check out my water block!

[grkM3]

I have this on the way and have 2 huge radiators before and after it.might try to tackle an ivy with it lol

 

[Idontcare]

You probably meant to post this in the Cases and Cooling subforum?

 

[Puppies04]

You can run niagara falls over an IB CPU, you will still get limited results. But i'm sure you already knew that.

 

[cmdrdredd]

You can run niagara falls over an IB CPU, you will still get limited results. But i'm sure you already knew that.

Actually it's been mentioned that heat, not voltage is what holds back some IB CPUs from high clocks.

 

[Puppies04]

Actually it's been mentioned that heat, not voltage is what holds back some IB CPUs from high clocks.

Correct, heat not temperature. Doesn't matter how fast you try to remove that heat the cores are so densely packed in IB that ambient cooling struggles way before you reach a voltage limit. Hence the fact people with full blown watercooling setups aren't getting much better temps (and therefore aren't getting much higher OCs) than high end air coolers.

 

[Kenmitch]

Strange looking. Who makes it?

 

[cmdrdredd]

Correct, heat not temperature. Doesn't matter how fast you try to remove that heat the cores are so densely packed in IB that ambient cooling struggles way before you reach a voltage limit. Hence the fact people with full blown watercooling setups aren't getting much better temps (and therefore aren't getting much higher OCs) than high end air coolers.

No way...water, good custom water, is way better than air. I know that. People on water can take a 3570k past my 4.6ghz mark easily because they can get way better temps. That is a 100% pure fact. Just ask Don Karnage here if you don't believe me.

 

[Kingkazma]

Correct, heat not temperature. Doesn't matter how fast you try to remove that heat the cores are so densely packed in IB that ambient cooling struggles way before you reach a voltage limit. Hence the fact people with full blown watercooling setups aren't getting much better temps (and therefore aren't getting much higher OCs) than high end air coolers.

i believe physics states if you remove heat faster than it's produced you get lower temperatures.

i may be wrong, but physics... it's never wrong.

 

[Diogenes2]

How do you remove it, if it doesn't make it to the IHS ?

 

[KingFatty]

How do you remove it, if it doesn't make it to the IHS ?

You may want to contact NASA for having a magical insulator, they could use that as thermal tiles on the next space vehicle.

 

[alyarb]

The heat flux vs. power profile through the IHS is still different because of the smaller die and higher apparent heat capacity of poly-tim vs solder and we were largely blindsided by that. This slightly narrows the gap between different heat exchangers compared to what we are used to, when even basic water made dramatic differences. The intermediate working fluid doesn't matter as much and will matter less over time.

I think you guys are just comparing a 120mm heat tower vs 6x120 radiators which will downplay the changing materials and geometry of your heat source.

Yes, at constant load the heat flux levels out, but during that plateau the temp is higher because the heat capacity is higher.

 

[OVerLoRDI]

What is that? Also who makes it?

 

[grkM3]

That's the best water block you can get its from 5-6 years ago and thermaltake made it

 

[OVerLoRDI]

That's the best water block you can get its from 5-6 years ago and thermaltake made it

Does it have a product name/part number? Quite curious to look this thing up.

 

[Puppies04]

No way...water, good custom water, is way better than air. I know that. People on water can take a 3570k past my 4.6ghz mark easily because they can get way better temps. That is a 100% pure fact. Just ask Don Karnage here if you don't believe me.

Really, way better? What like 50% or something? Plenty of people running 6.9ghz on their water cooling setups are they, oh sry maybe you meant 20% better surely there are tons of people running 5.4 on water... whats that oh you meant less than 10% better. Your choice of words do you a disservice. IBs heat limitations are well documented as a combination of smaller more dense cores and a possible negative performance of the IHS and thermal compound compared to previous chips which was changed due to increased thermal cycling because of the denser cores (according to intel). If you want way better performance you need to go sub ambient and normal water cooling doesn't do that.

 

[BonzaiDuck]

Correct, heat not temperature. Doesn't matter how fast you try to remove that heat the cores are so densely packed in IB that ambient cooling struggles way before you reach a voltage limit. Hence the fact people with full blown watercooling setups aren't getting much better temps (and therefore aren't getting much higher OCs) than high end air coolers.

Someone had mentioned something before about "IB showing different profiles" than earlier processors. What you say here now makes it understandable as to why.

What is the upper bound to a "safe" voltage range on the 22nm Ivy Bridgers, anyway?

Per "cmdrdredd" and remarks. It doesn't matter how much you can cool down a processor; you're still going to twist up the voltage to get those high clocks, and it will still lead to electro-migration and processor damage, even at 10C load temperature.

 

[Puppies04]

What is the upper bound to a "safe" voltage range on the 22nm Ivy Bridgers, anyway?

You will have to ask someone with an IB chip that question, keeping in mind that while a lot of the people on this forum are very knowledgeable what you will be getting is an opinion based on results from previous generations of CPUs with completly different transistors (in plain english, nobody knows how fast/slow tri-gate transistors degrade when overvolted compared to previous generation transistors).

Per "cmdrdredd" and remarks. It doesn't matter how much you can cool down a processor; you're still going to twist up the voltage to get those high clocks, and it will still lead to electro-migration and processor damage, even at 10C load temperature.

