« first day (1710 days earlier)      last day (3603 days later) » 

00:03
*psyco geek*, a customer just told us your review was helpful to them while
shopping on Amazon. Okkaaay where is my cut ? just send it to my paypal.
 
1 hour later…
01:21
13
Q: I'm but a speck of fairy dust

McMagisterA simple riddle of mine, focused towards meter and rhyme. I'm but a speck of fairy dust A glitt'ring jew'l, but only just So small I scarcely can be seen A trinity in me convenes And yet without me you can't see Fine art and words are made of me I'm resolute; through me you can...

Tangential but relevant: "I shrink along with passing years": We're at 577 ppi (Galaxy S5 LTE-A) and still going. — DragonLord Dec 13 '14 at 9:14
Now we're approaching 750 ppi: liliputing.com/2015/04/…
Pixel density has been getting more and more insane over these last few years.
overkill, for the phones, and the extra processing needed. for phones make the 10X battery first.
that is a cool puzzel riddle thing.
> I shrink along with passing years / My number grows, I gain more peers
Over the last five years, pixel density has been increasing at such a rate that there seems to be no ceiling.
I think we're finally starting to approach the technical limits of resolution and density as displays with resolution above 2160p aren't commercially viable just yet.
01:38
i am still trying to figure out how me video card is going to game at 4K , or how much it would cost to have it do so. although interpolation (with so many pixels) should be clean looking.
the phone chokes on a 3D rendered game with the 1920, given 4x more pixels it would have to be using a doubling method. Doubling was also used back when they went from 340 type reses to 640 type reses.
rendering the smaller res, then splatting that up on the high res, was little extra work, vrses actually rendering at the full res.
01:59
@Psycogeek heh. Wondering the same
@Psycogeek It's certainly doable with current technology—Maxwell was designed to cope with gaming up to 4K and beyond.
You would want to have a TITAN X (or several), though...
At US$1000 a pop, this is not something John Q. Public can afford.
I have avoided 2 video cards for quite some time. too many added issues, games that wont use em right. but i might be ok with a dual gpu on one board.
Supposedly a 970 can handle it.
Or pascal :p
@DragonLord good 4k monitors aren't cheap either. Maybe if you took the TV route....
 
1 hour later…
03:41
2 PBW in, now waiting a week to test data retention.
Background: UK-based networking consulting company subjects a 1 TB Samsung SSD 850 PRO to an extended endurance test.
When the drive finally stops working, I'm pretty sure Samsung will have something to brag about.
 
4 hours later…
Bob
Bob
07:27
@Psycogeek depends which game :P
The laptop with an 860M will be a "4k" screen
...I really wish they'd hurry up
hopefully I'll get it next week
downside of Clevo: ridiculous build times
 
