Root Access

12:51 AM

@JourneymanGeek Some concerns over the hinge collapsing in a bag though

@Bob: I'm not so much. Laptops are actually pretty tough

and I'd stick it in a sleeve anyway

@JourneymanGeek Problem is the gap there when it's closed. Puts a lot more pressure on the hinge.

Granted, in a traditional laptop that pressure is spread over the screen, so this might be better for all I know :P

Though with these it might be less of an issue

aluminium unibody + glass.

1:10 AM

Well, do I really want to run a desktop on the same server that houses my website?

I'm removing Xfce once again.

I might maintain a separate instance for desktop stuff.

I don't know why you want to run a desktop remotely at all.

It's just an experiment.

I've left twm installed, but that's pretty much it.

I don't want to have too much software on my production web server.

1:38 AM

hm. G4 on vibrate is a lot quieter than my S4.

So I guess the Pebble was a good idea :P

Michael Frank

@Bob My Nexus 5 never seems to vibrate.

Make it really annoying when I go into DND mode.

@MichaelFrank Yea, the watch really helps.

Hard to ignore :P

I could hear my S4 vibrate.

Both good and bad.

2:41 AM

@Bob hits / (hits+misses)

@allquixotic I meant, how did you get "104,493,479 ARC hits, 10,376,665 ARC misses, 268,479 L2ARC hits and 10,108,126 L2ARC misses"

JakeGould

3:25 AM

So here is a question I want to post somewhere but don’t know how to do it or even if the topic might be covered. I am on Mac OS X 10.9.5 (Mavericks) right now and have avoided Mac OS X 10.10 (Yosemite). But now with Mac OS X 10.11 (El Capitan) out, I am wondering if upgrading to Mac OS X 10.10.5 (note the patch level) is considered a good path. I’ve heard tons of stuff from people who went to 10.10 but then wanted to go back to 10.9.5 early on. Is that still the case now? Is 10.10.5 stable now?

3:41 AM

@JakeGould You might get better responses over at Ask Different or their chat.

IIRC most regulars in this room don't frequently use OS X.

4:07 AM

I think I know why my X forwarding at home is so slow. The DSLAM (second hop on the traceroute) is really slow.

Ping statistics for 10.32.6.1:
    Packets: Sent = 34, Received = 34, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 35ms, Maximum = 63ms, Average = 46ms

Result of a quick ping test. Pinging the router itself averages less than 5 ms.

The path from local router to DSLAM is the #1 source of latency in the whole Internet route to my server.

@JakeGould: as a question it might be a tad bit opinion based.

(Yeah, I know, if its crap its crap, but a lot of these bugs are somewhat localised)

@DragonLord Keep in mind that it may simply consider ICMP responses very low priority.

and by it I mean the specific release.

tcping might be useful

(not the OS. I don't want to get into an OS religious war ;))

4:22 AM

@Bob Can't tcping the DSLAM, ports are closed.

There isn't much I can do about it.

I'm still hoping we can make the switch to cable internet soon.

(Our area has Time Warner Cable with Maxx upgrades completed.)

I have a DOCSIS 3.0 8x4 modem (Arris SURFboard SB6141) ready for installation when the time comes.

JakeGould

5:12 AM

@JourneymanGeek I know, I know. The reality is that I might just pull the trigger to get up to Yosemite at home. But might also be landing a gig where I might be using that on their office machine; so let that thing crash as I learn about it! ;) Just one of those things I wonder about. Scared about El Capitan and SIP stuff so wondering what my “oldest newest” personal Mac OS X might be in the long run.

Downgrading the main production Linode to free up funds for a new instance dedicated to cloud desktop usage.

after my last 30 mins of trying to get tomcat to accept a fucking file upload I'm about ready to smash my screen

this is infuriating

there's probably some obscure setting remaining in a config somewhere

damn thing is cutting the connection after 2MB, which is a documented default limit that I've already overridden

annnnnnnd I'm out of time. Guess I'll just have to do this demo with a limited dataset.

5:37 AM

@Bob maxPostSize?

@allquixotic: Just spun up a new instance. The main production server isn't being run to capacity so I've decided to downgrade it.

