|
I did note that HardOCP and other reviewers are noticing poor manual overclocking performance and needing to use auto-overclocking to get best results, this may point to the need for some BIOS tweaking and optimization for best results.
|
#
?
Jun 10, 2014 20:04
|
|
|
|
| # ? May 8, 2024 14:48 |
|
|
The winners of Intel's overclocking competition at Computex did delid their winning chip: They also had quite a few to choose from.
|
|
#
?
Jun 10, 2014 20:12
|
|
|
MisterAlex posted:And have y'all really been simultaneously virtualizing a bunch of systems on an i7 rather than a Xeon? That's not exactly the right tool for the job. Besides ECC, and the previously unavailable VT-d extensions: where is the benefit of using a E3 xeon w/ hyperthreading have over an i7 for a virtualizing box? I'm not trying to turn this into the virt thread, but for the typical office 2 XP, 2000 and linux instance even i5 box should be a fine and acceptable choice for anything except HPC and storage. The extra ommph of devil's canyon might even make it attractive to put a older OS in that can't make use of multiple cores but need single threaded speed.
|
|
#
?
Jun 10, 2014 20:19
|
|
|
Alereon posted:I did note that HardOCP and other reviewers are noticing poor manual overclocking performance and needing to use auto-overclocking to get best results, this may point to the need for some BIOS tweaking and optimization for best results. Yes I kind of "discounted" the HardOCP review in my head after saying they couldn't get any manual value stable period. Probably something else happening there. Also the Tom's review got above average numbers at least for vcore. I kind of wish they just pushed it ... 1.31 is pretty lame considering the excitement over this chip. But all in all, the idea of 5.0 haswell for everyone seems to be dwindling at the moment.
|
|
#
?
Jun 10, 2014 20:29
|
|
|
More logical CPUs available for ESXi to schedule jobs against mostly helps with multiple vCPU situations. ESXi is quite aware of the differences between physical and logical CPUs but a good example of when it definitely helps performance is when you have a VM with 2 vCPUs. You could hypothetically schedule concurrent workloads to two different physical CPUs, but you'd be better off scheduling them both to the same and getting some memory locality advantages. With two different VMs' vCPUs, you're probably better off with using separate cores first. I'd consider an i7 like that for use as a development machine without having to spring for a full blown Xeon system. A possibly use case (home that is) besides that is to repurpose an older machine as a server by grabbing one of these instead of building an entirely new machine.
|
|
#
?
Jun 10, 2014 20:31
|
|
|
Welmu posted:The winners of Intel's overclocking competition at Computex did delid their winning chip: Interesting how delidding was so promonently displayed at an intel sponsored event. It's like saying they care about overclockers to organize the event but still won't solder the IHS for them on the unlocked parts. 'Dont try this at home kids'
|
|
#
?
Jun 10, 2014 20:33
|
|
|
Shaocaholica posted:It's like saying they care about overclockers to organize the event but still won't solder the IHS for them on the unlocked parts. Closest credible explanation: Tom's Hardware, reviewing 4790K posted:We don’t know much about the “Next-Generation Polymer Thermal Interface Material” Intel is now using on its Devil’s Canyon parts. But we do know Intel worked uncharacteristically fast to implement it. In fact, the speed at which the company moved is cited as one of the reasons it couldn’t switch back to the solder that figured so prominently in Sandy Bridge overclocking successes.
|
|
#
?
Jun 10, 2014 20:35
|
|
|
SpaceBum posted:Besides ECC, and the previously unavailable VT-d extensions: where is the benefit of using a E3 xeon w/ hyperthreading have over an i7 for a virtualizing box? My point was simply that the lack of the advanced virtualization features on those chips should not be a reason to suddenly feel let down about one's existing Core i-chip. It did (and continues to do) what it was meant to.
|
|
#
?
Jun 10, 2014 20:38
|
|
|
i7s and Xeons overlap in the workstation world. Especially if you go to dual socket workstations you'll have to use Xeons.
|
|
#
?
