Xeon E3 1231v3 vs. i7-4790K

Status
Not open for further replies.

kingofthecrimsons

Reputable
Jul 2, 2015
19
0
4,510
After about a year i'm feeling that my Xeon E3 1231v3 isn't as great as I hoped it to be. Thinking about selling it before it drops in price too much and buying an i7-4790K. I think spending the $60 extra might be worth it. I use it for gaming and do a tiny bit of CAD here and there. I have a Z97M OC and a rajintek adios so i'd definitely be overclocking if I get the i7. Would I see much of a performance difference between the two? And also where can I get the best price for selling my E3 1231v3? Also does a high clock rate increase performance?

Specs
CPU: E3 1231v3
MoBo: Z97M OC
GPU: GTX 970
RAM:18gb
CPU fan: Rajintek Adios
 
Solution
Selling is a crapshoot. I'd try the Tom's classifieds or Craigslist. Either of those is where you're likely to find interested buyers.

You'll definitely see improvement with the 4790k over the 1231v3, based on clock speed if nothing else. Since you can get an average of 4.5Ghz out of the 4790k, that's a full 1Ghz faster than the 1231v3. Some devices including laptops are still using processors not much faster than that, and probably slower than a Haswell Refresh chip at 1Ghz. So yes, you are going to see a difference.

Selling is a crapshoot. I'd try the Tom's classifieds or Craigslist. Either of those is where you're likely to find interested buyers.

You'll definitely see improvement with the 4790k over the 1231v3, based on clock speed if nothing else. Since you can get an average of 4.5Ghz out of the 4790k, that's a full 1Ghz faster than the 1231v3. Some devices including laptops are still using processors not much faster than that, and probably slower than a Haswell Refresh chip at 1Ghz. So yes, you are going to see a difference.

 
Solution
BS. A 1Ghz clock speed different is going to net a larger improvement than that. Especially in non-gaming applications, but even for gaming there's gains to be had. If there wasn't, no processor would be clocked higher than 3.5Ghz and nobody would bother to overclock at all.
 
I have both of them in different systems. This is the synthetic benchmark information I collected.

jl44eo.png


I now run the i7 at 4.6Ghz
 
So then, as previously stated, even at stock speeds it has some clear performance advantages over the lower clocked Xeon, and at it's overclocked speed we can assume that to likely scale with the clock speed. As is generally the case, otherwise, as said before, nobody would ever bother.
 
Do you want me to PM you a copy of the graphic or the tinypic link? (Although you could use Moderator Superpowers and just do it.)

This was a project for me to find out some actual data about the benefits of overclocking, the differences between Haswell processors and the benefit of 2400Mhz memory. I had my school kids work on it as a science fair project and got a LOT of data.
 
For most games, you'd be wrong on the improved performance. My 6700K is at 20/30% utilization in most games. I have it overclocked to 4.6, but it doesn't even run at more than 2.2GHz in the Witcher 3 (excluding menus, loading screens and playing Gwent, which is basically a menu). So only if games can REALLY use 100% of the processor would you see the improvement.

That Xeon is more than fine for gaming, and not really worth upgrading, unless you can sell it for a decent price and basically get a really cheap 4790K.
 
Of course the Xeon is fine for gaming, but YOU would be wrong about the potential for performance gains. Multiple benchmark testing and actual gameplay by a variety of forum members and moderators has shown that there IS significant performance gains involved. Just because your sample of one, that may well be mostly GPU bound due to settings or how that title is optimized, does not mean that for all scenarios your summation is correct.

I don't even recall anybody mentioning Witcher 3, but even if they had, that's just one title and not one that's seriously CPU intensive. If you're running at the upper range of settings, then the GPU is doing most of the work in that game. Other games, and other applications, such as the CAD he mentioned, will show improvements.

All that being said, what he has now is fine, nobody questions that, but to say that an extra 1000mhz will make no or very little difference is ludicrous. And I'd be rechecking your configuration as something sounds off about your utilization. That or you're not understanding that it's actually using almost 100% of one core, but very little of other cores, which ends up equating to only about 25% utilization of the entire CPU's potential including all cores.

All the more reason to want the one core that is being utilized by the game title to be as fast as possible.
 
I used the Witcher 3 as an example. It's the same for GTA V, Dirt Rally... So games that would utilize more CPU power, like Starcraft 2 or Arma III might get up there, but still, the performance gains from the Xeon to the I7 would only be meaningful with a seriously high-end graphics card and the need for 120 or 144 fps.

For almost all triple A titles that are out now, you wouldn't see any performance gain that's noteworthy, all of them are fine running on 60fps+ depending on the graphics card of course.
 


My point was that it's only a noteworthy upgrade if you have a high-end GPU, and then it's still not a very big deal for most games. Almost all new triple A titles will run 60fps+ on a Haswell-R I5.

Yes, obviously clockspeed matters, that wasn't my point.
 


...No? The fact that my CPU doesn't even run more than 2.2GHz and still runs fine means that clockspeed has diminishing returns, so it's not as important, and it won't be an upgrade for most games if you get a higher clocked CPU, since it doesn't even get utilized in games. So higher clockspeeds for non-CPU bound games are redundant, hence it wouldn't be a noteworthy upgrade for gaming.
 
Like I said before, just because it doesn't on YOUR system, doesn't mean your sample of one applies to everybody. The rest of us have had different results, like I showed in the hard data seen in the test results I linked to. There are a lot more results where those came from. A sample of one means nothing in the bigger picture, whether it's a sample of one system, or a sample of one game. It is not relevant as being representative of the effects of higher clock speed across the board.
 


But that would be a difference of over half the clockspeed on 8 threads, plus my I5 4590 runs most games as well as my 6700K, while it's only 4 threads and 3.3GHz, it just has higher usage (obviously). It still never ran at 100% for almost all triple A games I've played.

That's what I meant with diminishing returns.
 
Status
Not open for further replies.