@Bob the kstat output

Hence, it'll still cost just $20 a month, but with full separation between the website and the experimental stuff I've been doing.

@JakeGould could always do what I did with win 10. Image then test.

@Bob reload/restart tomcat?

Two separate instances should give me greater flexibility and eliminate the need to worry about experimental stuff messing up my production server.

If I need more performance, I know I can always upgrade to a larger instance.

6:00 AM

https://www.youtube.com/watch?v=qH3lPzk_LVk&feature=youtu.be the video and audio are safe for work, but the title of the video contains a swearword, so sliiiightly NSFW - some impressive photoshop editing in time lapse

Just scaled down my main web server and started a new instance for running a "cloud desktop". The process was seamless. Thanks, @Linode!

Tweeted by fierydragonlord on October 8, 2015 at 6:01 AM

@allquixotic That was the one I changed.

Also double-checked the multipart-config stuff, which all looks right but does not work

probably some stupid thing to do with servlet spec versions

urk

6:31 AM

80ms ping, disproportionately slow downstream speeds. Not exactly great.

Is there an issue with your ISP?

@DragonLord Nah, I'm on mobile. Currently on a bus so it's a bit flakey.

Was there a glitch?

I've never seen download speeds so high on anything I own...

Must've been stopped in a particularly good spot for the second test.

Meanwhile git clone is proceeding at 20KiB/s :\

Installing LibreOffice on the cloud desktop server.

This is going to get interesting...

6:36 AM

s/interesting/slow/

It's isolated from the production web server.

I've spun up a dedicated instance for this stuff.

Adjusting the Xfce desktop to reduce the number of X draw calls (turning off shadows in the Xfwm compositor helps)

Keeping compositing proper enabled—this will increase performance because the full contents of each window are kept in memory, reducing the need to redraw windows.

This is almost certainly a bug in the "multiarch" patches to GCC. The directory /usr/include/x86_64-linux-gnu is being treated as a system include directory and should be included in the search-path list printed by gcc -v. I am not sure how someone managed to achieve that bug; if I remember correctly, the most obvious way to add system include directories does add them to what is printed by -v. (I wrote ~50% of GCC's preprocessor, but that was 15 years ago, so I may be misremembering something.) — zwol Oct 3 at 23:02

That's pretty impressive ;p

7:49 AM

I feel dumb

I JUST worked out one of the 4 USB ports on my monitor is always on

Quick question about the big toroidal coil on the left adjacent to the USB Standard-A jack.

@JourneymanGeek Even when power is disconnected from the monitor?

Is this a transformer or an inductor? The coil is marked L on the PCB which indicates an inductor; however, there's four leads and two differently-colored wires on that coil, which would suggest a transformer.

@DavidPostill: No, but when the monitor is off or on power saving mode.

@DragonLord: or two inductors on the same coil?

If this is a transformer, why are the wires interleaved on the coil?

7:52 AM

There's no use for a transformer in a DC/DC circuit

Ah ... I thought you had discovered free energy for a moment ;)

Q: What is the difference between a transformer and a coupled inductor?

I think I get it.

5

Transformers and coupled inductors seem very similar. Is there a difference in construction? Or only in use? This question asks something similar, but the answers don't address my question: Coupled inductor vs an actual transformer?

It's a coupled inductor.

@DavidPostill: On the other hand, that port is easier to reach, and you can charge a phone off it

That would explain the L on the PCB. A transformer is designated T.

or power something like a rpi or one of those stick PCs

Which would be nice for a low powered system for running long running tasks

7:55 AM

ok, need a second opinion for a PC build: docs.google.com/spreadsheets/d/…

what should be changed, if anything?

What do you intend to use it for?

gaming (for brother)

You might want to select a better SSD than this. (I'd suggest the Samsung SSD 850 PRO.)

Everything else about this build looks good.

Most of the issues I would find are matters of choice. I do like samsung SSDs given a choice.

@DragonLord that's additional €50

7:58 AM

I'd go for nvidia over AMD, but once again, matter of preference.