Jun 10, 2014 20:44
|
|
|
MisterAlex posted:The Xeon is intended to be a server-level chip. That's what it's made for. And Core i-chips were and are intended for user-level applications (e.g. workstations, consumer home PCs, etc.). That's just a fact. You can use either for the other task, but so can you pound in nails with a ball-peen hammer. Either way, the problem people were having is that the non-k i5 and i7 supported it, when the k's did not. Besides isn't the Xeon 1240v3 exactly the same as the i7-4770 minus igpu and with added/turned on features? I think you'd be hard pressed to find much of a functional difference between comparable Xeon and I7 cpu's, unless you needed ECC memory support, etc. There is no reason you can't just stick it in one of many regular 1150 motherboards and have it perform exactly the same despite being made for servers.
|
|
#
?
Jun 10, 2014 21:00
|
|
|
Ignoarints posted:Besides isn't the Xeon 1240v3 exactly the same as the i7-4770 minus igpu and with added/turned on features? I think you'd be hard pressed to find much of a functional difference between comparable Xeon and I7 cpu's, unless you needed ECC memory support, etc. There is no reason you can't just stick it in one of many regular 1150 motherboards and have it perform exactly the same despite being made for servers. That's correct. In fact if you're building a system with a 4770 that's going to have a discrete video card anyway might as well save a few bucks and get the 1240v3.
|
|
#
?
Jun 10, 2014 21:07
|
|
|
Krailor posted:That's correct. In fact if you're building a system with a 4770 that's going to have a discrete video card anyway might as well save a few bucks and get the 1240v3. Wait, are they socket compatible? Unlocked multiplier?
|
|
#
?
Jun 10, 2014 21:09
|
|
|
Shaocaholica posted:Wait, are they socket compatible? Unlocked multiplier? Same socket, no unlocked multiplier. It's 100 mhz slower than a 4770, but also cheaper (well, was, no idea about now). There is no igpu (save a whole 4 watts). It is as far I remember simply a 4770 otherwise. It was a pretty common question a little while ago whether or not to get one over a 4770 and there was very little reason not to but the fact it said Xeon on it turned many away. edit: http://www.newegg.com/Product/Product.aspx?Item=N82E16819117315 updated it to match the 4790 . So you can see the appeal still
|
|
#
?
Jun 10, 2014 21:12
|
|
|
Ignoarints posted:Same socket, no unlocked multiplier. It's 100 mhz slower than a 4770, but also cheaper (well, was, no idea about now). There is no igpu (save a whole 4 watts). It is as far I remember simply a 4770 otherwise. It was a pretty common question a little while ago whether or not to get one over a 4770 and there was very little reason not to but the fact it said Xeon on it turned many away. That's identical to a 4771 for $40 less. $30 less than the SLOWER 4770, and of course that's locked! For some reason I hadn't thought of Xeons, but I think people definitely should, is there some reason this hasn't featured more prominently in the parts picking thread? vv Yeah, Xeon is a more high end brand than Core HalloKitty fucked around with this message at 21:37 on Jun 10, 2014 |
|
#
?
Jun 10, 2014 21:27
|
|
|
Xeon brand has a negative connotation? They've always been super baller to me. Especially since its the only want to get 8/10 core Intel parts.
|
|
#
?
Jun 10, 2014 21:31
|
|
|
HalloKitty posted:That's the same clocks as a 4770K for $60 less. Obviously it's locked, but still, Xeons. $30 less than the SLOWER 4770, and of course that's locked! For some reason I hadn't considered them, but I think people definitely should.. Woops I meant "match the 4790" as in, 100 mhz less, like the 1240v3 was 100 mhz less than its 4770 counterpart. But because of the refresh it makes it even more interesting now. I wasn't sure they'd release a refresh version of that as well but apparently they did. One caveat is make sure the motherboard lists Xeons as supported. I don't think I ever found one that didn't, but I didn't look all that hard either. edit: The Xeon name certainly means a lot more once you go up the chain. But when we're talking about these comparable ones I'm pretty sure we're talking apples to apples. I'm also positive they clocked it 100mhz less on purpose just to slot it "below" the i7 here. Ignoarints fucked around with this message at 21:36 on Jun 10, 2014 |
|
#
?
Jun 10, 2014 21:31
|
|
|
HalloKitty posted:That's identical to a 4771 for $40 less. $30 less than the SLOWER 4770, and of course that's locked! For some reason I hadn't thought of Xeons, but I think people definitely should, is there some reason this hasn't featured more prominently in the parts picking thread? Integrated GPU is invaluable in diagnostics and a cheap hedge against having no computer if you end up with a bum video card.
|
|
#
?
