Sign in with
Sign up | Sign in
Your question

Radiator Sizes Area Comparisons

Last response: in Overclocking
Share
March 18, 2012 5:16:21 AM

If my thinking is correct a 1x200mm rad is better for cooling than a 2x 120mm rad.
My thought is the area of the 200mm is 40000mm^2 and the other rad is 28800mm^2 (just a basic area formula of LxW)
Am I correct? I have a HAF 922 and the top of the case could support either of these without mods so I am wondering what the best solution is.
Any thoughts and insights are appreciated.
March 18, 2012 5:48:34 AM

Depends too one Fin spacing and What speed airflow its designed more, naturally bigger is better but that's just one of the things to consider
March 18, 2012 3:03:09 PM

So in what case is the 2 x 120mm rad more efficient?
Related resources
a b K Overclocking
March 18, 2012 3:23:15 PM

Don't confuse efficient with being 'good', while a higher fin count improves efficiency it also restricts airflow aka more fan noise. One of the major advantages to water cooling is lower noise, and obviously dissipating heat more efficiently with water vs air.

I recommend 20 FPI radiators over higher FPI e.g. 30FPI, plus low RPM and thermally controlled fans. The radiators >40mm also restrict too much and require higher CFM/RPM fans. The idea is 3 dimensional plus fin count. To cool a CPU and 2~3 GPUs I'd look at two 20FPI 3x120's (127x412x38).
March 18, 2012 3:44:32 PM

Okay, but with the same fin arrangement and thickness and fans higher Area will cool more efficiently?
a b K Overclocking
March 18, 2012 5:13:14 PM

I love "Okay, but" -- it's like everything you say afterwards was for nothing and voids what was said.

You are 'stuck on' efficiency, if you're deaf then sure it's about efficiency of an exchanger. It's all about dBA to cooling.

Most Exchangers (Rads if you want) will post their Wattage Disposition, what I posted above was 2000W. IF sized properly and with thermostatically controlled fans the vast majority of the time the fans won't even spin unless under high load.

I all too often see and worst hear incorrectly done cooling loops with gigantic (thick) 30FPI exchangers and fans running @ 1500~2000 RPM -- sure it cools but it completely defeats the dBA to cooling.
a c 78 K Overclocking
March 18, 2012 7:26:00 PM

hey jaquith, hows it going? are you jumping into the watercooling bandwagon?

@ OP - your thinking isn't correct, most of the questions you've asked and said "okay, but" to are answered in the watercooling sticky in the top of the section and in my sig.
a b K Overclocking
March 18, 2012 8:48:24 PM

@Lutfij - I just was bored. Migrating to this section - I seriously doubt it.

Most folks have interesting ideas and attitudes of exchangers and somehow think more is better and it's not. I find the vast majority of the exchangers, at least how I manage them, to be nearly six of one half a dozen to the other. Showing graphs with out dBA is meaningless, >2000 RPM ditto, excessive (LPM/GPM) and then 'efficiency' for low RPM almost identical.

I've seen 'skinneelabs' data and others before, and you have to look very carefully. As I recall they once tested (slipped in) an AirPlex Revolution (140mm) against several 120x360's; so unless you knew better it's comparison is meaningless.

IMO it all boils down to dBA, size, FPI, wattage disposition and very important the metals/platting. Most folks forget galvanic reaction and it's never a thought in their minds to presumably 'cherry pick' components on much less important things like efficiency. I've been known to use 10/11 FPI exchangers which are great for low resistance airflow and noise; it again depends on what I'm trying to cool and how much space I have to use.

There, I felt it is more important that the OP focuses on more important attributes.
a c 78 K Overclocking
March 18, 2012 9:03:06 PM

hmm, fair point. but this was also covered on some of the links - silence means alot of things to many people. in fact i'm nitpicky about the ticking of the motor, which is almost absent on some scythes and deltas.

about galvanic corrosion, martinslab conducted a test - showing some rads having fins soldered to the channels another reason for the gunk up loops.

FPI -
low fpi=less noisy setup
high fpi=nosiy to windtunnel setup

heat dissipation, true

apart from dba - all what you said were covered in the sticky.
a b K Overclocking
March 18, 2012 10:13:54 PM

Lutfij said:
low fpi=less noisy setup
high fpi=nosiy to windtunnel setup

apart from dba - all what you said were covered in the sticky.

