« first day (1661 days earlier)      last day (3374 days later) » 

12:13 AM
I've noticed that a lot of new web apps that work on Chrome are simply not supported on Firefox.
Is Firefox really this far behind?
It's as if Mozilla is critically underfunded and unable to make substantial progress on Firefox
The last major Firefox update was on January 13.
Okay, a new release is expected in three days.
This is the release history of Mozilla Firefox. == Mozilla Firefox timeline == == Current and future releases == === Current supported official releases === Firefox 31.4.0 ESR Firefox 35.0.1 === Current supported test releases === Firefox 36.0 Beta Firefox 37.0 Developer Edition Firefox 38.0 Nightly === Future official releases === Firefox 36.0 Firefox 37.0 Firefox 38.0 Firefox 39.0 Firefox 40.0 Firefox 41.0 === Future test releases === Firefox 37.0 Beta Firefox 38.0 Developer Edition Firefox 38.0 Beta Firefox 39.0 Developer Edition Firefox 39.0 Beta Firefox 40.0 Nightly Fir...
Firefox is about 2-3 years behind Chrome overall
E10s is stalled
Mozilla is grasping at straws at this point. I might have no choice but to switch.
 
12:44 AM
@DragonLord No, it's really not. It's just that Chrome has a huge market share advantage, so there are many web apps coming out that are targeting Chrome and not Firefox, and blacklisting the Firefox user agent without any justified cause. Or maybe there's one little trivial feature that they think they absolutely must have that only Chrome supports, but FF will support in a couple weeks.
Or maybe it's that a lot of people tend to run old versions of Firefox (e.g. in enterprises, or with the Enterprise Stable Release series), whereas anyone running Chrome gets auto updated without even being prompted to update
so if you target Chrome you know that it'll work because no one in their right mind runs outdated Chrome, but if you target Firefox, they might be still running Firefox 3.6 or some ancient crap like that
and Chrome is more generally willing to incorporate bad things like DRM and proprietary plugins and patent-encumbered media codecs, whereas Firefox at least has a conscience and resists doing that somewhat (though as of now they do support patent-encumbered codecs and DRM, but it took a long time)
 
1:20 AM
@BGM not too sure. Pre post is hard to troubleshoot. Remove and reseat everything?
 
BGM
@JourneymanGeek Could there be a jumper or something missing?
 
Maybe but doubtful.
 
BGM
What can I do to test that the board is receiving power?
 
Hm. Donno. Multimeter at CPU power?
 
BGM
When I plug in the power supply, the fan is supposed to spin up and stop.
 
1:25 AM
HAMMERTIME!
 
BGM
Ah, I see what's wrong now.
Some of the chips are smashed.
It's not something that can be repaired.
Thanks for your help, @JourneymanGeek
 
stop
 
HAMMERTIME!
 
Oo
@BGM actually if its a vrm rather than a chip set....
But at this point we're talking a massive time/effort sink
Basically you could find out what chips they are, check on say mouser, and replace them
 
Bob
2:19 AM
@DragonLord Uh, no?
 
I'm still using Firefox primarily because it's a bit more powerful than Chrome (more features, tab groups, better plugin support, etc.)
I'm getting increasingly tempted to switch, though.
 
Bob
@DragonLord e10s was briefly put on the backburner for a couple years because they had more important things to focus on - with great results. FF is now far slimmer on memory usage than Chrome. They've recently restarted dev on e10s in the last half-year (and it's already made its way into default on Nightly).
Honestly, for the most part FF is stable enough that e10s wouldn't make a great deal of difference.
OMTC alreay helps with that.
End of the day, the biggest issue was addon and plugin compat, which is mostly resolved by the compat breaks in recent versions of FF with a new addon SDK
 
Crashes happen, but they're weeks apart in my experience
 
Bob
I have no idea where you pulled a "2-3 years behind" figure out of.
 
