i5 4690k vs e3 1231 v3 processors

[Rhezner]

So i did some research and i found the e3 1231 v3 processor and i compared it with the 4690k which is 10$ cheaper and here are the differences.

xeon pros:
- hyperthreaded
- has workstation technologies
- better stock performance
- more reliable
- lower tdp by 8 watts

Xeon cons:
- not overclckakable
- no integrated GPU(not necessarily a con because you aren't paying for something you probably wont use.
- 10$ more.

4690k pros:
- 10$ cheaper
- overclockable
- has integrated graphics (not a pro if you have a dedicated GPU)
- higher tdp by 8 watts

4690k cons:
- weaker stock performance
- not hyperthreaded
- doesnt have workstation capabilities
- not as reliable


honestly, in the long run the xeon will cost about the same as the 4690k due to lower tdp, so think of them as the same price.

The real question is would you rather have hyperthreading and some other technologies or be able to overclock

 

[synphul]

I think the point was missed. How do the basics of math apply that 4.6ghz is possible on an i7 with ht (which consumes slightly more power) and somehow only 'realistic' to hit 4.1ghz on an i5? It was proving the point that 4.6ghz on a haswell is quite possible and a reality. Trying to say 4.1 is all someone will reach is a misleading conclusion with not much to back it up in an attempt to downplay the i5. It doesn't have an algorithm in the board or chipset or vrm that says oh well this board can only do 200mhz over, regardless of the starting point. It doesn't work that way. 4.6 is 4.6 is 4.6.

Of course there are better boards. There are better boards for the xeon as well. And better cpu's than the 4690k or the xeon.. not sure what that's saying really. Just pointing out that 'oh, overclocking is so hard' or 'oh, overclocking takes a $200 mobo' or any of the other unfounded myths are just that. Myths, continually perpetuated around from thread to thread.

If someone can only get 4.1 (200mhz over turbo) they either got one of the most rare and horrible overclocking chips ever or they don't know what they're doing when overclocking. It's not some mystic black magic voodoo, it's really pretty simple.

 

[CTurbo]

I think you are missing the point that the 4790k runs 4.4ghz on the stock Intel cooler whereas the 4690k runs at 3.9ghz at roughly the same temps on pretty much the same cooler. So just because the i7 4790k can run 4.6ghz doesn't mean the 4690k can. Some 4690k's never see 4.6ghz even on the best motherboards. By your logic, the 4690k will run 4.4ghz on it's stock cooler... because 4.4ghz is 4.4ghz is 4.4ghz


You are downplaying the motherboard's importance in overclocking. It is easily just as important as the cpu in determining maximum overclock.

 

[BiggerBluer]

This maybe not a big deal but I think the xeon cpu comes with a better stock fan. If you need me too I can look up the video that said this but again this is probably not a big deal to most just a thought

 

[synphul]

It does take slightly more vcore to push the 4690k than the 4790k to the same speeds. It's a slightly lower binned chip and costs $100 less. As for power draw, I don't see the 4690k requiring more power.

http://www.digitalstorm.com/unlocked/intel-devils-canyon-i5-4690k-review-and-overclocking-benchmarks-idnum301/
4690k oc'd to 4.6ghz, 151.6w full load.
4790k @ 4.4ghz, 144w full load.
4790k @ 4.8ghz, 219w full load.

Close in both clock speeds and power draw, slightly higher clocks, slightly higher power draw. If the power draw going through the vrm's can't support 151.6w under full load with the 4690k oc'd, how does it manage to support the i7 at stock?

Everything from vcore and system power seems to fall in line clock for clock here as well.
http://www.anandtech.com/show/8227/devils-canyon-review-intel-core-i7-4790k-and-i5-4690k/2

Power draw is what it is, a wall outlet doesn't care that you're using a vacuum cleaner or a hair dryer if they're pulling the same power. If you turn a hair dryer to 'high' and it pulls the same power as the vacuum, the plug doesn't suddenly explode because of a component being on 'high' settings while continuing to power the vacuum just fine. If that were the case, the z97 anniversary should come with a warning - not compatible with i7 cpu's. Of course there are better boards with more features and better quality for a higher price, that's true of anything. I wouldn't expect the z97 anniversary to have all the bells and whistles, but doesn't mean it's incapable.