Where did I say anything to the contrary regarding FPI?
March 19, 2012 2:23:59 AM

I have read all the stickys before and understand the theories and facts. I may have slightly misworded the question. Basically I will try to get as close to apples to apples as possible for a comparison. So here it is which will give the best overall performance, taking into consideration noise to cooling ratio and all other aspects.

A 200mm 20FPI that is 40mm thick with a 200mm mega flow fan with 110CFM 19dBA and 700rpm http://www.newegg.ca/Product/Product.aspx?Item=N82E1683...

vs

A 2 x 120mm 20FPI that is 40mm thick with two cooler master 120mm fans with 19.3dVA 39.77cfm and 1200rpm http://www.newegg.ca/Product/Product.aspx?Item=N82E1683...

I know these are not the best fans but going for apples to apples here with same price point and noise. I know the 120mm fans are cheap fans with low CFM and the mega flow is an average fan. Which is the best scenario?
a c 78 K Overclocking
March 19, 2012 4:07:48 AM

jaquith said:
Where did I say anything to the contrary regarding FPI?
:)  you didn't, ofc ... only stating to OP about the fpi and acoustic setup of the WC'ing venture
. and i misread the noise part in :
Quote:
low resistance airflow and noise


@ OP - if you're going ahead with the CM fans on your rads, then its clear you haven't read the stickies on any forum. You could google static pressure and the motor type of the fan. (bearing,sleeve, etc). sickle flows won't give you apples to apples comparison. rather the fans you chose won't do much but move very lil air through the rad and generate more noise than scythes AP's.

rule: don't just look at cfm to get yourself a fan for a rad - if that were enough, a leafblower would've been plenty to your needs. Or a 7950GT blower style cooler.

I'm not trying to be a prick, if you claim you read the sticky (and links for the fans) - you're not showing it.
* there's one here done by 4Ryan6. one for fans and one for TIMs.

**
Quote:
I know these are not the best fans but going for apples to apples here with same price point and noise
huh...
a c 324 K Overclocking
March 19, 2012 12:52:26 PM

To help benefit everyone that reads the sticky on radiator size and delta-T calculations, I am working on some equations and an overall average to determine cooling potential by total volume based on LxWxH and fin density of most radiators commonly used as well as adjusting for more realistic delta based on power consumption vs. power production of components at 100% load. This will likely be about 15% lower than our typical TDP/delta-T calculations, but either way, this calculation should still be in the ballpark.
March 19, 2012 1:06:23 PM

All radiator are graded via how much heat they can disipate per hour or BTU. The higher the better.
a c 324 K Overclocking
March 19, 2012 2:11:41 PM

Radiators aren't exactly rated in BTU, but watts could be converted into BTU...although this would just add another step in the computation. However, these numbers rely on several factors: fans used, ambient air temps, heat load applied, flow rate and fin density.
a b K Overclocking
March 19, 2012 2:51:33 PM

chris2341 said:
If my thinking is correct a 1x200mm rad is better for cooling than a 2x 120mm rad.

The HAF 922 grilling is ideally sized to a 2x120 exchanger. However, since it seems unlikely that a 3x120 exchanger will fit you're probably stuck with a 'fat' (50mm). To use a 200mm radiator then IMO you're going to need to do a lot of drilling; it appears like the 200mm fan has a lot of blockage.

As far as fans, it all depends upon the FPI and height of the exchanger used.

Frankly, if it were 'me' I'd either 'rework' the case or replace it. Keeping it then remove the 3.5" drive bays (you might need to remove rivets) 2x(2x120) or have (2) exchangers on top and rear externally mount a 3x120 or 4x120. The externals are okay, though be it noisier since the case isn't muffling any noise.

Though I'll get grief for it, maybe look at a Corsair H100 in push/pull.

Before doing anything MEASURE.
a c 324 K Overclocking
March 19, 2012 2:57:50 PM

Quote:
Though I'll get grief for it, maybe look at a Corsair H100 in push/pull.


Not necessarily grief from me- if the OP has no desire to pursue watercooling as a hobby or build project, this would be a decent suggestion. However, for the cost, it's difficult to validate for a CPU only solution, when there are very good air coolers for less.

