|
Install Windows posted:An alkaline AA battery holds 8100 joules. The iPhone 5 battery holds 19,260 joules (the 4s and 4 were 19,000 joules). edit: v Right right, Ah is not a measure of energy. Josh Lyman fucked around with this message at 03:55 on Sep 12, 2013 |
#
?
Sep 12, 2013 03:53
|
|
|
|
| # ? May 8, 2024 08:13 |
|
|
Josh Lyman posted:I was referencing the iPhone 5 have a 1440 mAh battery and a AA having 2700 mAh. 1440 mAh at 3.8 volts is significantly higher capacity than 2700 mAh at 1.5 volts. An AA is 44% of the power stored as an iPhone 5 battery, not the other way around.
|
|
#
?
Sep 12, 2013 03:55
|
|
|
Gonkish posted:I'm honestly kind of amazed that AMD is still going at this point. They've been outmaneuvered so many times now that it's just sad. At least they have the consoles for the next few years. Good thing AMD is a starfish while Intel is a whale or whatever stupid analogy that PR guy tried to spin.
|
|
#
?
Sep 12, 2013 05:56
|
|
|
Install Windows posted:1440 mAh at 3.8 volts is significantly higher capacity than 2700 mAh at 1.5 volts. An AA is 44% of the power stored as an iPhone 5 battery, not the other way around. This is still impressive imo.
|
|
#
?
Sep 12, 2013 14:54
|
|
|
WhyteRyce posted:Good thing AMD is a starfish while Intel is a whale or whatever stupid analogy that PR guy tried to spin. AMD is being prodded by beachgoers with a stick while Japan is trying kill Intel?
|
|
#
?
Sep 12, 2013 14:54
|
|
|
SRQ posted:AMD is being prodded by beachgoers with a stick while Japan is trying kill Intel? AMD has only one orifice it must use as a mouth and anus while Intel is the only company able to dive deep enough to fight giant squid (Asian Foundries)? Back on-topic though, I have an X120e that I got in first half of 2011 and it was an excellent machine in a space with no Intel competition. With the $300 Haswell chromebooks available now, and these new Atoms creeping up in the tablet & even lower end space I see no room for AMD anywhere. I'm also optimistic that Intel will find a way to remain highly profitable in a world of declining PC shipments and a huge market for commodity tablets.
|
|
#
?
Sep 12, 2013 15:28
|
|
|
The sun never sets on the Intel Empire. 
|
|
#
?
Sep 12, 2013 15:46
|
|
|
http://reviews.cnet.com/tablets/asus-transformer-book-t100/4505-3126_7-35827544.htmlquote:The Asus Transformer Book T100 is a Windows 8.1 tablet/laptop hybrid that runs on Intel's "designed for tablets" Bay Trail architecture. The very familiar-looking -- if you've been paying attention to Asus for the last two to three years -- tablet (and included keyboard) is hitting US stores on October 18 starting at $349. That's a pretty good price for a 10" non-RT Windows tablet that comes with the keyboard. Of course it's made by Asus so I'm expecting some kind of fatal flaw with the GPS/Wifi and poo poo NAND. http://blogs.wsj.com/digits/2013/09/11/intel-places-its-chips-in-new-google-chromebooks/?mod=yahoo_hs quote:The companies on Wednesday–in an announcement pegged to Intel’s annual developer forum in San Francisco–disclosed that new Chromebooks are on the way that are being built with Haswell, the code name for the latest version of Intel’s Core line of microprocessors. Price isn't given but Haswell in Chromebooks just seems interesting but weird to me. WhyteRyce fucked around with this message at 16:49 on Sep 12, 2013 |
|
#
?
Sep 12, 2013 16:40
|
|
|
WhyteRyce posted:Price isn't given but Haswell in Chromebooks just seems interesting but weird to me. Seems reasonable enough, Javascript needs a burly CPU to perform well, and it's not like they're putting in Core i7's. That kind of speed/battery life balance could be neat in a $300 laptop. Chrome OS is still silly though.
|
|
#
?
Sep 12, 2013 19:41
|
|
|
Phantom Limb posted:Seems reasonable enough, Javascript needs a burly CPU to perform well, and it's not like they're putting in Core i7's. That kind of speed/battery life balance could be neat in a $300 laptop. Chrome OS is still silly though. It would just be weird seeing a HSW in a <$300 product. Although I guess the <$300 thing is just an assumption since there were Chromebooks that cost more
|
|
#
?
