|
I was at the PCI SIG plugfest (woo!) last week and the Intel guy whose Ivy Bridge/Panther Point boards I was plugging stuff into said March of 2012 for release. Granted he was an engineering manager and not a marketing scuzzbag.. I should have asked when the Patsburg stuff is going to show up because we could use a 2P for interop testing.
|
#
¿
Sep 2, 2011 01:40
|
|
|
|
| # ¿ Apr 26, 2024 22:55 |
|
|
I've been really impressed by the video encoding capabilities of the Sandy Bridge (I have some background doing hardware for h.264/vc1 encoders). I don't own one yet but from messing around with ones we have at work that quick sync video is the bee's knees. Definitely planning on getting an Ivy Bridge in my next upgrade cycle.
|
|
#
¿
Oct 10, 2011 05:55
|
|
|
Yeah it'll definitely need some quality software developed for it, but the offload capabilities in the CPU is a pretty nifty thing. It's just kind of mind blowing considering what a decent hardware encoder was like 5-6 years ago (just like all computer related stuff, I know  ). Decent motion estimation alone took up a high end and bloody expensive FPGA and all that.
|
|
#
¿
Oct 12, 2011 04:44
|
|
|
Longinus00 posted:Well, if it can be done on an FPGA then you could also do it on a regular ASIC. I'm pretty sure you could write a smoking encoder targeting the PS3s cell chip if you wanted. It's economics that dictate what route you take. Well that's for sure, this was for a low volume/high margin broadcast video encoder so FPGAs made more sense at the time than a full blown ASIC development. It's just neat that Intel is pushing so many features into the CPU is all.
|
|
#
¿
Oct 12, 2011 06:33
|
|
|
Haha, it's true, video offload stuff is crammed into pretty much everything these days. I had picked up one of those Broadcom crystal HD things in order to put it in an apple tv v1 but nothing came of that (at least it was cheap). I miss working on AVC stuff, that was fun! Except for CABAC, that poo poo will give you such a headache
|
|
#
¿
Oct 12, 2011 07:21
|
|
|
That article is much less exciting than the headline 
|
|
#
¿
Oct 21, 2011 00:17
|
|
|
Yeah I was going to hold out til March for IB but even then who knows what quantities will ship right away etc.. Might as well just go for an i5/i7 SB now. 2600K or 2700K, hmm..
|
|
#
¿
Nov 2, 2011 19:09
|
|
|
This makes me glad I have no capacity to wait and ended up getting a 2500K  What a great cpu, love the setup.
|
|
#
¿
Nov 30, 2011 20:07
|
|
|
movax posted:At work we have a top-of-the line XPS with a 15" 1920x1200 screen, and 3.4GHz Pentium 4 CPU that we use as a random machine for running test scripts and such. I remember those, they're hilariously gigantic and heavy. "Semi-portable" would be more apt. Were those the first ones to have the LEDs that would cycle different colours? 
|
|
#
¿
Dec 15, 2011 02:58
|
|
|
I've had issues with Gigabyte boards in testing our production PCIe card, their slots don't adhere to the spec all that well. It had some seriously weird behaviour that was a beast to characterize. It's been the only manufacturer we've had problems with, not even the cheaper Asrock stuff has given any issues.
|
|
#
¿
Jan 17, 2012 21:55
|
|
|
mobby_6kl posted:New Broadwell Xenons. The E-5 2699 is looking good Wow, kind of surprised. I had figured they would go straight into Skylake/Purley E5s after the Broadwell delays. Basically I want Purley E5s with PCIe SRIS support already.
|
|
#
¿
Mar 31, 2016 21:22
|
|
|
BangersInMyKnickers posted:Does that thing have some kind of ribbon cable that connects to a PCIe slot? I've only seen the M.2 and the conventional expansion card ssd's, that one is new to me. This form factor (sff-8639 or U.2 as Intel is calling it) is mostly for enterprise backplanes etc.. The cables mechanically identical to the somewhat defunct SATA-Express so we might see some motherboards with ports but I think it'll be rare, slot and m.2 are gonna be the vast majority of consumer grade for a while. Also those nvme drives get stonkin' hot, they need a real good airflow.