Bob
2:23 AM
If you look at ES6, FF is far ahead of Chrome (granted, it's still all marked experimental)
> No Multi-Process Architecture
> CPUs are gaining more and more cores, becoming capable of doing more work in parallel.
Blatantly misleading.
There's this little thing called multiple threads
 
The typical consumer CPU has at least two cores and can concurrently handle at least four threads.
 
Bob
(Which, again, OMTC helps with a lot)
You don't need multiple processes and OOP tabs/sites to render on multiple threads.
 
Yes, Firefox is multithreaded, but it's still slower than Chrome.
 
Bob
Though, practically, they were tied down a bit due to the need for addon JS to execute on the same thread. That was an issue.
 
A busy script can tie up the entire browser.
 
Bob
2:25 AM
They've taken the compat-break steps now anyway.
My only current gripe with FF is the lack of a 64-bit build. But that won't be an issue for much longer.
Chrome... I have a whole list of random bullshit it throws at me.
Starting with broken font rendering since the initial release.
As a dev that has to target all three major browsers (and all of IE from 7 to 11 -_-), Chrome and FF are tied for special-cases we need to make.
Though with FF it's often just picking a different prefix or using a slightly different property.
With Chrome, it's screwing around with rendering, having to redraw charts, etc. ... heck, IE8 handles those cases better than Chrome.
Oh, there was that stupid filetype prediction too.
Oooh, look, there's this application/octet-stream with the name a.ext... it looks like a zip file! Better rename to a.zip proceeds to mangle the filename so the user can't load it properly
They seem to have fixed that at some point, actually... hm. Might have to remove that crappy workaround now.
Still, all browsers (IE11 included btw) are ahead in some ways and behind in others. That's just how this works.
Then again, it's also wrong to call non-standard features "ahead". Just different.
It could end up being dropped further down the line (see: Web SQL)
 
 
1 hour later…
3:52 AM
Apparently a four letter scifi themed word links these images i.stack.imgur.com/dXxOV.png
Halp
 
4:07 AM
woof
 
Tried that, didn't work :P :D
 
;p
clearly you got the pronunciation wrong
 
lol
 
hmm. Credit card... who's the green dragonball guy?
walking stick...
 
Picolo
I have no idea how science fiction links all this :(
and I'm not supposed to take it literally -_-
 
4:11 AM
lol
or seriously?
 
seriously :/
 
seriously.
;p
cane?
 
nope
I tried several hundred words :D
Only then i figured it was prolly something technical
 
;p
card?
 
Nope and the worst part is I get memes evertime I enter a wrong answer
 
4:18 AM
lol
wait, is this an anti-bot measure? ;p
 
lol no :D
It's like a passwordish thing to unlock goodies
 
Bob
Do a reverse image search on that image.
 
I did :|
To the entire one and also the smaller ones
I can't fing link them
Me thinks it's related to dragon ball lore
 
 
2 hours later…
6:08 AM
oh look Microsoft sent me a 50 mb zip file with the title How to write a resume !
 
Look there's a video file inside
Much smart
very microsoft
 
 
1 hour later…
7:32 AM
@HackToHell master dragon can (cane) code
 
7:44 AM
@Psycogeek But the answer is a single four letter word
 
8:14 AM
Just trying to clarify on authorative DNS servers.. a DNS server is authorative if it knows something that's a little more specific info on the next host. For example, my local computer wants the IP of my comp in the CS dept, mycomp.cs.wvwc.edu (provided publicly routable ) - a dns query would go to the authorative server for wvwc.edu, which would go to the authorative server of my department,
cs.wvwc.edu, and that would be the authorative server for my hostname... is this correct, although a bit complicated example?
@JourneymanGeek
 
8:28 AM
derived from somehow?
Card
Lore
Cane
Code
3Cs one L 3 Es but one D . "Lord" "Care" . could be anything, with only 4 letters brute force :-)
 
8:48 AM
@ekaj: Authoritative nameservers only answer queries about its own zone. Recursive name servers ask an authoritative nameserver before answering a query about a zone
as I understand it.
 
thankya =p I also pinged you because I was wondering if you would like some old-ish hardware, depending on if shipping isn't completely ridiculous
 
@Psycogeek I am :D
There's just too many words
And I can't automate
 
Whatever the green dude is might fit in better with the other 3. but i will never know who it is, even if i had once seen him. way to many pokeman :-) people (anime) to know.
 
yes no? =p
 
8:55 AM
maybe?
 