@tereško I don't see that big a price delta.

yeah :(
I am looking at local market

@tereško: gets you a MUCH longer warranty, better performance

Shouldn't be more than €25 difference.

also, I'd throw in a HDD!

SSDs are nice but bulk storage can end up being handy

7:59 AM

hdd is not mandatory .. he can buy it later

and there is actually enough space

yeah

at worst throw in a 1tb laptop drive.

I'm assuming you checked the power supply is adequate?

WD has high-capacity 2.5-inch Green drives with a 15mm z-height (enterprise form factor). These may be worthwhile (assuming it fits).

yes, PSU is compatible with the case and has enough juice

it probably would

Oh, I didn't consider the size, that one takes in a standard ATX...

Oh

checked the length of the GPU?

(if I'm nitpicking stuff like that, its not a bad build ;p)

the, GPU length is ok (though it would fail with Node 304, which is an alternative option for the case)

8:04 AM

as an nvidia fanboy, I'd suggest checking out the 960, or 970 as a possible alternative, and maaybe shop around a little on the ram

without onboard video or overclocking, ram choice is somewhat less than critical IMO

960 looks a lot less powerful compared to 380 .. and 970 is almost double the price

hm.

k

Nothing else major to nitpick at first glance ;p

mobo recommendation would be welcome :(

If you're not overclocking, that's probably fine

I'm running an asus z97 and no major complaints. Only issue I've had was fatpawing the processor, and damaging my first board, which I managed to get replaced.

@Psycogeek Any idea why some electric radiators make clicking noise? Delonghi are known for this.

Q: How can I stop my radiators "clicking" when they're turned on or off?

8:09 AM

The CPU on that system isn't going to support overclocking. GPU overclocking should always be considered an open possibility; a larger power supply may be needed for this.

1

I have a central heating system on a timer, and whenever the timer turns on or off, for around ten / fifteen minutes, some of the radiators in my home make regular clicking noises, kinda hollow-sounding, coming from inside. I've done some research and it seems that it's something to do with the ...

heating central-heating radiator noise-reduction water-circulating-heating

On a mini itx platform?

Even with a good case, its probably going to run a little warm

Gotta go in a few minutes. It's past 4 AM here in New York.

granted, cosairs typially are roomy

@JourneymanGeek the airflow is good according to stuff I read, that GPU is actually not bad for overclocking

@tereško: well, is serious gpu overclocking a primary consideration for that build?

8:12 AM

nope

it's more like added benefit

(I don't OC ever... so, once again, matter of taste)

.. and it not like there are only two options for overclocking: "none" and "over 9k"

@DragonLord It's central heating system. I asked about electric radiator.

@tereško if I totally need to nitpick...

realtek lan ;p

yeah :(

8:13 AM

(just noticed that)

but meh, its likely to be alright

do you even know a miniITX board with intel LAN?

nope

and something that's not €200+

I don't typically build mini ITX unless its SOC

Too small for my paws.

let me fire up pc parts picker

meh, there's no option for sorting by ethernet chip ;p

hmmm ... Gigabyte GA-Z97N-WIFI has an Intel LAN, but that's +€40

8:16 AM

er... check if that's the family with the bad VRMs

That's why I didn't go with gigabyte on the build

Otherwise, I'd recommend them, their desktop motherboards are tanky

well, it's also the Z97 board and those are all over €100

Its not a major thing

Realtek's have been entirely reliable for me other than one bloody pci/gig-e card

and most of the fun stuff with intel cards need 2 or more ports to do

Gigabyte GA-B85N looks like has Intel LAN too

much better selection of ports too

tho, ROFL at that single gold plated port for DACs

Yeah, that looks like a better option

agreed

8:32 AM

@JourneymanGeek DC/DC would use a transformer in a SMPS

@Bob: 0_0

I thought DC/DC conversion was typically solid state

Q: What makes the transformer in a SMPS smaller?

That's a transistor for the switching.

1

I'm reading up on how SMPS's work, and all the videos says that after the high-volrage chopping stage, a smaller transformer can now be used, because of this high frequency. My question is, how does higher switching frequency result in a smaller transofmer? In other words, what make switching 50...

transformer switch-mode-power-supply

@JourneymanGeek ^

From memory, an SMPS is basically a transistor switching very quickly and feeding that into a transformer input.

