Sign in with
Sign up | Sign in
Your question
Closed

I5 3570K vs 2500k

Last response: in Overclocking
Share
July 25, 2012 1:38:12 AM

My birthday is coming up and I am hoping to get a new CPU/Motherboard.

The motherboard that I have picked out is "ASRock Z77 Pro4 LGA 1155 Intel Z77 HDMI SATA 6Gb/s USB 3.0 ATX Intel Motherboard" http://www.newegg.com/Product/Product.aspx?Item=N82E168...

I am hoping to get my first overclockable CPU though and am drawn to the i5 2500k and the i5 3570k. The 3570k is only $10 more.

Which one is the better CPU? Also which one can overclock better on stock fans? I do not plan on doing water cooling.

Also how much of an upgrade will I get from my current system which is a Core 2 Quad Q9300 at 2.5GHZ?

Thanks!

More about : 3570k 2500k

a c 186 à CPUs
a b V Motherboard
a c 150 K Overclocking
July 25, 2012 1:44:11 AM

You won't get that much of an upgrade form a Q9300, a better upgrade would be waiting a few more months for Haswell.
Score
0
July 25, 2012 5:22:08 AM

You should invest a few extra bucks in an aftermarket CPU Cooler. I just finished my new build and I snagged a Cooler Master Hyper 212 Plus for $20.00 after rebate from Newegg. It is keeping my CPU considerably cooler than the stock fan. You will want an aftermarket cooler, especially if you plan on overclocking. Intel stock fans are garbage. Just make sure if you decide to get an aftermarket cooler it will fit in your case, some of them can be quite bulky.
Score
0
Related resources
a b à CPUs
a b V Motherboard
a b K Overclocking
July 25, 2012 7:04:04 AM

An i5 would be almost twice as fast as your Q9300, maybe significantly more with overclocking, but whether or not that would be even a significant increase in gaming performance is a different thing to consider. This would also depend on your graphics, your display(s), the games that you play, and at what settings you're playing at.

An after-market cooler is necessary for overclocking with Intel. Intel's stock coolers, as matt1990 said, are crap.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 12:52:57 AM

One thing is for sure is don't waste your money on any dual core these days.

I got an i5 system as well a 775 build. An i5 quad is a lot faster than any 775 era quad but if you can overclock your current build then tell us your complete specs.

As for Ivy Bridge don't waste your money as their overclocking isn't great without removing the ihs.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 1:05:46 AM

^seriously? Ivy does OC reasonably well and because it is a more efficient cpu/clock Ivy is a solid cpu choice for decent OCs and it uses less power. It may run hotter but with a good cooler, a good computer can be built. It will also give you USB 3.0 and PCI-e 3.0 if paired with a z77 mobo.
-Bruce
Score
0
July 27, 2012 2:01:51 AM

amuffin said:
You won't get that much of an upgrade form a Q9300, a better upgrade would be waiting a few more months for Haswell.


Haswell is just months away? Ivy was just released and I am not happy with the performance.
Score
0
July 27, 2012 2:04:10 AM

Don't OC in stock cooler. Buy 2500k and spend in some cash to get Hyper 212 Evo and OC.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 2:06:37 PM

AyuX said:
Haswell is just months away? Ivy was just released and I am not happy with the performance.


Ivy was never supposed to be a big leap over Sandy. Haswell is supposed to be a big leap over Sandy. Ivy was just a test-drive for Intel's 22nm node whereas Haswell will be the new architecture on the 22nm node (Ivy and Sandy are almost the same architecturally).
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 2:30:36 PM

If you want better CPU performance just buy a cheap cooler that will fit in your case and start overclocking!
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
When i looked some reviews of your processor i saw it can go as high as 3.4GHz!That more than enough to drive any single GPU graphics card!

Or if you dont want to overclock and you still want more CPU performance wait for Haswell.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 3:23:54 PM

Kamen_BG said:
If you want better CPU performance just buy a cheap cooler that will fit in your case and start overclocking!
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
When i looked some reviews of your processor i saw it can go as high as 3.4GHz!That more than enough to drive any single GPU graphics card!

Or if you dont want to overclock and you still want more CPU performance wait for Haswell.


That's not even close to being enough for a Radeon 7970 or GTX 670/GTX 680 in any CPU limited games. An overclocked 7950, 7870, and when ti comes out, probably also the GTX 660 TI, would all be far to much for a Core 2 Quad (even the 45nm versions) at 3.4GHz in several, perhaps many, games. It would be enough for most single-GPU graphics cads, but not all of them in all games.
Score
0
July 27, 2012 3:46:03 PM

2500K vs 3750K

Really the only difference here is PCI 3.0. The 3750K gets a bit hotter than the 2500K, and the 3750K gives you a 3% performance increase. This mostly comes down to personal preference. I went with the 2500K, but generally I have been recommending the 3750K. Truthfully you cannot go wrong with either one unless you are going to buy a PCI 3.0 GPU.

