|
As an aside, can anyone identify the person responsible for USB having a rectangular form factor, that you can almost, but not quite plug in upside down, when you're feeling around the back of a case with limited access? I have a... List. A list I need to put them on.
|
#
?
Feb 19, 2015 22:49
|
|
|
|
| # ? Apr 25, 2024 19:00 |
|
|
Darkpriest667 posted:If USB 3.1 has a different plug and controller it will take forever for adoption to occur. Everyone remember firewire? USB C (the new plug everyone is hyping) is completely optional and in fact independent of USB3.1. There will be USB A (the old rectangle hole) 3.1 plugs, and they'll likely be around for a long time to come. The controller is just different in the sense that a USB 3 controller can't run things at 3.1, just how a USB 2 controller can't run things at 3 speeds. Sure it is a burden in the meantime, but the industry is used to periodically adding a new standard for USB. I have a feeling that USB 3.1 will replace 3 but USB 2 will instead still be seen in parallel to 3.1, instead of the usual "this generation and the generation before are common" that we see with USB adoption.
|
|
#
?
Feb 19, 2015 23:00
|
|
|
WhyteRyce posted:Firewire didn't fail because it had a different plug and controller It was superior in transfer speeds to IDE transfer AND USB transfer. So why exactly did it fail? I had 2 motherboards that supported firewire and it never was adopted. BurritoJustice posted:USB C (the new plug everyone is hyping) is completely optional and in fact independent of USB3.1. There will be USB A (the old rectangle hole) 3.1 plugs, and they'll likely be around for a long time to come. The controller is just different in the sense that a USB 3 controller can't run things at 3.1, just how a USB 2 controller can't run things at 3 speeds. Sure it is a burden in the meantime, but the industry is used to periodically adding a new standard for USB. Ahh Thanks for the info, I haven't read up MUCH on USB 3.1, but everything I heard was that it was a completely different pin amount and plug type.
|
|
#
?
Feb 20, 2015 00:13
|
|
|
Darkpriest667 posted:It was superior in transfer speeds to IDE transfer AND USB transfer. So why exactly did it fail? I had 2 motherboards that supported firewire and it never was adopted. More complex (hence expensive) to implement, storage-focused vs usb, and Apple had a $1/port licensing fee for a while, USB was good enough for most users.
|
|
#
?
Feb 20, 2015 00:45
|
|
|
MrYenko posted:As an aside, can anyone identify the person responsible for USB having a rectangular form factor, that you can almost, but not quite plug in upside down, when you're feeling around the back of a case with limited access? Be sure to add people who produce non-compliant cables that don't clearly mark the "up" side with the USB logo to that same list.
|
|
#
?
Feb 20, 2015 01:08
|
|
|
JawnV6 posted:Be sure to add people who produce non-compliant cables that don't clearly mark the "up" side with the USB logo to that same list. 90% of the time I have no problem blindly plugging in cables. The 10% of the time is non-compliant garbage 
|
|
#
?
Feb 20, 2015 01:13
|
|
|
PCjr sidecar posted:More complex (hence expensive) to implement, storage-focused vs usb, and Apple had a $1/port licensing fee for a while, USB was good enough for most users. Thanks I never knew I just knew it wasn't adopted! I appreciate the knowledge spreading =)
|
|
#
?
Feb 20, 2015 02:29
|
|
|
JawnV6 posted:Be sure to add people who produce non-compliant cables that don't clearly mark the "up" side with the USB logo to that same list. Be glad it's not a hub with upside-down ports. Also, one of the big things about FireWire, Thunderbolt, and ExpressCard is the way they're implemented allows anyone who can touch the port to read and change the entirety of your computer's memory. USB doesn't have that problem. That's also why FireWire was more widespread in high-end A/V work; direct memory access meant it performed tons better than USB, and your equipment is probably expensive enough that you aren't letting other people touch it.
|
|
#
?
Feb 20, 2015 04:31
|
|
|
dpbjinc posted:Also, one of the big things about FireWire, Thunderbolt, and ExpressCard is the way they're implemented allows anyone who can touch the port to read and change the entirety of your computer's memory. USB doesn't have that problem. That's also why FireWire was more widespread in high-end A/V work; direct memory access meant it performed tons better than USB, and your equipment is probably expensive enough that you aren't letting other people touch it.
|
|
#
?
Feb 20, 2015 15:55
|
|
|
MrYenko posted:As an aside, can anyone identify the person responsible for USB having a rectangular form factor, that you can almost, but not quite plug in upside down, when you're feeling around the back of a case with limited access? https://www.youtube.com/watch?v=qfHzzy6T9to
|
|
#
?