|
|
#
¿
Apr 11, 2016 16:02
|
|
|
Skandranon posted:I believe it's the controller chips that ends up producing most of the heat, when doing constant read/write activity. The actual storage chips barely heat up at all. Yes this, the controllers are dumping something like 14W I believe. The NAND itself is negligible. NVMe is a bit more intensive than SATA as each drive is a full on PCIe endpoint along with they usually have some significant firmware running on the disk itself. They'll get cooler in the future it was just a shock picking one up that was under load after being used to sata drives and saying "holy poo poo that's loving hot"
|
|
#
¿
Apr 11, 2016 16:53
|
|
|
There are also things like this: http://serialcables.com/largeview.asp?cat=349&tier=264&id=1669 ($$$ though due to low volume)
|
|
#
¿
Apr 11, 2016 21:07
|
|
|
With all the NUC chat does anyone know of any more business/industrial models that would be good to check out? I'm interested in finding something in an i5 with dual nics and vga output that is pretty small and monitor mountable. Intel has one NUC with vga but it was a pentium iirc which would not cut it. I suppose I could go hdmi to vga adapter and use a usb ethernet dongle for the 2nd nic but something more tidy would be great. It's for test automation setups in an electronics lab and just want it tucked away mostly to be rdp'd into.
|
|
#
¿
May 16, 2016 22:50
|
|
|
Sidesaddle Cavalry posted:That's probably best solved with a ***-to-VGA dongle of your choice, if you're considering new equipment. Yeah, it's likely we'll have to start migrating over to hdmi or DP at some point. In the meantime I found the hp prodesk 400G2 which looks pretty decent, going to try one of those out. It even has a db9 serial port! (Also useful in a lab environment although we usually use serial to usb adapters now)
|
|
#
¿
May 16, 2016 23:09
|
|
|
nostrata posted:If you are just rdp-ing in do you even need a monitor connected? Just use it for the initial setup and then put it in place with just power and data. It'll be probably 75% RDP but still have a chunk of time when you may want to direct access it. We could go 100% headless though. Definitely an option.
|
|
#
¿
May 17, 2016 01:50
|
|
|
Got an HP Prodesk 400 G2 Mini in at work, the i5-6500T CPU. It is a really nice little machine, not quite as small as the NUCs but still very compact. One gotcha I found is that windows 7 will not seem to recognize the USB keyboard/mouse when I boot up, and I think that has something to do with EHCI/xHCI. Ended up installing windows 10 off a usb stick and that worked flawlessly. The drat thing boots so fast, from powering it on to login screen it is faster than the monitor coming out of power saving mode. Crazy!
|
|
#
¿
May 26, 2016 08:15
|
|
|
AVeryLargeRadish posted:Windows 7 does not have native USB 3 drivers so you would need to install drivers for the USB 3 ports if it has those. Interestingly it has a couple of black usb ports which I figured were 2 or lower but plugging mouse & keyboard in those didn't work either. The keyboard worked fine for the safe mode menu and anything cli based (on any of the ports, 2 or 3) before booting to windows but once the gui loaded up, nada. A good excuse to force us to start moving to win 10 anyway.
|
|
#
¿
May 26, 2016 15:46
|
|
|
That got me looking up mine, $219CAD in November 2011. What a ridiculously good value that has been, man. Think I'll wait til the fall to make it an even 5 years and then get a Kaby Lake system.
