|
DZgas Ж
|
|
Meow
Oh it works
|
|
2024-10-25 12:28:51
|
what
|
|
2024-10-25 12:29:43
|
You literally did as before, not as I said
|
|
|
Meow
|
2024-10-25 01:01:36
|
The results are different on my Discord
|
|
2024-10-25 01:02:39
|
The earlier one is stretched
|
|
|
TheBigBadBoy - 𝙸𝚛
|
2024-10-25 02:22:11
|
yeah bc you only used one backstick
|
|
2024-10-25 02:23:29
|
also with triple-backstick you can mark the language of the code
```c
int main(void) { return 0;}
```\`\`\`c
int main(void) { return 0;}
\`\`\`
|
|
|
spider-mario
|
2024-10-26 10:50:30
|
https://youtu.be/FIk-8QkxCOg
|
|
|
_wb_
|
2024-10-28 08:09:38
|
ugh I upgraded to macOS 14.6.1 and now my Thorium stopped working
|
|
2024-10-28 08:12:12
|
some simple pages can still be loaded but anything a bit fancier (say github or gmail) causes it to crash completely, not just the tab but the whole browser
|
|
2024-10-28 08:20:34
|
```
[1028/091656.129226:WARNING:crash_report_exception_handler.cc(235)] UniversalExceptionRaise: (os/kern) failure (5)
1237024622016:error:1000007d:SSL routines:OPENSSL_internal:CERTIFICATE_VERIFY_FAILED:../../third_party/boringssl/src/ssl/handshake.cc:392:
[6658:12291:1028/091656.132803:ERROR:ssl_client_socket_impl.cc(878)] handshake failed; returned -1, SSL error code 1, net_error -3
Trace/BPT trap: 5
```
|
|
|
Fox Wizard
|
2024-10-28 08:29:30
|
Always fun when you look at clothing online and you get low quality images like this <:sadCowBoy:911197082944147476> https://res.cloudinary.com/teepublic/image/private/s--ev1hlUiz--/t_Resized%20Artwork/c_crop,x_10,y_10/c_fit,w_465/c_crop,g_north_west,h_626,w_470,x_2,y_-131/g_north_west,u_upload:v1462829015:production:blanks:mtl53ofohwq5goqjo9ke,x_-395,y_-456/b_rgb:eeeeee/c_limit,f_auto,h_630,q_auto:good:420,w_630/v1604618211/production/designs/15804478_0.jpg
|
|
2024-10-28 08:30:26
|
Hm, let's see how the fabric looks. Heh, seems like it's made out of webp compression artifacts <:KekDog:884736660376535040>
|
|
|
jonnyawsom3
|
2024-10-28 08:37:21
|
q_auto:good:420
Well at least you can edit the url
|
|
|
|
_wb_
|
2024-10-28 08:41:56
|
not in this case, they use signed urls
|
|
|
|
Fox Wizard
|
2024-10-28 08:42:20
|
Yeah, tried that and any editing causes a page not found error
|
|
|
|
_wb_
|
2024-10-28 08:51:04
|
anyway you can tell from the url that it's not a real photo anyway, they overlay the print design on top of an image of an empty t-shirt
|
|
|
|
jonnyawsom3
|
2024-10-28 09:02:47
|
Very common these days unfortunately
|
|
|
|
Fox Wizard
|
2024-10-28 09:18:42
|
They do that with both print design and t-shirt color which I don't mind. It looks accurate to the real product and makes it easy to compare products
|
|
2024-10-28 09:18:53
|
And they can't take real pictures anyways, because they print on demand
|
|
|
AccessViolation_
|
2024-10-28 02:48:14
|
I'm using Tor to access the internet in Turkey because they block Discord but now I can't read the JXL modular mode blog from Cloudinary which I saved up because Cloudinary blocks Tor <:PepeHands:808829977608323112>
|
|
|
|
spider-mario
|
2024-10-28 03:03:32
|
archive.is?
|
|
|
|
AccessViolation_
|
2024-10-28 03:26:45
|
I mean I can just open it in normal Firefox, cloudinary.com specifically isn't blocked here. I was just already using Tor when visiting the site and was sad to see that it didn't work (the last thing people circumventing censorship need is more blocks to get around)
|
|
|
|
Meow
|
2024-10-28 03:33:03
|
https://www.apple.com/newsroom/2024/10/apple-introduces-new-imac-supercharged-by-m4-and-apple-intelligence/
|
|
2024-10-28 03:33:13
|
https://www.apple.com/newsroom/2024/10/apple-intelligence-is-available-today-on-iphone-ipad-and-mac/
|
|
2024-10-28 03:33:33
|
An intelligent morning
|
|
|
|
AccessViolation_
|
2024-10-28 03:35:00
|
We finally get the virtual dumbass we've been promised
|
|
|
w
|
2024-10-28 03:44:36
|
cloudinary HATES privacy
|
|
|
|
Meow
|
2024-10-29 03:30:29
|
https://www.apple.com/newsroom/2024/10/apples-new-mac-mini-is-more-mighty-more-mini-and-built-for-apple-intelligence/
|
|
|
lonjil
|
2024-10-29 03:57:48
|
that's so tiny
|
|
|
|
spider-mario
|
2024-10-29 04:04:29
|
slightly reminds me of https://youtu.be/BGGOn-H7s3Q
|
|
|
|
Meow
|
2024-10-29 04:08:31
|
Possibly inspired by Hawking's input device
|
|
|
|
lonjil
|
2024-10-29 04:10:25
|
Oh, they put 16GB of RAM in the $600 base model! Finally
|
|
|
|
Meow
|
2024-10-29 04:13:48
|
And three Thunderbolt 5 ports for M4 Pro model
|
|
|
|
DZgas Ж
|
2024-10-29 05:47:06
|
M4 $600, M4 Pro — $1400
|
|
|
CrushedAsian255
|
|
AccessViolation_
I'm using Tor to access the internet in Turkey because they block Discord but now I can't read the JXL modular mode blog from Cloudinary which I saved up because Cloudinary blocks Tor <:PepeHands:808829977608323112>
|
|
2024-10-29 11:54:32
|
oh yeah i heard about that
|
|
|
|
Meow
|
|
DZgas Ж
M4 $600, M4 Pro — $1400
|
|
2024-10-30 02:48:02
|
As the M4 Pro one provides the performance possibly similar to M3 Max
|
|
|
|
DZgas Ж
|
2024-10-30 08:11:13
|
buy two M4 — 1200$ 👍
|
|
|
|
Fox Wizard
|
2024-10-30 02:25:39
|
Nice effort to hide a neighbor's package <:KekDog:884736660376535040>
|
|
|
|
Meow
|
2024-10-30 05:59:18
|
https://www.apple.com/macbook-pro/
|
|
2024-10-30 05:59:45
|
Some people should buy the M4 Max one and test how fast cjxl runs on it
|
|
|
|
afed
|
2024-10-30 08:10:00
|
https://www.tomshardware.com/tech-industry/artificial-intelligence/google-now-uses-ai-to-write-25-percent-of-its-new-code-alphabet-ceo-sundar-pichai-underlines-the-companys-role-in-the-ai-industry-amidst-strong-q3-24-financials
|
|
|
RaveSteel
|
2024-10-30 08:58:59
|
<@1259949633195737132> did you use an older build of ffmpeg? My testencode of Test1.mkv is only 100MB, yours is ~130MB
|
|
|
|
JKUser4592
|
|
RaveSteel
<@1259949633195737132> did you use an older build of ffmpeg? My testencode of Test1.mkv is only 100MB, yours is ~130MB
|
|
2024-10-30 10:07:30
|
My version is ffmpeg version 7.0-full_build-www.gyan.dev
|
|
|
|
DZgas Ж
|
|
afed
https://www.tomshardware.com/tech-industry/artificial-intelligence/google-now-uses-ai-to-write-25-percent-of-its-new-code-alphabet-ceo-sundar-pichai-underlines-the-companys-role-in-the-ai-industry-amidst-strong-q3-24-financials
|
|
2024-10-31 12:25:04
|
nah... i gen 99% of code <:Stonks:806137886726553651>
|
|
|
|
RaveSteel
|
2024-10-31 11:44:33
|
For my next PC i'll get 128GB RAM smh, those +1GB JXL images are not kidding around
|
|
|
|
DZgas Ж
|
2024-10-31 08:02:45
|
jxl bomb
|
|
|
Demiurge
|
2024-11-01 02:51:48
|
It should be theoretically possible to write a decoder that uses a fixed amount of memory. Ideally it would not allocate its own memory but leave memory allocation to the caller.
