« first day (1724 days earlier)      last day (3286 days later) » 

12:00 AM
...and yes, it is an XL 1900.
I would suspect these are Cherry MX Black switches.
The keys are pretty stiff.
 
newegg.com/Product/Product.aspx?Item=N82E16823816024 Corsair, not getting great reviews .
"Keys are starting to loose their tactile feel and click. I bought the MX Blue because I like the click sound and feel but I have a couple keys that already lose the feel and click sound. " That sucks, I didnt even know that can happen, as it can't happen with the buckling spring.
 
The Cherry MX Black keyswitches have been around since 1984.
 
I could not buy that corsair :-( i could put up with the cute lights being a pain, but not when that many people report (minor) issues with the keys. Uhh Key board , Key being important here :-)
Neweggers can be pickey computer geeks.
 
Just typed a bit more on the typewriter, and suffice it to say it's very fatiguing to type on it.
 
newegg.com/Product/… the rosewill also using clone keyswitches .
 
12:39 AM
mechanicalkeyboards.com/shop/index.php this place seems to have more Mech selection in one place, although the prices might not be best.
 
I ultimately want a Roccat keyboard to match my Roccat mouse.
As I've said, I'm currently using an Isku.
The Ryos has the same key layout, but uses mechanical keyswitches with a choice of Cherry MX Black, Blue, Brown, or Red.
The Ryos MK Pro is the highest-rated keyboard on PC Magazine.
It's also expensive, however, at about US$160-170.
 
that still beats $300 , for the price that a great keys fully configurable RGB backlighting. That is the price when these things first came out.
 
Was thinking about how to use the per-key lighting.
 
At $300 i was not going to budge. now at 100-150 it might be worth looking into. I am not doing so well seperating the "backlight" and "cool effect" with Different color for different key.
 
On first thought, it would seem to be little more than mere eye candy, but in practice, it gives you the ability to color-code specific keys or groups of keys for a significant ergonomic advantage.
 
12:56 AM
Then how transparent, or even using white keys, would a person want that, as 50% of the light acts as an underglow for a mexican porche :-)
Many of the well designed color schemes show that much of the light is "between the keys"
I see they have part transparent black keycaps, and we know that white could pass more light through.
 
Before I thought about this, I only considered simple backlighting to be a requirement when choosing a keyboard.
I had not realized that per-key color lighting is a very useful feature. It is by no means just eye candy.
 
I thought the LCD keycaps were kinda cool :-) Or OLED as it would be.
My caps already have moronically huge fonts on them :-) soo i can find H (my med kit) easily, right before i die again.
 
Okay, the Ryos MK Pro isn't RGB color, but it allows individual keys to have their backlighting turned on or off
It's still a very useful feature.
 
From what i am reading, the light designs (location of led) did not even take into account the angle that a KB is viewed at , which is ??? Hunched over looking straight down? 30* offset ?
 
Okay, I'll pony up the $170 when I have the money. A good keyboard is an investment. My Isku ran me $70 and is already showing some wear.
Well, according to my Best Buy receipt, I purchased it on May 28, 2014. Time flies...
 
1:07 AM
Wear as in filth? or wear as in edges rubbing off? or keycap printing wearing off?
 
The keycap printing is still okay but the key surface is smoothed considerably.
 
oh
 
This keyboard was originally purchased to reduce the need to type on my laptop's keyboard. I feared I would prematurely wear down the built-in keyboard especially during heavy gaming.
Eleven months in, the Isku is still fully functional, but the wear and tear is starting to show.
 
hmm that does not seem to happen here, there is some kinda texture that has for the most part stayed, not sure how important that was. Oh noo, add another design (and longevity) feature to wonder about
 
1:38 AM
I would not want my keyboard to be doing most of the cute lighting effects , while using it. More like do that stuff in keyboard standby mode :-)
I can understand that they did the effects because that is what you do with leds :-) rope light disco effects and all, but what practical application would that have while using it? To show off great, but who would want a pulsing light going when typing?
I can think of 101 ways where it could be practical. Brighter as the CPU use or GPU use goes up. Snake lighting to show progress of actions (like install and copying) For the game to light up that stupid QTE button your supposed to be pushing :-)
But color codeing functions of the keys, that i can get behind. to special color all the keys that a game uses.
Coloring WSAD (only) , that is sorta lame, because anyone who has gamed more than 5 hours, the hand just moves into that position without any assistance :-) Had to know they would start with that though, Ohh lookie gamers we can highlight WSAD , wooptidy doo.
 