|
|
#
¿
Jun 6, 2016 07:17
|
|
|
So I'm looking at options for a FreeNAS build using a skylake chipset that supports ECC, C236 in this case. Why is it most of the motherboards use unbuffered ECC (DDR4), and why is it that it seems that registered is much more plentiful and cheaper? I was all set to get the C236M from MSI but I looked at the limited selection of unbuffered ECC DDR4 and cringed. Is it just that not much stuff used unbuffered DDR4 previously and the product pipeline hasn't gotten filled up yet?
|
|
#
¿
Jun 18, 2016 07:24
|
|
|
I don't know if I can hold off til Skylake-X/Kaby Lake-X to upgrade from my 2500k. Not because anything needs more cpu HP, just because I want something new and exciting.
|
|
#
¿
Jul 19, 2016 17:02
|
|
|
God, who cares what resolution monitor someone uses. Stop gooning it up, goons.
|
|
#
¿
Jul 19, 2016 18:00
|
|
|
I am disappointed the dell ultrasharps are moving away from 16:10, I have a 2412 and I love it. 16:9 is like what my lovely samsung work monitor is and so I hate it. It's less about the resolution and more that the ultrasharps don't seem as premium as they once did.
|
|
#
¿
Jul 19, 2016 19:51
|
|
|
I have a lenovo work laptop and I always forget to use the pointer thingie.
|
|
#
¿
Jul 20, 2016 17:06
|
|
|
canyoneer posted:Well, it can be hard to find. 
|
|
#
¿
Jul 20, 2016 17:16
|
|
|
BIG HEADLINE posted:I'm not sure what to take from this: Looks like a 6, 8 and 10 core Skylake-X! And since the Skylake-X uses the same PCH as the Kaby Lake-X it could be up to 24 PCH PCIe lanes plus 44 CPU lanes.. Crazy! Yeah I'd love to get the 8 core too. Gonna be tough to hold off on the upgrade til 2H 2017 though.
|
|
#
¿
Jul 21, 2016 07:07
|
|
|
What's the best AIO brand these days? I have a Corsair H50 or something in my current build and it gets my i5-2500k to 4.4GHz with no fuss. Wondering if they're still the standard for my next build.
|
|
#
¿
Aug 21, 2016 16:47
|
|
|
DrDork posted:They are still the standard. There are some other players out there that are also good, but they tend to be more hit or miss, and certainly don't have the availability and support that Corsair does. Swiftech has some neat ones that are easy to drop additional loop modules into, which is nice. Cool. I had seen Fractal Design has some but they aren't very available in my area. Corsair is a little confusing with all the different H1XX models for the 280mm radiator, but I've heard outside of if you want the "i" features just get whatever, it doesn't really matter.
|
|
#
¿
Aug 21, 2016 21:05
|
|
|
Hm, hopefully they do a X-series 6/8 core eventually but I don't know if I could hold out for that The 7700K would be pretty sweet. Are the Z270 boards supposed to launch at the same time or will they be before/after?
|
|
#
¿
Aug 23, 2016 06:28
|
|
|
Got an Intel SDK in with LGA 3647 processors  They are humongous. I'm so used to 2011s I was totally confused when I took the heatsinks off at first..
|
|
#
¿
Sep 8, 2016 20:31
|
|
|
Paul MaudDib posted:Note that chips marked "ES"/"Engineering Sample"/"Intel Confidential" are not official chips intended for retail consumers and there are likely to be compatibility issues. You really need to look at the specific model of mobo and BIOS to be sure they'll work. Technically they are all supposed to be Intel property and It's not that much of a mystery, we have a poo poo-ton of prototype intel hardware that is marked confidential with paperwork telling us to return it and when we're done with it we call them up and they say "eh you can just chuck it out". Now multiply that by thousands of suppliers/vendors/system makers.. Definitely true that compatibility with retail BIOS can be a problem though, they're usually for either their own platforms (S2600 series) or are going on boards that run funky not for public consumption BIOS.
|
|
#
¿
Sep 10, 2016 04:32
|
|
how they ended up on eBay (possibly ghost shifts at the fab, which can make quality even worse).