1 hour later…
08:56
@BenRichards Remember I told you about a guy (stupid guy) who returned it to me claiming it is shutting down after 2 hours? So far I understand it works after you did a few initial tests? I prefer to gift it to you since I don't know what to do with it (I am outside of US) and as a way to thank you for your time.
09:20
@Bob a 660 is the minimum GPU nvidia says is 4k capable :p
I'll give it a try :p
Bob
Bob
10:13
@JourneymanGeek really depends what you're doing
I don't exactly play many brand-new AAA titles
@Bob: I don't usually either
Bob
Bob
though, I won't be trying that with Insurgency any time soon o.O
I barely get 60fps as it is (1920x1080, GTX 560)
and heavy smoke (in game) causes heavy smoke (real life)
Wonder if smoke is ever configurable, sometimes premium shadows are a deal breaker, ya just tweak them down to high or medium, shadows really assist in realism. The smoke :-) when it kills the frames that badly is overdone sometimes. Could be it qualifies to exist on a battlefeild during a bombing
anything else smoking that badly should be dead by now :-)
Bob
Bob
@Psycogeek FPS => smoke grenade => dense smoke is literally the intended purpose
they don't normally let you configure the smoke density, otherwise that could be unfair
10:27
Do new york city sewers have that much vapor comming out of them.
@Bob then solution that by putting you into "slow motion" when you get smoke grenaded, instead of horrid balking frames.
some 3d games i was checking (with frapps) would drop to 15FPS in a billow of vapor, when the rest of the frames were pulling a solid 60. Ya'da think they would have optomised that aspect of it.
Bob
Bob
@Psycogeek nah, smoke blocks things off visually
you're thinking more stun (not in this game)... I suppose flash has a similar effect but not as heavy
nearby explosions do slow you though
like i wasnt already having enough trouble seeing.
Bob
Bob
@Psycogeek turn off vsync
How did you know i usually have v-sync on?
My current monitor is pretty crap, so my 660 tends to run everything at pretty close to max settings ;p
10:34
I dont much see the point in rendering 135Frames when the monitor only shows 60 of them.
(while the gpu cooks)
Bob
Bob
@Psycogeek then restrict it in-game
vsync is not the right way to limit to 60
the problem with basic vsync is, it has to be an even fraction... if it can't make 60, it has to do 30 (55 not allowed), if it can't make 30 it has to do 15 (25 not allowed)
used to be games would tear pretty badly, not as many have that issue anymore.
Bob
Bob
at least that's what I vaguely recall, might be wrong o.O
triple-buffering helps IIRC
tripple buffering also seemed to delay the picture longer. another few MS before what was rendered makes it to the screen, and it rarely fixed horrid frame rates, it might balance a few, but balks in smoke are not a few frames going slow, it is all of em.
What is a "rail shooter" in referance to style of game, beyond being linear closed worlds ?
10:52
Man, I haven't heard that in ages.
Something like rebel assault 2
You aim and shoot, and pilot in some limited fashion, but you don't have full control
one of those Lame and shoot games :-)
youtube.com/watch?v=2DJT5EXARBk <--rebel assault 2 , or look how far we have come in 17 years :-)
I think i have actually played some version of star wars games similar to that.
11:16
Re newyork sewers: Sometimes: Yes
They still use steam via underground pipes
Those pipes are ancient. If they break and it gets into the sewer then you get a lot of 'smoke' from them.
but 1 out of 5 regular manholes , say on a really cold day , and the chineese laundry ? One wouldnt expect at night to walk the street and see many of them with a vapor comming out?
1 out of 5 is more than I expected
pic from newyork metropolitan art metmuseum.org
11:33
R.A. 2 seems a tad .. predictable.
Assault on a reactor...
Bob
Bob
11:46
@Hennes first thought: Red Alert 2
Ah.
Oh. 34:356. Lets do another reactor.
and again.
"Well, kid, it loos as if you saved us all. Again".
Never mind the NPC. She did not do anything ?
 