Add rectifier to get from AC to DC, add capacitors for smoothing

Further optimisations possible but that's it at the core.

an SMPS is typically for AC? ;p

Eh, the rectifier is not an integral part

anyway

Switched-mode power supply

8:45 AM

A switched-mode power supply (switching-mode power supply, switch-mode power supply, SMPS, or switcher) is an electronic power supply that incorporates a switching regulator to convert electrical power efficiently. Like other power supplies, an SMPS transfers power from a source, like mains power, to a load, such as a personal computer, while converting voltage and current characteristics. Unlike a linear power supply, the pass transistor of a switching-mode supply continually switches between low-dissipation, full-on and full-off states, and spends very little time in the high dissipation ...

> If the SMPS has an AC input, then the first stage is to convert the input to DC. This is called rectification. A SMPS with a DC input does not require this stage.

a good way to solve this would be to identify the chip and look up the datasheet ;p

Psycogeek

9:08 AM

@Boris_yo if it is a steam system , steam collapse powerfull stuff, air bubbles in the system, common from poor maintance, everthing expanding and contracting. temperature regulators switching. electric ones click mostly from the alum expanding contrating, and the themal switch does not make much noise. Compared to a fan blown space heater, a radiator type of heater should be less obnoxious. baseboard electric heaters are like almost silent :-) if you have heard the alternatives.

Like a big fat blower kicking in

@Bob and beat the ever living crud out of the way underneed capicitors in them. The wiki claims of efficiency are beloted too. there are almost as many losses as a magneitc linear transformer. The design of a good transformer style will live for 30 years, because the caps are just minor cleanup. in switching the caps are abused :-) and there is much loss in charging caps and heat comming off torrid coils and Mosfets (much like a transformers losses).

They also assemble switching PS with cheap capicitors usually, wheras the "switching" (Dc-Dc) that goes on on the motherboards they learned the hard way the need for extreeme MTBF capacitors.

I still hate the replacement of beastly live forever transformer PSes, as the limited life of the switching supply reveals itself in much of this cheap lightweight stuff.

While reading data sheets and doing the math, i am thinking "how the heck is that going to work" with switching of 1000htz and capacitors that have 1-10million lifetime. sounds like it would cope fine. the capacitors are "worked" (what?) 1.4million times a day.

general old school electronics also used 2X need parts, you could weld something with a car battery charger :-) today the mere act of shorting out the stuff , and you achieve elemental seperations. the blue smoke and the yellow flames :-) But who really cares? When i tell people here that crap is only going to work corrrect for 2-3 years, they say "That is a Loooonnnggg time" "it will all be obsolete by then" and "who cares it is cheap" a few years later the quote change to.

"What a piece of crap that was" and "this new stuff is soo much better"

9:55 AM

@Psycogeek Eh. Perhaps for the always-on stuff, but things like phone chargers that are sparingly used are usually fine.

@allquixotic It's sad how removable batteries are now advertised specs.

Psycogeek

@Bob but it is those and the lappie power packs that people are loosing to poor quality. meaning a switching PS could have extra goodies mitigating the low lifetime. i have lost more cheap USb wall chargers, while old beasly wall warts from the Last Century :-) are still working.

@Psycogeek Depends what you mean by cheap.

If you mean the really cheap ones with questionable safety and performance, yea, they're shit.

If you mean the cheap-ish good-quality ones? They tend to last a while.

Psycogeek

yea that kind of cheap, the it works today cheap, lets sell it, make the bean counters and the investors happy , , , today , cheap.

11:59 AM

in JavaScript on Stack Overflow Chat, 1 min ago, by Jan Dvorak

Have you considered a console instead?

well ... that's waht I get for asking opinions on my regular haunts

rofl

I hope we were of more help? :)

Oh and in many cases, stock cooling is good enough on intels

my core i7's been using the standard heatsink for the past... 2 years?

airy case tho

ChatBot John Cavil

I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.

in JavaScript on Stack Overflow Chat, 4 mins ago, by Jan Dvorak

@tereško Gaming consoles with specialised hardware and lightweight OS can easily outperform PCs with neither.

snort

Oh, yeah, and you can upgrade your console GPU when you need more power right? Oh waaaaitttt.