I'm not sure if this thread is about a full loop design, or just cooling a CPU. I think it would be a good idea to determine what the full plan is in order to help determine the best solution.
March 19, 2012 2:59:28 PM

With both radiators being equal in all things except dimensions, yes, 200mm by 200mm is (theoretically) better (bigger) than 120mm x 240mm. (obviously)

The biggest difference are the fans. With 120mm (or 140mm) you have dozens of manufacturers producing hundreds of models in all price ranges. With 200mm you can consider yourself lucky if you find 10 different models availabe to you (and then you will find out that 5 of them are actually the same oem model rebranded/marketed differently).
Manufacturer specs concerning dBA and CFM are to be considered marketing BS (roll of the dice) until proven otherwise. The only thing remotely reliable about fan specs are RPM (and those can be off by 15%).

That still does not mean the larger rad will cool your system more efficiently - the difference might be less than 1°, and you will not even notice that and will not be able to overclock your system any higher.
a c 324 K Overclocking
March 19, 2012 3:07:08 PM

Quote:
With both radiators being equal in all things except dimensions, yes, 200mm by 200mm is (theoretically) better (bigger) than 120mm x 240mm. (obviously)


This isn't necessarily true...some radiators are much thicker than others, some have different tubing thickness or more internal tubes. Some are more than double the FPI of other radiators, thereby bumping up total heat-dissipation surface area. So, it isn't simply just about the size of the rad in the fight.
a b K Overclocking
March 19, 2012 4:10:39 PM

rubix_1011 said:
I'm not sure if this thread is about a full loop design, or just cooling a CPU. I think it would be a good idea to determine what the full plan is in order to help determine the best solution.

Either way, and depending on the OC, desired temps, tolerable noise, etc I still don't like a 2x120 or square 200mm for the CPU even for itself. Sure you can adequately cool it (CPU), but the cost is a lot of noise. I can adequately cool with an H100. However, I prefer 50C~58C with very low noise vs H100 68C~71C hear it in the next room.

If going through the aggravation and expense then IMO do it right, and in most cases doing it right doesn't include kits nor poor dBA/CFM fans i.e. CM. Doing it right: decent and long lasting pump, adequate sized reservoir, 3x120 exchanger, quality block e.g. EK/Koolance.

Kits like XSPC Raystorm RX240 are pretty much useless. I've even seen their 360 kits only beat the, as simple as it gets, H100 by <5C~8C. Nice comparision -> http://www.overclock.net/t/1164638/my-corsair-h100-revi...

I AM NOT PUSHING THE H100, the idea is to know what to expect. You need to look at the 'Big Picture' and realize Pro/Cons/Costs. It ain't cheap to have a mid-50C's and utlra-low noise setup.

So it depends on how 'Loopy' you decide to get.
a c 324 K Overclocking
March 19, 2012 4:23:17 PM

Quote:
If going through the aggravation and expense then IMO do it right, and in most cases doing it right doesn't include kits nor poor dBA/CFM fans i.e. CM. Doing it right: decent and long lasting pump, adequate sized reservoir, 3x120 exchanger, quality block e.g. EK/Koolance.


Agree- a custom loop has the potential to perform better and quieter depending on components chosen.

Quote:
Kits like XSPC Raystorm RX240 are pretty much useless. I've even seen their 360 kits only beat the, as simple as it gets, H100 by <5C~8C. Nice comparision -> http://www.overclock.net/t/1164638 [...] ustom-loop


5-8C can be a decent amount of change depending at load or idle. I would definitely suggest a Rasa/Raystorm kit over an H100, but really only when considering similar price points: ~$100-$130. If you go beyond that, I really think strong consideration should be given to some of the entry Swiftech kits or the Rasa/Raystorm kits with the D5 pump.

And yes, a single 120 or 140 rad can cool a CPU, depending on your setup. If you are going this route, though, you are better off with a good air cooler unless you need a LCS cooler like an H70 due to space and mounting constraints. There really is a gradual step of cooler choices once you get to $75 and above, but each really should be considered for what you actually are trying to accomplish.

Quote:
So it depends on how 'Loopy' you decide to get.


