|
Install Windows posted:Not in the sector of the market where Intel CPUs are used though. Mac sales flattened out last year. iPad sales haven't really gone anywhere either. 
|
#
?
Nov 4, 2013 22:40
|
|
|
|
| # ? Apr 26, 2024 01:50 |
|
|
JawnV6 posted:
Picturing some non-OCR'd lovely 200x200dpi scans of some old revision of orange/red books as the "documentation" (maybe it's a nice EDS, who knows)
|
|
#
?
Nov 4, 2013 22:54
|
|
Uhh... did they also open source a few gigs worth of documentation required to touch any significant part of it?
|
Bob Morales posted:Mac sales flattened out last year. iPad sales haven't really gone anywhere either. To be fair, we'll see a major swing on iPads sales as it no longer weighs as much as a MBA.
|
|
#
?
Nov 5, 2013 08:34
|
|
|
Bob Morales posted:Mac sales flattened out last year. iPad sales haven't really gone anywhere either. It would be interesting to compare the sales revenue plot against a profits plot. One would expect Apple's profit margins to drop due to increased competition from other manufacturers.
|
|
#
?
Nov 5, 2013 12:45
|
|
|
Hendrik posted:It would be interesting to compare the sales revenue plot against a profits plot. One would expect Apple's profit margins to drop due to increased competition from other manufacturers. I doubt Apple's margins are dropping on the iPad. They still charge $100 to go from 16gb to 32gb on all of their devices, and we all know that flash memory is cheaper than it was 3 years ago when the iPad launched, or even back to the iPhone 3g. iOS devices are Apple's cash cow. Edit: right from today's news macrumors posted:IHS Suppli has released its estimate of the component costs involved in building the new iPad Air, performing a virtual teardown based on information revealed by Apple and industry knowledge. According to IHS estimates, the component cost of the iPad Air is between $274 and $361 depending on the model, with the base model's components actually totaling $42 less than that of the entry-level third generation iPad last year despite significant technology improvements to reduce size and weight while improving performance. IHS iSuppli did not perform a cost analysis on the fourth-generation iPad, which was released in late 2012. http://www.macrumors.com/2013/11/05/ipad-air-component-costs-estimated-to-begin-at-274-roughly-13-cheaper-than-ipad-3/ mayodreams fucked around with this message at 17:29 on Nov 5, 2013 |
|
#
?
Nov 5, 2013 16:52
|
|
|
incoherent posted:To be fair, we'll see a major swing on iPads sales as it no longer weighs as much as a MBA. I originally took this as meaning "Sales reps place as much value in having an iPad as an MBA (as in the degree)." Sadly, it works both ways.
|
|
#
?
Nov 5, 2013 18:47
|
|
|
Am I right to assume that all the 4th Generation Intel® Core™ i5 Processors offer roughly the same gaming performance as one another? In other words it wouldn't make any difference to my experience if I bought the i5-4570 instead of the i5-4670K?
|
|
#
?
Nov 8, 2013 11:30
|
|
|
It's around a 6% difference. For ports, a 2500K with a modest overclock (4200 MHz) will still run them handily (and at stock it'll still almost certainly do the job). For native games, they still have to at least try to accomodate the poor souls hidebound to their Q6600 (which even overclocked by 50% is about 2/3 of a 2500K at stock at some tasks and as low as 2/5 at others). The 4570 is as powerful as that OC 2500K. And since the consoles are x86-64 rather than exotic POWER variants, and get to use 6x1600MHz* of instruction sets that are very nearly all a clean subset of Sandy Bridge or APU-specific (and the rest will be worked out either by the engine's integrated porting process or by the fact you can't count on everything being there on a general PC), it's going to take a lot less time for developers to either discover they have more than enough CPU or max the CPU out - at which point performance is up to your GPU anyway. Odds are pretty good that a 4570 will carry you this generation, and where it couldn't the 4670K won't do much better. Getting the K matters if you want to overclock, and you're looking at a minimum of $70 in aftermarket heatsink+fan to guarantee a jump worth overclocking for. At this point I don't even recommend non-reference GPUs unless it's something like an R9-290X (which pretty much calls for a 4-slot shotgun cooler or something mad to go with something equally mad) and a CPU is involved in just about everything, not just rendering and the occasional vector calculation, so you can infer my opinion on overclocking from that. *Yes, I know Xbox One is closer to 1800MHz now. Thanks to the Kinect reserve it's still pretty much level. dont be mean to me fucked around with this message at 13:09 on Nov 8, 2013 |
|
#
?