The ONE advantage a console has is standardised hardware to aim for.

I suspect I got trolled

yeah

12:06 PM

and there are actually few people who know HW stuff .. I probably just dumped the question in bad time

(I've never owned a console, and this is the one period in my life where I've got a great machine ;p)

hell .. most of europe is on dinner break or happily eaten

@tereško Really? In the UK dinner is an evening meal. Perhaps you meant Lunch? And although I enjoy eating I draw the line at being "happily eaten" ;)

well .. in Latvia the employed people eat some time between 12:00 and 14:00

Meals and Mealtimes in Britain - What Are They Called?

12:11 PM

@DavidPostill: I'm sad to see the brits have dropped the wonderful institution of "tiffin"

What time is dinner? A guide to meals in the UK

@JourneymanGeek LOL. That would be because we are no longer the British Empire ;)

(we still do it, oddly. Its a snack in the afternoon ;p)

@JourneymanGeek Makes sense. "In the British Raj when the local Indian custom of taking a light meal superseded the British practice of an afternoon tea, tiffin became the word used to describe this practice."

Tiffin

Tiffin is an Indian English word for a light midday meal (luncheon), When used for "lunch", it is not necessarily a light meal. == History == In the British Raj when the local Indian custom of taking a light meal superseded the British practice of an afternoon tea, tiffin became the word used to describe this practice. == Today == In South India and in Nepal, tiffin is generally an in between-meals snacks: dosas, idlis, etc. In other parts of India, such as Mumbai, the word mostly refers to a packed lunch of some sort. In Mumbai, it is often forwarded to them by dabbawalas, sometimes known as...

(and my folk are pretty anglophile in denial)

;p

India still has "tiffin-wallahs, or packed lunch boys" who deliver your food - "Every day, like the subaltern heroes in a James Joyce novel, some 4,000 tiffin-wallahs or packed lunch boys set off across Bombay's far-flung and verdant outer suburbs.

Parking their bicycles outside a succession of middle-class tower blocks, they collect up to 160,000 home-cooked lunches. They take the tiffins to suburban railway stations. There, they sort them out by destination on the platform. The lunches then travel southwards into the centre of Bombay. Here, the dabba-wallahs deliver them by 12.45pm sharp…

A Bombay lunchbox

12:22 PM

lol. I brought a semi modern version of one of those to work.

rofl

""Your stomach can get off if you eat outside food," Mrs Gavai pointed out sagaciously. "In restaurants you don't know what oil they use or how they prepare. You get infections." "

My mom would SO say that.

"Forbes awarded the humble dabba-wallahs a 6 Sigma performance rating, a term used in quality assurance if the percentage of correctness is 99.9999999 or more. In other words, for every six million tiffins delivered, only one fails to arrive. This error rate means in effect that a tiffin goes astray only once every two months.

It is a rare day indeed when a customer's deep-fried rotis fail to turn up. The sigma rating was the same as that given to the top bluechip company Motorola - not bad considering that most dabba-wallahs are illiterate."

right down to the bit about oil.

HackToHell

Tiffin is going strong here in india

1:09 PM

@JourneymanGeek Ha.

Said "specialised hardware" is literally underpowered versions of the same x86 chips and GPUs found in PCs.

Well specialised hardware was once true

Advantage of uniformity across devices, true. And relatively cheap.

But that also comes with a nice helping of lock-in.

yup

@JourneymanGeek once

There needs to be a once macro

in the case of the current generation of consoles, you don't even have the thing where folk super-optimise for that hardware

which was cool

since any optimisations you do would work on the pc and You likely licenced the engine.

1:17 PM

@JourneymanGeek Thing is, with the current generation of PC hardware even a moderately-optimised game could probably outperform a well-optimised game targeting older console hardware.

Standard PC hardware has made massive leaps in the last decade or so.