|
|
2024-11-01 02:54:04
|
That's basically the gold standard when writing safe decoders for the web
|
|
|
|
CrushedAsian255
|
|
Demiurge
It should be theoretically possible to write a decoder that uses a fixed amount of memory. Ideally it would not allocate its own memory but leave memory allocation to the caller.
|
|
2024-11-01 02:55:04
|
How much ram does it need max?
|
|
2024-11-01 02:55:18
|
Doesn’t it have to store previous frames for patches? Those can get arbitrarily big
|
|
|
|
Demiurge
|
2024-11-01 03:01:23
|
You would need a really intimate understanding of the format to know how much pre allocated working space a decoder would need, but I am pretty sure there is a theoretical limit to how much working space a decoder needs to decode a JXL image.
|
|
2024-11-01 03:02:58
|
And for multiple threads the memory requirement would obviously increase since each thread is basically independent.
|
|
|
|
CrushedAsian255
|
2024-11-01 03:04:55
|
the max would probably be something like 1 billion px width * 1 billion px height * 4 bytes * 4099 channels * 4 reference frames
|
|
2024-11-01 03:05:10
|
as it has to cache the reference frames for future patches
|
|
|
Tirr
|
2024-11-01 03:06:38
|
memory usage for decoding single frame is theoretically bounded. of course memory requirement goes larger linearly with multiple frames blended together, but jxl spec also defines max decoded pixels for a keyframe iirc, not sure though
|
|
|
|
CrushedAsian255
|
2024-11-01 03:06:52
|
i guess it could possibly seek backwards and re-decode the reference frames ....?
|
|
|
|
Tirr
|
2024-11-01 03:08:22
|
well now I think it might not grow even with multiple frames
|
|
|
|
Demiurge
|
2024-11-01 03:33:50
|
But the jxl format is chunked
|
|
2024-11-01 03:34:20
|
There's a theoretical limit to how much you actually need to have in memory at one time to decode
|
|
2024-11-01 03:34:35
|
Not including the output buffer
|
|
2024-11-01 03:39:35
|
Just the minimum you need to start getting pixel data
|
|
2024-11-01 03:40:06
|
in the worst case scenario jxl from hell
|
|
2024-11-01 03:41:30
|
So I guess the question can also be rephased, "what's the maximum amount of memory a supported/conformant jxl is allowed to require"
|
|
2024-11-01 03:42:12
|
there are complexity limits on how crazy you're allowed to get with blending multiple frames and layers
|
|
|
|
CrushedAsian255
|
|
Demiurge
So I guess the question can also be rephased, "what's the maximum amount of memory a supported/conformant jxl is allowed to require"
|
|
2024-11-01 03:43:27
|
do you mean a supported/comformant jxl *decoder* ?
the actual jxl could be a 1x1 image with a 16 EiB box of random data
|
|
|
|
Demiurge
|
2024-11-01 03:45:17
|
I mean a conformant decoder will consider the file to be corrupt if it surpasses certain complexity thresholds
|
|
|
|
CrushedAsian255
|
2024-11-01 04:39:52
|
I will always prefer headphone jacks and Apple, you can take the adapter that you made me buy and stick it up your butt
|
|
2024-11-01 04:40:13
|
The one that came with ny phone broke
|
|
|
VcSaJen
|
|
Demiurge
It should be theoretically possible to write a decoder that uses a fixed amount of memory. Ideally it would not allocate its own memory but leave memory allocation to the caller.
|
|
2024-11-01 04:48:10
|
You mean using a memory-mapped file?
|
|
|
|
Demiurge
|
|
VcSaJen
You mean using a memory-mapped file?
|
|
2024-11-01 05:43:09
|
I mean a lot of libraries are not allowed to allocate their own memory and you have to pass them a preallocated working space for them to use.
|
|
2024-11-01 05:43:40
|
Dynamic memory allocation is sometimes not allowed in certain situations.
|
|
2024-11-01 05:43:54
|
And it's nice to have a library that is usable in those situations
|
|
2024-11-01 05:47:41
|
Plus then you eliminate a whole class of possible failures by not having any memory allocation
|
|
|
|
_wb_
|
2024-11-01 09:48:37
|
In principle you could make a multi-pass decoder that stays in bounded memory by not storing full reference frames but decoding those on demand instead.
|
|
2024-11-01 09:50:10
|
We will probably at some point define a lightweight profile aimed at hardware implementations, that profile will not have patches so it will be easier to do things with bounded memory.
|
|
|
|
spider-mario
|
2024-11-01 10:04:31
|
|
|
2024-11-01 03:45:30
|
https://github.com/reactjs/react.dev/issues/3896
|
|
|
|
TheBigBadBoy - 𝙸𝚛
|
2024-11-01 07:54:00
|
old but gold
|
|
|
|
CrushedAsian255
|
2024-11-02 10:52:43
|
who didn't edit the paragraphs
https://www.nebula.llc/solutions
|
|
|
|
Demiurge
|
|
_wb_
We will probably at some point define a lightweight profile aimed at hardware implementations, that profile will not have patches so it will be easier to do things with bounded memory.
|
|
2024-11-02 10:10:08
|
Defining a profile without patches seems like a risky idea?
|
|
|
|
_wb_
|
2024-11-02 10:32:28
|
Its main use case would be hardware decoders that can only decode the result of hardware encoders. Basically for on-camera display of the photos taken on a camera. Software decoders would still support the full Main profile, but for hardware decoders that is too expressive to allow a reasonable implementation.
|
|
|
|
DZgas Ж
|
2024-11-02 11:23:07
|
What a funny meme**.**
|
|
|
|
spider-mario
|
2024-11-02 11:28:17
|
US is not YYYY-MM-DD, though, it’s MM-DD-YYYY
|
|
|
|
lonjil
|
2024-11-02 11:47:29
|
how often is YMD used in y'all's countries?
|
|
2024-11-02 11:48:57
|
Here in Sweden, DMY is still pretty common in hand writing, but YMD is the only national standard and is used for pretty much everything (alas, EU regs force DMY for some stuff)
|
|
|
|
Demiurge
|
2024-11-03 12:01:47
|
I really wish we didn't use stupid units like feet and miles and ounces and pints and gallons
|
|
|
Quackdoc
|
|
lonjil
how often is YMD used in y'all's countries?
|
|
2024-11-03 12:06:30
|
i've given up and knowing what date format something is
|
|
2024-11-03 12:06:37
|
if I can't tell at a glance I smile and nod
|
|
2024-11-03 12:07:13
|
I've worked with so many international people that I legitimately can't even tell what canada uses anymore [av1_kekw](https://cdn.discordapp.com/emojis/758892021191934033.webp?size=48&quality=lossless&name=av1_kekw)
|
|
|
Demiurge
I really wish we didn't use stupid units like feet and miles and ounces and pints and gallons
|
|
2024-11-03 12:08:31
|
I like both imperial and metric systems, a really weird "canadianism" is that a lot of people mix metric and imperial often in the same sentence it really confuses people we talk to [av1_kekw](https://cdn.discordapp.com/emojis/758892021191934033.webp?size=48&quality=lossless&name=av1_kekw)
|
|
|
dogelition
|
|
lonjil
how often is YMD used in y'all's countries?