On my Isku, the W key has a tactile bump like the one on the F and J keys.
Even better, they say, is the Corsair Gaming K95 RGB, but I would prefer to stick with one brand (especially because of the Roccat Talk functionality).
I want Roccat to add per-key RGB lighting on a future flagship model.
The Corsair software and firmware isn't great according to customer reviews.
I have had some firmware issues with my Roccat Tyon, but they have been sorted out.
My experience with Roccat software is very positive and I would prefer to stick with a brand that has a proven track record.
Roccat hardware does have somewhat spotty QC, but it's not a huge deal to me.
Gaming peripherals are more than just the devices themselves, it's also the software that goes with it.
 
2:13 AM
Browns sound cool, all the beump without the noise, blues to stick with the noise i already have, but all those designs look like they would be subject to "wear over time". I switched to mechanical after many mushball keyboards failed over short time, and my buckling spring has been 99% the same today as it was 10+ years ago. and because the design for the bumping (and clicking) is a Default action of the design, not Rubbing stuff together, I am not sure i could change switches.
 
My father in laws Macbook is coming up with a cert error every time he opens iTunes. Anyone know how to get new certs on OS X?
 
With the membrane keyboards ONE stickey or obstinate key could turn the whole keyboard to trash,
 
This is stupid.
I just updated the Intel HD Graphics driver and it's crashed on me twice so far. Twice in the same session.
I've just rolled back the driver in Device Manager—this should address the issue.
 
Bob
@DragonLord wear down??
I type a lot, and I have yet to wear down any keyboard
or even come close
 
A laptop keyboard is much harder to replace than a desktop keyboard. That's why I'm concerned.
I have had lots of issues with my old laptop's keyboard. HP has had to replace it twice over a three-year timespan.
As a graduate student, I cannot afford any downtime with my laptop.
5
Q: My keyboard is typing by itself, could there be a ghost in it?

DragonLordThe keyboard on my laptop seems to be typing by itself: It sometimes types characters like 134, whether by itself or when I hit certain keys. Repeating characters may appear unexpectedly. The system volume may change by itself, or windows like the Print dialog or Web browser may appear. This ...

-1
Q: How can a CPU have a lower clock but be faster?

Tired GonzaHaving seen various people test powerful emulators on their computers, I have come to surprise hearing about how one CPU can have triple the clock speed but still have worse single-threaded performance; or how one CPU can use 16 threads but one forced to use a single-thread can wipe the floor usi...

Perfect. I'm studying computer architecture right now and can compose a high-quality answer to this question.
 
Bob
2:38 AM
@DragonLord pretty sure I've answered that question in the past...
 
@DragonLord Then you should look for the DUpe first.
 
Bob
Oh, here => not a direct answer to that particular question, but the last section pretty much addresses it...
> How are modern CPUs so much faster than older ones at the same clock speed?
 
I'll explain it in more detail. It's far more than just pipelining.
 
Bob
Not sure if there's a more specific dupe somewhere.
@DragonLord Of course, I barely touch on that - the answer was quite long enough as it was. And was orthogonal to the original question.
 
Out-of-order execution, superscalar architectures, branch prediction, caches, the list goes on.
 
Bob
2:41 AM
@DragonLord A laptop keyboard typically costs ~$40, and is usually pretty easy to replace.
I still think there might be a dupe elsewhere though.
2
Q: Whats the deal with these new lower clock speed CPUs on Laptops? Can they still run apps well?

NateI'm looking at some new laptops, and I'm seeing the AMD Neo CPUs at 1.5GHz, and some of the new Intels at 1.3-1.8GHz. The laptops they are on come well supplied in terms of RAM (up to 8GB) and Video Cards (ATI's with 2GB shared memory). My question is, how do these laptops perform running heavy...

3
Q: Which property of CPUs is good for what?

erikricBack in the days clock frequency used to be THE criteria to compare CPUs (or perhaps you had to take into account whether it was a DX or SX). The world was simple back then, but in these fast evolving modern times, it's not so easy to figure out to best spend your hard earned cash on a new CPU. ...

 
P key just bounced on me.
 
Bob
55
Q: Why have CPU manufacturers stopped increasing the clock speeds of their processors?

learnerI have read that manufacturers stopped concentrating on higher clock speeds and are now working on other things to improve performance. With an old Desktop machine with Intel® Xeon® Processor E3110 with clock speed of 3.0GHz and a new server with AMD Opteron(TM) Processor 6272 with clock speed...