1 hour later…
13:17
@allquixotic: The server/vps I run ttrss on seems down. I'll give it a poke if I can find the keys (coulda sworn its on my laptop) but if the main thing is down, well, you might need to know.
ahh found it
Yeah, server seems down
Bob
Bob
13:42
!!info
@JourneymanGeek Probably :P (bot's hosted on same physical server)
speaking of - @allquixotic I seem to have forgotten to save the creds for the SU VPS :\
Also, I need to figure out a better way to store those than in the PuTTY list, cause that list isn't really portable
Hmm... PuTTY's CLI is alright... perhaps DIY one
13:54
lol
I was convinced it was on Athena, It was on nyx
I probably should set it up on phoebe
@JourneymanGeek damnit, I let the payment expire and OVH "suspended" the server (shut it down) :P now I need to wait for them to bring it back up
kept saying I need to pay, I need to pay
Happens ;p
good news is I'm rebooting into a new kernel version
and a new host OS environment with security fixes
thank god for PXE boot
13:59
XD, which means I probably should make sure the update script is running.
whoa - did it actually bring up the Cavil VM automatically so robustly that the bot just came back on its own?
!!info
0_0
so apparently they didn't shut it down, they just pulled the NIC
14:00:04 up 257 day(s), 4:10, 1 user, load average: 0.19, 0.14, 0.12
Looks like it
your ttrss should be reachable then
!!info
might have to restart Cavil from the looks of it
14:04
Thanks.
my fault :P at least I don't have to figure out the launch procedure to bring everything back up again on reboot
this is the type of box that you never reboot
because doing so is a catastrophe
cavil should be restarting soon
I'm pretty glad it didn't restart, because I'm having a major library linkage issue on my solaris container right now
it's not able to find the symbol ILLUMOS_0.8 in libc.so.1, but some libs depend on that symbol
(post-upgrade woes)
14:18
._.
Bob
Bob
@allquixotic heh
root@gaia:~# uptime
 14:18:28 up 140 days, 20:47,  2 users,  load average: 0.19, 0.06, 0.06
The last time that went down was due to DC issues.
Was a massive pain to bring it back up.
wait, two users? o.O
...oh, I had another session open
@Bob "Who else is on my server?!?!? D:"
"Oh. Me."
I'm pretty close to the point where other than my ISP being stupid, I can probably host most of the stuff I want to run myself ;p
Bob
Bob
@allquixotic The problem is, one of the VPSes on there... I think it was the Gitlab one? Well, the gitlab install autostart is stuffed cause my host is old deb init, guest is supposed to be systemd but is a container so it actually isn't, and the gitlab installer couldn't cope with that. Attempting to bring it up just tosses a bunch of errors in your face if you don't follow some precise sequence
I really should figure out how I managed to start it last time and script it
14:26
really ought to do periodic database dumps on puppy, but I rarely have issues ;p
and RSS feeds are mostly transient anyway, as long as I have the most recent opml feed list I'm good
Bob
Bob
hmmm
something's not right
@JourneymanGeek @Bob do you mind terribly if I do reboot my physical box in a minute here to try and fix some problems?
@allquixotic: not at all
I'll bring up all the VMs, including yours, once it's up again
Bob
Bob
@allquixotic Huh, I don't think I'm using your host
14:35
@Bob well, Cavil :P
Bob
Bob
51 mins ago, by Bob
speaking of - @allquixotic I seem to have forgotten to save the creds for the SU VPS :\
only ever used it for hacking on Cavil
ah okay
Bob
Bob
but lost it after the OVH move anyway
right now I'm trying to figure out networking issues on my WI host :P
> - SmartOS Live Image v0.147+ build: 20140515T211957Z
that's from before; let's see what it is after I reboot
heh. My brother is moving out in a few months. Thinking of seeing if I can talk him into hosting my brix, and paying him 50 dollars so we can get a static IP at his place ;p
14:36
why d'you need him to host?
@allquixotic: He wants to use myrepublic - they give you a static IP for a one off payment
ah, okay, nice
(and I can do quicky stuff there, pending either moving out, or switching to MR here)
Bob
Bob
ok, this is really weird
I can't make any outgoing connections from my host
14:38
Its really good value if 1) you want to self host. 2) Want to use them as an ISP anyway
not even http?
Bob
Bob
but the VPSes are fine
@allquixotic not even ICMP!
not even DNS!
> - SmartOS Live Image v0.147+ build: 20150320T034246Z
Bob
Bob
I think I stuffed up iptables at some point
but it's been ages
May 2014 -> March 2015 :D
14:39
(and it gives me an excuse to upgrade the home brix ;p)
0_0
Rebooted?
yup. free kernel upgrade thanks to PXE
cavil, puppy, and my personal zone are all "running" now
mine is going to take a few minutes to manually restart all my services ;p
OOH
rebooting fixed my problem with the library linkage! :D
Bob
Bob
Chain INPUT (policy ACCEPT)
target     prot opt source               destination
DROP       all  --  0.0.0.0/0            x.x.x.131
Hmmm
now why is that there...
but... the host uses x.x.x.130
.131 is for the VPSes o.O
apparently there are some files in SmartOS container zones that are "injected" into a zone automatically by the host, depending on the version of the host
so @JourneymanGeek your GNU libc (and other stuff) has been upgraded on puppy :D
whether you wanted it or not :P
now I need to remember how to run the upgrade script ;p
*update
pkgin up; pkgin fug
that'll upgrade all your packages
14:44
there we are
@allquixotic: oh, for ttrss
I need to run it in screen, as the www user and run a varient which IIRC you modded for smartos ;p
Bob
Bob
O_O WTF
the host seems to have picked up one of the guest private IPs as its default
how the...
@JourneymanGeek not sure what I modified O_O that was a while ago
lol
Won't work otherwise
want me to take a peek?
lol
its cool if you want to. I probably won't be touching ttrss until I get around to doing a full update
(which isn't as easy as it used to be)
Bob
Bob
14:54
oh god
Linux networking is an utter disaster
Installing windows, after chugging away with Almost Done on screen for a while, it now just has a blue screen, blank
Has been like this for 30 mins
Is this normal?
@fredley not in my experience, but what version of Windows?
8.1
I pressed start, and that seemed to work
Bob
Bob
ok
leap of faith: restarting networking while connected via SSH
15:02
Holy crapware batman
Bob
Bob
[ ok ] Deconfiguring network interfaces...done.
[....] Configuring network interfaces...
Hmm... I kinda expected it to disconnect
or maybe it did and I just didn't notice?
oh yup it's gone
Question: if you're using an SSD, does the same "keep a small part free" guideline apply as you have for HDD?
Bob
Bob
@NateKerkhofs I've never heard that guideline for HDDs.
damn
ok, the host is fine now but the vservers are all dead
fml
and it looks like that's for defrag purposes
so I think it's not needed for SSD
Bob
Bob
...
that's terrible and outdated advice
also, you don't want free space on a HDD, you want free space on the filesystem; unallocated space would be useless
Bob
Bob
you're right in that an SSD won't benefit much from defragging
Number 4 is "if you fill it up, write speed will plummet
Bob
Bob
there's actually a good reason for leaving free (unallocated) space on an SSD though: better wear levelling
but it's not a practical concern for most people
unless you're very write-heavy
I got an SSD for gaming. Is 16 GB on a 128 GB disk enough?
Bob
Bob
@NateKerkhofs Don't bother. Allocate everything.
Or you can leave a bit out and expand the partition later, if you want. Doesn't really matter.
I'd say it's not a practical concern, assuming TRIM support: if you need the space for data, it'll be used; otherwise it won't.
(If you don't have TRIM support then this doesn't apply and leaving unallocated space can actually help)
15:11
It's a 4 month old MSI gaming laptop, so I'm fairly sure it has TRIM
Bob
Bob
@NateKerkhofs TRIM support primarily depends on the OS, not the hardware.
Windows 8.1, so I think that's not an issue either
AFAIK, the only OS that doesn't support TRIM by now is Windows XP, and if you run that, you got bigger problems to worry about
Bob
Bob
O_O
@allquixotic gitlab somehow autostarted...
ok, I have no idea what's going on with this server anymore
CTX   PROC    VSZ    RSS  userTIME   sysTIME    UPTIME NAME
0      225 570.9M  33.1M   1h43m02   5h56m43 140d21h37 root server
40006   11   4.8G  19.5M  12m59s82  10m40s30 123d04h31 phoebe
40010    9 754.1M  16.2M  10m18s43  10m25s38 123d04h31 leto
40011    9 925.5M  24.2M  17m42s97   9m26s25 123d04h31 web
40012    6   1.3G  85.3M   6m59s74   2m14s13  54d02h19 blog
40014   11 910.5M 138.9M  25m13s28   3m31s21 123d04h34 firefox
40015   62     7G   1.2G   9h03m36   4h56m44 123d04h33 clio
that was before the networking restart, which apparently requires a restart of all guests
...speaking of, I should probably update the host while I'm at it... nah
you can see how my naming conventions are entirely inconsistent :P
you have one called firefox? O_o
Bob
Bob
@allquixotic That would be my ff-sync server :P
Bob
Bob
16:24
 
