent..

[watercooled]

It depends what you consider quiet - the stock coolers wouldn't find themselves in that ball park by my standards.

They can be quiet-ish when idle, but the fan speed climbs a lot under load, and subjectively they're more audible than a decent 120mm fan do to the higher RPM, and it varies between samples but I've found a lot of stock coolers to make a very audible and annoying buzzing noise even when idle.

If the noise from a decent quality 120mm fan and the pump are annoying the OP, I suspect the stock fan would be worse. And it's not going to be as efficient at keeping temps down either.

[Willzzz]

My friend has a stock 2500k and it is silent.

If you really don't like the idea of the stock cooler then a Freezer 13 would be ideal, just not free.

Going down to an i3 is likely to have very little effect on noise unless the noise is only present at very high loads.
If you are frequently at high loads then you should probably stick with the i5.

[watercooled]

No fan is silent. Yes I'm being pedantic, but quiet really is a subjective term, and it's also more complex than the single number often given in reviews. I've tested a few stock coolers, the low and high TDP versions of the 1155, and I think it was a high TDP 1150 (it's possible it's the same cooler supplied with 1155 CPUs). In fact I still have the 'large' 1155 cooler, which came with a Celeron, in my parts box. They're not awful at idle, but there's better; again it's down to the individual, whether it's used in a sound dampened or an open case, and where the case is situated. And they do ramp up the noise when put under load.

But yes I agree about the swap - if the noise doesn't fade when idling, then there's already room to improve as both CPUs use a very similar amount (low single digits) when idle, so the i3 wouldn't do any better here.

[CAT-THE-FIFTH]

I didn't find any of the Intel stock coolers that quiet in my SFF systems. Especially during summer when gaming,they did really ramp up and that was with the 65W TDP Core i3 2100 and the 80W TDP Xeon E3 1220. I was using the cooler with the copper core which was rated for 95W TDP CPUs. The same goes with the AMD stock coolers. Maybe in larger steel cases they are OK,but not for the aluminium SFF cases I use. I switched over to a modded Xigmatek Praeton which was quieter and eventually got a cheap Corsair H40 with undervolted Corsair fans.

My current system is as follows:

Xeon E3-1220
8GB DDR3 RAM
Gigabyte H67N mini-ITX motherboard
GTX660
XFX PRO 450W
Cubitek Mini Cube aluminium mini-ITX case
Corsair H40 with one AF120 fan as inlet to the cooler
One AF120 on the bottom of the case as inlet fan

At idle and low load,the CPU runs at around 30C to 40C and around 60C to 65C when running games,or anything CPU intensive.

When at idle or low load the system is generally quiet.

This is what the case looks like:

http://cdn.eteknix.com/wp-content/up...2/DSC_8284.jpg



The picture is from a review BTW.

The Bitfenix Prodigy has better cooling and is made of steel,so should not only be getting better temperatures but lower transmitted noise too.

I would also not downgrade to a Core i3 from a Core i5. I went from a Core i3 to a Xeon E3 since the former was starting to be a limiting factor in some games I was playing.

[watercooled]

I have one of those Akasa CPU coolers from the picture!

IIRC the Prodigy has a fairly open (vented) side panel though, which could allow noise to escape, especially if it's facing towards the user.

[MaverickWill]

Right, chaps, for those who wanted to know what happened:

I tried to stall making the switch, by moving stuff around in the case, strapping a second fan on to the H60, removing the front drive bays... none of it did much to temper the noise (although some made the CPU run cooler, notably the dual fans).

I got hold of the i3, and ran it through some games and their built-in benchmarks. Bioshock registered barely any difference - think there may have been a frame or two difference in minimums at some point during the benchmarks, but nothing game-changing. Tomb Raider's benchmark, ironically, registered slightly faster minimums (4-5 extra FPS) on the i3 than the i5, and did so every time the benchmark was repeated (I can only theorise the higher IPC of Ivy Bridge helping it toe over the line here). Bit of a mixed bag overall, but I couldn't tell a blind bit of difference when in full-on firefights, so that's good enough for me.

Best of all, the noise has dropped through the floor. The whole case can be cooled with a single 120mm PWM fan on the H60 set to quiet. Happy isn't the word.

Thanks to all for your input.

[watercooled]

TBH I'd probably attribute at least some of that to simply remounting the cooler - as has been explained, i3/5/7 all use a negligible amount of power when idle, and all practically the same, unless some idle settings have been messed with in BIOS. If fan speed has changed at idle, it's nothing to do with it being an i3. Similar story for gaming load - with GPU-limited, CPU-light games, it's unlikely there'd be a significant difference (read: enough to make a difference to fan speed) in power consumption (therefore heat) between the two CPUs.

Just for reference, my Thuban CPU (which produces more heat than your i5) barely causes the fan speed on my Hyper 212+ to move from the idle setting of 1000 under *full* load e.g. when transcoding, and CPU temps sit around 45C. I would've expected a significantly more expensive closed-loop water cooler to be able to cool a lower power CPU at least as effectively...

Either way, you may not notice much of a difference unless you play some more CPU heavy games, or upgrade your graphics card. IMO it's best to be sure there's not some other problem than pay for a downgrade and regret it down the line.