 
Is it time to get a mechanical keyboard?
 
Even the gonza repeated much of hisself superuser.com/questions/906198/… second question from same user using the same theme.
 
Bob
^ closed as too broad
 
2:44 AM
Edited the title:
0
Q: What factors other than clock speed influence the performance of a CPU?

Tired GonzaHaving seen various people test powerful emulators on their computers, I have come to surprise hearing about how one CPU can have triple the clock speed but still have worse single-threaded performance; or how one CPU can use 16 threads but one forced to use a single-thread can wipe the floor usi...

 
Bob
@Psycogeek Ok, that one is close to an exact dupe of this one
@DragonLord Then it becomes a list question.
Which isn't good.
I think the original title would've been more acceptable for this site.
 
I like original titles, only for the purpose of "How People Search"
 
Bob
Oh, here. Perfect dupe (cc @DragonLord @Psycogeek):
16
Q: Why are newer generations of processors faster at the same clock speed?

agzHow come, for example, a 2.66 GHz dual-core i5 will be faster than a 2.66 GHz Core 2 Duo (also dual-core)? Is this because of newer instructions that can process information in less clock cycles? Other architecture changes?

 
Or how the idiot (like myself) searches for something, when they do not already know the correct terms.
 
Okay, voting to close. I'll compose the answer for that question.
 
2:55 AM
youtube.com/watch?v=SEEzQKJfNE0 <--- lego keyboard video, aka i hope somone is paying you to do that :-)
I like the pictoral designs of some of the keys .
 
Bob
@Psycogeek re: your comment, to be fair it took me a while and a few pages sorted by votes to find the dupe
and I was pretty sure it was there
can't really blame him for that one
 
I am betting there are 4 questions of that type, but yea finding them. I just remember them being asked.
Thing about questions like that on regular forums, it ticks off the regulars, because they have already, been there, done that, seen that many times. So the people with no life other than the forum , will piss and moan about it being asked again. But the users , and other users dont mind answering it again, with pretty much the same info.
It is the circle of life :-) I mean the brain to brain transfer of information, student becomes teacher.
The more the students teach what they learned the better they become at (things like) answering other questions here. and the big circle goes round and round. With the grumpy old chimpanzees in the corner :-)
 
My answer will be a huge undertaking. I will post it partially completed and expand it over time.
 
3:51 AM
0
A: Why are newer generations of processors faster at the same clock speed?

DragonLordDesigning a processor to deliver high performance is far more than just increasing the clock rate. There are numerous ways to increase performance beyond clock rate, which are instrumental to the design of modern processors. Clock rates can't increase indefinitely. At first glance, it may seem...

Keep in mind this is a work in progress.
 
Dudes are asking simple questions, if you be putting all them complicated math in there and all that how to code info , are they just supposed to upvote it because "well it has a load of technical shit in it " :-)
 
It's conceptual in nature.
I'm just getting started. Right now, I'm writing about hazards.
In the domain of central processing unit (CPU) design, hazards are problems with the instruction pipeline in CPU microarchitectures when the next instruction cannot execute in the following clock cycle, and can potentially lead to incorrect computation results. Typical three types of hazards are data hazards, structural hazards, and control hazards (branching hazards). There are several methods used to deal with hazards, including pipeline stalls/pipeline bubbling, operand forwarding, and in the case of out-of-order execution, the scoreboarding method and the Tomasulo algorithm. == Background... ==
 
Dumb it down, dont just copy paste technobabble :-) how am I supposed to read it?
And dumbing it down does not require 40 paragraph analogy stories ;-o Either.
If it is a technobabble contest then your could potentially end up with 160Page Intel PDF poked into it, when a quick summary of what that means to a user would be finished reading faster.
 
4:16 AM
Clock rates can't increase indefinitely ??? true statement, but if they are 10,000 angstroms away from being atomic still, there is room for faster and way less power. through the ages they have always had limitations , back when they were 8 times slower they percieved all the limitations.
Then surpassed them again.
The processors are still being organically etched , so they can pump out 50,000 of them in a month. But don't you figure in some backroom at intel there is some guy who could (20 years later) atomically master you a single processor that would walk all over todays processors , fit on the head of a pin, and be cooled by it :-)
that for 20billion dollers they could use the linear accelerator , to etch out some transisters that will flip states if you blow on them?
 