I concur- this is definitely the meat and potatoes of the discussion we are intending to explain.
a b K Overclocking
March 19, 2012 4:48:15 PM

You are mixing together (2) different Kits. I said " even seen their 360 kits only beat the, as simple as it gets, H100 by <5C~8C." not to be at all confused with their 2x120 '240 kits', in that case it's a toss-up and maybe a few C's less BUT basically the same loud noise.

edit/Internally the OP cannot fit a 360 kit! There's only enough room for the 240 kit...etc. I therefore made suggestions about having (2) 240's to cool and reduce noise.
a c 324 K Overclocking
March 19, 2012 4:57:58 PM

Yeah, I saw that in the link you had before. When it comes to cooling CPUs, it doesn't take as much to cool a CPU when comparing with other options, which is why watercooling is more a luxury and hobby than a necessity when it comes to overclocking current-gen processors. However, it really starts to shine when you start incorporating GPUs into the loop as well, as they benefit far more from the cooling capability of watercooling...as much as 30-40C difference of temps.

Regardless, when it comes to CPU-only cooling, it's hard to really justify many high-end options that are liquid based with current CPUs. This is why the XSPC 360 and the H100 really do perform relatively in the same ballpark.
a b K Overclocking
March 19, 2012 5:32:48 PM

I happen to have an H100 on my i7-3930K and under full load 4.8GHz @ 1.4v 67C~72C but loud (AIDA64/Prime95), whereas on other X79's with 3/4-WAY that I've built 2x360 (min) (depending on the case/some have 3 exchangers) 55C~60C with ultra-low noise. Both work. **

However, if you're a gamer then IMO full block. In my case, I have one rig for gaming full Koolance so it's fully blocked. In contrast, the i7-3930K is all about work and not for gaming i.e. enterprise SQL testing/coding so the GPUs just sit there. Even with a 4~5 hour job not making too much noise.**

** So it depends entirely on use.
a c 324 K Overclocking
March 19, 2012 5:42:32 PM

Yeah, I usually turn my fans way down when I'm just surfing the web, etc., but will beef them up on high when I need to. Even when gaming, I don't need to run my fans past 50% (2000 and 3000 rpm Ultra Kaze's) as they are pretty strong fans. However, for long runs of BOINC/SETI/Milkyway @ home, I turn them up on high just to let them blow dust out of my rads.

I'm not the biggest SQL guy, but work with queries on a daily basis, so learning it. I'm more a OS/network/hardware guy. It's always interesting to hear what other members do for a living and even as hobbies.
March 20, 2012 3:26:58 AM

Okay so to address a few of the comments and concerns. This is going to be mostly for a hobby and project as I already have my q6600 overclocked to 3.0 idling at 23 -28 on the four cores and my GPU will be upgraded when kepler arrives so that is a non issue. However I would like to watercool it.

I have no problems modding my case and have done some minor mods already. Here is an easy way I have found to make room for a 3 x120mm rad:
http://forums.legitreviews.com/about22692.html

The H100 to me seems like a lot of money when my Cm 212+ is giving me good results already and the H100 has not hobby/fun aspect and is pricey. I personally would not buy one when you can get the above mentioned kits for 200-300 dollars.

'Silence' is not a huge issue for me but quiet during idling/web surfing would be important I do not mind turning up fans for gaming.

Since you all seem to be very educated in this area. It would be great if you could recommend possible kits or just parts to build a custom loop in the 200 dollar price range and one in the 300 dollar price range and outline the added benefits.

I live in Canada so shipping can be a problem with some sites.

This is something I have put together not sure if it is any good but after reading some threads and charts I think it is okay.

Pump Swiftech MCP655 $90
http://www.frozencpu.com/products/2128/ex-pmp-27/Swifte...

Rad Swiftech MCR320-XP $73
http://www.frozencpu.com/products/13547/ex-rad-263/Swif...

Swiftech Apogee HD waterblock $75
http://www.frozencpu.com/products/14365/ex-blc-1000/Swi...

Then I will need a res the micro REV I heard is good and only about $25

I will also need fans, tubing and fittings which I think will come to about $60
Any recommendations here would be great.
This totals $323 dollars before taxes. I believe there are some spots such as the waterblock that I could spend less. But you all know from experience.

Compared to an all in one kit from various companies. With the all in one kits I see decent performance/value and the ability to upgrade parts progressively as I have more money to spend on the system. As well as compared to something like the H100 it has better value moving forward for upgrade ability.
Here are a couple kits I have seen, I know some of you do not like the RASA kits but I will put it in for discussion.

XSPC RASA 750 RX360 $190
http://www.frozencpu.com/products/14183/ex-wat-182/XSPC...

EK H30 360 HFX $320
http://www.frozencpu.com/products/15036/ex-wat-200/Ek_H...