|
|
2024-11-03 12:40:33
|
anything other than D.M.Y is extremely uncommon in austria, but when i fill in forms i usually use YYYY-MM-DD because it's based
|
|
|
|
w
|
2024-11-03 01:25:15
|
MDY is good in speech, dmy is just ridiculous
|
|
|
|
CrushedAsian255
|
2024-11-03 01:34:52
|
I use Y-M-D for everything even though its not norm where i am
|
|
|
w
MDY is good in speech, dmy is just ridiculous
|
|
2024-11-03 01:35:03
|
"Fourth of July"
|
|
|
|
w
|
2024-11-03 01:35:28
|
that's just dm
|
|
|
|
lonjil
|
2024-11-03 01:37:03
|
In British English, they speak DMY
|
|
|
|
w
|
2024-11-03 01:37:32
|
US is more metric than british
|
|
|
|
lonjil
|
2024-11-03 01:37:35
|
In Swedish, it's not even possible to use the MDY order in speech
|
|
|
|
Demiurge
|
2024-11-03 01:45:04
|
I like Fahrenheit
|
|
2024-11-03 01:45:28
|
But miles, ounces, gallons etc is stupid
|
|
|
|
CrushedAsian255
|
2024-11-03 01:55:42
|
I kelvin
|
|
|
|
dogelition
|
|
lonjil
In Swedish, it's not even possible to use the MDY order in speech
|
|
2024-11-03 02:10:01
|
same for german
|
|
|
|
VcSaJen
|
|
lonjil
how often is YMD used in y'all's countries?
|
|
2024-11-03 04:19:37
|
Japan, AFAIK
|
|
|
Quackdoc
i've given up and knowing what date format something is
|
|
2024-11-03 04:22:22
|
I wish they would standardize at least separators, so that it's always YYYY-MM-DD, MM/DD/YYYY, and DD.MM.YYYY. In reality separators are also mixed and matched, making it even more confusing.
|
|
|
Oleksii Matiash
|
|
w
MDY is good in speech, dmy is just ridiculous
|
|
2024-11-03 07:04:20
|
For non-English people mdy is ridiculous, at least for Slavonic
|
|
|
lonjil
In Swedish, it's not even possible to use the MDY order in speech
|
|
2024-11-03 07:06:03
|
Same for Slavonic, at least for eastern
|
|
|
|
Quackdoc
|
2024-11-03 07:07:19
|
why is it not possible? I only know english and really rust french
|
|
|
|
Oleksii Matiash
|
|
Quackdoc
why is it not possible? I only know english and really rust french
|
|
2024-11-03 07:13:43
|
I probably can't explain. It is not impossible actually, because order of words does not *really* matter for the sentence meaning here, but wrong order just sounds weird. Like Yoda. Normal is d (of) m (of) y, where (of) does not really exist, but 'encoded' in the d's, m's, y's endings\suffixes (not sure how it is called)
|
|
|
|
Quackdoc
|
2024-11-03 07:24:29
|
ah I see
|
|
|
|
spider-mario
|
2024-11-03 07:45:34
|
my preference is to spell out the month, which makes the order almost irrelevant
|
|
2024-11-03 07:46:03
|
23 Oct. 1975
Oct. 23rd, 1975
|
|
|
|
_wb_
|
2024-11-03 08:52:18
|
Oct. 31 is the same as Dec. 25 though, the old joke goes
|
|
2024-11-03 08:57:01
|
Here in Belgium we use dd/mm/yy(yy), though for archiving (like file or folder naming) it is also common to use yyyy-mm-dd since that is convenient for sorting.
|
|
2024-11-03 08:57:08
|
|
|
2024-11-03 08:57:17
|
(from https://en.m.wikipedia.org/wiki/Calendar_date#Date_format)
|
|
|
|
spider-mario
|
2024-11-03 08:58:32
|
truncated yy is the source of so many problems
|
|
|
|
_wb_
|
2024-11-03 08:59:08
|
Also interesting is how in some places the street number goes before the house number and in other places the house number goes first
|
|
|
|
spider-mario
|
2024-11-03 09:00:00
|
ah, yeah, that took some getting used to (France puts the number before the street, Switzerland puts the street before the number)
|
|
2024-11-03 09:00:20
|
17 rue Charles de Gaulle
vs.
Brandschenkestrasse 110
|
|
2024-11-03 09:03:05
|
also, Swiss streets can be very long
|
|
2024-11-03 09:03:16
|
Badenerstrasse in Zürich is about 5km long https://maps.app.goo.gl/aDMny2g3joaURVbr5
|
|
|
|
_wb_
|
|
spider-mario
ah, yeah, that took some getting used to (France puts the number before the street, Switzerland puts the street before the number)
|
|
2024-11-03 09:05:53
|
In the Dutch-speaking parts of Belgium we do it the swiss way, in the French-speaking parts the french way...
|
|
|
|
spider-mario
|
2024-11-03 09:10:05
|
oh, interesting feature of France: a street can cross city boundaries
https://maps.app.goo.gl/67QSJuJSc4AKvmUP6
|
|
2024-11-03 09:11:44
|
(or maybe Maps’ city boundaries are not quite accurate)
|
|
|
|
_wb_
|
2024-11-03 09:12:26
|
Streets can also cross country boundaries
|
|
|
|
spider-mario
|
2024-11-03 09:12:48
|
oh, it seems that the boundary actually coincides with the street https://maps.app.goo.gl/im1mxsLFEpqqTK6A6
|
|
2024-11-03 09:12:50
|
as in, runs along it
|
|
2024-11-03 09:13:13
|
that sounds very dubious
|
|
2024-11-03 09:13:21
|
Maps is probably wrong
|
|
|
|
_wb_
|
|
_wb_
In the Dutch-speaking parts of Belgium we do it the swiss way, in the French-speaking parts the french way...
|
|
2024-11-03 09:13:25
|
Actually I just checked and the official way in all of Belgium is the swiss way, though I am pretty sure I have seen french speaking people do it the french way too
|
|
|
|
spider-mario
|
2024-11-03 09:14:24
|
maybe I should knock on those people’s doors and ask them “excuse me, which city do you live in?”
|
|
|
|
_wb_
|
2024-11-03 09:14:56
|
https://maps.app.goo.gl/mtsYpYyBSQfVraFz6
|
|
2024-11-03 09:15:47
|
There are houses there which have a country border running through their house. They can have their bedroom in Belgium and kitchen in the Netherlands
|
|
|
|
DZgas Ж
|
|
DZgas Ж
What a funny meme**.**
|
|
2024-11-03 09:40:02
|
wow meme is real fun 😁
|
|
|
_wb_
|
|
2024-11-03 10:07:28
|
Hm... this is strange. and it is written in wikipedia in English and Russian. that Russia has 2 time standards. I decided to dig around and found out that this is not the case, not really. throughout Russia, without exception, the DD.MM.YY format (in docs DD.MM.YYYY) is used everywhere according to "GOST R 7.0.97-2016". however, "GOST R 7.0.64-2018" (same as ISO 8601:2004) has recently been introduced, which is made primarily for computer processing. but in reality it is not "used"
|
|
|
|
Quackdoc
|
2024-11-03 10:35:15
|
From now on, I'm just going to suffix everything. xD
1990y-08m-02d
|
|
|
|
lonjil
|
|
_wb_
|
|
2024-11-03 10:51:05
|
this graphic hides large differences in practice. Like Norway and Sweden are both listed as DMY + YMD, but in actuality, Norway's official format that is used for almost everything is DMY, with YMD only allowed for "technical purposes", while in Sweden YMD is the official format and DMY is only occasionally allowed.
|
|
|
_wb_
Also interesting is how in some places the street number goes before the house number and in other places the house number goes first
|
|
2024-11-03 10:51:54
|
and in some places, street numbers are counting numbers, while in other places, they're measurements of distance!
|
|
|
|
spider-mario
|
|
Quackdoc
From now on, I'm just going to suffix everything. xD
1990y-08m-02d
|
|
2024-11-03 10:54:12
|
or… prefix?
|
|
|
|
lonjil
|
|
lonjil
and in some places, street numbers are counting numbers, while in other places, they're measurements of distance!
|
|
2024-11-03 10:55:06
|
(Like the famous 666 N LSD in Chicago, which is located at the 666th foot along Lake Shore Drive. I have a friend who lived in what was apartment #13 on floor 13 in that building before they renumbered it.)
|
|
|
|
Quackdoc
|
|
spider-mario
or… prefix?
