It's like the executives scrutinize every bit of the rollout: either it has a positive ROI or it won't be done.
The big issue with corporations today is that they care about making money now more than they care about delivering good business to customers, even though the latter is more likely to give you repeat business and a sustainable revenue stream!
It is said that absolute power corrupts absolutely. While I see no basis for making a sweeping generalization that effectively claims that not getting corrupted by absolute power is 100% impossible (which would imply loss of free will on attaining or staying in power), human nature makes it exceptionally difficult in practice to stay benevolent in a position of power.
A firm belief of mine (and I'm adamant about this) is that one can move up in society without becoming corrupt. Human nature doesn't have to get the best of us. If you know what you're doing, you don't have to fall into traps like this.
The problem is that the higher you are in society, the harder it is to continue doing the right thing.
I stand by my word but what's possible in theory is not necessarily what actually happens in practice.
@bwDraco I don't think it's harder to continue doing the right thing. But rather that the higher you get, the harder it is to get the right things done. Suddenly everyone needs a favour in return, and that's where corruption starts to seep in.
@Bob I'm not sure if updating to the NDIS 6.x version of the RNDIS driver will fix my random dropping out, but it certainly fixed my XCOM 2 issues, and it also fixed my Network Connections control panel applet issues (severe freezing / slow loading).
I think there was some bug in some subsystem (perhaps attempting to send "broadcast" packets) of the NDIS 5.x RNDIS driver.
Or maybe the mere action of loading such an ancient (XP 64-bit era) driver on Windows 10 would cause the problem due to a core kernel networking stack problem.
@Bob Meh. My maximum investment to get a brand new 128 GB iPhone 6S Plus, regardless of the reason for the loss or damage of my phone, is $70 USD.
AppleCare is a pretty good program, all in all.
If my phone ever becomes damaged beyond just cosmetic scrapes or whatever, I have a golden parachute to a replacement, no questions asked.
Even loss of the phone (left it on the subway or something) is covered.
I have no reason to seek out third party repairs.
There's also an Apple store within convenient driving distance from my house.
Hardware question: if my headphone seem to be having trouble (distortion) at the base end of the range, is it more likely to be (a) the headphones (b) the sound card?
Sorry, my vocab is failing me. Or my power of description.
I've been considering whether to replace my sound card, so I'm trying to find evidence it is actually falling down on the job. It's an ancient Sound Blaster Live! 5.1
I know exactly how to detect whether it's your hardware or your sound card :D
This sounds ridiculous but it's a PERFECT diagnosis method :D
Record the sound you are hearing with the most evident distortion
Then, either listen to that sound using another sound device (your phone or something), or send it to someone else
Then describe what you hear, and if the other person doesn't hear it, it's your headphones
This method will eliminate the possibility of it being anything on the software or "digital" side, but unfortunately it can't eliminate the possibility of the DAC on the sound card going bad.
@allquixotic I was looking at EC2/Azure/GoogleCloud in Sydney... the smallest instance is actually pretty cheap, especially with Google, but wow data transfer is expensive...
I'm sure someone's already asked it before and have gotten an answer for their problem which is different than mine even though the problem is the same...
I was hoping to get ahold of allquixotic because he helped me get the IRST working
Basically run the installer again and tell it to "upgrade" your current installation which should keep your current files but reinitialise the OS
Sounds like you've repaired it not booting, though I've no idea why changing the processor would stop it booting unless you knocked a hard drive cable out.
But now your problem is Windows itself hates you, so you need to fix that
I'm really getting into scripting these days. It beats the art of high- or low- level programming hands down when it comes to quick development work. You are taking what other people have already done and reusing it. And if --help, help. info. and man don't tell you enough, there is always the...
@DavidPostill I tried, I honestly tried, but that holier-than-thou gibberish gives me the screaming heeby jeebies and I feel icky just looking near it.
This is going to be a long one. If you read it through, maybe you will understand why. If you just want the problem part, page or scroll down to the very end and work back up a paragraph or two. I always start with a full disclosure of the facts as I see them, and some people want just the sum...
For other readers: fstab (all lowercase) stands for File System TABle, and is a file located in directory /etc (short for etcetera), so its path and name together are /etc/fstab. This is a text file, so easily read and changed with a text editor like gedit or nano, two of the most popular, beca...
A discussion about fstab goes on to Pi and encryption ...
HDMI sound too High pitched, notebook/laptop speakers work fine :
Turn "PITCH SHIFT" to OFF in my Toshiba, just clik Start, then goto control panel, then Sound, then to your device , clik on your device, then it should show you some tabs such as Playback, under playback tab choose Output Tab...
Disguised spam. (look at the last two lines of the post)
Ok, this is probably an idiot question, but are LCD tvs and LCD monitors substantially different? I.e. can I substitute one for the other? I think the connectors might be different - not sure
I disagree on the point of cutting RAM to 4GB. That's like cutting off your arm to save your leg. Swapping to even a very fast SSD is going to suck just as much as swapping to an HDD. Trust me, I've done it.