On the note of a CPU cooler, the Hyper 202+ or evo is a great choice and will get you anywhere you want to go to the point where you need a $200-$400 custom water cooling system for benchmarking.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 4:15:57 PM

DarkOutlaw said:
2500K vs 3750K

Really the only difference here is PCI 3.0. The 3750K gets a bit hotter than the 2500K, and the 3750K gives you a 3% performance increase. This mostly comes down to personal preference. I went with the 2500K, but generally I have been recommending the 3750K. Truthfully you cannot go wrong with either one unless you are going to buy a PCI 3.0 GPU.

On the note of a CPU cooler, the Hyper 202+ or evo is a great choice and will get you anywhere you want to go to the point where you need a $200-$400 custom water cooling system for benchmarking.


IB is not a 3% increase over SB. The performance increase depends greatly on the workload. Please don't try to simplify what is too complex to be simplified because doing so doesn't work. The performance increase varies from about 2% to over 10%, depending on the workload.

Also, for gaming, PCIe 2.0 versus PCIe 3.0 is hardly any difference except in extremely high end systems even if you use a PCIe 3.0 video card. Also, video cards and GPUs are not synonymous either. We don't call computers CPUs, so why would you cal a video card a GPU? A GPU is a part of a video card just as a CPU is a part of a computer. Doing things like this is contributing to computer ignorance. Yes, the 3570K and the 2500K are generally very close in performance, even if overclocking is considered for them and yes, the 3570K gets hotter in most situations, so it can be said to come down to personal preference.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 4:34:43 PM

Quote:
CORRECT


INCORRECT


You tell me that I'm incorrect after agreeing that IB is not only a 3% increase over SB in all workloads. Tom's tested this too. Go ahead and explain how that makes sense.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 4:44:32 PM

Quote:
not worth arguing or even correcting you dude, your a lost cause..
I said your incorrect and you are, just to blind to see it or believe it.
the charts and benches say so as well but I guess you out rank them huh.?
not.


http://www.tomshardware.com/reviews/ivy-bridge-benchmar...

The charts disagree with you.
Score
0
a c 186 à CPUs
a b V Motherboard
a c 150 K Overclocking
July 27, 2012 7:28:28 PM

OK FINE, 3.7% :p 
Score
0
July 27, 2012 7:44:16 PM

amuffin said:
OK FINE, 3.7% :p 


CRAP....we were .7% wrong! We are such failures....
Score
0

Best solution

a b à CPUs
a b V Motherboard
a b K Overclocking
July 27, 2012 8:00:43 PM

So to answer your question, OP, there's no noticeable difference in performance (overclocked or not) between the two. Neither will OC on stock coolers, but with a decent air cooler both will hit ~4.6GHz.

The 3570k has the advantage of having native USB3 (not really a big deal) and PCIe 3 (not really a big deal right now, maybe eventually it will be). The 2500k has the advantage of running cooler and being $10 cheaper.

Choose whichever you think is best, because those are really the only noticeable differences (which aren't terribly noticeable). Neither will get you better gaming performance over the other.
Share
July 27, 2012 8:11:08 PM

mousseng said:
So to answer your question, OP, there's no noticeable difference in performance (overclocked or not) between the two. Neither will OC on stock coolers, but with a decent air cooler both will hit ~4.6GHz.

The 3570k has the advantage of having native USB3 (not really a big deal) and PCIe 3 (not really a big deal right now, maybe eventually it will be). The 2500k has the advantage of running cooler and being $10 cheaper.

Choose whichever you think is best, because those are really the only noticeable differences (which aren't terribly noticeable). Neither will get you better gaming performance over the other.


Correct. I also want to throw out there to the OP that you may or may not hit 4.6 with either chip. Every chip is different, and the only thing we know for sure in reguards to overclocking both of these chips is that the 3750k will be hotter. Statistically this means that the 2500K chip can be clocked higher without water cooling, but there is that chance that you could get a really really good 3750K chip. Both chips -should- be capable of hitting 4.5, which seems to be the most common overclocking speed.
Score
0
July 27, 2012 11:46:55 PM

Thanks for all of the info guys!

If anyone is wondering my GPU is a Radeon HD5770.

I cannot overclock my core 2 quad CPU because my motherboard is a Gateway custom board that came with my pc and the bios do not allow for overclocking.

Some more questions:

1. Some people have said that the 3570k runs hotter. How much hotter does it run, since I thought that the 22nm process is cooler that what the 2500k is built on?

2. Also is the performance of the 3570k really double that of my current CPU? I know that on paper is might be but what about real world tests? Are there any benchmarks of the 3570k/2500k vs the Core 2 Quad Q9300 that you guys know of?