|
|
2024-11-03 11:13:18
|
yeah thats prolly better
|
|
|
|
Meow
|
2024-11-03 12:55:08
|
ISO 8601 uses YYYYMMDD or YYYY-MM-DD
|
|
|
|
spider-mario
|
2024-11-03 03:40:37
|
let's please have separators
|
|
2024-11-03 03:40:59
|
I beg you
|
|
|
|
lonjil
|
2024-11-03 03:48:22
|
RFC3339 wisely requires separators
|
|
|
|
VcSaJen
|
|
Oleksii Matiash
I probably can't explain. It is not impossible actually, because order of words does not *really* matter for the sentence meaning here, but wrong order just sounds weird. Like Yoda. Normal is d (of) m (of) y, where (of) does not really exist, but 'encoded' in the d's, m's, y's endings\suffixes (not sure how it is called)
|
|
2024-11-03 05:12:50
|
It's maybe possible if you separate it to several sentences. Like year in the first sentence, then month and date in another.
|
|
|
Quackdoc
From now on, I'm just going to suffix everything. xD
1990y-08m-02d
|
|
2024-11-03 05:19:25
|
For me it's weird that the full date in English (November 03, 2024) does not include "year" word or it's abbreviation.
|
|
|
|
Oleksii Matiash
|
|
VcSaJen
It's maybe possible if you separate it to several sentences. Like year in the first sentence, then month and date in another.
|
|
2024-11-03 05:25:45
|
Does not matter, if I got your idea. It is said like "what1 (<belongs to> what2)". where (<belongs to> what2) is actually month name with different ending. You can't say (<belongs to> what2) what1
|
|
|
|
VcSaJen
|
2024-11-03 05:28:18
|
I was thinking in the lines of "2024. November 3rd", meaning "It was year 2024. November third, to be specific". But I dunno, I'm not an expert in European languages.
|
|
|
|
Oleksii Matiash
|
2024-11-03 06:49:19
|
"November 3rd" this one is wrong, only "3rd November*th*"
|
|
|
|
spider-mario
|
2024-11-03 06:57:16
|
in German, you can use years as adverbs, pretty much
|
|
2024-11-03 06:57:29
|
“I was 2024 born.”
|
|
|
|
Oleksii Matiash
|
2024-11-03 08:16:08
|
Hmm, here it is like "I [am of] 2024th". What is in [] is meant, not spoken. But it does not sound the same as if I'm telling that I'm 2024th in queue, for example, different endings of "..four<>" used
|
|
|
|
jonnyawsom3
|
2024-11-03 09:06:04
|
Probably not too interesting as a Brit born and raised, but we can just say "I'm 2024" "when?" "November 3rd" or "3rd of November, 2024"
|
|
|
Cacodemon345
|
2024-11-04 08:51:36
|
https://youtu.be/kG92vmsYI54
The Most Mysterious Song on the Internet has been found.
|
|
|
|
Posi832
|
2024-11-04 08:56:42
|
This is so trippy and somewhat anxiety-inducing
|
|
|
|
CrushedAsian255
|
|
Cacodemon345
https://youtu.be/kG92vmsYI54
The Most Mysterious Song on the Internet has been found.
|
|
2024-11-04 09:09:45
|
…and an edit war ensues
https://en.m.wikipedia.org/w/index.php?title=The_Most_Mysterious_Song_on_the_Internet&action=history
|
|
|
|
Cacodemon345
|
2024-11-04 09:12:50
|
Wikipedia editors in full denial mode.
|
|
|
|
CrushedAsian255
|
2024-11-04 09:19:01
|
Let’s get the TMMS people on these cases
https://listverse.com/2020/02/03/10-baffling-disappearances-that-remain-unsolved/
|
|
|
|
DZgas Ж
|
|
|
TheBigBadBoy - 𝙸𝚛
|
2024-11-05 12:34:47
|
want a little episode?
here you are
https://media.discordapp.net/attachments/437863525965365249/1220931955961368626/RDT_20240323_0355203337994736527586312.gif
|
|
|
|
DZgas Ж
|
2024-11-05 03:40:47
|
https://camp2.rectangle.zone/index.php?title=Rotating_CSS_Tesseract_but_every_plane_is_a_Rotating_CSS_Cube still works
|
|
|
|
Quackdoc
|
2024-11-05 04:45:39
|
anyone know any good email clients for android that are Foss, webclient or not as long as its generic.
fairemail is too resource intensive and Mozilla destroyed k9
|
|
|
|
AccessViolation_
|
|
Quackdoc
anyone know any good email clients for android that are Foss, webclient or not as long as its generic.
fairemail is too resource intensive and Mozilla destroyed k9
|
|
2024-11-05 06:59:32
|
Thunderbird for Android is now available
|
|
2024-11-05 07:00:24
|
I haven't tested it myself, though, I just know it was recently released
|
|
|
|
Quackdoc
|
|
AccessViolation_
Thunderbird for Android is now available
|
|
2024-11-05 11:14:33
|
that is what I am reffering to, Thunderbird is the new k9 and its horrid
|
|
|
|
lonjil
|
2024-11-05 11:17:35
|
but they just renamed K9 to Thunderbird, it's still the same app
|
|
|
|
Quackdoc
|
2024-11-05 11:21:23
|
and destroyed the UI
|
|
|
|
lonjil
|
2024-11-05 11:25:51
|
in most respects it's nearly identical though?
|
|
|
|
Quackdoc
|
2024-11-05 11:42:32
|
they completely broken one handed use since now the tray is way worse
|
|
2024-11-05 11:42:41
|
drawer
|
|
|
|
lonjil
|
2024-11-05 11:57:19
|
I only ever use one hand in K9 and I can't notice a difference in that regard
|
|
|
|
jonnyawsom3
|
2024-11-06 06:52:27
|
https://youtu.be/-Ln-8QM8KhQ
|
|
|
|
_wb_
|
2024-11-11 08:37:48
|
https://x.com/rob_heighton/status/1855552740706931111?t=8GSOdVdzejHVuekX_lYrRA&s=19
|
|
2024-11-11 08:38:24
|
https://x.com/rebane2001/status/1544774351777665026?t=mqLr5kNAUCeVRVxbZkmVug&s=19
|
|
2024-11-11 08:40:02
|
That's pretty weird downscaling
|
|
2024-11-11 08:42:45
|
Or is it just nearest neighbor?
|
|
|
|
spider-mario
|
2024-11-11 09:41:22
|
https://youtu.be/040ejWnFkj0
|
|
|
|
AccessViolation_
|
2024-11-11 10:08:13
|
Oh does Cloudinary support Tor now? 👀
|
|
2024-11-11 10:10:08
|
Oh nope, I guess that particular exit node wasn't added to the blocklist yet
|
|
|
|
jonnyawsom3
|
|
_wb_
Or is it just nearest neighbor?
|
|
2024-11-11 10:57:56
|
Looking at the comments it's just a 2x2 average
|
|
|
|
_wb_
|
2024-11-11 10:58:23
|
How does that turn yellow into green?
|
|
|
|
CrushedAsian255
|
2024-11-11 12:29:10
|
Chroma subsampling?
|
|
|
|
jonnyawsom3
|
2024-11-11 12:32:27
|
No, it's serving palette PNGs
|
|
2024-11-11 12:32:29
|
https://x.com/TikkaQrow/status/1855734542457516466
|
|
|
|
_wb_
|
2024-11-11 01:50:45
|
oh I think I understand what happens. The original image is a palette image with this palette:
|
|
2024-11-11 01:53:20
|
it downscales the image in full color and then uses the same palette again. The yellow lines that become darker due to downscaling then will use the green instead, since it is the nearest available palette color...