Jun 10, 2014 21:51
|
|
|
Is the Xeon iGPU just turned off or actually absent?
|
|
#
?
Jun 10, 2014 21:57
|
|
|
Again, this is all irrelevant. My point extends to any purchase situation. If you buy product ABC and are satisfied with the purchase, it's petty to later exude an air of entitlement and complain that, "Wait a minute. I should've received XYZ, too!"
|
|
#
?
Jun 10, 2014 21:58
|
|
|
Shaocaholica posted:Is the Xeon iGPU just turned off or actually absent? Probably turned off but it might as well be absent.
|
|
#
?
Jun 10, 2014 22:00
|
|
|
^^^ Having an igpu backup is indeed cool. Is it worth 12-15% more for just that purpose? I dunno. HalloKitty posted:is there some reason this hasn't featured more prominently in the parts picking thread? It says Xeon on the box, truly. I just realized this now though. Before, a 4770 and a 4770k was essentially the same if you didn't overclock, making the 4770k a pointless expense. Now that's not so much the case anymore. I'd honestly feel kind of weird putting down $315 on a CPU when the same exact chip 10% faster is "just" another $25. However, if I were to consider the 1241v3, that $25 becomes $70 and, well, that's more of a difference at the very least and less appealing, especially if I didn't really need 4.0 ghz anyways. Maybe they'll be more popular now MisterAlex posted:Again, this is all irrelevant. My point extends to any purchase situation. Well... like I said, the concern before was the cheaper, slower versions of the same exact processors (not even talking about xeons here) had features, particularly virtualization , that were disabled in the more expensive, faster, unlocked versions. That kind of thing is annoying. Just because virtualization is a "server" oriented thing doesn't make any of that any less lame. And now it is suddenly enabled for k versions. Ignoarints fucked around with this message at 22:05 on Jun 10, 2014 |
|
#
?
Jun 10, 2014 22:01
|
|
|
Shaocaholica posted:Is the Xeon iGPU just turned off or actually absent? Turned off, usually because of a fab flaw. They laser off its connections so you can't try to force it back on.
|
|
#
?
Jun 10, 2014 22:01
|
|
|
HalloKitty posted:is there some reason this hasn't featured more prominently in the parts picking thread? Consumer mobo's are hit and miss when it come to supporting xeons and it's features. ASrock and Asus were unofficaly supporting them on their q87 and z87 chipsets On top of that without a specific need for xeon's features just going with a K version and a quality Z*7 chipset mobo means the user can squeeze out a year or two of acceptable performance by overclocking. Makes better sense in the long run. Being clocked 100Mzh lower doesn’t mean much in real world performance results when turbo boost is being used.
|
|
#
?
Jun 10, 2014 22:03
|
|
|
Shaocaholica posted:Is the Xeon iGPU just turned off or actually absent?
|
|
#
?
Jun 10, 2014 22:29
|
|
|
Alereon posted:...drivers certified for professional workstation applications... Does this really mean anything to real people/businesses? I've worked at a big corporate video game company and a VFX company and it seems like it doesn't really matter. Maybe its just marketing.
|
|
#
?
Jun 10, 2014 22:39
|
|
|
Shaocaholica posted:Does this really mean anything to real people/businesses? I've worked at a big corporate video game company and a VFX company and it seems like it doesn't really matter. Maybe its just marketing.
|
|
#
?
Jun 10, 2014 22:50
|
|
|
HalloKitty posted:For some reason I hadn't thought of Xeons, but I think people definitely should, is there some reason this hasn't featured more prominently in the parts picking thread? Same reason that i7's aren't "featured" in the parts picking thread - most people building general use/gaming machines are fine with an i5 which is still ~$50 cheaper than that Xeon.
|
|
#
?
Jun 10, 2014 22:58
|
|
|
beejay posted:Same reason that i7's aren't "featured" in the parts picking thread - most people building general use/gaming machines are fine with an i5 which is still ~$50 cheaper than that Xeon. There are i5 equivalent Xeons too, but I grant you there aren't too many price points where this makes sense. Still, it's an option.
|
|
#
?