Nov 8, 2013 11:49
|
|
|
Sir Unimaginative posted:Odds are pretty good that a 4570 will carry you this generation, and where it couldn't the 4670K won't do much better. Thank you!
|
|
#
?
Nov 8, 2013 12:11
|
|
|
What is the vring / ring speed setting in my bios. Is that the memory <-> cpu bus speed? It's set to the same frequency as the cpu but is independent of it -- I can clock either one up separately.
|
|
#
?
Nov 9, 2013 09:11
|
|
|
regulargonzalez posted:What is the vring / ring speed setting in my bios. Is that the memory <-> cpu bus speed? It's set to the same frequency as the cpu but is independent of it -- I can clock either one up separately. That's an internal bus, a ring-shaped network that ferries data around a loop of the CPU cores, L3 cache, the GPU, the memory controller, and the QPI link(s) to the PCH and, if applicable, to other CPUs. Touching it is optional, assuming you're using Haswell - in Sandy and Ivy it was clock-locked to the CPU cores. Haswell doubled the bandwidth and unlocked its clocks, so you can leave it be and it won't be a bottleneck, as well as hopefully not a frequency wall as well.
|
|
#
?
Nov 9, 2013 15:30
|
|
|
Factory Factory posted:That's an internal bus, a ring-shaped network that ferries data around a loop of the CPU cores, L3 cache, the GPU, the memory controller, and the QPI link(s) to the PCH and, if applicable, to other CPUs. Ah thanks, yes it's a Haswell. I bumped it up from 3400 to 3800 when I started in with overclocking my 4670k and it seems fine so I'll just leave it as-is for the time being.
|
|
#
?
Nov 10, 2013 08:24
|
|
|
So I just unboxed my 4570 and the heatsink has what appears to be thermal paste pre-applied in three strips on the contact point. Just wanting to confirm that this is what it is and I don't have add any more.
KingEup fucked around with this message at 13:05 on Nov 11, 2013 |
|
#
?
Nov 11, 2013 13:02
|
|
|
KingEup posted:So I just unboxed my 4570 and the heatsink has what appears to be thermal paste pre-applied in three strips on the contact point. Just wanting to confirm that this is what it is and I don't have add any more. You're ready to rock and/or roll!
|
|
#
?
Nov 11, 2013 13:42
|
|
|
Sir Unimaginative posted:It's around a 6% difference. For ports, a 2500K with a modest overclock (4200 MHz) will still run them handily (and at stock it'll still almost certainly do the job). For native games, they still have to at least try to accomodate the poor souls hidebound to their Q6600 (which even overclocked by 50% is about 2/3 of a 2500K at stock at some tasks and as low as 2/5 at others). How big of a difference is the Ivy Bridge and Haswell processors compared to an i7 920 in terms of both games and just general Windows stuff?
|
|
#
?
Nov 28, 2013 05:27
|
|
|
fookolt posted:How big of a difference is the Ivy Bridge and Haswell processors compared to an i7 920 in terms of both games and just general Windows stuff? Ehhhh. AnandTech's Bench utility can tell you some real world tasks. But if you want to rule of thumb it, then the per-clock performance difference is 15% up for Sandy Bridge, 5% up for Ivy Bridge, and 10% up for Haswell, generation to generation. That makes a a 2.66 GHz Nehalem equivalent to a 2 GHz Haswell chip (though even the dinkiest mobile Haswell quad-core is faster than that).
|
|
#
?
