Okay I was just wondering what the point is of some of the Xeons that exist, let's take a look at the Xeon e5 1650 identical to the i7 5930k in every way except it supports ecc ram and supports Vpro.
How benefitial is this as I always see people saying the regular "ECC Ram corrects errors that would make a normal computer crash and is very necessary for servers" Okay the server part yes it makes sense but for workstations is it ever going to help you I always see people talking about ECC ram but nobody has mentioned if it helped and how it did. Also this CPU is not compatible with the Intel c612 chipset so no dual CPU.
So it is a non over clockable i7 with ecc ram support, Vpro support and it costs more. I have heard that Xeons last longer than i7's so that's something to think about.
Well in my day to day travel as a sysadmin ive never seen a server with an i7 To make it quick Xeons are "Server" Parts, rated for 24/7 use, lasting longer than most i7's Gamers dont ECC Ram so why put it on an i7, server make use of error correcting data in ram Xeons are really meant for business/servers and you cant buy an 18 core i7 well not yet at least kinda like low end quadro and the titan are the same
Some xeons like the 1231 v3 is a better choice than an i7 4790, they're virtually the same but the xeon is cheaper
If your working with big data sets, large 3d models, or big video projects, ecc can be very handy. Say your an hour into a two hour video render, and there is an error in the memory. That whole hour or rendering is wasted. It will crash and you will be forced to start over.
Imagine having about a project being simulated for 200 hours. Due to one small error in one byte, every calculation is off by a little bit. Nobody notices it until someone notices a bend in the final product. 24/7 is also something consumers normally do not need. Nobody expects Xeons or Opterons to be overclocked, so no support for OCing.
ECC Support (but ECC not required) Able to address waaaaaaaaay more memory Rated for 24/7 use Usually rated for much higher temperatures Differences in instruction set support, and various performance or security features
People might tell you that you can't overclock on a Xeon also, but that is not true. If you're using a dual CPU motherboard, it's going to be really hard and you're not going to be able to overclock far at all, but for the chips that are similar to their i7 counterparts, you should be able to overclock just fine (assuming a single CPU mobo that supports overclocking). I'm running an overclocked Xeon right now, though it's an older X58 chip. It overclocks very well.
I forgot about more memory, also yes they use the Intel64 instruction set (very similar to amd64) I think I am aware that you can overclock but the multiplier is still locked on these CPU's
My VMWare ESXi servers have lots of RAM, storage and many CPU cores. None of them particularly fast but just lots of them. Plus I have over a hundred servers running 24/7. I've had maybe two DIMMs in the last 5 years or so go bad compared to probably 10s of HDDs. So far I don't think I've had a CPU go bad just go obsolete!
OP, I think that the real answer to your question is branding. From an enterprise or professional perspective, there are the couple of minor differences outlined above that may come into play in specific workloads, but from a consumer perspective, there is little difference. Architecturally, there is almost nothing to distinguish, say a Xeon 1650 v3 from an i7 5930K, and they are even cut from the same die; the differences in the capabilities of the chips are essentially artifically imposed by Intel ex post facto. Under the vast majority of consumer or even prosumer usage scenarios, you would be equally well served by either chip. But this is not to say that Intel is being willfully deceiving: when you occupy the market position Intel does and cater to different clienteles, branding is exceedingly important. Put it this way: enterprise customers come to Intel and say "We need a single-slot 6-core server Xeon," and it offers them the 1650 v3, whereas an enthusiast comes to Intel and says "I want to run four video cards with an i7, but can't afford a 5960X," and it has the 5930K. Insiders know that they're essentially the same chip, but the branding difference helps different kinds of buyers to make purchasing decisions. Xeons can use regular, unbuffered DDR4 just fine, and the 1650 v3 overclocks like a champ, so from a functional standpoint, Intel could axe the 5930K from its lineup tomorrow and it wouldn't make a difference, but it would definitely confuse some buyers. "Huh, there's no i7 between the $400 5820K and the $1000 5960X?!!"
All that having been said, however, I do think that Intel needs to bring some sanity to its product nomenclature. The mere fact that we are here on a forum full of pc enthusiasts and IT professionals struggling to tease out the differences between two products over a year after release should offer some indication that the naming scheme is less than ideal. I also think that this problem is yet another example of what happens when Intel has no effective competition in the high-end market. When you don't have a meaningful competitor, you get these bloated lineups with the same product being re-labeled again and again, or these artificial distinctions meant to disguise the lack of innovation. AMD, Qualcomm, IBM: somebody, please stand up to Intel.
Two xeons is a lot faster than any single i7. If you're only doing basic things and playing video games than an i7 would be best. But for rendering dual xeons is far better.
Heck even if you don't render things , you can set the affinity so one cpu plays your games while another run videos or browsers or capture software.
Didn't know silicon could have grades. so cheap cpu's would have silicon where the atoms weren't lined up vs expensive cpu's where they are lined up better? lol
They are all made the same and are binned based on testing. So your xeon and i7 are both cut from the same silicon wafer. Generally the silicon in the center of the wafer has less impurities than that of the outside so chips cut from the center end up as xeons and higher end i7s. After testing they enable or disable features within the chips, set stock clock speeds and brand them as whatever, but physically they are all the same.
People who choose Xeons for home use often do claim to do so because they are manufactured to a higher standard, that they are made to be more stable, and so on, and of course Intel plays these issues up in its marketing materials, but I wonder whether this is actually the case.
In my experience, i7s do run into issues more frequently, but I would attribute that to the fact that people overclock them and do other things to introduce instability. If you used them in exactly the same way with otherwise identical setups, I suspect that an i7 would last just as long and perform just as well as its Xeon counterpart, but one would have to do extensive testing to be sure. Twenty years ago, CPU failure was something to worry about, but these days, even the mainstream Intel chips are quite hardy. I am continuously amazed by my Nehalem-Sandy Bridge-era systems, which take a ton of abuse and just keep plugging away.