Build PC over "Intel Xeon E3-1231v3"

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Xzeon

Reputable
Jul 15, 2014
36
0
4,540
hallo people, I am trying to build new PC for me, I will heavy use it for 3D Production and 1080p Video Editing, heavy Rendering...soft which I will use: 3DsMax with Vray + RT, Cinema 4D, After Effects, Premiere, PS and etc. I will also make some simulations using Phoenix FD..(Gaming we can Ignore)...I have HDD, Case (ATX, MiniATX, MicroATX) and OS..

I made my choice with CPU,GPU and I need to resolve other parts around it

253$ - Intel Xeon E3-1231 v3 (If here is someone who knows difference good/bad with E3-1231 v3 and E3-1230 v3, please describe it. I know 31 is new and +100MHz, but what means TSX I do not know)
240$ - MSI GTX 760 OC 2GB

this is parts which I think best performance vs cost. if there is better solution for 250$ will be happy (i5 4690K is also good but new Z97 is costs more and it has 4 core 4 thread over 8, and without overclocking i5 it cant catch up with E3 so I decide no Overclock and buy Xeon with stable cool experience)...also I do not need iGPU

so my main complication is Motherboard and Memory (I want 16GB overall). I do not know I need ECC or Buffered, and I need help. also I do not know how many Watt PSU I need and which brand is good (80+ Gold is good I know and modular would be nice)

what about 107$ Gigabyte GA-H87M-D3H ?? but It does not have PCI Express 3.0 right?
it is from HERE

thanks everyone, waiting solutions :)
 
Solution
Xzeon,

Sorry, sorry, I somehow completely missed your July 16 post.

Yes the difference in the CPU's is that the E3-1231 v3 is 3.4 /3.8GHz and the E3-1230 v3 is 3.3/3.7GHz, and tsx seems to be a very complex protection of multiple thread integrity that defines thread initiation/ completion. However I haven't taken time to understand it properly. This reads as though it's a benefit on multi-threaded applications which rendering often is.

I think your parts list in general is a good direction but I wasn't able to find mention of the ASUS H97-Plus supporting Xeon E3. It's possible, but if ASUS doesn't list it on their "Supported CPU List" they may well not help you if you have problems.

In the future when upgrading to a...


You just blew my mind!
Yes, that was the computer I was considering...the 6 cores and the 24 gb' s of ram enticed me.
But this new system you showed me with the video card... I'm so confused...here's a question, is there a performance difference between buying something used, compared with a new item? Like if I removed 16 gb of ram and installed brand new 16gb ram, would there be a dramatic difference?
 
djClaw wrote:

"You just blew my mind!
Yes, that was the computer I was considering...the 6 cores and the 24 gb' s of ram enticed me.
But this new system you showed me with the video card... I'm so confused...here's a question, is there a performance difference between buying something used, compared with a new item? Like if I removed 16 gb of ram and installed brand new 16gb ram, would there be a dramatic difference?"



djClaw,

The RAM in a Lenovo S20 or Precision T3500 is going to be limited to the RAM the chipset limits. The T3500 can use up to DDR3-1333 ECC so changing it alter alter the performance- it will be the same thing. If you put in a faster speed like DDR3-1600, it will still run at 1333. If you put in 1066, it will run at 1066.

However, new graphics cards are in general better in every way- performance, features,and lower energy use- and therefore less heat. If you bought the T3500 with a Quadro 5000- which is a double height, $2,200 GPU from 2010-11, it would cost $350 for a new Quadro K1200 to equal it. My point in bringing up that particular system was that the Lenovo would need the GPU changed on the first day and the price difference with the T3500 of $200 would not be enough to pay for a new card as good as the Quadro 5000. I've never had a Lenovo, but I tend to think of Precisions as having very high build quality and really good support- they're better than HP too.

Cheers,

BambiBoom
 


 
Hey so what's the deal with AMD processors?http://www.newegg.com/Product/Product.aspx?Item=N82E16819113284
This is an 8 core, really inexpensive processor...why such a price difference compared to Intel?
 


djClaw,

I admire AMD in many ways as they produce some good performance at reasonable prices. The Firepro workstation GPU's are very good for the cost and in the upper end, more calculation-oriented than Quadros. NVIDIA, in search of specialized, optimized perfromance created the Tesla co-processors whereas AMD retains both functions in Firepro. I tend to think that Quadro drivers have been quite a bit better, but Firepro is catching up and every so often they leapfrog Quadros, or in the Radeon, top GTX, and then it switches back. For visualization work, I'd always have a Quadro but for business and analytics with large datasets- Firepro.

With the CPU's, there a difference in the way AMD and Intel cores are characterized. The FX 8350 has a high clock speed and is called an 8-core, but this is equivalent to an Intel hyperthreading 4-core. On Passmark, the CPU score for an FX-8350 is 8975 and rated No.131. For comparison, an i7-3820 (4-core, hyperthreading @3.6 /3.8 from Q1/ 2012) scores 9005 rated 129. So, by the numbers, there is similar performance from the FX-8350 and a 4-core, hyperthreading Intel i7.

The prices are of course quite different and this is reflected in the relative refinement. Higher performance from 3 years ago from a lower clock speed i7 means the Intel architecture is producing more cycles per second through a higher calculation density than the FX-8350. I'm not an expert in the architecture, but my guess is that the nature of hyperthreading, which looks for opportunities to 'fill in" the data streams for each core, is more efficient than running separate cores with discontinuous data streams. The AMD is holding it's own by a higher clock speed, but in the end, the throughput of the hyperthreading has a wider bandwidth. Given the billions of calculations per second, small efficiencies in calculation density add up. Historically, AMD has required more power for the same calculation density but the last iteration of the FX-8350 reduced power consumption and it's now rated as 125W. For comparison, a Xeon E3-1245 v3 (4-core @ 3.4 / 4.0 GHz) scores 9522 at No. 99 and is rated to use only 84W but typical consumption from the E3 is 68W and the FX-8350 is 159W *** which suggests the Intel is much more efficiently scalar to the data loading.

*** http://cpuboss.com/cpus/Intel-Xeon-E3-1245-v3-vs-AMD-FX-8350

By the way, a Xeon X5690 scores 9233 rated at No, 117.

Given the price difference between AMD and Intel CPU's for similar performance, the AMD the cost /performance is extraordinarily good, but an "8-core" AMD needs to be compared to a 4-core hyperthreading Intel.

Cheers,

BambiBoom