4:35 AM
they already changed up to be closer to your ram, to improve speeds, if people plunked down enough money, and the gpu had not come into play, could thier next move be to slap huge quantities of ram right on the die. What if it did require an extra inche of space to do so? The chip is so puny now, they could put more stuff under there.
If heat was such a big deal, why did they choose to run the voltage regulators on the die? They gained control, but ended up with quite a bit of heat that did not need to be there.
 
1
A: Why are newer generations of processors faster at the same clock speed?

DragonLordNote: This answer is a work in progress and will be expanded over time. Designing a processor to deliver high performance is far more than just increasing the clock rate. There are numerous ways to increase performance beyond clock rate, which are instrumental to the design of modern processor...

Feedback so far?
 
4:51 AM
ok hazzards how were they improved between the 2 processors?
for pipelining you have "While modern processors often have significantly more. . .."
 
Huh?
 
The question asks why a 775 processor is slower? so each of your layers of work that the processor has to do , has been Improved in the modern processor?
Cant just describe stuff that has gone on since pentium II (which was able to process 64bit instructions) without indicating how that is different somehow?
Some might think the chart (clock speed) is archaic , beings processors sell that start at 4.0 and turbo to 4.5 and overclock to 5 (or realistic 4.8-4.9)
 
Bob
> This video contains content from Fox International Channels, who has blocked it in your country on copyright grounds.
 
5:09 AM
@DragonLord you have "different" covered just fine in the "branch prediction" section also "modern processors use more more sophisticated techniques for higher prediction accuracy."
 
I'm rather confused as to what you want changed.
 
@DragonLord I do not even know (first), but you have sections were you cover processor work, in the hazzard section you do not explain "how that has changed to be faster" The other sections each explain how that is Different.
 
The advances in these techniques will be discussed at the end of the answer. I'm drafting that section now.
 
 
1 hour later…
6:38 AM
1
A: Why are newer generations of processors faster at the same clock speed?

DragonLordDesigning a processor to deliver high performance is far more than just increasing the clock rate. There are numerous ways to increase performance beyond clock rate, which are instrumental to the design of modern processors. Clock rates can't increase indefinitely. At first glance, it may seem...

The answer is mostly complete. Ping me with any suggestions.
 
7:07 AM
@DragonLord ok now which one literally doubbled the speed at the same basic clocks?
Wasnt there like 2X as many transisters in there? so general execution capability?
doubbled the capability (not speed)
 
Hang on. I'm adding Moore's law to the answer.
Done.
1
A: Why are newer generations of processors faster at the same clock speed?

DragonLordDesigning a processor to deliver high performance is far more than just increasing the clock rate. There are numerous ways to increase performance beyond clock rate, enabled through Moore's law and instrumental to the design of modern processors. Clock rates can't increase indefinitely. At fir...

 
What happened to FSBs ? (front side bus) the (i guess you would call it) frequency by which all communications occured at ?
Wonder if that changed much in those 2
 
@Psycogeek Memory access speed is what's relevant here. Caches reduce this limitation.
I've addressed caches in my answer.
 
The wiki says that FSB speeds have been represented for thier capability , not just the actual "speed" changes that occured. Much like when they double speeded other stuff by having signal on the high and the low.
Because with the new stuff, they did increase the speed of the I/O lanes (like PCIe) going in and out. that offered up a good boost in overall computer speed, vrses just that computational stuff.
 
7:24 AM
The emphasis is on the main processor, the CPU, not coprocessors such as GPUs.
 
yea i dont use their gpu , so that i dont care so much about.
And counting the transisters used for the gpu , as if that changed the CPU, wouldnt be right.
I do know though that the old 775 was dependant on the rest of the chipsets northpole and southpole :-) to do much of the communications junk, and that was all obsoleted , and paths were reduced, and the cpu tookover (again) and is doing that stuff more direct.
 
I'll come back to this tomorrow. It's 3:30 AM here in New York.
 
Plus the more direct memory, putting the controller for it on the die, which put my memory speed up full 2x at least.
 
The integration of buses onto the CPU is something I'll look into, even though this is really tangential to the subject.
 
@DragonLord Well dont let me bug you.
I can stay up till the rooster crows with no pro. . . ok soo that aint true .
I agree reaching out to the PCI lanes is beyond the question., but nobody listed the memory controller, and that was a sure win.
 