In that case, it isn't even so much "waiting" as "continuing to live my life" -- it'll be years before we see that tech available even to enthusiasts and enterprises
I'll check back in 2025 and see if I can get something with XPoint for less than $1000
@qasdfdsaq Yeah, and 2010, 2011, 2012, 2013, 2014, 2015 and 2016 are the year of the Linux desktop :P seriously, though, I'm extremely skeptical that the prices and volume will be anywhere in the ballpark of what will be necessary for people like me to afford it (and I'd totally be willing to throw $800-900 at something new and crazy fast like that)
I dunno about prices. But when Micron and Intel are the ones talking release dates, instead of theoretical researchers, it's far more likely to actually happen.
Intel may have their delays, but they rarely announce products that aren't going to materialise for 10 years.
not saying the tech isn't real; I just don't think the volume and affordability factors are even a little mature at this point
for example, the Intel Many Integrated Cores project's Xeon Phi is "available", but you have to shell out like $2k to get a base model, or much more for the top end
do you see home enthusiasts buying those? not really, unless they're a researcher for a company that plans on using them
and Crosspoint is aimed at being (eventually) deployed to every single segment of the computing market, from smartphones to craptops to high-end gaming desktops to dedicated servers to HPC... which of those market segments has the most potential for providing the companies the highest profit per unit sold? it's definitely not consumer
you go to consumer last to get high volume sales (which, in order to be able to do that, comes with enormous up-front costs) at very low margins per sale... it's a lot cheaper to have a small assembly plant that can churn out on the order of a few thousand chips per month, and mark up the cost ridiculously... HPC will buy them up and eat out of your hand
the cost of "NAND" is highly variable, though -- the cost of enterprise NAND in 4 TB PCIe cards is way different than the cost of consumer NAND in a TLC 128 GB SSD that's barely faster than an SD card ;p
I'm skeptical of a very advanced and expensive non-volatile storage technology being used in lieu of state of the art DRAM for the purposes DRAM is used for today; memory latency and throughput are quite important for many workloads, so it really depends on how fast XPoint can get without being more expensive than DRAM
again, their description was very qualitative; if "slower than DRAM" means "slower than all DRAM, even DDR1", then nobody's going to want to use it for the purposes we use DRAM for today.
if it's, say, 95% as fast as standard DDR3, maybe that'd work.
> Guy Blalock, co-CEO of IM Flash, told the EETimes yesterday that sample 3D XPoint chip availability is "just around the corner", however he added that it could take 12 to 18 months to get Xpoint into mass production.
Well they've given claimed performance figures already
"With NVMe-interfaced 3D XPoint memory chips capable of delivering more than 95,000 IOPs at a 9ms latency, compared to 13,400 IOPs and 73ms latency for flash"
12 to 18 months from some time in 2016 means it's definitely not coming in 2016 in any meaningful way, even if they rush their schedule (which, frankly, is unlikely for such a new and unique technology; I'm betting it'll be 18-24 months instead)
@qasdfdsaq as long as the chips use the same communications protocol (and IIRC DDR4 is serial so, there's probably a chip that negotiates these things) it might be a drop in replacement
yeah -- I'm quite pessimistic when it comes to totally unproven, never before seen technology that's only been outside the laboratory phase for about 5-6 months, and I'm not ready to accept that a CEO who wants his stock price to rise due to perceived future value is quoting realistic timeframes, and the early part of that timeframe will actually be met (on, again, an unproven product).
if we were talking about 4 TB consumer SSDs and they said 12-18 months, sure, as long as it's the same basic tech as we've been doing, I would totally say 12 months is reasonable
Well if we compare it to timescales for Intel CPUs when they announce "Chips sampling" and "mass production", it could be soon. And again we're not talking any old company here, but two of the biggest industry giants.
the closest thing to totally experimental, fundamentally new process tech that's come down the pipe in recent years was Intel's 22nm tri-gate transistor stuff, and that stuff never really got the attention of mass media or consumers until a year before it was ready; it was in development for about 8 years prior to that
my perception of this situation is that we're basically looking at 2016 as the equivalent of 2008-2009 and we're looking ahead at "wow, this thing called Ivy Bridge is coming in 2012 and it's going to have 3d transistors!"
Point to note is there are several other companies developing similar technology, aimed at 2017/2018 release as well, so it seems to be something that all the major producers are investing in.