Feb 20, 2015 16:02
|
|
|
phongn posted:IIRC, Apple virtualized FireWire's DMA to make sure that external devices couldn't easily snoop on memory. Eventually we should be able to use IOMMU to do the same with Thunderbolt and ExpressCard, but nobody seems to have done that yet. Well, there's that whole fact where still pretty much only Apple products have Thunderbolt, and ExpressCard-capable laptops are getting ever fewer and farther between (and that's a shame because it works really well).
|
|
#
?
Feb 20, 2015 17:23
|
|
|
dpbjinc posted:Be glad it's not a hub with upside-down ports. Randomly? Like every other one? Or is the thing sitting on your desk day in day out with the exact same orientation and you're still unable to make clearly marked in-spec cables go in the right way without 3 or 4 tries? This is the worst complaint. It's not like PS2 was any better and I've had to look at VGA and DVI cables more frequently than USB. There's an "up" side on the cables and the things you're plugging them into don't tend to move or flip around a whole lot.
|
|
#
?
Feb 20, 2015 18:57
|
|
|
DVI cables seem to have this quality where even when they're oriented correctly they still need a couple of tries.
|
|
#
?
Feb 20, 2015 18:59
|
|
|
phongn posted:IIRC, Apple virtualized FireWire's DMA to make sure that external devices couldn't easily snoop on memory. Eventually we should be able to use IOMMU to do the same with Thunderbolt and ExpressCard, but nobody seems to have done that yet. Huh, didn't know they did that -- smart and forward-looking of them, it's a fairly large "security hole", though building devices that can take advantage of that type of access (at least in hardware; modifying firmware of a device with that access is much easier) can be though. Boiled Water posted:DVI cables seem to have this quality where even when they're oriented correctly they still need a couple of tries. Yeah, this happens to me -- I don't think it's shape is quite polarized enough, it's a very gentle 'D'.
|
|
#
?
Feb 25, 2015 07:59
|
|
|
movax posted:Huh, didn't know they did that -- smart and forward-looking of them, it's a fairly large "security hole", though building devices that can take advantage of that type of access (at least in hardware; modifying firmware of a device with that access is much easier) can be though. Nothing is virtualized. DMA is just disabled, sometimes .http://www.frameloss.org/2011/09/18/firewire-attacks-against-mac-os-lion-filevault-2-encryption/ It's not limited to apple in any way of course, but they're just the most ubiquitous devices with firewire. Thunderbolt also allows all sorts of fun stuff. https://trmm.net/Thunderstrike Longinus00 fucked around with this message at 19:40 on Feb 25, 2015 |
|
#
?
Feb 25, 2015 19:34
|
|
|
So mid-2015 desktop Broadwell is confirmed. http://techreport.com/news/27911/socketed-intel-desktop-broadwell-coming-mid-year quote:GDC — In a press conference at the Game Developers Conference in San Franscisco today, Intel offered a few new details about its plans for a desktop version of its 14-nm Broadwell CPUs. The firm plans to release a socketed version of Broadwell in the middle of this year, and this CPU will play to Broadwell's strengths by offering Iris Pro graphics and fitting into a tidy 65W power envelope. No need to freak out though, as this does fit with the previously leaked slide, so hopefully Skylake is roughly on track as well. 
|
|
#
?
Mar 5, 2015 19:44
|
|
|
Too bad broadwell-E won't be out until sometime in 2016, I need a new workstation PC before then.
|
|
#
?
Mar 5, 2015 20:47
|
|
|
Intel has released 14nm Broadwell SoC Xeons: Anandtech article.
|
|
#
?
Mar 11, 2015 14:18
|
|
|
Interesting followup to their Arm/Atom/Xeon-E3 scaleout benchmarking article. These Xeon-Ds should be a pretty damned beastly CPU for those applications.
|
|
#
?
Mar 11, 2015 18:20
|
|
|
Dual 10Gb NICs built-in is pretty cool.
|
|
#
?
Mar 11, 2015 21:57
|
|
|
My Rhythmic Crotch posted:Dual 10Gb NICs built-in is pretty cool. We may finally be seeing the start of the cheaper 10GbE networking trend.
|
|
#
?
Mar 11, 2015 22:45
|
|
|
Welmu posted:Intel has released 14nm Broadwell SoC Xeons: Wait, 10Gb networking without extra bullshit? As in, 10Gbit interfaces on the motherboard? Edit: ^ what they said
|
|
#
?
Mar 12, 2015 01:55
|
|
|
.
sincx fucked around with this message at 05:55 on Mar 23, 2021 |
|
#
?
Mar 12, 2015 02:08
|
|
|
I think that I read that the Supermicro mobo with the 10gbe PHY is expected to be about $100 more than the gigabit-only model. Not unreasonable, but not a lot cheaper than a X520. If you want a laugh, price out a 100gbit switch port and an optic.
|
|
#
?