8:21 AM
http://ark.intel.com/compare/29765,75123
a core 2 quad compared to a haswell quad, new instruction sets, boost of the claimed system buss speed, turbo boost & HT , completly different bus technology, (VT-x with EPT Extended Page Tables ), TXSs (Transactional Synchronization Extensions) , and double the memory bandwidth.
From the same line (but not in that compare) it went from 65 nm to 22nm
core 2 duo shows as having OMG 167 million transisters , glad they didnt ask me to count em.
 
8:34 AM
a haswell quad may have had about 1.4-1.6 billion transisters (unknown if they are also counting gpu)
 
8:51 AM
oh it was there, a Core 2 Duo Quad shows 582 million transisters (and did not have a gpu)
 
 
1 hour later…
10:20 AM
Mogges
 
The south bridge and the north bridge are actually quite easy to explain.
Dump all the stuff which needs to be fast in on chip (Communication with memory, with the CPU and with the EISA/MCA/PCI/PCI-X/VESA/AGP/PCI-e lanes).
Dump the rest on another chip (keyboard IO, serial, paralel, ...)
 
10:39 AM
According to "Dennard scaling" transistor dimensions are scaled by 30% (0.7x) every technology generation, thus reducing their area by 50%. This reduces the delay by 30% (0.7x) and therefore increases operating frequency by about 40% (1.4x). Finally, to keep electric field constant, voltage is reduced by 30%, reducing energy by 65% and power (at 1.4x frequency) by 50%.
Therefore, in every technology generation transistor density doubles, circuit becomes 40% faster, while power consumption stays the same
 
Bob
@DragonLord "Moore's law" isn't a law at all.
 
Wirth's law – generally is referred to as bloat and is the principle that successive generations of computer software increase in size and complexity, thereby offsetting the performance gains predicted by Moore's law. In a 2008 article in InfoWorld, Randall C. Kennedy formerly of Intel, introduces this term using successive versions of Microsoft Office between the year 2000 and 2007 as his premise. Despite the gains in computational performance . . .
Office 2007 performed the same task at half the speed on a prototypical year 2007 computer as compared to Office 2000 on a year 2000 computer
Lol, such is my premise, if you give them computational speed, they will blow it out your butt, with cute grafic, unnessisary computation, and fluff and bloat.
So an old intel worker who has busted his hump to deliver to us all 4 times speeds, can look over at the software messes and wonder what all that was about.
How many pieces of software have come out , in the same fassion as the CPU upgrades, touting major optimisations of code , reduced failure rates, smaller footprints , lower power . When the "market" still feeds off of More junk added to the features list.
 
11:05 AM
Its more of a suggestion ;p
 
If the first real powerfull computers took up half the room, who is to say there is not some person out there who wants a CPU that goes into a computer that still takes 1/2 the room
Nobody would want that :-)
 
11:22 AM
 
Bob
@OliverSalzburg ...why do I get the feeling the content of that particular site isn't particularly SFW?
 
@Bob I got it because I own dirty-motherfucker.org and .com
I occasionally get these and I find them hilarious
 
Oh, also, my Gandi domain coupons are expiring
.website free, .info free and 85% off on a .press
If anyone is interested...
americanex.press is available
Oh, now it switched to unavailable :(
american-ex.press works!
Could be useful for some nice phising
Could also make for some sweet email aliases
 
11:38 AM
american-ex-pres ? is that Nixons old domain, or clinton?
 
@OliverSalzburg Thinking up on domain names is too hard
 
12:12 PM
Somehow this reminds me of a console game where you have to press some insane combination on the controller:
1
Q: 2005 Toyota Avalon how to reactivate vehicle's ECU

Beth2005 Toyota Avalon FOB stopped working, got replacement, now believe accidentally deactivated vehicle's ECU, by turning ignition on/off 5 times during process of adding new FOB. So now how can I get receiver in car to accept signal codes from remote? It is not a smart key system!

 
I have had to do that before. Insert key in ignition, turn it but do not start, press the lock key switch on the door 5 times, press the FOB button, remove key.
There are other fun codes too, that involve combinations of buttons and pedals to reset parts of the computer, but none of it is in the manuels.
 
Time to start googling for service manuals for cars. So far I only needed to do that for computers.
 
