|
Latest Intel roadmap leak. Notably it lists Skylake-K for Q2/Q3 2015. source
|
# ? Feb 2, 2015 17:10 |
|
|
# ? May 9, 2024 00:01 |
|
Hopefully it'll be a worthwhile upgrade from Sandy Bridge, are there any upcoming SATA-Express drives?
|
# ? Feb 2, 2015 17:29 |
|
Tab8715 posted:are there any upcoming SATA-Express drives? Jam this into an M.2 slot for ≥ twice the speed of SATA-Express.
|
# ? Feb 2, 2015 17:43 |
|
Welmu posted:Latest Intel roadmap leak. Notably it lists Skylake-K for Q2/Q3 2015. How will they sell Broadwell, except as a budget line (somehow, with a different socket)?
|
# ? Feb 2, 2015 17:46 |
|
Tab8715 posted:Hopefully it'll be a worthwhile upgrade from Sandy Bridge, are there any upcoming SATA-Express drives? There was a 4tb hdd and 128gb ssd hybrid drive prototype at ces http://www.pcper.com/news/Storage/CES-2015-Storage-Visions-Western-Digital-SATA-Express-SSDHDD-Spotted it looks interesting but I'm not sure what problems it can solve better than 2 separate drives. Since it is still a prototype I wouldn't be surprised if 2tb - 4tb ssds are available at reasonable prices by the time it launches. Meaning it might already be a bit obsolete when it launches.
|
# ? Feb 2, 2015 17:54 |
|
blowfish posted:How will they sell Broadwell, except as a budget line (somehow, with a different socket)? It's possible that Broadwell isn't bringing the performance improvements that desktop users care about, so Skylake-k showing up wouldn't change much. In that case, it makes sense to push broadwell for low power segments while they can and let its desktop release die a quick, quiet death.
|
# ? Feb 2, 2015 17:58 |
|
Welmu posted:Latest Intel roadmap leak. Notably it lists Skylake-K for Q2/Q3 2015. The X-processor being two Intel Generations behind the latest S-processor for at least half a year is getting me the most. Goodness knows people will still be trying to shell out for Haswell-Es next year thinking that they'll be the one true elite machine for video games.
|
# ? Feb 2, 2015 18:28 |
|
Man, I hope Skylake-K doesn't make my SB-E completely irrelevant. I mean, yeah, it's an old chip, but it's still fast for current stuff. I'm just banking on it still being fast enough a year from now.
|
# ? Feb 2, 2015 18:33 |
|
Sidesaddle Cavalry posted:The X-processor being two Intel Generations behind the latest S-processor for at least half a year is getting me the most. Goodness knows people will still be trying to shell out for Haswell-Es next year thinking that they'll be the one true elite machine for video games. People play games on Xeons? e: no wait, the i7 thing. The old fab has to build something, I guess. suck my woke dick fucked around with this message at 18:48 on Feb 2, 2015 |
# ? Feb 2, 2015 18:37 |
|
Welmu posted:Jam this into an M.2 slot for ≥ twice the speed of SATA-Express. Are there any limitations with this kind of device?
|
# ? Feb 2, 2015 19:14 |
|
Tab8715 posted:Are there any limitations with this kind of device? We'll know from reviews when it ships around March-April.
|
# ? Feb 2, 2015 19:50 |
|
blowfish posted:People play games on Xeons? Edit: I am a retard and don't read necrobobsledder fucked around with this message at 22:28 on Feb 2, 2015 |
# ? Feb 2, 2015 20:36 |
|
blowfish posted:People play games on Xeons?
|
# ? Feb 3, 2015 01:34 |
|
Yeah I was going to say, gamers buy Xeons when they're rebadged as i7s, I had a knee-jerk reaction when I saw the blingy red HEDT up on the top row of that roadmap.
|
# ? Feb 3, 2015 01:54 |
|
Combat Pretzel posted:E3 Xeons are cheaper i7s.