Blah, I got another clue
passive listening
How does that even fit in :{
 
Lets see Green pointey eared dude who does spells , descibes 500 different charachters in games google.com/… the search proving that
(with a previous head injury)
 
@Psycogeek he's picolo from Dragon ball manga
He apparently has hypersensitive ears and can pick up stuff from miles away
 
was that before or after the head injury :-)
 
i think that's part of his dress(???). I don't know
The wiki doesnt say anythjing
 
9:06 AM
4pics1word.ws/4-letter-words Oh my , there is even a web site for this
 
I wonder how long it is before neural networks can solve this faster than humans
 
Soo there is no huge puzzel, all the pictures could be defined with the same 1 , 4 letter word?
 
yeah but loosely
SQL injection isn't working
hmm
 
Heck i can get all the others :-) you picked the hardests one
Credit master card plastic bank money logo
Green , Strike Lore dragon ears ball wind
Cane Walking stick curve round wood twig knot
Code Hand computer stat screen font
nope nothing :-)
as a side note Font would make a good quad too, with its other definitions being

font (fònt) noun
1. A basin for holding baptismal water in a church.
2. A receptacle for holy water; a stoup.
3. The oil reservoir in an oil-burning lamp.
4. An abundant source; a fount: She was a font of wisdom and good sense.
from Latin, fountain
after you get it , you will have one last 4 letter word, starts with S and ends with T. :-)
 
9:37 AM
This is what I got for entering shit :D :D
 
3 out of 4 "Push"
"Aids" money crutch, super hearing, walking crutch and a computer
 
10:21 AM
That's the emmc on my HP stream
 
Bob
10:48 AM
@JourneymanGeek Better than the eMMC on my tablets :P
 
@Bob: actually, better than the random spinning rust drive I tested
stuck a SD card for slow/bulk storage
going to give linux a shot off a USB key
0_0
Debating throwing a ipv6 tunnel software on the new laptop
since its following me around, and I can set AAAA records for things at home
 
 
2 hours later…
12:37 PM
blah
UPDATES
 
12:56 PM
answer is yoda
cc @Psycogeek and jmg
i have no fucking clue how
 
1:09 PM
Ahh
Cane, green skin, pointy ears....
Master....
 
oooh
Now i get it :D
what about master card ??
and computer
 
Bob
@HackToHell typing backwards <=> speaking backwards
 
These people think too much
 
Says the guy who spent most of the day thinking about it?
master card ;p
I don't remember what was on the screen on the computer ;p
 
1:24 PM
@JourneymanGeek ^_^
 
could you link the image again?
 
yup
typing backwards he is.
3
 
related :D
 
 
7 hours later…
8:34 PM
Out of curiosity: Will I ever find a use for a machine with more than 8 cores or 16 threads in a client environment?
I'm playing with the idea of building a dual-socket Xeon workstation using type EP processors.
Assume budget is unlimited.
Once again, this is in a client environment.
 
8:50 PM
Anyone?
 
Bob
@DragonLord As always, depends what you do.
There's no answer for everything.
Web browsing? Nope. Graphics editing? Probably nope. Watching videos? Nope. Editing videos? Nope. Playing video games? Probably nope.
Unless you have a specific need for it, the answer is probably no.
In fact, such a machine might perform worse than the equivalent enthusiast machine, wince you have to deal with shared memory, GPUs, etc.
 