I think like what nutjob thinks the car is a casio watch
 
anger. rising
 
12:28 PM
did you try turning it off then on again :-)
 
Oh, I'm gonna turn it off alright!
Also, I just installed an additional printer+scanner combo. The magenta ink is empty apparently. I don't care, I only need the device to scan. But it constantly starts blinking, to alarm me that I MUST replace the ink ASAP to continue being able to print
What an annoying POS
Also, the accompanying software package created 3 desktop shortcuts and then started to tell me that I should register my printer, I declined. Then it asked me to register the software package, no thanks. Then it asked me to register my IRIS OCR component...
Seriously, wtf is wrong with these manufacturers
I'm gonna send them an angrily worded email from ceo@american-ex.press
 
lol
Last multi-function printer i installed, i spent an hour, clarifying which driver i would require at the time. leaving the bundle on the disk, for that eventual need (never).
Free OCR is cool.
 
My sister gave me this device because she had no need for it and I was like, wth, free scanner
I should just get a proper document scanner and ebay this POS
 
all-in-ones have always come with huge bundles, and multiple notification icons, and software that 3X tries to update (when the net is off). they really are amasing tools for the price though.
If it no longer prints you have this huge scanner that still works :-) and when that fails, you can still send out low res faxes :-)
 
This thing is already driving me insane
The lid on the scanner opens to like a 90° angle. So the slightest imbalance causes it to slam shut
If you're wondering how often that happens, the answer is: all the fucking time
 
12:46 PM
stick a fork in it
 
@OliverSalzburg And it fukcing stays there half the time
Until I kill Skydrive.exe
 
Damn. Document scanners are expensive
 
lol
What brand printer?
 
@JourneymanGeek The ones I have right now? Both HP
 
@OliverSalzburg And btw what causes those smallish graphs there ?
 
12:56 PM
@HackToHell Process Hacker
 
I know canon and dell with STFU if you tell em to
 
This one is in blinking alarm mode again already
If you scan a page, it goes back into blink mode
If you confirm the alert, it waits 5 minutes, then goes back into alarm mode
Like, they have an exclamation mark shaped indicator on the front that is blinking constantly to annoy you so you'll buy ink
 
they need to make the money somewhere
 
1:00 PM
 
Oh whee cute icons
 
$100 printer, $75 dollers to replace the print carts.
 
The arrows aren't right, but you can see the exclamation mark
I'm watching an HP 9200C on Ebay now
That could be a nice addition to the setup
WTF @ price O___o
Oh yeah, that'll probably do then. If the price is right :P
 
a deal at twice the price.
 
shudders
 
1:04 PM
They go for ~300 on ebay
 
I have used HP multifuctional before. That was always a abd experience
starting with install times for 'the driver' which were 30-60 minutes on then modern hardware
As well failing to install most of the time. Even on a freshly installed OS
 
"because users can troubleshoot problems themselves" which is only 1/2 the fun
 
Troubleshoot answer #1: "it does not do anything!"
 
1:26 PM
Okay, great. It seems like that latest Apache2 update just broke our whole setup :P
Ah, 2.4 has some breaking changes
Nice
> Ubuntu and Debian: Virtual host configuration files in the sites-available directory must now use the .conf extension
 
1:53 PM
Ugh and apparently one of our clients websites is written with short PHP open tags :\
 
Bob
!!info
 
@Bob I awoke on Thu, 23 Apr 2015 18:44:27 GMT (that's about 2 days ago), got invoked 9 times, learned 96 commands, teleported 58 goats
 
Bob
!! s/short (.*?) open tags/$1/
!! s/awoke/died/
!!s/learned/forgot/
@allquixotic Cavil has amnesia...
 
2:09 PM
!!wiki something
 
Something is an indefinite pronoun. It may refer to: == Philosophy and language == Something (concept) Something, an English indefinite pronoun Existential quantification, in predicate logic, the predication of a property or relation to at least one member of the domain; described in layman's terms as "something" == Music == === Albums === Something (Chairlift album), 2012 Something (Lost and Found album), 2001 Something (Shirley Bassey album), 1970 Something (Shirley Scott album), 1970 Some Things, the debut album by Lasgo === Songs === "Something" (Beatles song), a 1969 song by Th...
 
!! s/a/b/
something with substitution
> Closure messed stuff up in minified eval.
and I'm using a closure-compiled version of the bot because I didn't understand why he got rid of closure
with SIMPLE_OPTIMIZATIONS, but still
 
Isn't the code optimized anyway by V8? Or whatever engine you're using
 
@OliverSalzburg PhantomJS uses WebKit JSCore, which is a lot slower than V8
!!info
 
@allquixotic I awoke on Sun, 26 Apr 2015 14:16:19 GMT (that's about 3 minutes ago), learned 96 commands
 
2:20 PM
!! s/Bob/Cob/
 
@allquixotic @Cob I awoke on Thu, 23 Apr 2015 18:44:27 GMT (that's about 2 days ago), got invoked 9 times, learned 96 commands, teleported 58 goats (source)
 
@Bob ^^
 
Greetings, SU!
 