|
# ? Feb 3, 2015 03:22 |
|
I have no idea how prices look now. It's just on release day, and a while after that, the Xeons are cheaper because they "don't ship" with IGP (i.e. disabled). So unless you really need the IGP, watch out for the E3 Xeons the next upgrade cycle. AFAIK these things run on the desktop chipsets just fine, so no need for a Cxxx mainboard.
|
# ? Feb 3, 2015 03:37 |
|
Hrm, I totally forgot that I could just try to get a Xeon E3 from the same Microcenter and it's the same value differential ($240 for an E3-1246V3 Xeon that's $274 on Newegg). In any case, it's not quite as cut and dry of a comparison as before when comparing an E3-1276v3 to an i7-4790k (roughly similar pricepoints and all that). http://cpuboss.com/cpus/Intel-Xeon-E3-1276V3-vs-Intel-Core-i7-4790K. For the relatively minor performance difference between a 1276v3 and 1246v3 though, might as well just save $40 and spend the rest on some beer and be happier with your purchases. The extra instructions supported by the 4790k may be handy from time to time but I have no idea if anything I use could use them or not (I don't do much A/V stuff besides batching like 500 movies through Handbrake). Maybe AVX2 is what really matters though and it's not a big deal to lose out on FMA3.
|
# ? Feb 3, 2015 04:11 |
|
Are there any real differences between a Xeon v3 and an i7?
|
# ? Feb 3, 2015 11:18 |
|
Other than the instruction set support, Trusted Computing bullshit, and ECC RAM nothing meaningful. Also, turns out FMA3 is a Haswell thing in general, not sure why some listings I saw before said E3v3 Xeons didn't do FMA3. Figured it was Intel doing some more market segmentation trickery.
|
# ? Feb 3, 2015 14:21 |
|
I just hope my Nehalem lives long enough that I can upgrade to Skylake without having to pick up another motherboard or proc before then. I think I'll end up with a faster proc with something like a 50 or 70% IPC improvement, which should be really handy.
|
# ? Feb 3, 2015 20:21 |
|
Lord Windy posted:Are there any real differences between a Xeon v3 and an i7? Not really even though almost nobody does it. Technically its not expressly supported on some motherboards (but it is on most), and even then I've never heard of it not working. It was more of a thing when the i7 was 3.5ghz but now since most go for the 4790k if they're going i7 at all its less cool.
|
# ? Feb 3, 2015 22:42 |
|
Methylethylaldehyde posted:I just hope my Nehalem lives long enough that I can upgrade to Skylake without having to pick up another motherboard or proc before then. I think I'll end up with a faster proc with something like a 50 or 70% IPC improvement, which should be really handy. I'm nursing mine along, as well. Starting to get impatient for Skylake.
|
# ? Feb 4, 2015 03:55 |
|
I didn't want to pay the DDR4 premium and bet on Broadwell. My 920 is disgusted in me.
|
# ? Feb 4, 2015 04:43 |
|
I'm on a 2600k that I haven't even overclocked yet. The only thing that has pushed my CPU (besides folding at home) is an unoptimized Star Citizen while recording video lossless. I think I will wait on Broadwell-E but DDR4 is nasty expensive for almost no real world benefit.
|
# ? Feb 4, 2015 11:57 |
|
Darkpriest667 posted:I'm on a 2600k that I haven't even overclocked yet. The only thing that has pushed my CPU (besides folding at home) is an unoptimized Star Citizen while recording video lossless. I think I will wait on Broadwell-E but DDR4 is nasty expensive for almost no real world benefit. 6 months should get the prices down from 'enjoy your vicious colon reaming' to 'at least we wined and dined you first, you big babby'. Still not as Ideal as buying ram for $6/GB, but 16GB for $300 isn't far off what I paid for my ram way back in the time before time when DDR3 2133 was hot poo poo bleeding edge stuff. Hell, unless you get the fancy gaming poo poo, it's barely more expensive than DDR3, an 8 GB stick of Crucial is $100, it's come way down since it first launched.