EK H30 360 LTX $237
http://www.frozencpu.com/products/15211/ex-wat-206/Ek_H...

Swiftech H20 320 Edge $270
http://www.frozencpu.com/products/14573/ex-wat-188/Swif...

Thanks for everybody's time as this is becoming very helpful.
a c 78 K Overclocking
March 20, 2012 5:18:54 AM

323 huh... go over to frozencpu and take a look at the xspc raystorm extreme 36o kit - is customizable with the tubings and you get a strong pump.

http://www.frozencpu.com/products/15140/ex-wat-203/XSPC... to which you can add more rads and gpu blocks.

and if you're in canada - dazmode has the same kit plus one more with another pump - but you miss out on the tubing option.

stay away from the EK's mate will thanks us later for not going with them.

use colored tubing, not dyes in the loop.
a c 324 K Overclocking
March 20, 2012 1:53:47 PM

I just came from a Q6600 @3.4ghz...it was a hell of a chip. A lot of grunt and personality in those old Core2 Quads.

I'll post back in later today- just getting some of that first-in-the-morning stuff taken care of.
a b K Overclocking
March 20, 2012 3:35:52 PM

IMO - don't waste money on this now, it's time to first upgrade your system. CPU, MOBO and RAM -- probably best to wait for the IB (May~June depending on which rumor mill you follow). Reading between the lines it seems like money is tight, and you're bored just wanting to do something. Water cooling is fine, but I see no sense in poor allocation of your resources (money).

Therefore, ass-u-ming you'll upgrade to LGA 1155 then only get components not some kit(s) and make sure they are compatible with LGA 1155 or 2011 if that's an option; think ahead.

Further, I would measure very carefully and do what I do on any new build -- create a cardboard mock-up of both the exchanger and its' fans, a simple rectangle is best. Then make damn sure it'll fit. Example, the LGA 2011 often have massive VRM's that extend past the MOBO's mainboard's dimensions of LxW.

Either EK or Koolance, pick one and stick to it. Yes, EK in the past had some platting issues but that's long since past. Often I see folks mixing components and mucking-up their loops. Pumps, they're all DELPHI and re-branded (Swiftech, EK, Koolance, Laing, Danger Den, etc.).

Just get a pump that looks like this (Molex + RPM):
a c 324 K Overclocking
March 20, 2012 3:42:48 PM

Quote:
Further, I would measure very carefully and do what I do on any new build -- create a cardboard mock-up of both the exchanger and its' fans, a simple rectangle is best. Then make damn sure it'll fit.


I very much concur.

Quote:
Either EK or Koolance, pick one and stick to it. Yes, EK in the past had some platting issues but that's long since past. Often I see folks mixing components and mucking-up their loops. Pumps, they're all DELPHI and re-branded (Swiftech, EK, Koolance, Laing, Danger Den, etc.).


You shouldn't have issues using components from different manufacturers. EK's nickel plated components are still under scrutiny by those who it initially affected, but copper obviously was not an issue.
a b K Overclocking
March 20, 2012 4:00:27 PM

rubix_1011 said:
You shouldn't have issues using components from different manufacturers. EK's nickel plated components are still under scrutiny by those who it initially affected, but copper obviously was not an issue.

I see it all of the time. Oddly as this may seem, I'm a Chemist by education and know all about corrosion, oxidation, galvanic corrosion (reactions), etc. - http://www.engineersedge.com/galvanic_capatability.htm

Koolance's product warranty does not cover the use of 3rd-party coolants, coolant additives, or corrosion. Koolance LIQ-702 or LIQ-705 coolants are strongly recommended to help avoid issues with mixed metals or biological growth. Additionally, do not use aluminum with bare (unplated) copper or bare (unplated) brass in the same system. Do not use nickel with silver in the same system. - http://www.koolance.com/technical/myths/gold_plating_co...

Most folks that mix blame -- the blocks, the coolant, the platting, you name it vs blaming themselves for doing something stupid. I have NONE of these issues because I Do Not Mix parts unless I 100% know all of the metals and platting used by ever single component. Mixing even the barbs/compression fittings can be a problem and ditto with many of the radiators.
a c 78 K Overclocking
March 20, 2012 4:08:02 PM

bloody, this jaquith guy amazes me with that brain of his - you're going to be the best addition to the Watercooling section ever :D 

just kidding -

Quote:
Mixing even the barbs/compression fittings can be a problem and ditto with many of the radiators.
yeah, i saw that on overcolock.net...or was it on realredraider :/  black chrome fittings with corrosion.
a c 324 K Overclocking
March 20, 2012 5:08:40 PM

Galvanic corrosion is a fact, regardless of components you run if metals aren't all the same alloy. Brass and copper are typically used in most fittings and radiators and copper is the primary metal of most blocks. Nickel plating also reacts with these same metals, but as we all know, aluminum is the bad apple of the bunch and can greatly amplify the window of time for corrosion to occur.