Bob
@allquixotic nice
 
now it's only minified by uglifyjs2, not closure
 
Bob
2:21 PM
big difference?
 
there's a difference, but I wouldn't call it big
something like 10 KB of source text difference
(uglifyjs2 is bigger)
 
Bob
...that's some massive bot source
 
uglifyjs2 only removes whitespace and shortens variable and function names; closure's WHITESPACE_ONLY does the same thing
SIMPLE_OPTIMIZATIONS actually makes code changes that should result in identical behavior but the code is perhaps reordered or obvious things like always true if statements are removed
 
Bob
@HackToHell oh hey, I haven't seen that stoplight icon in a few years
 
2:26 PM
@allquixotic Ah, interesting
 
@HackToHell wow, Hubble's controllers use Windows 7; their software uses Java Swing; and they have Adobe Vulnerability Reader XI installed
I would wager however that those computers are not connected to the public Internet.
I'm a little disappointed that they aren't using something like Solaris or RHEL for the UI -- it's not like they need advanced 3D graphics or anything -- a little 2D card could easily drive Java2D/AWT
Windows 7 is not exactly the stablest of OSes. What if Aero causes the GPU to TDR during a critical point in the mission?
 
I was sorta expecting Linux
 
some of those boxes have the Windows icon for "Internet connection", too, which scares me
 
Hubble is older than me ._.
 
their font crosses their zeroes :D
 
Bob
2:35 PM
@allquixotic As far as stability goes, vanilla Win7 can be rock solid on the right hardware (read: without any dodgy drivers)
I'm more concerned about the hardware.
More likely, any actual commands are queued up on a server somewhere and this is just the UI.
 
@Bob I'm sure that's true
 
Bob
 
@Bob Yeah, but its correct functioning is not certified, is it?
The Evaluation Assurance Level (EAL1 through EAL7) of an IT product or system is a numerical grade assigned following the completion of a Common Criteria security evaluation, an international standard in effect since 1999. The increasing assurance levels reflect added assurance requirements that must be met to achieve Common Criteria certification. The intent of the higher levels is to provide higher confidence that the system's principal security features are reliably implemented. The EAL level does not measure the security of the system itself, it simply states at what level the system was tested...
Windows 7 isn't even listed; Windows 8 and 8.1 are "PP Compliant" which is weaker than even EAL1. commoncriteriaportal.org/products_OS.html#OS
SUSE, RHEL, Solaris and IBM z/OS are all EAL4+.
You'd think being able to send commands to such an expensive piece of equipment (which could also cause a slight problem on the ground if pieces of it were to reenter the atmosphere in the wrong place) would bring with it a desirable assurance level, if not for security then for stability.
a lot of EAL is about security, which wouldn't matter very much if they have no path to the Internet, but you can't be secure if shit is broken, usually, so it also affords a degree of stability if you know it's EAL certified to a fairly high level.
@Bob damn :/ Fox International Channels not allowing a fox to view a non-fatal ACI is just downright criminal
 
2:52 PM
@allquixotic: or centos 6.5 ;p
 
@JourneymanGeek no; specifically, RHEL 6.2 is EAL4+ certified.
 
Cause 6.6 is apparently too newfangled for anything important ;p
 
CentOS's modified background images and /etc/issue could be a security vulnerability.
Hey, you never know.
 
lol
(work insists on 6.5 and gnome. With annoys me. I prefer KDE)
 
 
2 hours later…
4:57 PM
3
A: Why are newer generations of processors faster at the same clock speed?

DragonLordDesigning a processor to deliver high performance is far more than just increasing the clock rate. There are numerous other ways to increase performance, enabled through Moore's law and instrumental to the design of modern processors. Clock rates can't increase indefinitely. At first glance, i...

I'm starting to venture into unfamiliar territory such as micro-operations.
 
 
1 hour later…
5:58 PM
There's something, let's say, magical about learning how modern processors work.
My answer is complete for the most part, but be sure to let me know what else may need to be covered or what corrections are needed.
This was the result of hours of painstaking research combined with nearly an entire semester's worth of graduate-level computer architecture content.
 
it actually had more info on CPU's than I previously had.
 