|
# ? Feb 4, 2015 12:50 |
|
Methylethylaldehyde posted:6 months should get the prices down from 'enjoy your vicious colon reaming' to 'at least we wined and dined you first, you big babby'. Still not as Ideal as buying ram for $6/GB, but 16GB for $300 isn't far off what I paid for my ram way back in the time before time when DDR3 2133 was hot poo poo bleeding edge stuff. The Ram I am using right now, Gskill 1600Mhz Cas9, was 16 GB for 108 dollars in 2010. I'd at least need to jump to 32GB if I moved to DDR4.
|
# ? Feb 4, 2015 13:04 |
|
I paid two thousand dollars for 8gb of DDR3-1600 a week or two after it came out. None of you get to complain. (In my defence, I was completely retarded and had no loving clue what the hell I was doing back then)
|
# ? Feb 4, 2015 13:19 |
|
Since DDR4 is point-to-point will consumer Skylake be capable of running memories in dual, triple, and quad-channel, depending on the number of RAM sticks installed?
|
# ? Feb 4, 2015 14:38 |
|
Welmu posted:Since DDR4 is point-to-point will consumer Skylake be capable of running memories in dual, triple, and quad-channel, depending on the number of RAM sticks installed? Essentially in laymen's terms yes. It's a little more technical, but each DIMM slot has it's own lane.
|
# ? Feb 5, 2015 00:23 |
|
there will be no noticeable difference between ddr3 and ddr4 for the next few years most of the difference is only noticeable at datacenter scale until ddr4 clock speeds surpass ddr3
|
# ? Feb 5, 2015 02:30 |
|
Malcolm XML posted:there will be no noticeable difference between ddr3 and ddr4 for the next few years No noticeable difference ever unless the way programs are accessed by RAM changes dramatically.
|
# ? Feb 5, 2015 02:37 |
|
Biggest difference is the lower voltage as standard and other power-saving features. Will help eek out that little bit more out of the battery. Standard voltage on DDR3L is 1.35V, DDR4 standard is 1.2, and DDR4 has additional sleep/etc functions.
|
# ? Feb 5, 2015 02:43 |
|
I thought lpddr4/WIO2 were going to be making some changes, some that might leak into the cpu memory access abstraction. Like what if the cache size buckets weren't monotonically increasing.
|
# ? Feb 5, 2015 02:47 |
|
Darkpriest667 posted:No noticeable difference ever unless the way programs are accessed by RAM changes dramatically. these words do not make sense programs generally have no idea how ram is laid out unless they designed for NUMA which is openly very specialized ones. NUMA sucks to program against cache hierarchy is what matters more than maybe a doubling in total bandwidth. I would rather have more ram than faster ram, because ram is quick enough. LiquidRain posted:Biggest difference is the lower voltage as standard and other power-saving features. Will help eek out that little bit more out of the battery. Standard voltage on DDR3L is 1.35V, DDR4 standard is 1.2, and DDR4 has additional sleep/etc functions. on desktops this is meaningless, maybe a few watts at best on laptop and mobile, perhaps we may see some decent gains.
|
# ? Feb 5, 2015 02:48 |
|
Malcolm XML posted:these words do not make sense I would counter what you said but basically you just said what I was thinking.. More is better than faster at this point. the leap from EDO and SDRAM to DDR was probably the largest leap we'll get in Random Access Memory (also known by laymen as RAM) we'll ever get unless the entire way programming is done is changed. I don't know what's so complicated or misunderstood let me explain the basics open program Program loaded from storage device to RAM <------------------------- until the storage device speed is faster than RAM RAM speed is largely irrelevant even SSDs are not fast enough Program loaded into RAM Use Program Close Program Program Dumped from RAM That's a quick lesson for you stupid people.