So unlike Rambus it seems that lots of companies are going the same way
all of the involved companies that I've seen have been pretty quiet about announcing openly which market segments they plan to target first, though -- if consumer / enthusiast or at least "run of the mill enterprise" (servers) isn't on their first year of mass production plans, the prices for the HPC-targeted units will be out of our league.
so: in 2016 they say "mass production won't start for 12-18 months"; that means we'll see lots of chips coming out in q2 2017, but for which segment? -- if it's not consumer, that means optimistically mid-year 2018 would be when they'd start opening it up to the other markets.
I might buy myself an XPoint something or other for Christmas 2018 -- that's about the earliest I can realistically fathom
I don't manage an HPCC at work, so my work definitely won't be getting them prior to 2020... they tend to lag behind what I can afford as an enthusiast consumer, even for their datacenters.
for instance, SSDs are almost completely absent from any purchases within my enterprise
for you, it doesn't really even matter which market segment they target first -- if the tech is as good as they claim, you'll be seriously considering it regardless
as long as the chips are physically available to buy
I should note, however, that SSDs that meet that 95000 IOPS figure (or even exceed it) exist in the market, and they're very expensive but HPC type people buy them
those SSDs don't have anywhere near the write endurance that XPoint will have, though.
my next main desktop upgrade is going to use 16 GB DDR4 modules... only question being whether I can afford two of them when I upgrade then two again later, or if I'll buy all four in one purchase
does anyone know if Samsung's NVMe cards are going to continue to lag behind the SATA 6 Gb/s drives in terms of storage capacity? I love the IOPS and throughput characteristics of the 950 Pro, but the idea of paying such a price premium for half the storage I could get with an 850 Pro kind of sucks
It's a Samsung MEX controller. Samsung make their own proprietary controllers, so I dunno if it's ARM or not
But I wouldn't be surprised if there was an ARM CPU on there (or multiple, I think its a triple-core controller)
> First up is the upgrade to the controller. Samsung’s naming scheme from the 830 onwards has been MCX (830), MDX (840, 840 Pro) and now the MEX with the 840 Evo. This uses the same 3-core ARM Cortex R4 base , however boosted from 300 MHz in the MDX to 400 MHz in the MEX.
Well, it's a 3-core ARM Cortex R4.
So while there's definitely spare space on the PCB, the 1TB drive is using 50% more area than a 2280 M.2 has available. So I guess yeah it's purely a space issue
@allquixotic Get one of those laptops with 3 M.2 RAID slots :-P
there's no downside to running your routers on i686 or x86_64, obviously, except generally higher cost and the need to roll your own antenna setup if you need decent external antennas (whereas a fully integrated router comes with its own, of course)
ugh. Hopefully I have time but the current plan is to get the beebox back up with the install on a spare external hdd RMAing the old drive (thank god I buy HGST. 3 year warranty)....
@qasdfdsaq that's where a nuc class machine might shine
mainly I just like having the external antennae pre-mounted on the unit and already optimized for a broad variety of use cases and room geometries and such... having the PHY working well out of the box is a lot less tinkering for me
@qasdfdsaq I don't know, but I know I'm doing it wrong; I bought a log periodic Yagi external antenna to connect to my Verizon Jetpack, and I currently have it precariously balanced on an old milk carton pointed toward the tower
That's fine tho, I'll be spending the day trying to ensure my old HDD is dead (can't really RMA it until CNY is over) and getting my brix beeebox back up
I get 5 to 12 Mbps down with the hotspot sitting on my desk and no external antennae; pretty consistent 12 Mbps down with the hotspot balanced carefully on a high-up shelf ledge (with a cable running down to my computer) and no external antennae; and consistent 25 - 30 Mbps (!) with the Yagi plugged in.
I had a good feeling about the prospect of increasing my signal strength with an antenna, because when I was eating dinner at a Pizza Hut right next door to a Verizon tower (the one the Yagi is pointed at) a few months ago, I got 35 Mbps on speedtest.net
On my Mid-2009 MacBook Pro, I have OSX 10.6.8 installed, Windows 7 (through Boot Camp), and 2 FAT32 partitions. The extra two partitions are not visible to Windows 7, but they are visible on the OSX side.
Below is the output of diskpart, trying to show the volumes:
DISKPART> list disk
Disk #...
I'm really getting into scripting these days. It beats the art of high- or low- level programming hands down when it comes to quick development work. You are taking what other people have already done and reusing it. And if --help, help. info. and man don't tell you enough, there is always the...
HDMI sound too High pitched, notebook/laptop speakers work fine :
Turn "PITCH SHIFT" to OFF in my Toshiba, just clik Start, then goto control panel, then Sound, then to your device , clik on your device, then it should show you some tabs such as Playback, under playback tab choose Output Tab...
On my Mid-2009 MacBook Pro, I have OSX 10.6.8 installed, Windows 7 (through Boot Camp), and 2 FAT32 partitions. The extra two partitions are not visible to Windows 7, but they are visible on the OSX side.
Below is the output of diskpart, trying to show the volumes:
DISKPART> list disk
Disk #...