The 4790k being able to hit 4.6ghz more commonly than the 4690k has more to do with silicon quality rather than voltage differences which were claimed to be the issue overclocking a 4690k on such a 'weak' motherboard. Intel also felt more comfortable shipping the 4790k with full warranty with a turbo of 4.4ghz rather than 3.9ghz of the i5. Part of that extra $100 cost factor. That's nothing to do with power phases or vrm's on the motherboard.

It just seems funny that despite conflicting facts the argument is made a realistic oc for the 4690k is only 200mhz over turbo. This has been proven untrue by various users and review sites. Is it a possibility a 4690k will only oc to 4.1ghz and nothing more? Sure. It's also been proven that sometimes ht hurts performance and it's better to turn it off. It's not the norm though.

First the argument was the 4690k takes too much power oc'd, the motherboard won't handle it. Not true. Then it gets argued that the 4690k can't be oc'd on the stock cooler to the same speeds as the 4790k (despite being 2/3 the cost of the i7). This may be true, but for the $100 difference I think a $25 cooler isn't too much to whine about. Also, how many users are pushing their stock 4790k's and having trouble keeping temps under control on the stock cooler? Many. Then the argmument is something else.

I never said heat would be identical or the stock cooler would do the same job, I said that when it comes to power draw it takes xyz amount of power to get the same clocks out of the same cpu chip whether it's been clocked that way from factory or done by the user. Proof in the links above.

Just like some people will argue overclocking only gets 400mhz faster speed and does nothing for performance yet at the same time advising people to get a faster locked core cpu over a slower model because it will perform better. As if overclocked mhz are somehow different. By that logic an i7 though 500mhz faster is a complete waste of time since the extra speed over the i5 isn't worth the money and has no impact on performance - which it clearly does. Sounds like digging for a reason to try and bash on the i5, when one tactic doesn't work try another. Maybe it's just me.

 

[CTurbo]

I never said anything about power draw. Raising voltage is very hard on a motherboard.

A 4790k at 4.4ghz is barely breaking a sweat.
A 4690k at 4.4ghz is working VERY hard.


Now you are saying that power phases and vrms are not important or a determining factor?





I'm done with this argument unless somebody else has something to say about it.

 

[synphul]

Apparently you missed the testing already shown in the links I provided. The 4690k at 4.4ghz 1.174v load volts. Coming through the power supply from the wall, through the vrm to the cpu, yes? The 4790k at 4.4ghz 1.199v load volts. Coming through the power supply from the wall, through the vrm to the cpu, yes? Are you seriously trying to say the LOWER volts at the same 4.4ghz consumed by the 4690k are magically more stressful than the 1.199v taken by the 4790k at the same clock speed? I surely hope not.

I never said power phases and vrm are unimportant, I'm simply showing you in plain math how flawed your logic is. Math doesn't lie.

I have a feeling I'm through arguing as well, it's like talking to a brick wall. Though this made me laugh a bit. "I never said anything about power draw. Raising voltage is very hard on a motherboard." Apparently other people draw power through magic purple clouds and I'm the lone ranger who's system draws power in the form of volts and amps. Thanks to ohm's law I think it's pretty safe to say that a particular mobo and its circuitry are static values with the same resistance. All else being the same, resistance being equal and voltage being equal from the same power source, amperage too must be equal or it would defy ohm's law.

Therefor for the sake of others, according to the values provided in my sources who have tested the voltage between the i5/i7, they can do the math for themselves. Rather than be conned into thinking that an overclocked vs stock cpu both using the same voltage and resulting in similar power consumption somehow mystically 'stresses' the motherboard any differently from one another. I would agree if the voltage required were higher, if the resulting amperage throughput were higher but this simply is not the case according to the figures obtained through testing rather than assumption.