Sep 12, 2013 20:52
|
|
|
WhyteRyce posted:It would just be weird seeing a HSW in a <$300 product. Although I guess the <$300 thing is just an assumption since there were Chromebooks that cost more Yeah, I agree that it's pretty weird, especially since I didn't even realize that Intel's gonna be making Pentium and Celeron branded Haswell chips. It looks like Asus's Chromebook is gonna be using the Celeron 2955U, which is priced identically to the Celeron 847 that's used in their previous $200 Chromebook. There's probably a subsidy from Google involved too to push adoption of the OS.
|
|
#
?
Sep 12, 2013 21:16
|
|
|
Haswell is cheaper to manufacture than Sandy Bridge, thanks to the process shrink, and it's far more power-efficient. Why not in a Chromebook?
|
|
#
?
Sep 12, 2013 22:12
|
|
|
You can spend more at Applebee's than it costs Intel to make an average Haswell...of course, there's R&D costs to recoup but the actual production cost of the chips is extremely low.
|
|
#
?
Sep 12, 2013 22:20
|
|
|
Factory Factory posted:Haswell is cheaper to manufacture than Sandy Bridge, thanks to the process shrink, and it's far more power-efficient. Why not in a Chromebook? I forgot Chromebooks had Sandy Bridge in them. I thought they were mostly Atom garbage.
|
|
#
?
Sep 12, 2013 22:32
|
|
|
WhyteRyce posted:I forgot Chromebooks had Sandy Bridge in them. I thought they were mostly Atom garbage.
|
|
#
?
Sep 13, 2013 01:16
|
|
|
Anandtech has an article up on memory bandwidth scaling with Haswell, from 1333Mhz to 3000Mhz. The upshot is what I've been saying for some time: There is no good reason to buy anything slower than DDR3-1866 or faster than DDR3-2133, and there is a significant performance penalty for dropping to DDR3-1333. It's not huge, so if you already have enough memory there's no reason to upgrade from DDR3-1333 to anything faster, but if you're buying new it just doesn't make sense to buy slower memory for the same price.
|
|
#
?
Sep 26, 2013 22:13
|
|
|
Coming out of IDF there were some questions about the Quark SoC; there's more info available now. It's a 32 bit Pentium-based core; they're releasing an Arduino board based on it, running at 400 MHz. Cost is under $60. http://www.anandtech.com/show/7387/intel-announces-galileo-quark-based-arduino-compatible-developer-board
|
|
#
?
Oct 3, 2013 17:10
|
|
|
Can anyone give me the dimensions of the stock heatsink/fan that ships with the new Haswell CPUs? I actually just need the height of the thing. I'm trying to figure out how tall it sits off the motherboard and I can't find this info using my google skills. From the photos it looks like about 6cm tall: 
KingEup fucked around with this message at 10:05 on Oct 16, 2013 |
|
#
?
Oct 16, 2013 10:02
|
|
|
KingEup posted:Can anyone give me the dimensions of the stock heatsink/fan that ships with the new Haswell CPUs? I actually just need the height of the thing. edit: Oops, this advice is for Ivy Bridge. Don't know if it still applies to Haswell, but it's sound nonetheless.
|
|
#
?
Oct 16, 2013 10:13
|
|
|
Josh Lyman posted:You really shouldn't use the stock HSF. Technically it will keep your CPU from burning up, but even a $10 HSF from Newegg will do much, MUCH better. I used the stock heatsink on my i7 920 for a while, was perfectly fine... Until one of the push pins just gave up one day and the whole thing popped up with no notice... I've got a nice screwed down aftermarket HSF now. In terms of actual cooling it was perfectly fine though at stock clocks.
|
|
#
?
Oct 16, 2013 11:59
|
|
|
KingEup posted:Can anyone give me the dimensions of the stock heatsink/fan that ships with the new Haswell CPUs? I actually just need the height of the thing.
|
|
#
?
Oct 16, 2013 13:09
|
|
|
Josh Lyman posted:You really shouldn't use the stock HSF. Technically it will keep your CPU from burning up, but even a $10 HSF from Newegg will do much, MUCH better.
|
|
#
?