Jun 10, 2014 22:59
|
|
|
MisterAlex posted:Again, this is all irrelevant. My point extends to any purchase situation. I don't buy this line of reasoning. I mean, actually, I would agree if that's what anyone were saying, but if you'll use your time machine (in this very thread, just keep on going back pages) you'll see people grumblingly just having to fuckin' deal with a lack of features, and a bunch of talk over whether this is going to have any impact on programming and wider rollout of the "exotic" extensions, especially TSX, since that's one where you'd conceivably see somebody bringing work home and the demographic overlap there suddenly had a hard wall for no clear or given reason that basically just said "you can either have the new extensions OR you can overclock, same chip otherwise, you pick." If it's a validation issue, it's a validation issue, but what in the gently caress in your estimation would cause those particular special instruction sets to suddenly stop working right at 4.4GHz instead of their previous max turbo ceiling? This is not a radical re-imagining of the chip, that much is clear, it seems very very likely that no actually it was just a market segmentation thing and even Intel decided maybe it was a bad idea after all (since their chip is faster now than stock chips were then, and since it was never explained at the time, you dig it?). It's very easy to acontextually gripe about people being "petty" and "exuding an air of entitlement" but that's not what happened if you've been paying attention.
|
|
#
?
Jun 11, 2014 00:29
|
|
|
So you're saying that you were upset about the lack of those features before you begrudgingly bought your existing Core i-chip?
|
|
#
?
Jun 11, 2014 01:12
|
|
|
Yeah. Believe it or not, people were extremely peeved that TSX and/or Vt-d were being omitted from enthusiast processors, because both those features can give significant performance boosts that enthusiasts might enjoy.
|
|
#
?
Jun 11, 2014 01:13
|
|
|
Fair enough. You won't hear another peep from me regarding the complaints about hamstrung features.
|
|
#
?
Jun 11, 2014 01:15
|
|
|
MisterAlex posted:Fair enough. You won't hear another peep from me regarding the complaints about hamstrung features. Intel have been erratic with enabling/disabling things like x64, VT-whatever, AES and so on over the last few years, presumably in the name of market segmentation. "Pentium" branded chips have had higher end features switched on than "Core i3" chips from the same generation, for example. For those of us old enough to have run dual CPU Pentium II systems, it's quite annoying.
|
|
#
?
Jun 11, 2014 01:53
|
|
|
Mr Chips posted:Intel have been erratic with enabling/disabling things like x64, VT-whatever, AES and so on over the last few years, presumably in the name of market segmentation. "Pentium" branded chips have had higher end features switched on than "Core i3" chips from the same generation, for example. For those of us old enough to have run dual CPU Pentium II systems, it's quite annoying. Since when has intel disabled x64 on a capable part?
|
|
#
?
Jun 11, 2014 02:20
|
|
|
MisterAlex posted:It's quite possible previous chips did not pass validation and those things were disabled. That happens all the time. Intel's manufacturing process is always improving; they may have decided that "Okay, we can now reliably produce i7 chips that pass these feature tests." It was always artificial market segmentation. The likelihood of some significant fraction of the die population working ok only if TSX is disabled is basically nil. I have no inside info, but I do know enough about this kind of thing to state that with high confidence, and I'll try to describe why. There's two categories of things to think about here, defects and process variation. Point defects are usually what's behind selling partially defective chips as good ones. They are what they sound like: a particle of some kind of contaminant fucks up a circuit wherever it happened to land. Harvesting dies that have defects is almost always about designing in redundant copies of block structures which are already being replicated for other reasons, or are somewhat standalone and not required by all customers. Got a memory array with 32 columns? Add a 33rd, some fuses, and a little bit of extra logic, and now you can make a good chip out of one that has a defect in one column. Got an entire GPU that not everyone needs? Sell chips with defective GPUs into markets that don't mind the whole thing being disabled. (Though I think it's more likely that Intel does this mainly to reduce power for those markets.) Note that harvesting requires extra design effort, including circuitry to completely isolate a defective block. It might be consuming excessive power even when not clocked, it might be pulling signals on a bus low or high. You have to depower and disconnect it. This is why it's not a realistic explanation here. If you have