|
I'm curious if Kaby Lake Motherboards (Z270 etc) are going to have onboard USB 3.1 gen2 headers for front panel connectors.. Haven't seen anything coming out from case manufacturers to indicate that from my limited googling.
|
|
#
¿
Sep 24, 2016 05:43
|
|
|
BIG HEADLINE posted:There is a new socket coming for Skylake-E, LGA-3647. It's loving enormous, but it might be 9 months out. We have one of the Skylake/Purley Intel SDKs at work and got sent the upgrade to Beta CPUs and when I swapped them out I was like what the fuuuuuck because the chips were so huge under the heatsinks. Also the heatsinks snap right on to the CPU and you have to carefully pry them off which is ridiculous. I thought I had snapped a pic of them but apparently not. They're almost the size of a playing card. Serve the home has a good writeup with pics: https://www.servethehome.com/big-sockets-look-intel-lga-3647/
|
|
#
¿
Oct 16, 2016 07:06
|
|
|
BIG HEADLINE posted:I do kind of *not* look forward to a $150 HSF or $250 CLC for them. Yeah no kidding. The plastic clips on the stock heatsinks are very strong but I was really worried I was going to break them especially when removing the CPU. They're definitely not very user friendly. Perhaps a 3rd party will come up with a clever retention clip. Also the screws on the heatsinks are gigantic torx style heads too which was also annoying, didn't have those lying around and had to hunt one down. As they are it wouldn't be a very good cpu for the "prosumer" crowd, they'd get a lot of RMA'd cpus and boards from tragic mishaps.
|
|
#
¿
Oct 16, 2016 07:51
|
|
|
I really like the Xeon-D and would love an 8 core for a home server/NAS. 4 core would be more than sufficient really, but 8 core, yessss.
|
|
#
¿
Nov 8, 2016 07:12
|
|
|
That's a great summary. I also highly doubt we'd ever see 3647 in anything consumer facing. I don't know if they'll even be doing a socket mount for it that doesn't require torquing some screws to get it down. Any kind of lever based retention like on previous sockets would have to be fairly ridiculous to keep those monsters in place.
|
|
#
¿
Nov 10, 2016 18:33
|
|
|
Twerk from Home posted:Is Xeon-D expected to get a Skylake refresh or operate on an entirely unique product cycle? It's more than a year old at this point. I'm wondering if it might be some kind of CPU/FPGA hybrid with the Altera stuff in there to change what peripherals it supports.
|
|
#
¿
Nov 10, 2016 18:43
|
|
|
Methylethylaldehyde posted:That would actually be really handy for certain kinds of scientific calculations, assuming you can get whatever wierdass algorithm into usable verilog code. FPGAs can be 8-20x as power efficient compared to a regular CPU for a lot of stuff that you can either parallelize out the rear end by putting 90 functional compute units on the FPGA, or involves a lot of chewing on small bits of data, like a hash or checksum function, where you can store it in the FPGA's local memory and then pipeline the crap out of the calculation process. I know certain implementations of the openCV spec on FPGAs would get real time 1080p Haar Cascade performance for like 30W, compared to the CPU+GPU at like 200ish. Absolutely, I am convinced a big reason to move to the giant 3647 sockets is so they can have FPGAs with the cpu as a multi chip module and they can do a lot of stuff that offload cards/GPUs do now. Makes sense to have some configurability on the smaller side too.
|
|
#
¿
Nov 10, 2016 20:48
|
|
|
|
| # ¿ Apr 26, 2024 22:55 |
|
|
SlayVus posted:ASRock makes TWO mITX, they make a server 2011-3 with 4 DIMM slots. Oh man it's adorable! Also saw this yesterday and thought it was neat: https://www.servethehome.com/intel-demonstrating-broadwell-ep-fpga-package/
|
|
#
¿
Nov 17, 2016 07:59
|
|