3. I do some video converting, game playing, and graphic design work. Also a lot of web surfing with a lot of tabs open.

I currently have 6GB of DDR2 PC2-5300 Ram and I often get laggy when on tumblr. Would getting 8-16GB of DDR3 PC3-12800 RAM be a huge improvement over my current setup? I do not think the ASRock Z77 Pro4 motherboard supports DDR2 RAM.


Thanks!
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 28, 2012 12:08:26 AM

shadowedge said:
Thanks for all of the info guys!

If anyone is wondering my GPU is a Radeon HD5770.

I cannot overclock my core 2 quad CPU because my motherboard is a Gateway custom board that came with my pc and the bios do not allow for overclocking.

Some more questions:

1. Some people have said that the 3570k runs hotter. How much hotter does it run, since I thought that the 22nm process is cooler that what the 2500k is built on?

2. Also is the performance of the 3570k really double that of my current CPU? I know that on paper is might be but what about real world tests? Are there any benchmarks of the 3570k/2500k vs the Core 2 Quad Q9300 that you guys know of?

3. I do some video converting, game playing, and graphic design work. Also a lot of web surfing with a lot of tabs open.

I currently have 6GB of DDR2 PC2-5300 Ram and I often get laggy when on tumblr. Would getting 8-16GB of DDR3 PC3-12800 RAM be a huge improvement over my current setup? I do not think the ASRock Z77 Pro4 motherboard supports DDR2 RAM.


Thanks!


1. How much hotter, quite significantly. It's hotter because of Intel using crap thermal paste instead of flux-less solder between the CPU die and the IHS of the CPU. It's almost a thermal insulator compared to the flux-less solder and it's so bad that it negates the lower heat output of the smaller process node.

2. It probably wouldn't be exactly double, but it could be around there. I don't know of direct comparisons between Ivy Bridge and Core 2, but there probably are some. If not, then there are definitely Core 2 comparisons to Nehalem and Sandy bridge and there are comparisons of Nehalem and Sandy bridge against Ivy Bridge, so you'd just need to compare the results of two reviews to get the info that you want if we can't find a direct comparison.

3. this is not a question.

Increasing the amount of RAM that you have probably won't make a difference unless you're already running out at 6GB. Increasing the frequency will help to an extent, especially coming from PC2-5300. You are correct in that Z77 does not support DDR2. No LGA 1155 motherboard that I've heard of supprots it. This is because the memory controller is located on the CPU and the CPUs for this platform only have DDR3 controllers, unlike AMD's AM3 CPUs that have a DDR3 and a DDR2 controller.
Score
0
July 28, 2012 1:13:44 AM

blazorthon said:
Ivy was never supposed to be a big leap over Sandy. Haswell is supposed to be a big leap over Sandy. Ivy was just a test-drive for Intel's 22nm node whereas Haswell will be the new architecture on the 22nm node (Ivy and Sandy are almost the same architecturally).


Maybe the fourth gen core i3 will give you the performance of 2500k! Expected release?
Score
0
July 28, 2012 1:15:52 AM

mousseng said:
So to answer your question, OP, there's no noticeable difference in performance (overclocked or not) between the two. Neither will OC on stock coolers, but with a decent air cooler both will hit ~4.6GHz.

The 3570k has the advantage of having native USB3 (not really a big deal) and PCIe 3 (not really a big deal right now, maybe eventually it will be). The 2500k has the advantage of running cooler and being $10 cheaper.

Choose whichever you think is best, because those are really the only noticeable differences (which aren't terribly noticeable). Neither will get you better gaming performance over the other.


10$ CHeaper? In our country it's 100$ CHeaper!!
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 28, 2012 1:26:33 AM

AyuX said:
10$ CHeaper? In our country it's 100$ CHeaper!!


What country has the i5-3570K $100 more than the i5-2500K and is this a different $ from $USD?
Score
0
July 28, 2012 1:40:56 AM

blazorthon said:
What country has the i5-3570K $100 more than the i5-2500K and is this a different $ from $USD?


India. Here corsair 800d is 400$, GTX 670 600$.... I am converting my currency into USD
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 28, 2012 3:36:07 AM

AyuX said:
India. Here corsair 800d is 400$, GTX 670 600$.... I am converting my currency into USD


Jeeze, and I thought that Australia had bad pricing.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 28, 2012 11:22:24 AM

Quote:
I vote for 2500K..A 2500K @ 4.8 on air is better than a 3570K that's limited to less than 4.5 on air due to temp issues.


Hardly. That less than 5% performance lead isn't very important and is not even close to being noticeable in-game.
Score
0
July 28, 2012 6:01:15 PM

blazorthon said:
Hardly. That less than 5% performance lead isn't very important and is not even close to being noticeable in-game.