For a high-end gaming machine, considering current hardware, I would actually prefer the i7-5930K because higher clocks matter more than core count.
Yes, you're right.
I was just considering some special-purpose machine for exceptionally compute-heavy workloads (read: Folding@home)
 
Bob
@DragonLord Well, the whole point of that kind of task is you'd get more out of the idle time of 3-4 home computers than a single high-power workstation.
If you wanted to donate all that power, sure.
If you want to actually use it yourself? Not so much.
Actually, I can't think of anything off the top of my head that would benefit significantly.
I mean, sure, some image and video processing algos will benefit. And maybe some specific physics-heavy games like ARMA 3.
 
The only valid use case I could think of is in a high-density server/datacenter environment.
That's what these Xeon processors are really for.
0
Q: Why does a Xeon processor cost much more than a similar Core i7 part?

yasharI want to know why the price of a Xeon CPU and a similar Core i7 CPU is so different. The CPU clock is more or less the same. I want to know this because I want to have an understanding of what kind of CPU is more suitable for me. I do heavy mathematical computations and simulations which may t...

 
Bob
But unless you're doing something extreme, you probably won't notice the time for the first two (and GPU would probably do better for video), while the games are relatively specific and even then rarely make use of more than four cores.
 
9:03 PM
I think I already know the answer.
4
A: Why does a Xeon processor cost much more than a similar Core i7 part?

DragonLordDifferent Uses and Environments Consumer-grade Core processors are designed for everyday desktop or gaming applications and are therefore optimized to operate at higher clock speeds. Most consumer applications cannot take advantage of more than a few processor cores and would benefit significan...

 
Bob
They develop those games to target typical consumer machines.
@DragonLord Yes, but you did specify client (assuming home) uses.
 
@Bob The expectation is that no normal consumer machine will have more than 4C/8T.
 
Bob
Exactly.
There's the possibility you'd run some expensive operation in the background, too.
Though, at that rate you might be better off getting a second machine - the cost of a Xeon CPU alone (let alone mobo and other parts) would be more than a full second consumer computer.
 
Probably better spent on a GPU array.
3-way SLI GTX 980 is much more effective for Folding@home.
There are very, very few applications in a client environment that could ever require more than 8C/16T anyway.
 
Bob
@DragonLord I honestly can't think of any over the top of my head.
Heck, for most people I can't even justify an i7 over an i5.
(Even me personally... the only reason I'd benefit from the former is because I tend to run VMs now and then.)
 
9:09 PM
On a system using a i7-5960X and three-way SLI GTX 980, you could run a CPU slot with 12 threads and three GPU slots, one for each of the three graphics cards.
At about $4,000, the cost of a machine of this sort is less than the cost of one Xeon 18-core processor.
 
Bob
@DragonLord The 18-cores are an extreme example, though.
More typical to see 6-core parts.
And even lower.
 
8-, 10-, and 12-core parts are pretty common in high-density datacenter environments, though.
 
Bob
And even with a Xeon E5-2680 V3 you're looking at $2k-$3k just for the CPU. Double that of the i7-5960X.
@DragonLord Most businesses wouldn't even have that.
Even in a business environment you're mostly hurting for RAM and I/O bandwidth and redundancy. Not raw CPU power.
 
@Bob It's about US$1,800. 12C/24T, 2.5 GHz.
 
Bob
@DragonLord shrug I'm quoting Aus prices.
Though now I see I picked the wrong part to compare :P
It really depends on your needs. Running AD? DNS servers? Database servers? You probably want anything, anything and more RAM/I/O respectively.
 
9:15 PM
@Bob The same holds for the vast majority of lighter consumer workloads. I/O bandwidth is what really matters, hence the move to SSDs.
 
Bob
An application server might want some more CPU power, but even then probably not that much unless you're really big or happen to be doing something particularly expensive.
@DragonLord I would say it's even more true for businesses.
A DB server eats RAM like there's no tomorrow.
 
@Bob My VPS is primarily CPU-bound (PHP is horrifically inefficient). There's a DB but it doesn't see very heavy use, only as needed for the blog.
Most of the waiting time a regular consumer experiences using a PC is due to inadequate I/O bandwidth.
 