It takes hard work to mess up a game on the PC
(see arkham knight) ;p

GemsFromStackExchange: An error message. By the shore?

Keeping in mind that the older console gen was released when the P4 was still common, and C2D was brand new.

End of the day, some people find consoles easier.

But for a techy person, they don't offer objective advantages (apart from platform-specific titles)

GFSE

posted on October 08, 2015

An error message. By the shore?

@Bob: I can hook up a TV and a console controller to my PC :p

@JourneymanGeek Some people don't want to fiddle with the OS.

Though modern OSes tend to make it a bit easier.

1:21 PM

@Bob: If you were one of those people, you might not be here ;p

1:46 PM

Should I leverage Android table and smartphone as additional help for my MP3 conversion needs to speed-up conversion of albums?

...probably not

Unless you're converting an absolutely ridiculous amount, it doesn't take that long.

And unless you have a very powerful tablet/phone, it probably won't speed up the process by all that much.

lol

You'd have better luck using your GPU if you wanted to speed stuff up

@JourneymanGeek Nup.

MP3 encoding is not well suited to GPGPU.

@JourneymanGeek How about CASIO calculator? I am expecting 1 track for 5 minutes. At least it's something.

Now you're just being silly @Boris_yo

@Bob I distinctly remember some GPGPU encoding options in cuetools

1:52 PM

@JourneymanGeek Being silly is hiring someone on Fiverr to do MP3 conversion...

of course, I only encode to flac.

@JourneymanGeek Huh? Flac track?

@Boris_yo Put it this way. A Pentium 4 in 2002 could do a ~170MB wav in ~120 seconds. A modern i7 is >20x faster.

You'd be wasting more time transferring your files to and from your other devices than just going straight ahead.

If you had a 2TB HDD completely filled with uncompressed PCM data, it would take under 24hrs to encode to MP3 with a modern desktop i7.

I can't even find any benchmarks newer than 2002

Probably because it's so damn fast now that it's pretty pointless to measure.

I think the limiting factor is needing to swap disks

@JourneymanGeek Nah, still slower than sequential R/W of a HDD.

1:57 PM

@Bob: actual work

@JourneymanGeek Hm?

you need to be at the pc, and remember to swap the disks ;p

> it appears unlikely that MP3 encoding of a single file will be made much more parallel without significant resource expenditures

@JourneymanGeek Oh, CDs. I was thinking transcoding from files on disk.

cs.berkeley.edu/~agearh/cs267.sp10/hw0/results/files/…

2010 paper

and I'd guess an optical drive is slower than a hdd

Constraint was maintaining bit-identical output with LAME

1:59 PM

meh. Let it run, go the **** to sleep ;p

They could barely parallelise it usefully on x86 SMP/SMT.

GPGPU would be a much harder problem.

(and darn, its been a while since I've had to transcode anything)

Same.

and I tend to download and delete cuetools whenever I need to do that ;p

o.O

I don't know what I'd use now.

Maybe just Foobar's built-in encoder.

2:03 PM

(its a portable app, works well, but I keep finding I have lots of spare copies ;p)

2:21 PM

@Bob I am trying to cram as much music as possible in 4GB Sony Walkman. Will I feel difference between 128Kbps and 160Kbps? How about 128Kpbs and 192Kbps?

I am re-converting back from 320Kpbs to lesser rate using CDex.

@Boris_yo First question to ask yourself is, do you really need to cram all that in?

I don't know if it does direct compression. I think it decompresses to PCM on the fly and then re-converts.

192kbps is generally accepted as the point where all artefacts are inaudible to humans. Unless you have magic pebbles.

@Bob Yes. I like varying music. Different music and there's many tracks.

@Bob Thanks. Might just go with no less than 160Kbps.

160kbps, 128kbps... I probably would be fine listening to that. Don't know about you.

2:23 PM

@Bob I know @JourneymanGeek won't be fine with it...

@Boris_yo He keeps FLAC mostly for archival/reencoding purposes, not for the listening quality :P

At least, that's what I remember.

128 might be a tad low depending on your listening equipment.