Oct 16, 2013 16:19
|
|
|
Broadwell, the 14nm shrink of Haswell, has been delayed for at least one quarter due to yield issues with the 14nm process. Originally it was planned for initial production late this year for a launch in 2014, but now may be pushed back until late 2014 or even 2015. Note that Broadwell is a "tick", meaning it will be offered in as a multi-chip module that is soldered onto motherboards, not a socketed processor ("tocks get socks"). The Haswell refresh next year should tie us over until Skylake in 2015.
|
|
#
?
Oct 16, 2013 16:25
|
|
|
td4guy posted:I just measured mine. 5cm from the bottom of the black legs to the top of the fan enclosure. Thank you for measuring! It is indeed for a SFF build.
|
|
#
?
Oct 16, 2013 22:29
|
|
|
Alereon posted:Broadwell, the 14nm shrink of Haswell, has been delayed for at least one quarter due to yield issues with the 14nm process. Originally it was planned for initial production late this year for a launch in 2014, but now may be pushed back until late 2014 or even 2015. Note that Broadwell is a "tick", meaning it will be offered in as a multi-chip module that is soldered onto motherboards, not a socketed processor ("tocks get socks"). The Haswell refresh next year should tie us over until Skylake in 2015. Wikipedia (yeah I know) is telling me there will be a variant "Broadwell-D" that is socket 1150. Has this been debunked?
|
|
#
?
Oct 17, 2013 00:02
|
|
|
Lolcano Eruption posted:Wikipedia (yeah I know) is telling me there will be a variant "Broadwell-D" that is socket 1150. Has this been debunked?
|
|
#
?
Oct 17, 2013 00:11
|
|
|
Is it just me or did Broadwell come out way sooner than expected?
|
|
#
?
Oct 17, 2013 02:55
|
|
|
Alereon posted:Broadwell, the 14nm shrink of Haswell, has been delayed for at least one quarter due to yield issues with the 14nm process. Originally it was planned for initial production late this year for a launch in 2014, but now may be pushed back until late 2014 or even 2015. Note that Broadwell is a "tick", meaning it will be offered in as a multi-chip module that is soldered onto motherboards, not a socketed processor ("tocks get socks"). The Haswell refresh next year should tie us over until Skylake in 2015. 
|
|
#
?
Oct 18, 2013 07:48
|
|
|
It's a new policy with Haswell.
|
|
#
?
Oct 18, 2013 07:56
|
|
|
Factory Factory posted:It's a new policy with Haswell.
|
|
#
?
Oct 18, 2013 14:16
|
|
|
Josh Lyman posted:If they plan to only sell a CPU integrated into the motherboard, what are desktop DIYers supposed to do until the next tock comes out? Broadwell seems to be geared towards mobile and low power use, so I assume you use Haswell + or whatever the name is for their improvement of Haswell at the same time.
|
|
#
?
Oct 18, 2013 14:27
|
|
|
Josh Lyman posted:If they plan to only sell a CPU integrated into the motherboard, what are desktop DIYers supposed to do until the next tock comes out? That depends. If it's just system builders, we're stuck with what evilweasel said. If they let the component manufacturers in, we could see essentially integrated motherboards. That does raise concerns like this user can't overclock and that user doesn't have all the neat features they want, but we're not too far from that now even with discrete processors and motherboards - and unless you keep unjustifiable spares, losing either part still leaves you down a computer for a while. I wouldn't be surprised if this is a litmus test for Skylake - if the CPU is generally good enough (arguably it was for overclockers two generations ago and is even for stock users now) and it doesn't stand in the way of typical addons, the number of people legitimately left out in the cold may be too small for Intel to care. dont be mean to me fucked around with this message at 14:49 on Oct 18, 2013 |
|
#
?
Oct 18, 2013 14:44
|
|
|
I guess it won't be too bad. We already do this already with k processor: z87 board and non-k : h87 board. We only want the illusion of choice, but we just make the same picks anyways.
|
|
#
?
Oct 18, 2013 17:51
|
|
|
Lolcano Eruption posted:I guess it won't be too bad. We already do this already with k processor: z87 board and non-k : h87 board. We only want the illusion of choice, but we just make the same picks anyways. My Ivy Brige setup is a 3570K and ASRock Z77 Extreme4. Assuming ASRock would sell a Z97 or whatever motherboard with an embedded processor, what are the chances something feature-rich like the Extreme 4 wouldn't only come with a more expensive CPU like the 3770K? Also, I got my CPU + Mobo for like $270 combine due to an insane bundling deal.
|
|
#
?