|
|
|
|
CrushedAsian255
|
2024-11-11 01:54:37
|
Wonder why it tries to use palette again for the thumbnail
|
|
|
|
_wb_
|
2024-11-11 01:56:21
|
probably because going from palette to full color is too likely to make the thumbnail file larger than the original
|
|
2024-11-11 01:56:22
|
https://x.com/jonsneyers/status/1855972492793229517
|
|
2024-11-11 01:57:01
|
my hypothesis seems to check out. I changed the green in the palette to purple and now Twitter does this:
|
|
|
|
jonnyawsom3
|
|
_wb_
https://x.com/jonsneyers/status/1855972492793229517
|
|
2024-11-11 02:00:01
|
And clicking on the embed makes the Discord preview cream. Strange
|
|
|
Kleis Auke
|
|
_wb_
https://x.com/jonsneyers/status/1855972492793229517
|
|
2024-11-11 03:41:49
|
`name=thumb` is even more curious - https://pbs.twimg.com/media/GcG9GcrXIAAJlFl?format=png&name=thumb
|
|
2024-11-11 03:52:06
|
If Twitter supports custom resize dimensions, you could test it against <https://github.com/jsummers/resamplescope> to determine which algorithm it uses.
|
|
2024-11-11 04:37:47
|
It looks like `name=thumb` is broken anyway, e.g. resizing [this image](https://user-images.githubusercontent.com/128982/71692048-f7558c80-2db9-11ea-956e-a8cc6e4bbfc0.png) to 150x150 would result in: https://pbs.twimg.com/media/GcHZot1XUAA9xUc?format=png&name=thumb (notice the difference in the last column).
|
|
|
|
DZgas Ж
|
2024-11-12 12:06:00
|
archive org - 2 years still can not open this page https://archive.org/details/gacha-life-archive-by-dzgas <:galaxybrain:821831336372338729>
|
|
|
|
CrushedAsian255
|
2024-11-12 12:13:27
|
maybe thats what caused the internet archive ddos
|
|
|
|
DZgas Ж
|
2024-11-13 12:40:57
|
||python can not break out of multiple loops||
|
|
|
|
AccessViolation_
|
2024-11-13 01:40:41
|
If you use a VPN you might like Mullvad Browser (can be used with any VPN, not just Mullvad). Mullvad commissioned the Tor Project to separate the Tor logic from the rest of the browser, so you have the fingerprinting resistance and other security features of Tor Browser without the Tor
|
|
2024-11-13 01:41:54
|
But yeah I agree, the captchas are annoying
|
|
2024-11-13 01:44:35
|
Some like hCaptcha do not discriminate between normal browsers and Tor Browser, while reCAPTCHA will immediately give a negative weight to your request if it detects you're using Tor Browser, and will also often decide "you are sending automated requests" and won't even let you attempt to solve a captcha to prove the contrary, while Cloudflare's Turnstile is just bad and will sometimes go into "bot fight mode" and waste your CPU until you close the tab
|
|
|
|
DZgas Ж
|
2024-11-13 01:44:55
|
internet must die now
|
|
2024-11-13 01:45:52
|
real https://en.wikipedia.org/wiki/Enshittification
|
|
2024-11-13 01:46:25
|
new term
new era
|
|
|
|
AccessViolation_
|
2024-11-13 01:47:37
|
Soon only Genuine iPhone® Devices with attestation will allow you to make web requests
|
|
|
|
DZgas Ж
|
2024-11-13 01:49:00
|
no tor
no vpn
no anonymity
do register
Don't forget write your number
and be sure to show your passport
|
|
|
|
AccessViolation_
|
|
AccessViolation_
Soon only Genuine iPhone® Devices with attestation will allow you to make web requests
|
|
2024-11-13 01:52:20
|
This is not a joke btw, Cloudflare's solution to "our bot detection is ass and we will show captcha windows that are only partly inside the viewport in mobile apps" was to basically create a new wonderful DRM system
|
|
|
|
DZgas Ж
|
|
DZgas Ж
real https://en.wikipedia.org/wiki/Enshittification
|
|
2024-11-13 01:53:39
|
hey say there are a lot of bots on the site.
in fact, they have made the site so overloaded and complex that it cannot handle more users. at the same time, they are all so greedy. all the data is publicly available, but when you say: oh, let me download it all! — nope It's just for us, ours, of course it's all open and accessible, but no, we won't give you anything.
|
|
|
|
AccessViolation_
|
2024-11-13 01:55:21
|
Yeah services really don't like it when you use them
|
|
|
|
DZgas Ж
|
2024-11-13 01:58:25
|
All YouTube videos are available, but you can't get a list of at least 1,000 of them. there is no search tool for this. database is full close.
do you want to index the Internet? you're late, you won't be able to open any website with a bot, because of cloudflare
a fully open archiveorg file library. Everything is free. Everything is open. by the way, you can't download all of it. because. (If we die you can do nothing👍 )
|
|
2024-11-13 02:04:25
|
But
You can download illegal gog full games dump <:PepeOK:805388754545934396> and all books and films, For all the time, use torrent illegal ( full illegal, according to the laws, literally 500 thousand life sentences )
|
|
|
diskorduser
|
2024-11-13 07:29:13
|
https://fxtwitter.com/Locati0ns/status/1856325749252272604
|
|
|
|
CrushedAsian255
|
2024-11-13 10:53:59
|
I’m learning how to use Amazon s3 api (well technically Backblaze b2)
|
|
|
|
Quackdoc
|
2024-11-13 01:39:35
|
whats the fastest way to generate a pixel hash of an image? currently I am using `magick identify -format "%# "` ffmpeg has a image res limit so that's not an option
|
|
|
Kremzli
|
|
diskorduser
https://fxtwitter.com/Locati0ns/status/1856325749252272604
|
|
2024-11-13 01:55:26
|
whats bad about that
|
|
|
A homosapien
|
2024-11-13 01:59:06
|
is the joke that France hates Belgium?
|
|
|
|
Quackdoc
|
|
Kremzli
whats bad about that
|
|
2024-11-13 02:01:58
|
a lot of people have a strong and irrational hatred of nuclear power
|
|
|
|
CrushedAsian255
|
|
Quackdoc
a lot of people have a strong and irrational hatred of nuclear power
|
|
2024-11-13 02:04:47
|
I love nuclear power as long as I don’t have to eat the nuclear waste, that stuff tastes bad 🤮
|
|
|
|
Quackdoc
|
2024-11-13 02:05:26
|
understandable lol
|
|
|
|
Kremzli
|
2024-11-13 02:05:38
|
you can lick the containers ngl
|
|
|
|
CrushedAsian255
|
|
Kremzli
you can lick the containers ngl
|
|
2024-11-13 02:05:55
|
I guess depends on the brand
|
|
|
|
Kremzli
|
2024-11-13 02:07:35
|
https://youtu.be/lhHHbgIy9jU
|
|
|
|
CrushedAsian255
|
|
Kremzli
https://youtu.be/lhHHbgIy9jU
|
|
2024-11-13 02:08:34
|
Anyone else here Nuclear-waste-sexual?
|
|
|
|
Kremzli
|
2024-11-13 02:09:08
|
Chris Hemsworth proves nuclear is safe
|
|
|
|
CrushedAsian255
|
2024-11-13 02:09:53
|
“But.. but.. but.. Chernobyl!!!”
Planes crash, we still fly on planes.
|
|
|
|
Kremzli
|
2024-11-13 02:10:41
|
When the fossil fuel reactors kill thousands daily its fine but one big boom killing like double digit amount of people or none thats too scary
|
|
|
|
Quackdoc
|
|
CrushedAsian255
“But.. but.. but.. Chernobyl!!!”
Planes crash, we still fly on planes.
|
|
2024-11-13 02:12:04
|
chernobyl was a massive collection of bad design and bad decisions, it was peak "ah yes, soviets" moment
|
|
|
|
Kremzli
|
2024-11-13 02:13:15
|
People also need to know about how waste storage is handled especially compared to other solutions, and how they can be recycled. How the green stuff is not going to fix things quick enough and their problems compared to nuclear. And just nuclear safety features in general. Its pretty interesting anyway, if someone can be interested in computers then nuclear is easy to be interested in especially in a format like Kyle Hill videos
|
|
|
|
spider-mario
|
|
CrushedAsian255
“But.. but.. but.. Chernobyl!!!”
Planes crash, we still fly on planes.
|
|
2024-11-13 02:43:03
|
incidentally, per https://xkcd.com/radiation/, flying on a plane from New York to Los Angeles exposes you to about 400 as much ionizing radiation as living for a year within 50 miles (80km) of a nuclear plant
|
|
|
|
Oleksii Matiash
|
|
CrushedAsian255
“But.. but.. but.. Chernobyl!!!”