1 hour later…
Bob
Bob
17:28
huh, there's an Engineering
 
4 hours later…
21:51
If this is true, AMD will have a real monster for enthusiasts to play with (although this looks like a server chip).
16C/32T in four modules.
More likely for the consumer space is 8C/16T in two modules for high-end parts and 4C/8T in one module for mainstream parts.
DDR4 and HBM support.
Pretty sure what is described in the linked article is a server chip (notice that specs indicate ECC support which is normally found only on servers and workstations)
16C/32T is useless for 99% of consumers, after all.
Not a whole lot of games can even fully utilize a 4C/8T processor like the one in my laptop, after all.
Same thing with the Intel HEDT platform: the 8C/16T i7-5960X tends to perform worse than the faster-clocked, 6C/12T i7-5930K in a lot of games
More cores is not always better for consumers.
It's a significant advantage for servers where more cores are more efficient than faster cores for the nature of the applications they run, but the vast majority of consumers simply cannot benefit from more than 4C/8T in any meaningful way.
Most games, even modern ones, don't meaningfully scale beyond four threads.
On anything short of a highly-overclocked liquid-cooled build, I would strongly prefer the 6-core part.
23:01
@DragonLord Yeah. For algorithms that "people like us" use, you either need 1 or 2 high-throughput cores at most, or it's so embarrassingly parallel that you throw it on a GPU
@Bob Maine Coon! <3
23:22
Morning
heh. Someone asked where I worked in a comment ;p
They need an Office IT Dog, willing to travel :-)

« first day (1710 days earlier)      last day (3603 days later) »