Mar 12, 2015 02:51
|
|
|
Has the 6100 or 6200 been tested/benchmarked yet? I'm interested from an academic point of view. I'm not looking to upgrade but the Iris Pro (from benchmarks on how it performs on the couple of games I play) is so close to being exactly the GPU I need.
|
|
#
?
Mar 13, 2015 03:50
|
|
|
PCjr sidecar posted:I think that I read that the Supermicro mobo with the 10gbe PHY is expected to be about $100 more than the gigabit-only model. Not unreasonable, but not a lot cheaper than a X520. These still close to a million dollars a port, or have they gone down?
|
|
#
?
Mar 13, 2015 04:21
|
|
|
Lord Windy posted:Has the 6100 or 6200 been tested/benchmarked yet? The brand new Macbook pro 13" has a 6100 (not pro), you'll find a benchmark of it surely. No 6200 pro yet afaik.
|
|
#
?
Mar 13, 2015 08:01
|
|
|
http://anandtech.com/show/9081/asrock-to-debut-mini-itx-haswell-e-at-cebit-x99e-itx-ac-with-usb-3-1 So powerful, they won't just let you put any old standard X99 heatsink on it! 
|
|
#
?
Mar 13, 2015 22:19
|
|
|
Looking at the motherboard size constraints could also be an issue.
|
|
#
?
Mar 13, 2015 23:49
|
|
|
KillHour posted:These still close to a million dollars a port, or have they gone down? Oh, it's much more reasonable than that now; $110K for an ER optic, ~$50K/port on a line card.
|
|
#
?
Mar 14, 2015 00:02
|
|
|
It comes with that cooler because its a non standard size if im reading the article right
|
|
#
?
Mar 14, 2015 00:22
|
|
|
1gnoirents posted:It comes with that cooler because its a non standard size if im reading the article right Yup, you got it man. But it's really nice thing!
|
|
#
?
Mar 14, 2015 00:37
|
|
|
Yay, ultra M2 on ITX. Nice work, Asrock. Hope they do this again with skylake!
GRINDCORE MEGGIDO fucked around with this message at 01:13 on Mar 14, 2015 |
|
#
?
Mar 14, 2015 01:11
|
|
|
Yeah, there are some good ideas here for how to get new tech onto a smaller form factor, but the practicality of working with a 140W processor in an enclosed ITX case is going to be a sticking point. What specific applications for this platform will need to be that small?
|
|
#
?
Mar 14, 2015 01:45
|
|
|
Sidesaddle Cavalry posted:Yeah, there are some good ideas here for how to get new tech onto a smaller form factor, but the practicality of working with a 140W processor in an enclosed ITX case is going to be a sticking point. What specific applications for this platform will need to be that small? This is for the LEET gamer market who thinks that the Extreme series processors are The Best and who want a cool looking mITX system to put their 390x/TitanX in.
|
|
#
?
Mar 14, 2015 01:52
|
|
|
PCjr sidecar posted:Oh, it's much more reasonable than that now; $110K for an ER optic, ~$50K/port on a line card. Considering the cost of 10x 10g ER optics (and the cost of running 20 stands of OM3 fiber ~70KM), that's actually not terrible.
|
|
#
?
Mar 14, 2015 04:25
|
|
|
Sidesaddle Cavalry posted:Yeah, there are some good ideas here for how to get new tech onto a smaller form factor, but the practicality of working with a 140W processor in an enclosed ITX case is going to be a sticking point. What specific applications for this platform will need to be that small? I don't need X99 but if I did I can see the appeal of wanting to use an mITX case. I'm just glad they released this product because I had seen a lot of people bring up the concept of mITX X99 before and it was always shot down as physically impossible, which it clearly is not.
|
|
#
?
Mar 15, 2015 18:46
|
|
|
The biggest head scratcher with mITX x99 is that you lose one of the advantages it has over the 97 chipsets; extra PCIE lanes. I guess one use case is that this allows you to stuff more cores per rack.
|
|
#
?
Mar 15, 2015 19:10
|
|
|
Krailor posted:The biggest head scratcher with mITX x99 is that you lose one of the advantages it has over the 97 chipsets; extra PCIE lanes. You'd think a mini-X99 board would include more NVMe-capable M.2 slots so you'd have to worry even less about where you'd put 2.5" SSDs in the case. Put them on the back side of the board, maybe.
|
|
#
?
Mar 15, 2015 22:16
|
|
|
|
| # ? Apr 25, 2024 19:00 |
|
|
My Intel i5 is seriously underperforming. I did some benchmarks, and this was the result:  Not sure what's wrong , since I've made sure I have the latest drivers.
|
|
#
?
Apr 3, 2015 17:17
|
|