Planes crash, we still fly on planes.
|
|
2024-11-13 02:46:27
|
Correct spelling is Ch**o**rnobyl, btw
|
|
|
|
spider-mario
|
2024-11-13 02:48:11
|
isn’t it Чорнобиль?
|
|
2024-11-13 02:48:25
|
(😉)
|
|
|
|
Oleksii Matiash
|
2024-11-13 02:48:36
|
Exactly
|
|
|
|
lonjil
|
2024-11-13 02:50:17
|
I hate that HBO miniseries for all the lies it told
|
|
|
|
paperboyo
|
|
lonjil
I hate that HBO miniseries for all the lies it told
|
|
2024-11-13 04:23:15
|
Do you mean [these](https://www.newyorker.com/news/our-columnists/what-hbos-chernobyl-got-right-and-what-it-got-terribly-wrong) kind of lies or [those](https://cancerletter.com/series/chernobyl/)? Despite both being a great and informative read, I still love the series. And Alexievich’s book too.
Pedantic (like about AVIF pears being clearly so much better): so it’s `Chornobyl` for a place and `Chernobyl` for the series – makes sense. But what is it for the NPP’s name itself?
|
|
|
|
Oleksii Matiash
|
|
paperboyo
Do you mean [these](https://www.newyorker.com/news/our-columnists/what-hbos-chernobyl-got-right-and-what-it-got-terribly-wrong) kind of lies or [those](https://cancerletter.com/series/chernobyl/)? Despite both being a great and informative read, I still love the series. And Alexievich’s book too.
Pedantic (like about AVIF pears being clearly so much better): so it’s `Chornobyl` for a place and `Chernobyl` for the series – makes sense. But what is it for the NPP’s name itself?
|
|
2024-11-13 04:56:53
|
It depends on language. Chornobyl is Ukrainian spelling and pronounce, Chernobyl is russian. As plant is located in Ukraine, correct is Chornobyl nuclear power plant or "Chornobylska atomna elektrostantsiia" if transliterated from Ukrainian
|
|
|
|
paperboyo
|
2024-11-13 05:08:03
|
🇺🇦 🙂
I say `a`. `Czarnobyl`. Been in Żarnowiec this summer. Not as cool, but cool.
|
|
|
|
jonnyawsom3
|
2024-11-13 05:14:19
|
I'd love to do a bit of urban exploration, but most places in the UK are either redeveloped or so rural they're inaccessible now
|
|
|
|
Oleksii Matiash
|
|
paperboyo
🇺🇦 🙂
I say `a`. `Czarnobyl`. Been in Żarnowiec this summer. Not as cool, but cool.
|
|
2024-11-13 05:21:18
|
Quite expected, czharny = chornyj 🙂
|
|
|
|
_wb_
|
2024-11-13 08:48:07
|
So why did France put that nuclear power plant in that location? It would make more sense to put it somewhere close to where the power is needed, no? Rather than in a pretty remote part of the country...
|
|
|
|
jonnyawsom3
|
2024-11-13 08:54:17
|
> Chooz A was an early pressurized water reactor (PWR) designed by Westinghouse , built and operated by French ( EDF ) and Belgian (SENA) grid operators.
Guess it makes sense
|
|
|
|
Quackdoc
|
|
_wb_
So why did France put that nuclear power plant in that location? It would make more sense to put it somewhere close to where the power is needed, no? Rather than in a pretty remote part of the country...
|
|
2024-11-13 09:02:42
|
france has a good number of reactors, in general, so this reactor is probably dedicated for exports
|
|
|
|
spider-mario
|
2024-11-13 11:20:27
|
relevant website: https://app.electricitymaps.com/
|
|
2024-11-14 09:35:48
|
https://en.wikipedia.org/wiki/File:En-us-sixths.ogg
stop doing English
|
|
|
|
Kremzli
|
2024-11-14 12:43:18
|
https://proton.me/blog/nl-dk-lu-politicians-exposed
|
|
|
|
A homosapien
|
2024-11-15 12:49:29
|
New oscilloscope music album just dropped!! 🔥
https://www.youtube.com/watch?v=BDERfRP2GI0
|
|
|
|
DZgas Ж
|
2024-11-16 11:05:44
|
generated
|
|
|
|
diskorduser
|
2024-11-17 01:45:50
|
https://www.threads.net/@ishowcybersecurity/post/DCcp0nhzis_?xmt=AQGz2Z4s5eP7bM_erLbHtD-bmfsG9Jg7c7eFiSezJoMWrQ
|
|
|
|
_wb_
|
2024-11-17 06:43:09
|
https://bsky.app/profile/rainda.bsky.social/post/3layl7ihvp226
|
|
|
|
spider-mario
|
2024-11-17 10:20:40
|
still a classic
https://youtu.be/jcZUPDMXzJ8
|
|
|
|
lonjil
|
2024-11-18 02:41:01
|
When I move to the US, I'll probably need a gun for self defence. Does this look reasonable as an every day carry weapon?
|
|
|
|
Quackdoc
|
|
lonjil
When I move to the US, I'll probably need a gun for self defence. Does this look reasonable as an every day carry weapon?
|
|
2024-11-18 02:43:31
|
now that is some nice taste [av1_perfect](https://cdn.discordapp.com/emojis/842099338904600599.webp?size=48&name=av1_perfect) the ammo for that thing tho is pure pain for the wallet
|
|
|
|
lonjil
|
2024-11-18 02:44:31
|
They haven't even started selling the .277 Fury variant yet :(
|
|
|
|
Quackdoc
|
2024-11-18 02:47:56
|
pain, I was looking into buying seeing what it would cost to buy a 277 rifle, but importing it seems to be the only option, and the costs to import rifles rn are absurd here
|
|
2024-11-18 02:48:10
|
the canada tax is pain
|
|
|
|
lonjil
|
2024-11-18 02:53:23
|
I think the only .277 Fury rifle currently available is the SIG Sauer CROSS bolt action
|
|
|
|
Quackdoc
|
2024-11-18 02:56:27
|
yup, and it's pretty much out of stock everywhere lol
|
|
2024-11-18 02:57:02
|
no chance I would ever be able to afford the spear lol
|
|
|
|
lonjil
|
2024-11-18 02:57:19
|
hey, look on the bright side, if you get the 7.61×51mm NATO variant of the Spear, you'll save on ammo!
|
|
|
|
Quackdoc
|
2024-11-18 02:58:35
|
[av1_dogelol](https://cdn.discordapp.com/emojis/867794291652558888.webp?size=48&name=av1_dogelol)
|
|
2024-11-18 02:59:10
|
well if it doesn't become illegal, my next firearm is going to be an mp40 because ar15 is banned and an mp40 isn't..... yeah...
|
|
|
|
lonjil
|
2024-11-18 02:59:46
|
lmao the ar15 is banned?
|
|
2024-11-18 03:00:14
|
Here in Sweden we recently concluded that AR15 style stuff is totally reasonable for civilians
|
|
|
|
Quackdoc
|
2024-11-18 03:00:21
|
yeah, the AR10, AR15 are banned here. not the AR180, but those are stupid priced
|
|
|
lonjil
Here in Sweden we recently concluded that AR15 style stuff is totally reasonable for civilians
|
|
2024-11-18 03:00:43
|
canada has a totally sensical ban on *specifc* guns, not "AR style" guns, but actually specific models of firearms
|
|
|
|
lonjil
|
2024-11-18 03:01:05
|
lmao
|
|
2024-11-18 03:01:19
|
that's so silly
|
|
|
|
Quackdoc
|
2024-11-18 03:01:55
|
any gun that has some decent amount of media recognition was banned, so now we have some rifles like the wk180c, which granted is actually a kinda great rifle https://www.cabelas.ca/product/157291/kodiak-defence-wk180-c-gen-2-semi-automatic-rifle
|
|
|
|
lonjil
|
2024-11-18 03:02:14
|
more realistically, maybe I should get a P365, 9mm is cheap. And then I can get an MPX-K as well, only have to buy one kind of ammo.