Bob
@DragonLord Where's your VPS hosted? Because you're probably getting a single core's worth, if that.
 
Linode, 2GB plan.
Two CPU cores.
 
Bob
It scales to an extent - you might see benefit up to 2, maybe 4, cores.
 
9:18 PM
Contention is pretty low on these servers.
 
Bob
But above that? No, you'll likely end up finding bottlenecks elsewhere.
 
Web server loads of this sort scale linearly with CPU core count.
 
Bob
@DragonLord You're also right about PHP being inefficient :P
 
Keep in mind Apache runs a separate process for every client.
Nginx and other web servers are mostly the same.
 
Bob
Not necessarily the core itself (though some language "features" also make that unavoidable).
PHP sites (Wordpress in particular) tend to be... ah... messy.
Not well optimised.
@DragonLord ...no, that's not true.
 
9:20 PM
It's Serendipity, which requires Apache.
 
Bob
Nginx uses an event architecture.
That's also available with Apache (event MPM).
But you're probably on the prefork MPM, which is, frankly, shit and only useful for mod_php (which is also pretty nasty).
 
@Bob I thought it's threaded nonetheless and therefore scales linearly with compute performance.
@Bob Yes, I'm on prefork
 
Bob
@DragonLord It's threaded, yes, but not a separate process per client.
 
Currently configured for 32 processes max
 
Bob
And it reuses workers.
Far more efficient.
 
9:23 PM
In any case, there's really no reason for me to build a desktop around Xeon processors.
i7-4790K is plenty of compute for the vast majority of use cases.
 
Bob
I still don't max out my i7-2600 most of the time.
Actually, I can't remember the last time I maxed it out.
If/when I upgrade, it'll be for newer mobo features more than the CPU.
 
On my laptop (i7-4800MQ, 4C/8T, 2.7 GHz + Turbo), I do run into CPU bottlenecks on some games.
A faster-clocked CPU would help.
But it's not a huge issue.
No game I've played so far genuinely requires more than four threads.
 
Bob
Oh, I'm sure my CPU is bottlenecking on games.
But it's not significant enough to really affect me. It can always run faster, but that doesn't mean running faster is noticeable.
And even then it's single-threaded bottlenecks.
(Game I play: runs at 100-200 FPS. Could potentially reach or exceed 300 FPS with better CPU and GPU. Do I need it? No. Do I want it? ...not really, actually, I'd probably never notice.)
 
I have two desktops on my mind, one ultra-high-end system, one regular gaming system.
I was considering the Xeons for the ultra-high-end build but now I don't even see a good reason to pick the i7-5960X because benchmarks actually show worse results than on the i7-5930K.
Once again, most games are not designed to use more than 4C/8T.
 
9:40 PM
At least I can put that thought to rest.
There's just no reason to get a processor with more than 6 cores, now or in the future.
 
9:59 PM
I just finished rebuilding my toshiba sattilite C655D-S5226 laptop after i Had to open it up to fix the screen hinges
my C655D is the laptop I am using to type this
 
10:13 PM
0
Q: Need Write Access to Physical Drives - Getting Access Violation

Nick McCamyPurpose: To clone a drive from another drive. Using Functions: CreateFile, WriteFile I have no problem reading all of my physical drives. But I cannot write to any of them without getting Error(5), Access Violation. I need full writing access to all of my physical drives. Obviously I could ge...

Why are you using the Windows API directly to do this? — DragonLord 28 secs ago
o_O
 
 
2 hours later…
11:57 PM
A few years ago, I read about AMD leaving BAPCo over biased benchmarks
It turns out SYSmark has historically been biased towards Intel
 
@DragonLord we have those sorts of crazy stuff at home work
and in general? No
 
Huh?
Oh, you're talking about the Xeon processors.
 
We do have workloads that tax a 12 core system with a shitton of memory
The nice thing is your machin will age well
 

« first day (1661 days earlier)      last day (3374 days later) »