While this is true keep in mind that the cooler you keep a CPU the less VCORE you have to apply to hit any given speed so good cooling can actually prolong the life of your chip in more than 1 way.

 

[notty22]

Does it have a product name/part number? Quite curious to look this thing up.

http://www.thermaltakeusa.com/product/liquid/upgrade/cl-w0037/cl-w0037.asp#overview

 

[T_Yamamoto]

Thermaltake+water cooler=hm....

 

[grkM3]

I looked all over the net for one of these and lost a sale 2 weeks ago on eBay for 25 bucks for a used one and found another in the UK for 59bp and 29bp for shipping for a new one.

It ended up costing like 140 bucks and a decent block goes for 65-75 anyways so the extra spent was ok to me.

This is basically a huge water block that is also a huge heat sink.you could run this thing without water and just use a few fans on it.

 

[cmdrdredd]

Really, way better? What like 50% or something? Plenty of people running 6.9ghz on their water cooling setups are they, oh sry maybe you meant 20% better surely there are tons of people running 5.4 on water... whats that oh you meant less than 10% better. Your choice of words do you a disservice. IBs heat limitations are well documented as a combination of smaller more dense cores and a possible negative performance of the IHS and thermal compound compared to previous chips which was changed due to increased thermal cycling because of the denser cores (according to intel). If you want way better performance you need to go sub ambient and normal water cooling doesn't do that.

Go read forums like overclock.net and xtremesystems.org and you'll see. People who go beyond bolt on solutions for cooling are getting good results. Yes way better...it's no secret that a high end water loop can achieve temps much lower than even my NH-D14 can. It's just a lot more expensive.

You can come here and talk all you want but you still don't know what you're talking about. My CPU can go much higher than 4.6Ghz, I've had it up to 4.9. It may have even been stable if I could get the temps down. I'm sure if I did some modding (de-lid) and did a custom water loop with all the high performance stuff I could get it 24/7 at 4.9 or maybe even higher. I'm just not willing to do that or to spend the cash for top of the line water.

All you IVB haters are the same.

Someone had mentioned something before about "IB showing different profiles" than earlier processors. What you say here now makes it understandable as to why.

What is the upper bound to a "safe" voltage range on the 22nm Ivy Bridgers, anyway?

Per "cmdrdredd" and remarks. It doesn't matter how much you can cool down a processor; you're still going to twist up the voltage to get those high clocks, and it will still lead to electro-migration and processor damage, even at 10C load temperature.

No cause there's people who ran LN2 and got high clocks below stock voltage. It's not always voltage. Even some reviews mentioned that with their IB chips they got stuck on temps before they got stuck on the chip just not running the clock speed. It can scale really well, but you need the cooling solution to do it.

 

[dma0991]

Do not open another thread till you come back with an Arqtik Quadtec instead.

 

[Puppies04]

You can come here and talk all you want but you still don't know what you're talking about. My CPU can go much higher than 4.6Ghz, I've had it up to 4.9. It may have even been stable if I could get the temps down. I'm sure if I did some modding (de-lid) and did a custom water loop with all the high performance stuff I could get it 24/7 at 4.9 or maybe even higher. I'm just not willing to do that or to spend the cash for top of the line water. All you IVB haters are the same.

That is a whole lot of drivel to explain that you think less than 10% (6% near as damn it if you are interested) is "way better" and for the record I don't have a problem with IB so stop putting words in my mouth. Maybe if you read what I wrote instead of twisting it into something that fits into your apparently limited knowledge then you might not misquote people.

No cause there's people who ran LN2 and got high clocks below stock voltage. It's not always voltage. Even some reviews mentioned that with their IB chips they got stuck on temps before they got stuck on the chip just not running the clock speed. It can scale really well, but you need the cooling solution to do it.

Thanks for confirming what I already said...

While this is true keep in mind that the cooler you keep a CPU the less VCORE you have to apply to hit any given speed so good cooling can actually prolong the life of your chip in more than 1 way.

Once again you need to understand the limitations of IB (and SB to some extent) with ambient cooling. If you want to bring LN2 into the conversation then thats fine but it is exactly what I have been saying this entire thread. It doesn't matter how much air/water you push past an IB CPU you will very soon reach a point with ambient cooling where the denser cores stop you from going any higher and any further gains will have to be achieved through sub ambient cooling techniques.

Thanks for telling me I don't know what i'm talking about then agreeing with me in the same post though lol.

 

[cmdrdredd]

Dude I'm not changing anything. I am saying you can get way better clocks with better cooling and a custom water setup is way better than a NH-D14 like I have. That's just fact you want to deny. You came here looking for a fight. There is no denying that custom water is going to get lower temps and thus higher clock speeds. You implied that you couldn't scale clock speed up on IB CPUs no matter what.

What is so hard to understand?

 

[grkM3]

Guys I have 2 huge radiators and a water block with a 300watt peltier on it in the loop.

Im sure I could get an ivy to break 5.5 for benching and I'll prolly pick one up to mess with.the first thing I'll do is pop the heat spreader off and slap this water block on it.