|
|
|
Quackdoc
any gun that has some decent amount of media recognition was banned, so now we have some rifles like the wk180c, which granted is actually a kinda great rifle https://www.cabelas.ca/product/157291/kodiak-defence-wk180-c-gen-2-semi-automatic-rifle
|
|
2024-11-18 03:02:47
|
so basically just keep starting fake shell companies that sell "new" models anytime they get banned, got it
|
|
|
|
Quackdoc
|
|
lonjil
more realistically, maybe I should get a P365, 9mm is cheap. And then I can get an MPX-K as well, only have to buy one kind of ammo.
|
|
2024-11-18 03:03:20
|
I've been looking at designing my own bullpup style 9mm since making your own rifles is totally legal here, minimum barrel length is however 18.5" so a bullpup shortens that up nicely
|
|
|
lonjil
so basically just keep starting fake shell companies that sell "new" models anytime they get banned, got it
|
|
2024-11-18 03:03:46
|
yeah, they did ban "platforms" but like, there are so many rifle platforms out there it's gonna take years lol
|
|
|
|
spider-mario
|
2024-11-18 08:56:18
|
https://www.painscience.com/blog/awful-power-of-cramps-demonstrated-by-rabies.html
this is quite instructive and also horrifying
|
|
|
|
jonnyawsom3
|
2024-11-19 11:56:43
|
I feel like I should mention something about us wanting to freely share the spec... But also that I really shouldn't... https://x.com/FFmpeg/status/1858654093985935740
|
|
|
|
CrushedAsian255
|
2024-11-19 11:59:14
|
More people should work on https://discord.com/channels/794206087879852103/1254012901816274956
|
|
|
|
_wb_
|
2024-11-20 07:57:50
|
You could point to https://docs.google.com/document/d/12Gmy2s4Nmkw6VDv2B6b5K1DLYhPrTUqSntrlmYzJpNw/edit?usp=drivesdk
|
|
|
Dejay
|
|
_wb_
You could point to https://docs.google.com/document/d/12Gmy2s4Nmkw6VDv2B6b5K1DLYhPrTUqSntrlmYzJpNw/edit?usp=drivesdk
|
|
2024-11-20 10:26:57
|
Someone should write an ISO standard on how to manage economic or political power without ending up subverting the intended goals and values
|
|
|
|
CrushedAsian255
|
2024-11-20 11:50:53
|
Since JXL is an iso standard I am now naming all jxl files .iso
|
|
|
Dr. Taco
|
2024-11-22 03:12:19
|
**News:** Google may be forced to sell Chrome to another tech company. Though most likely they will just fight to keep it.
**Me:**
|
|
2024-11-22 03:12:36
|
https://tenor.com/view/dumb-and-dumber-lloyd-theres-a-chance-jim-carrey-gif-11098164
|
|
|
|
TheBigBadBoy - 𝙸𝚛
|
2024-11-22 04:04:22
|
>there's a chance for me to buy it!
|
|
2024-11-22 04:04:25
|
[⠀](https://cdn.discordapp.com/emojis/867794291652558888.webp?size=48&name=av1_dogelol)
|
|
|
|
diskorduser
|
2024-11-22 05:57:18
|
What if Microsoft buys chrome
|
|
|
|
A homosapien
|
2024-11-22 07:16:24
|
I don't even want to think about that <:SadCheems:890866831047417898>
|
|
|
|
Quackdoc
|
|
diskorduser
What if Microsoft buys chrome
|
|
2024-11-22 07:20:41
|
~~they remove avif and add jxl~~
|
|
|
|
afed
|
2024-11-22 07:43:47
|
jxr <:kekw:808717074305122316>
|
|
|
|
CrushedAsian255
|
2024-11-22 08:58:45
|
vvcif incoming
|
|
|
juliobbv
|
|
Quackdoc
~~they remove avif and add jxl~~
|
|
2024-11-22 09:03:32
|
~~they'll make a tiny breaking change to either avif or jxl and call it wmp~~
|
|
|
|
Quackdoc
|
|
HCrikki
|
2024-11-22 09:45:37
|
unlike the scaremongering (including from mozilla) implies, it will make no material difference to google
|
|
2024-11-22 09:47:22
|
chromium still exists and the only reason google people have high % of contribution is because the company *gatekeeps*
|
|
2024-11-22 09:50:00
|
wether its wearing a chrome or chromium mask, change needs to drastically slow and become saner - no more of that constant engine/browser changing to the point noone can keep up because google forces or encourages sites to immediately deploy the very latest shiny thing they deployed in chrome so autoupdate pushes its "adoption" to 75% overnight
|
|
2024-11-22 09:55:27
|
what google wants exclusive, they put in chrome
what google wants spread, they put in chromium instead (regardless of how useless, harmful or rejected it could be)
what google wants prevented, theyll just gatekeep and block from being committed - sounds familiar ?
|
|
|
|
jonnyawsom3
|
2024-11-24 10:39:30
|
Not sure if it's because the socket here is only 1A, but my phone is hitting 96% and then starts dropping while 'charging'
|
|
|
|
spider-mario
|
2024-11-24 10:50:00
|
is it doing something power-intensive?
|
|
2024-11-24 10:50:19
|
the “saturation charge” (roughly 80%+ iirc) of lithium-ion is done at a lower current
|
|
2024-11-24 10:50:23
|
|
|
2024-11-24 10:50:33
|
(illustration from https://batteryuniversity.com/article/bu-409-charging-lithium-ion)
|
|
|
|
jonnyawsom3
|
|
spider-mario
the “saturation charge” (roughly 80%+ iirc) of lithium-ion is done at a lower current
|
|
2024-11-24 11:32:41
|
My phone is actually Lithium Polymer, the only reason it's lasted 8 years.
And no, just sitting locked. Yesterday I shut it down and used my own 2A charger and it hit 100%, but for some reason never logged as having fully charged. Then my charger started making a buzzing noise when I got in later so I decided to go back to 1A :P
|
|
|
|
spider-mario
|
2024-11-24 11:50:11
|
lithium polymer is a subtype of lithium-ion
https://batteryuniversity.com/article/bu-206-lithium-polymer-substance-or-hype
> Charge and discharge characteristics of Li-polymer are identical to other Li-ion systems and do not require a dedicated charger.
|
|
|
|
jonnyawsom3
|
2024-11-24 12:45:43
|
Ah, good. Couldn't remember the exact differences but I thought they had different curves at the same voltages
|
|
2024-11-25 11:04:00
|
Well, I got it to 100% this morning, although the logging still says it never fully charged. Either way, another flight in a few hours, so I guess I'll be reading the ISO spec a second time
|
|
|
|
DZgas Ж
|
2024-11-28 02:14:10
|
Encoded text in random circles: *I havent found anything like this, it looks like no one has invented it yet.*
|
|
|
|
CrushedAsian255
|
2024-11-28 03:10:27
|
how does it work
|
|
|
|
DZgas Ж
|
|
CrushedAsian255
how does it work
|
|
2024-11-28 07:14:31
|
use seed (it is not written here, but it can be placed in several predefined circles)
generate table of random coords
use same seed for geneare table of random circle size
noise is data.
if circle size same as table, = 00
if size +5% = 01
if +10% = 10
if +15% = 11
(bruteforce diff circle image)
I have already created other circle barcode, but I did not publish it. the information load is 60 bytes maximum, the circles can safely cross each other and overlap
so it's just an interesting idea. an analog version of encoding in noise. not practical
|
|
2024-11-28 07:28:58
|
<:AngryCry:805396146322145301> I'm actually trying to come up with a way to convert a photo in such a way data (like 2d fft) that it can be printed on a black and white(only b&w) printer and then decoded back into an image. I have not found this so far. but I decided to do something like this
|
|
|
|
_wb_
|
2024-11-28 02:47:33
|
just print the R,G,B channels as three grayscale images?
|
|
|
|
DZgas Ж
|
|
_wb_
just print the R,G,B channels as three grayscale images?
|
|
2024-11-28 04:19:38
|
I don't want the image to be obviously visible. I'm trying to come up with something like how to recreate the sound from the spectrogram. Only in a different sense
|
|
|
|
_wb_
|
2024-11-28 05:04:54
|
I guess you could apply an image-sized DCT and print the coefficients as an image...