|
# ? Feb 5, 2015 04:03 |
|
Darkpriest667 posted:I would counter what you said but basically you just said what I was thinking.. More is better than faster at this point. the leap from EDO and SDRAM to DDR was probably the largest leap we'll get in Random Access Memory (also known by laymen as RAM) we'll ever get unless the entire way programming is done is changed. I don't know what's so complicated or misunderstood let me explain the basics not really no modern memory managers will actively page in/out + compress ram to avoid the very slow ram <-> nonvolatile storage executables are mapped into virtual memory just like any other file and pages can be in and out of physical memory. the largest perceptible speed increase in the last decade was SSD storage to vastly scale down the memory mountain; processor <-> DDR3 ram is already fast enough for everything sans 3D graphics (cf: XBox One) but RAM <-> HDD is still extremely slow, but RAM <-> SSD is much faster. and the best way to get 99% of the gain without spending a ton more is to just cache accesses, which is what Samsung RAPID does among others, and is why the eDRAM L4 cache/ eSRAM caches are not particularly dumb ideas
|
# ? Feb 5, 2015 05:46 |
|
I can't wait until we have some new storage that is both RAM and Harddisk. Maybe Flash Memory will one day get fast enough. What does 400mb/s translate to in RAM land? Although 160ms latency is essentially forever in computers.
|
# ? Feb 5, 2015 09:26 |
|
Lord Windy posted:I can't wait until we have some new storage that is both RAM and Harddisk. Maybe Flash Memory will one day get fast enough. What does 400mb/s translate to in RAM land? Although 160ms latency is essentially forever in computers. Google "Memristor".
|
# ? Feb 5, 2015 09:38 |
|
Malcolm XML posted:not really no Right, but Samsung is already having some controller issues (or at least I think it's NAND controller issues) which severely degrades the speed of accessing older memory blocks. They said they fixed it but it's rearing it's ugly head again. Basically what we need is large RAMdisks. We need to eliminate storage and RAM as separate and combine them into one thing. That's what I am saying about how programs are accessing RAM. If they were IN RAM and not loaded into RAM from storage. That's the main slowdown. If DDR wasn't so goddamned expensive now because of a B.S. shortage that wasn't even real I'd have bought more this past year. RAMdisk is really nice for loading stuff, but unless DDR4 comes down quite a ways it's really stupid to upgrade unless you need x99 for video editing and computational stuff. I do both so it's double annoying for me. I would never thought I would AGAIN live to see RAM more expensive than my CPU since the mid 1990s, but somehow it will be! Instant Grat posted:Google "Memristor". Basically this needs to be way forward yes.
|
# ? Feb 5, 2015 12:28 |
|
|
# ? May 9, 2024 00:01 |
|
The bigger problem I see is that our software written at present is incapable of handling super high speed without rewrites and completely rethinking networking. Here's a good example of what is required to handle the network hardware coming down the pipe at 100 gigabits - it's NOT easy, and ironically enough it's somewhat gated by how fast your CPU can work: https://lwn.net/Articles/629155/ The penalty for getting something wrong in sending data is really severe for network applications and any form of high performance computing historically. So improving the memory hierarchy's latency as mentioned above is likely to provide a lot more throughput than simply doubling that theoretical bandwidth. Sure, bandwidth helps for peak performance, but that's an idealized view of the world. This is exactly how Intel has done so well in the past 10 years - clock speed doesn't matter, smarter cache, smarter branch prediction, more efficient TLBs, etc. have been far more helpful than just blindly scaling down transistors and putting small nuclear reactors into our homes (that won't work well anyway due to current leakage to begin with). The question that's unanswered is whether we'll hit a wall even on how smart we can be about this general programming paradigm. Even multi-core / parallel programming won't save us at a point if what we're doing requires serial processing like what's typical in most games because well... most game programmers aren't going to do threads everywhere just from handling overhead alone and guaranteeing some form of hard realtime guarantees that are what people demand from their games (although nobody does hard realtime in practice I'd say because nobody's going to die if you lost a couple ms worth of frames or something during a CS:Go match). Lord Windy posted:I can't wait until we have some new storage that is both RAM and Harddisk. Maybe Flash Memory will one day get fast enough. What does 400mb/s translate to in RAM land? Although 160ms latency is essentially forever in computers. Darkpriest667 posted:We need to eliminate storage and RAM as separate and combine them into one thing.
|
# ? Feb 5, 2015 16:26 |