So...what your basically saying is that the 3.7% performance lead that the 3750k has over the 2500k is even less important...lol :pt1cable: 
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 28, 2012 6:16:41 PM

DarkOutlaw said:
So...what your basically saying is that the 3.7% performance lead that the 3750k has over the 2500k is even less important...lol :pt1cable: 


You're trying to simplify the performance difference again despite the fact that this is too complex of a subject to simplify in such a way. Just because it averaged at 3.7% doesn't mean that depending on the application, it could vary extremely from 3.7%. You also continue to call the 3570K the 3750K despite the fact that no such CPU model number exists with any desktop SKU of Ivy Bridge. There's also the fact that the 2500K, when overclocked, is still usually slightly faster than the 3570K, not the other way around, granted as I said the difference is too minimal to matter either way.

That Ivy Bridge has slightly to somewhat higher performance at the same frequency doesn't change the fact that it usually has lower real-world performance with overclocking because the paste between the CPU die and the IHS keeps the Ivy Bridge CPUs running too hot when overclocked. I never said that the increase of Ivy over Sandy was very important. I simply said that it wasn't what you said it was.
Score
0
a b à CPUs
a b V Motherboard
a b K Overclocking
July 28, 2012 8:34:58 PM

Offensive post has been hidden by moderator, and so the quote of that post is removed. -Proximon

I'm impossible when it comes to other people being wrong and then trying to call me wrong when what I said is right, especially when they're saying that I said things that I didn't even say in a futile attempt to prove their incorrect point. I suppose everyone who disagrees with you must have a mental disability as far as you're concerned, especially when they're not backing down on something that they're correct about. Someone wants to say that Ivy is 3.7% faster than Sandy?

Fine, maybe I'll be there to point out that even Tom's own benchmarks clearly show the performance difference at the same frequency to range from 0% and 7.6% in their tests in their very easy to read chart. 0% is not 3.7% and 7.6% is not 3.7%. Averages don't work here because they are only useful when comparing larger data sets; averages aren't a one-size-fits-all answer, especially when someone tries to apply them in a real-world scenario. You can't accurately expect everything to be 3.7% faster just because that is an average. You can accurately expect several real-world scenarios similar to those tested by Tom's to average out to around a 3.7% boost in performance at the same frequency, but each application could vary widely from that average in the performance difference, granted the performance itself is still similar due to he fairly small numbers in this.

When it comes down to the individual tasks, Ivy is clearly shown to not only be 3.7% faster. It can be the same or it can be more than double the 3.7% number. The margin of error can stretch 100% behind the average and apparently a little over 100% higher than the average. That's a huge margin of error when it's 3.7%, give or take about 3.7%. Acting as if these aren't there, even if they aren't very impacting differences in actual performance due to the numbers being fairly small, is promoting the spread of intentional ignorance and short-cut taking that society as a whole seems intent on doing regardless of the stupid sacrifices made in doing so.

This is a public forum and nobody here is required to read through lies and over-simplifications that lose the actual meaning of what they were intended to simplify when they ask questions. I admit that I don't speak for anyone other than myself when I say this, but if I ask a question to people who are supposed to be able to answer it, and they give answers, then I don't just expect their answers to be accurate rather than gross over-simplifications, that is exactly what I should receive in those answers just as it is what any other person asking a question here should receive in an answer if they get an answer. I'm fairly sure that people don't come here to be lied to by someone who is willing to write an answer, yet not willing to give them a correct answer.

Heck, even my initial numbers from earlier in this thread were shown to be a little optimistic and thus wrong when I posted a link to Tom's chart. I'm not saying that I'm infallible. However, the numbers are right in front of you. Whether or not anyone cares is their problem, especially if they try to incorrectly simplify them for what I can only see as convenience. So, yes, I'll be impossible to agree with when the opposing side of the argument is shown to be quite clearly wrong by reputable sources. I don't know about any of you, but I consider Tom's to be an at least generally reputable source and this chart seems to be quite proper. It clearly shows that what DarkOutlaw said is not quite true.

http://www.tomshardware.com/reviews/ivy-bridge-benchmar...

Beyond that, no self-respecting person that I know would take kindly to someone saying that they said something that they didn't and that someone attempts to use this as an apparent personal attack. Not even typing the proper model number of the CPU even once is just ridiculous.
Score
0
a b à CPUs
a b V Motherboard
July 28, 2012 9:43:43 PM

OP, you have had more than enough advice here. Despite the lack of respect that some people show others that disagree with them. Ignore those parts and all that is left is lots of data and good advice... perhaps it's all a waste of air though.

All you needed was the simple post I have selected as best answer.
Score
0
a b à CPUs
a b V Motherboard
July 28, 2012 9:45:14 PM

Best answer selected by Proximon.
Score
0
a b à CPUs
a b V Motherboard
July 28, 2012 9:46:19 PM

This topic has been closed by Proximon
Score
0
!