Nov 28, 2013 05:38
|
|
|
Factory Factory posted:Ehhhh. AnandTech's Bench utility can tell you some real world tasks. But if you want to rule of thumb it, then the per-clock performance difference is 15% up for Sandy Bridge, 5% up for Ivy Bridge, and 10% up for Haswell, generation to generation. That makes a a 2.66 GHz Nehalem equivalent to a 2 GHz Haswell chip (though even the dinkiest mobile Haswell quad-core is faster than that). Hm, my i7 920's at 3.36GHz. That's probably not worth the ~$400-500 for transitioning right? I'm just tired of Windows 8 being slow as gently caress for the first 5 minutes of loading up.
|
|
#
?
Nov 28, 2013 05:47
|
|
|
That's a storage issue. Get an SSD.
|
|
#
?
Nov 28, 2013 05:49
|
|
|
Factory Factory posted:That's a storage issue. Get an SSD. I've got a 512GB SSD 
|
|
#
?
Nov 28, 2013 06:10
|
|
|
Something else isn't quite right...an i7 920, especially over clocked is plenty of CPU for everyday tasks and gaming. What type of SSD?
|
|
#
?
Nov 28, 2013 06:42
|
|
|
movax posted:Something else isn't quite right...an i7 920, especially over clocked is plenty of CPU for everyday tasks and gaming. What type of SSD? I've got a Crucial M4. I've also got 24gb of memory. When I have the task manager open on the performance tab, the ssd is at 100% activity during that period of starting up and the computer being mega slow.
|
|
#
?
Nov 28, 2013 07:37
|
|
|
Someone in the Win 8.1 thread recently did some troubleshooting. You have Win 8, right? Open the task manager (ctrl-shift-escape) and check the startup items for things with "high" impact and/or fishy duplications.
|
|
#
?
Nov 28, 2013 07:53
|
|
|
Factory Factory posted:Someone in the Win 8.1 thread recently did some troubleshooting. You have Win 8, right? Open the task manager (ctrl-shift-escape) and check the startup items for things with "high" impact and/or fishy duplications. Thanks; I'm sorry about the derail. I just want to make sure it wasn't my CPU (it's kind of amazing how long it has lasted me.
|
|
#
?
Nov 28, 2013 08:04
|
|
|
fookolt posted:Thanks; I'm sorry about the derail. I just want to make sure it wasn't my CPU (it's kind of amazing how long it has lasted me. There's no way a 3.36GHz i7-920 with 24GiB RAM and a 512GB SSD should be "slow for 5 minutes after booting". There's definitely something fishy with your Windows install.. HalloKitty fucked around with this message at 14:12 on Nov 28, 2013 |
|
#
?
Nov 28, 2013 11:06
|
|
|
If your SSD is more than ~75% full it does suffer performance degradation but even that wouldn't account for 5 minutes of slowdown. I'd disable unnecessary poo poo in msconfig and check Device Manager for unknown / conflicting items.
|
|
#
?
Nov 28, 2013 13:56
|
|
|
Does your BIOS have your SATA ports set to native or IDE? I've seen big speed problems with that and Win7+ set to compatible/IDE. I"ll concur on a full drive and/or Windows Bit Rot that a clean install may fix things. I do one every 6 months or so. 
|
|
#
?
Nov 28, 2013 18:40
|
|
|
There's a thought. If it's in IDE mode, your SSD may have filled itself up, as TRIM wouldn't work. You'd have to re-enable AHCI mode then image the drive, secure erase it, and restore the image to get your performance back.
|
|
#
?
Nov 28, 2013 21:00
|
|
|
Either my google-fu is failing me or what I'm looking for is hard to find....Any suggestions on where to look for benchmarks/reviews of the Haswell ULT SKUs (specifically the i5-4300u and i7-4600u)? I'm looking at getting a T440s as a new work notebook and those are the two CPU options, so I'm trying to get an idea of if the i7 is worth the extra cost. The closest thing I've been able to find is this Sony VAIO Pro 13 review on Anandtech. The VAIO Pro has an i5-4200U, and one of the machines it's compared against is an Acer with the i7-4500U. However, the Acer also has a discrete GT750M GPU and the review is mostly gaming/synthetic 3D benchmarks, so it doesn't really give a clear picture.
|
|
#
?