Oct 18, 2013 18:23
|
|
|
MicroCenter would have to offer a different loss-leader, that's for sure.
|
|
#
?
Oct 18, 2013 19:12
|
|
|
I don't think Intel really wants to screw over enthusiasts, rather they're trying to address both size and power, as well as recognizing that the pace of change in the desktop space has slowed down. On the power/size front, Intel wants to integrate as much as possible onto the CPU module to consolidate TPD headroom. For example, if your chipset isn't drawing its full TDP, the CPU could turbo up an extra couple bins. For small form factor and embedded systems you don't have room for a separate chipset module from the CPU module, so having a combined module is important. Finally, with smaller differences between each successive generation, skipping generations makes sense for the desktop where the enhancements won't be needed. In the lower power markets you do need every little bit of performance and efficiency you can get.
|
|
#
?
Oct 18, 2013 19:39
|
|
|
The enthusiast sector could use a break anyway, I'm hoping that engineering toward efficiency kills overclocking dead for good at some point in the near future because there just isn't any headroom to be had on any variation of current processes. It's too much hassle for too little reward, and they provide solutions that are appropriate to various goals at understandable price points. If overclockers get left out in the cold, and I count, I've got a shitload of 200mm case fans and a big, three-fan NH-D14 and all that jazz, fine, gently caress it, I'll come inside where it's warm and stop wasting money to wrestle performance out of parts that only have the overhead sometimes and be happier to buy fully warrantied and validated chips with higher core counts, etc., for future projects that call for more processing power than can be supplied by a stock configuration. Computing has changed a lot since the introduction of GPGPU, we've got little supercomputers stuck inside these boxes anyway, and they're just going to get better at the range of tasks appropriate to them while CPU manufacturers - so Intel, I guess, in the desktop sector, for however long that lasts - and software developers, and everybody else grudgingly, finally gets on with it and starts solving some of the tougher problems with multithreading common workloads and matching software efficiency to hardware efficiency for an overall at-least-as-good and probably better computing user experience.
|
|
#
?
Oct 18, 2013 20:06
|
|
|
With the furious attention to power management, I'm surprised enthusiasts are still able to wring correct performance out of overclocks. Intel's very good at characterizing speed paths and binning. Overclocking is built on the assumption that they screwed up one of those. Or, as I suspect, the speed path that put a part into a lower bin isn't hit by the OC stress test.
|
|
#
?
Oct 18, 2013 21:24
|
|
|
JawnV6 posted:With the furious attention to power management, I'm surprised enthusiasts are still able to wring correct performance out of overclocks. Intel's very good at characterizing speed paths and binning. Overclocking is built on the assumption that they screwed up one of those. Or, as I suspect, the speed path that put a part into a lower bin isn't hit by the OC stress test. I seriously question the incomplete validation tools available to overclockers anyway. I don't think most overclocks are STABLE stable, just stable enough to pass tests and do stuff. The distinction might not matter, but I trust Intel's validation over a bunch of software recipes that people have come up with as a folk validation process for running parts out of specification. It helps in some sense that companies like Asus do work with higher quality tools for shaping some broad expectations regarding performance but on the whole I welcome the final, blessed end of overclocking as it solves a problem virtually nobody will actually have going into the future. My opinion. Really Good Overclocking pretty much ended with the 2600K, by my guess. Now if you want more performance, just be okay with paying for it and you'll have a much more likely to be genuinely stable system for it. But this perspective is also informed by my own 2600K losing stability after a couple years at 4.7GHz and me not being able to quantify a difference after dropping it down to 4.5GHz, and then thinking, well gently caress, in all quantitative analyses of CPUs this thing at stock settings is already coming in as close as possible to the two generations since barring special instruction sets it lacks, what have I been killing this chip for this whole time?
|
|
#
?
Oct 18, 2013 21:35
|
|
|
|
| # ? May 8, 2024 08:13 |
|
|
JawnV6 posted:Or, as I suspect, the speed path that put a part into a lower bin isn't hit by the OC stress test. Bingo.
|
|
#
?
Oct 18, 2013 21:41
|
|