a defect in TSX-NI silicon, that's a defective CPU core. No way Intel designed that as an isolate-able function -- by its nature, it has to be part of too many critical timing paths in the CPU core to move it out into its own little island. Process variation is shorthand for all the things which can cause variance in transistor performance. Most of these issues are regional in nature: all the chips from some region of the wafer don't perform as well, or this entire core doesn't have as high a Fmax as the others in the same die, and so on. Usually this happens thanks to reasons such as slight but non fatal misalignment between process steps, or variations in the thickness of deposited films. These variations occur over relatively large areas, not tiny little spots. Once again the nature of the disabled functions tells us that harvesting of some sort isn't what's going on. For example, AMD once harvested 3-core CPUs from quad core dies because sometimes they'd have one of the four cores have a significantly lower Fmax, or use too much power at high frequency. But with tightly integrated functions like TSX-NI there's no way for those circuits to test out slow without the rest of the core also testing out slow (and vice versa). Alereon posted:Most Xeons feature Intel HD Graphics P4600/4700, which has drivers certified for professional workstation applications. This can be a pretty compelling benefit since even the shittiest Quadro and FirePro cards are pretty expensive. This too is an artificial distinction, including Nvidia and AMD standalone GPUs. It's literally that it costs a lot to support some of these pro applications (usually because they have been around so long that they use caveman 1990s style OpenGL, which requires a lot of fiddly work to emulate on any modern GPU core). So they include fuse bits which ID the same exact GPU core as the "pro" version, sell the cards at prices where they can actually make a profit despite the support costs and smaller sales figures, and prevent customer cheating by having the pro drivers look for those IDs. Intel's rationing of features like TSX-NI is actually motivated by much the same thing, extracting more money from smaller user bases which need difficult to validate features. It seldom makes economic sense to manufacture a wholly different product line without the whizbang feature, since in the silicon world the cost for taping out and validating fundamentally different variants is so high. Personally I think Intel goes a bit overboard with it, but there's a pretty strong economic argument that if this kind of practice went away entirely in the chip industry (note: everyone does it) we'd see less innovation and new features.
|
|
#
?
Jun 11, 2014 02:27
|
|
|
Shaocaholica posted:Since when has intel disabled x64 on a capable part? Atom.
|
|
#
?
Jun 11, 2014 02:30
|
|
|
BobHoward posted:This too is an artificial distinction, including Nvidia and AMD standalone GPUs. It's literally that it costs a lot to support some of these pro applications (usually because they have been around so long that they use caveman 1990s style OpenGL, which requires a lot of fiddly work to emulate on any modern GPU core). So they include fuse bits which ID the same exact GPU core as the "pro" version, sell the cards at prices where they can actually make a profit despite the support costs and smaller sales figures, and prevent customer cheating by having the pro drivers look for those IDs.
|
|
#
?
Jun 11, 2014 02:35
|
|
|
Alereon posted:Oh it certainly is, I'm just saying that since the graphics comes free with a Xeon CPU, you can get out of paying the nVidia/AMD workstation graphics tax, which is one reason why the IGP is valuable even if you don't plan on using the compute performance. Didn't mean that as a counterpoint; I used your post more as a jumping off point to say more  about differentiation. Over time more workstation application code bases are getting modernized, so I do wonder how sustainable the workstation GPU thing is. Sooner or later workstation certified is just going to mean "not built with all the unstable bleeding edge AAA game-title specific optimizations", which IMO is why Intel GPUs have become a reasonable option here (Intel is not so good at the GPU driver thing). Both Nvidia and AMD are trying hard to pivot into GPU compute as the next big driver of high end sales, with Nvidia the biggest winner so far.
|
|
#
?
Jun 11, 2014 02:49
|
|
|
Sir Unimaginative posted:Integrated GPU is invaluable in diagnostics and a cheap hedge against having no computer if you end up with a bum video card. Also you can use it as an extra monitor, or for QuickSync H.264 encoding.
|
|
#
?
Jun 11, 2014 04:56
|
|
|
|
| # ? May 8, 2024 14:48 |
|
|
There's also some specific feature lock-outs. One that was big a few years ago was that a Quadro could draw a 2D anti-aliased line in hardware, and a GeForce could not, for example.
|
|
#
?
Jun 11, 2014 05:10
|
|