I think you've made a good decision there. Nice balance between price and performance.
Critique my upcoming build please
|
07-05-2013, 02:10 PM
(This post was last modified: 07-05-2013, 02:11 PM by moosehunter.)
(07-04-2013, 11:18 AM)admin89 Wrote: i5 4670k is superior at overclocking , it can reach 4.6GHz with 1.2Vcore (lower Vcore is better and safer , <1.35Vcore is recommended )What?! I want one of these chips. My first chip was stable with 4.1GHz at 1.25Vcore, and I haven't gotten around to thoroughly testing my second one yet, but it'll boot at 4.4GHz at 1.25Vcore. I haven't fully tested it's stability at that voltage and frequency yet, but it did just fine under 30ish minutes of non Prime95 stress. Point is, it depends on what you get. Haswell doesn't overclock as well as the past two generations did. 07-05-2013, 02:17 PM
(07-05-2013, 02:10 PM)moosehunter Wrote:(07-04-2013, 11:18 AM)admin89 Wrote: i5 4670k is superior at overclocking , it can reach 4.6GHz with 1.2Vcore (lower Vcore is better and safer , <1.35Vcore is recommended )What?! I want one of these chips. I'd settle for 4.1 or 4.2. That's still a LOT of clock speed compared to the AMD 5050e that I currently run. Based on my reading, 4.6 is probably a bit ambitious for the 4th gen CPU, as you've pointed out. What I'd like to know is: what is different between the CPUs that overclock well and the ones that don't? 07-05-2013, 02:40 PM
(This post was last modified: 07-05-2013, 02:42 PM by NaturalViolence.)
@rokclimb15
A rosewill psu!?!? Why? Have you considered a full atx htpc chassis or sugo chassis? How much space do you have for this thing below your tv? What you're asking for doesn't sound that unreasonable if you have the money. @admin89 @moosehunter I don't see how it's possible for a haswell cpu to be stable at 4.6GHz 1.2v with every application. Likely just a really really lucky chip. rokclimb15 Wrote:What I'd like to know is: what is different between the CPUs that overclock well and the ones that don't? What do you mean? Do you mean why do some CPUs overclock better than others? Do you mean why do some architectures overclock better than others?
"Normally if given a choice between doing something and nothing, I’d choose to do nothing. But I would do something if it helps someone else do nothing. I’d work all night if it meant nothing got done."
-Ron Swanson "I shall be a good politician, even if it kills me. Or if it kills anyone else for that matter. " -Mark Antony Quote:Haswell doesn't overclock as well as the past two generations did.Yup , but 4.6GHz with 1.2Vcore is true Proof (post#22) Laptop: Mini PC :: 07-05-2013, 02:51 PM
(This post was last modified: 07-05-2013, 02:55 PM by rokclimb15.)
(07-05-2013, 02:40 PM)NaturalViolence Wrote: @rokclimb15That Rosewill PSU is 80+ silver certified and has some great ratings. It is probably due to larger sample size, but many more failures were reported on the OCZ, Cooler Master PSUs. If you'd asked me yesterday I would have said no way based on brand name, but it looks pretty good on specs/reviews, so I'll try it. 7 year warranty as well. I considered a full ATX HTPC case, but most were either very flimsy looking (and sounding according to the reviews), or in excess of $150 which is way too much for a case IMHO. None of them looked like they had exemplary cooling properties. I had considered a giant hack involving a dremel tool and a Zalman cooler, but I just don't have the time/energy to do that kind of stuff anymore. I have no idea what a sugo is. To be honest, I just had an idea in my head that it would fit on the shelf, but there's plenty of room around the thing to put another mid tower case. When I started this build spec, I didn't realize I'd probably be overclocking which I haven't done since an Athlon XP 1700 JIUHB. I might need more cooling or room for a larger CPU cooler, so an HTPC case seems risky. The limiting factor is height, which is about 12" on a shelf of my entertainment center. Depth and width are not a constraining factor. What I mean is, why are some chips "lucky" and some not? As in, the same model, just different serial # of chip. Last I actually did an OC, all 1700s with JIUHB serials were known to be unlocked and OC like madmen. They varied a little, but that mostly had to do with cooling and board voltage stability. Edit: it was the 1700, not 2500. This is ancient history and my mind has failed me. 07-05-2013, 03:26 PM
(This post was last modified: 07-05-2013, 03:33 PM by NaturalViolence.)
rokclimb15 Wrote:What I mean is, why are some chips "lucky" and some not? As in, the same model, just different serial # of chip. Last I actually did an OC, all 1700s with JIUHB serials were known to be unlocked and OC like madmen. They varied a little, but that mostly had to do with cooling and board voltage stability. CPUs manufactured closer to the center of the wafer generally have fewer defects after the ionization stage of manufacturing due to [insert long and complicated explanation dealing with chemistry and physics here]. In other words they are closer to the reference design at every level. Including down to the atomic level makeup of the transistors themselves. rokclimb15 Wrote:The limiting factor is height, which is about 12" on a shelf of my entertainment center. Depth and width are not a constraining factor. In that case (pun intended) you're limited to either a micro-atx/mini-itx build (which will restrict you to stock clocks), a very expensive chassis, or a poorly made chassis. So your assessment is correct. Sugo is a very reputable line of htpc chassis by silverstone: http://www.newegg.com/Product/ProductLis...isNodeId=1 They have become the most common choice for htpc builds these days.
"Normally if given a choice between doing something and nothing, I’d choose to do nothing. But I would do something if it helps someone else do nothing. I’d work all night if it meant nothing got done."
-Ron Swanson "I shall be a good politician, even if it kills me. Or if it kills anyone else for that matter. " -Mark Antony 07-06-2013, 12:20 AM
(07-05-2013, 03:26 PM)NaturalViolence Wrote:Interesting about the CPU wafer. Never knew that. I guess it makes sense since those 1700s varied by 100 MHz on average, and at double that clock speed you could expect double or more the difference between units.rokclimb15 Wrote:What I mean is, why are some chips "lucky" and some not? As in, the same model, just different serial # of chip. Last I actually did an OC, all 1700s with JIUHB serials were known to be unlocked and OC like madmen. They varied a little, but that mostly had to do with cooling and board voltage stability. Glad to hear I got that one right. I think I'll stick with a tower until some stock chip is powerful enough to run the settings I want on Dolphin without OC. Those are some good looking cases! If I were going to pick one, I'd probably pick one of those or the one I posted above. |
« Next Oldest | Next Newest »
|
Users browsing this thread: 1 Guest(s)