Coolants that are corrosion inhibitors are suggested by manufacturers because this allows them to limit their liability and state their specific terms when honoring RMA's on damaged or defective products. There is no way they can 100% confirm the components you are running in a loop and they won't ask you to send everything to their labs simply to decide whether to honor an RMA or not. While there is truth to using a corrosion inhibitor in a loop, the main reason for their claims are to develop a user loyalty toward their products and 'strongly suggest' only using their products...including their coolants. If Koolance and EK are so strongly supportive of their claims to only use their products and coolants, I would expect them to supply their coolant free of charge with any of their block purchase. I would assume that their argument would be that their coolant will protect their products and should be used 100% of the time to ensure warranty and RMA terms. I am by no means disagreeing with the concepts of metal reaction from different manufacturers, I am disagreeing with Koolance and EK and their claims that their products are only superior if you follow their guidelines...while other manufacturers with like products do not appear to have the same issues (nickel plating corrosion).

I do agree that manufacturers will tend to use the same types of metals in many of their components, thus lessening the chances you'll have components that chemically react with one another. This does beg the question of why there are so many products that DO NOT seem to have a problem in a mixed manufacturer loop. I've watercooled for almost 10 years and haven't had corrosion on any of my components and I've only run distilled water, biocide, killcoil and on occasion, dye. The last time I ran a coolant, it was Swiftech Hydrx that came with my H20-80 kit, which was my first watercooling setup...I ran that coolant for 3+ years.

Quote:
bloody, this jaquith guy amazes me with that brain of his - you're going to be the best addition to the Watercooling section ever


I completely agree; he adds a significant amount of technical insight and knowledge.
March 22, 2012 2:32:11 AM

rubix_1011 said:
I just came from a Q6600 @3.4ghz...it was a hell of a chip. A lot of grunt and personality in those old Core2 Quads.


I know they are a good chip especially for their age. What did you upgrade to and what kind of performance increase did you get?


Quote:
http://www.frozencpu.com/products/ [...] ?tl=g30c83 to which you can add more rads and gpu blocks.


http://dazmode.com/store/index.php?main_page=product_in...
This kit is $30 dollars cheaper but has a different pump, and both kits have blue LEDS that I believe can be switched for red, can anyone confirm these to points? And maybe comment on the pump differences.

Quote:
Further, I would measure very carefully and do what I do on any new build -- create a cardboard mock-up of both the exchanger and its' fans, a simple rectangle is best. Then make damn sure it'll fit.


Very good idea I will make sure I do this. And especially if I get a new motherboard. It may be a tight squeeze with a 50mm rad.
a c 324 K Overclocking
March 22, 2012 1:33:07 PM

Quote:
http://dazmode.com/store/index.php [...] ts_id=1874
This kit is $30 dollars cheaper but has a different pump, and both kits have blue LEDS that I believe can be switched for red, can anyone confirm these to points? And maybe comment on the pump differences.


D5 pump is a great pump, very good performer; much better than the stock X20 750 pump from XSPC. I have been using the same D5 for over 6 years.

Quote:
I know they are a good chip especially for their age. What did you upgrade to and what kind of performance increase did you get?


I went from the Q6600 @3.6 to an i7 2600 (non-K, although I kick myself for that).

When it comes to gaming and most stuff, both chips are very good and fast. Once you get to very processor core and thread-based applications like Folding, BOINC/SETI or video encoding, photo editing...the i7 shows some real brute strength. Even an i5 is a fantastic chip. Depending on your budget, and what you do, you might consider only upgrading graphics cards (if you game or perform a lot of CUDA processing). When it comes to sheer CPU to CPU comparisons, the i7 is better and has more power, but against an overclocked Q6600 at 3.4 or 3.6, for most everyday applications, it still holds it's own. You aren't going to see a lot of real-world difference until you throw a ton of work at a new Core chip.
!