Modern YouTube videos use 126k AAC, btw.

@Bob I only know about codec they use - H264 or something. As for their audio I thought it is MP4...

Well bit of both

I prefer lossless since space is relatively cheap and I'm pretty sure I can tell the difference.

@Boris_yo "MP4" is the container format H.264 video and AAC audio are often stored in.

@JourneymanGeek When's the last tine you double blinded yourself? :P

Well, I suppose single blind.

Meh.

Love this one:

"most people don't like to hear the truth,
so they elect politicians to hide it from them."

2:30 PM

@Bob: I'm fine with the possibility its psycological ;p

I'm not going to get annoyed at people with 128 kbps music, and occationally randomly listen to youtube music at unknown bitrate

But if I'm going to store stuff, I ought to store it at the best, most flexible format possible ;p

@JourneymanGeek YT is mostly 126k AAC, with some higher-bitrate Opus thrown in

@Bob: I'm pretty sure the source of some of the wierd old stuff that tuned up on a random playlist is VHS

@Bob I haven't asked about audio files with variable rate. Should I leave them alone? For no reason I always thought that those files will turn out not as good as audio files with normal bitrate.

hm

VBR is sensible

I just think variable rate audio files are not consistent.

2:34 PM

@Boris_yo Generally, VBR is just as good quality-wise as 320kbps, while averaging around 200kbps in terms of space.

Modern VBR encoders are very good.

you basically set a maximum bit rate, and it drops down only wjen needed.

@Bob And if I encode 228Kbps vBR to normal 128Kbps bitrate?

._.

@Boris_yo That "228" is probably the average.

Re-encoding is generally to be avoided.

2:35 PM

You'll reduce size by about 40% but lose some quality.

128 is around listenable but imperfect

You might be able to tell the difference in a side-by-side test.

But by itself it's usually fine.

No massive distortions, etc.

192 is generally considered identical to 320 for listening. worse for re-encoding.

@Bob In terms of quality will there be more degradation from re-encoding VBR than normal bitrate audio file? That's what I am wondering?

@Boris_yo Ah... I'm not sure. Probably not much.

@Bob I should hire @JourneymanGeek to tell the difference. Only he has superpowers.

@Boris_yo: Basically lossy audio does a few tricks to keep filesize down

At least I should avoid re-encoding to rate higher than audio's original i.e. no 128Kbps to 320Kbps...

2:38 PM

if you re-encode it, you might introduce artifacts.

@Boris_yo Some good info here => sound.stackexchange.com/questions/26222/…

Don't re-encode at all if possible.

@JourneymanGeek And if it first deconverts to PCM and then converts to target bitrate? It's not like it converts 228Kbps to 128Kpbs. It first deconverts to original source. Wouldn't that prevent majority of artifacts?

A: Are there any audible differences between 192 and 320 kbit/s .mp3 files?

@Boris_yo: THe changes are to the sound itself.

9

I find it comes down entirely to the type of music and the speakers you listen to it through: I can happily listen to pop music on an ipod through earphones at low bitrate (well, I say happily - not really a fan of pop music, but what I mean is it sounds as it should) If you play the same mp3 t...

I like this answer.

theverge.com/2015/2/19/8068923/… is entirely irrelevant but cool

@JourneymanGeek What kind of paranormal activity is in that video? I think I saw ghost and moving dragonfly...

Are they generated from unecessary things that VBR cuts out?

2:46 PM

essentially that's the sounds cut out when you convert a uncompressed wav into mp3

but when you reconvert, that process is run again

I just checked audio files for one of albums. The bitrate varies from file to file and it's most likely VBR. What I see is average or highest?

highest.

185Kbps, 153Kbps, 207Kbps, 212Kbps...

There's actually also an average bit rate option, but no one ueses that

@JourneymanGeek I think Bob told earlier it's average.

2:49 PM

nope

@JourneymanGeek I will round down/up to closest standard bitrate.

why?

@JourneymanGeek Because I want to re-encode to standard rates like 128Kbps, 160Kbps, 192Kbps...

No, that's what you want to do.

at the end of the day there's no real benefit to doing that