Most people see a CPU as a black box that executes instructions as efficiently as possible.
I've always wondered how CPUs work and how they're able to do so much work in every clock cycle.
 
Ditto.
 
I never thought I would go this far down the rabbit hole to explain modern CPU architecture.
And here I am, with a huge answer (even by Stack Exchange standards) explaining all this stuff.
 
I got as far as the cache levels vs speed. Pipelines, flushing pipelines. HT, communication between CPUs to other CPUs or to the north bridge or to memory. After that it used to be somewhat fuzzy
 
6:06 PM
Pipelines, reservation stations, branch prediction, reorder buffers, this stuff is freakin' complicated.
 
Lets add memory interleaving.
And how EDO works (and why the heck a 6502 already did sort of that thing at 980KHz)
 
This is also the culmination of my experiment with Thaddeus's answer format, to explore how well this format would scale to very long answers.
I think I could add more images...
 
6:45 PM
@DragonLord It's a really great post, and like Mokubai said, very nicely suited for a blog post, but I'd really appreciate if you could stop making trivial edits to the answer.
 
@slhck I'm trying to do edits in batch, but what constitutes trivial in an answer this long? Each edit is intended to cover more than a few words.
Is revision 20 trivial?
 
Take your time, edit it offline, then submit a new version. Like I said, it's already been flagged for too many owner edits.
 
So the problem is quantity and frequency of edits more than the size of each edit?
Just trying to clarify.
 
@DragonLord That's the main issue. It pushes the post to the front page every time. It wouldn't be such a big problem if there were significant changes. But for example, rev 19 is just two images, rev 20 I'd say is quite trivial. Rev 18 is a 1-character edit. Rev 13 is just 12 characters deleted. Those were all within a very small time frame. Would be nicer to just combine them.
 
The scale of the answer led me to take an unusual approach to writing it, building it out in stages rather than drafting a complete piece in one shot before submitting.
Revision 18 involved moving a paragraph of text from one section to another.
 
6:51 PM
Yes, I see. But the purpose of edits pushing posts to the front page is that there should be something new to see. Perhaps also something to review.
 
I'll try to reduce the quantity of edits in future answers. An answer this large with such specialized content poses special challenges in composition and maintenance, but I will keep this advice in mind.
My apologies if this is disruptive.
 
No worries. I know it's hard to write the "perfect" post when it's so long. Thanks for understanding!
 
7:20 PM
Anyone here who has experience with something called 'origin'
Can I install it, download the installer for games and then deinstall it?
And use the downloaded games whenever I want ?
Or is it too much like steam and annoyingly restrictive?
 
Not familiar with the Origin client, but I believe it needs to be kept open. Origin does allow for offline play, though.
 
Meh. In that case I will just ignore some of the games from the Humble Origin Bundle 2
I already registered most of those on steam and one client is more than enough.
 
 
2 hours later…
9:10 PM
Hernia (the portable XT) and core (i7-920) next to each other
 
@Hennes Origin is balls.
 
I should have realised that I needed to wait on GOG before buying games.
(I accept steam, but only grudgingly. I hate the idea that they can turn off my games at will.)
 
10:14 PM
@Hennes zat case.
 
@JourneymanGeek roar
 
base or acid ?
 
Have you seen my CPU architecture answer?
 
Yup
At least, this mornings version of it
 
The idea of an extremely long answer covering the topic in great detail was partly inspired by this answer:
8
A: What medium should be used for long term, high volume, data storage (archival)?

gaborous/EDIT: I will soon post a follow-up with more information I have gathered and better solutions. Meanwhile, you can check about error-correcting codes and file fixity. I also developped a set of opensource and simple python scripts (to be used in complement with other solutions like DVDisaster or ...

Gotta get back to work soon, but let me know what you think.
 
 
1 hour later…
11:42 PM
@JourneymanGeek I am aware that you like spam, here's some chat.meta.stackexchange.com/transcript/message/3363553#3363553
 
It's been a long time since I've written a blog post...
> Processors don't just execute one instruction at a time, they scan ahead and find out what instructions can be executed together without changing the result, and then try to run as many instructions as possible all at the same time. Some processors are smarter than others at doing this, and this is why some processors perform better than others even at the same clock rate.
 

« first day (1724 days earlier)      last day (3286 days later) »