Dec 17, 2013 22:26
|
|
|
chizad posted:Either my google-fu is failing me or what I'm looking for is hard to find....Any suggestions on where to look for benchmarks/reviews of the Haswell ULT SKUs (specifically the i5-4300u and i7-4600u)? I'm looking at getting a T440s as a new work notebook and those are the two CPU options, so I'm trying to get an idea of if the i7 is worth the extra cost. You can get a good idea by just looking at their base and turbo frequencies on ARK. The i5-4300 is ALMOST as fast as the i7-4500. The i7-4600 should be somewhere between zero and 10% faster than the i7-4500.
|
|
#
?
Dec 17, 2013 22:40
|
|
|
fookolt posted:I've got a Crucial M4. I've also got 24gb of memory. When I have the task manager open on the performance tab, the ssd is at 100% activity during that period of starting up and the computer being mega slow. Sorry to derail just a little bit more, but I have encountered something similar. Have you tried to rune things like HDTune to see the performance of your SSD? In particular, if the tests takes forever to finish (i.e. you get odd pauses during the tests), then you probably have a bad drive. I actually had a bad intel ssd that took me forever to isolate the problem to the drive instead of something else in the system (since I never had reliability issues with intel ssd before so I resisted the idea that the drive is bad).
|
|
#
?
Dec 19, 2013 03:58
|
|
|
JawnV6 posted:
I wonder if this is in part related to SteamOS? I know that Intel graphics were not part of the initial rollout, and to be honest, the Nvidia drivers blew in my testing. Just a thought.
|
|
#
?
Dec 31, 2013 06:15
|
|
|
lethial posted:Sorry to derail just a little bit more, but I have encountered something similar. Have you tried to rune things like HDTune to see the performance of your SSD? In particular, if the tests takes forever to finish (i.e. you get odd pauses during the tests), then you probably have a bad drive.  Well poo poo, thank you for the suggestion. Time to get a new SSD
|
|
#
?
Dec 31, 2013 06:45
|
|
|
JawnV6 posted:
Time to make my speculation about where we'll be at by HD7000 with Intel's highest end integrated GPUs way less baseless! Assuming I can make heads or tails of that wikileaks-esque infodump!
|
|
#
?
Dec 31, 2013 06:52
|
|
|
fookolt posted:
uhh, what's with that CPU usage? That's a single core pegged on a quad core.
|
|
#
?
Dec 31, 2013 07:09
|
|
|
fookolt posted:
Boy is that SSD having a hard time.
|
|
#
?
Dec 31, 2013 11:01
|
|
|
chizad posted:Either my google-fu is failing me or what I'm looking for is hard to find....Any suggestions on where to look for benchmarks/reviews of the Haswell ULT SKUs (specifically the i5-4300u and i7-4600u)? I'm looking at getting a T440s as a new work notebook and those are the two CPU options, so I'm trying to get an idea of if the i7 is worth the extra cost. Not sure if this is what you need, but the new 13" MacBook Pro base models use 4258U and 4288U SKUs.
|
|
#
?
Dec 31, 2013 14:48
|
|
|
David Kanter has posted a summary of Knights Landing details on Real World Tech, a follow-up to his previous article on Knights Landing CPU speculation.
|
|
#
?
Jan 3, 2014 19:27
|
|
|
So what's up with these new Bay Trail Atoms? Why are they so damned good? I've got one in my tablet and it's blowing my expectations out of the water.
|
|
#
?
Jan 4, 2014 20:19
|
|
|
|
| # ? Apr 26, 2024 01:50 |
|
|
Alereon posted:David Kanter has posted a summary of Knights Landing details on Real World Tech, a follow-up to his previous article on Knights Landing CPU speculation. Totally solid on the first page. The second... ehhhh. I don't think QPI is a slam dunk. It's quite heavy and hadn't quite been banged into an IP. Assuming that the EX line will be sharing architecture and resources is another stretch. Overall solid, just a little wishful on some of those features.
|
|
#
?
Jan 5, 2014 00:35
|
|