|
|
|
|
DZgas Ж
|
|
_wb_
I guess you could apply an image-sized DCT and print the coefficients as an image...
|
|
2024-11-28 05:15:41
|
I don't want to use the digital method at all
|
|
|
|
_wb_
|
2024-11-28 05:17:36
|
Maybe some kind of holographic process then?
|
|
|
|
DZgas Ж
|
|
_wb_
Maybe some kind of holographic process then?
|
|
2024-11-28 05:20:53
|
I'm not sure... But I think I'll figure something out. I am currently watching the details of the black-and-white television device to understand the essence
I know that there are no problems for sound. whether it's a drawn wave, or a transformation via FFT.
there is 2d FFT technology but it is not what I would like
|
|
|
MSLP
|
2024-11-28 05:46:59
|
Well, going by the path of least resistance, if audio is not a problem, you could encode the image using <https://en.wikipedia.org/wiki/Slow-scan_television> and then the audio waveform for that
|
|
|
|
DZgas Ж
|
|
MSLP
Well, going by the path of least resistance, if audio is not a problem, you could encode the image using <https://en.wikipedia.org/wiki/Slow-scan_television> and then the audio waveform for that
|
|
2024-11-28 05:49:02
|
It looks funny. But still. These are megahertz frequencies. And what I'm thinking about is limited to a piece of paper
|
|
|
|
MSLP
|
2024-11-28 05:49:55
|
The unmodulated SSTV should fit in 4kHz band, only when modulated for radio transmissions it goes into MHz band
|
|
2024-11-28 05:50:35
|
Here's a sample https://commons.wikimedia.org/wiki/File:SSTV_sunset_audio.ogg
|
|
2024-11-28 05:50:51
|
(beware, it's loud)
|
|
|
|
DZgas Ж
|
2024-11-28 05:52:15
|
It sounds interesting. I think I'll leave it to the very end.
|
|
|
|
MSLP
|
2024-11-28 05:52:57
|
the beginning is like single colour, at the end there is more happening
|
|
2024-11-28 05:58:18
|
here's the example as-it decodes: https://youtu.be/tuyEwI9fJGk?t=12
|
|
|
|
DZgas Ж
|
2024-11-28 05:59:57
|
I understand. And even in this sense, there will be no problems with sound, even if you record the sound wave literally as it is. As in my recent project
|
|
2024-11-28 06:00:28
|
|
|
2024-11-28 06:00:51
|
but I want something else. I have a couple of ideas while I'm thinking about them
|
|
2024-11-28 06:00:57
|
<:galaxybrain:821831336372338729>
|
|
|
|
MSLP
|
2024-11-28 06:02:28
|
Nice, it's like a vinyl record unwound and print on paper.
|
|
|
|
DZgas Ж
|
2024-11-28 06:02:51
|
yep
|
|
|
MSLP
Nice, it's like a vinyl record unwound and print on paper.
|
|
2024-11-28 06:05:50
|
In fact, in the USSR of the 1930s, it was literally invented and implemented. A tape on paper. for the music. Alas, it sounds disgusting.
|
|
2024-11-28 06:07:06
|
But in my work I took what is used on film
|
|
|
|
MSLP
|
2024-11-28 06:11:00
|
Right, so in fact you could fit th 1930 tape in the typical audio magnetic cassette format, if they had gone with "printed" celuloid tape, instead of magnetic tape for consumer audio.
|
|
|
|
DZgas Ж
|
|
DZgas Ж
I'm not sure... But I think I'll figure something out. I am currently watching the details of the black-and-white television device to understand the essence
I know that there are no problems for sound. whether it's a drawn wave, or a transformation via FFT.
there is 2d FFT technology but it is not what I would like
|
|
2024-11-28 07:58:13
|
well
|
|
2024-11-28 08:29:00
|
This is the classic DFT. But it didn't work out very well for me
|
|
|
|
RaveSteel
|
2024-11-28 08:35:17
|
would it not be easier, at least for this image, to use lossless, since it would have fewer artifacts?
|
|
|
|
DZgas Ж
|
2024-11-28 08:35:54
|
I have another task point
|
|
|
Smegas
|
2024-11-28 08:57:33
|
A little play with Gigapixel AI 🙂
|
|
|
|
DZgas Ж
|
2024-11-28 09:00:12
|
<:SadCheems:890866831047417898>
|
|
2024-11-28 09:09:16
|
lowercase analog wave decoding method
|
|
2024-11-28 09:11:07
|
each line is a block. the brightness of the pixels inside the block itself. after a white pause, the values obtained are evenly distributed across the field. Unfortunately, can't make a square image here. Tape Again
|
|
2024-11-28 09:11:38
|
but it shows the essence of analog television well.
|
|
2024-11-28 10:52:02
|
circle Haar wavelet
|
|
2024-11-28 10:53:11
|
Unfortunately not applicable for analog use, any shift or half-pixel rotation and this all breaks
|
|
2024-11-28 11:23:19
|
Okay, I'm out of ideas. Alas. Primitive methods convey almost no information, and complex formulas break down at the slightest movement. Eh, an analog is an analog.
|
|
|
|
diskorduser
|
2024-11-29 02:52:50
|
https://fxtwitter.com/nixcraft/status/1862494722821874075
|
|
|
username
|
2024-12-01 10:32:55
|
https://bfcm.shopify.com/
|
|
|
yoochan
|
2024-12-01 11:14:04
|
As long as there is oil for trucks, planes and boats....
|
|
|
|
DZgas Ж
|
|
MSLP
here's the example as-it decodes: https://youtu.be/tuyEwI9fJGk?t=12
|
|
2024-12-01 03:43:25
|
I studied the algorithm and the principle of operation, unfortunately very primitive, nothing outstanding. Alas. this example is literally identical in principle like https://discord.com/channels/794206087879852103/806898911091753051/1311801111048622141 . You can see the quality
|
|
|
|
jonnyawsom3
|
2024-12-01 03:56:50
|
Just woke up to a Bulgaria login to my Spotify... At the start of the week it was a US login, so I already changed my password, twice. Honestly have no clue how they're getting in so consistently
|
|
|
|
MSLP
|
|
DZgas Ж
I studied the algorithm and the principle of operation, unfortunately very primitive, nothing outstanding. Alas. this example is literally identical in principle like https://discord.com/channels/794206087879852103/806898911091753051/1311801111048622141 . You can see the quality
|
|
2024-12-01 04:13:46
|
Ye, It's very simple, and there's no error correction, so if there is noise in radio transmission noise appears on the image. If you'll be able to find something good for this purpose please share.
|
|
|
|
RaveSteel
|
|
Just woke up to a Bulgaria login to my Spotify... At the start of the week it was a US login, so I already changed my password, twice. Honestly have no clue how they're getting in so consistently
|
|
2024-12-01 05:06:32
|
Does spotify not offer 2FA?
|
|
|
|
jonnyawsom3
|
|
RaveSteel
Does spotify not offer 2FA?
|
|
2024-12-01 05:14:34
|
Nope, only alternative login methods through Google and Facebook
|
|
2024-12-01 05:15:15
|
|
|
2024-12-01 05:15:17
|
|
|
2024-12-01 05:15:42
|
Hmmm
|
|
|
|
RaveSteel
|
|
Nope, only alternative login methods through Google and Facebook
|
|
2024-12-01 05:34:48
|
that's unfortunate
|
|
|
|
DZgas Ж
|
|
DZgas Ж
circle Haar wavelet
|
|
2024-12-01 11:02:20
|
One of the fun ideas
circle scan
|
|
|
|
Fox Wizard
|
|
Just woke up to a Bulgaria login to my Spotify... At the start of the week it was a US login, so I already changed my password, twice. Honestly have no clue how they're getting in so consistently
|
|
2024-12-02 10:10:58
|
Had the same thing last week lmao
|
|
2024-12-02 10:11:40
|
Fun thing, my Spotify password was a new password that's barely used in other places, so guess there was a recent data breach <:KekDog:884736660376535040>
|
|
2024-12-02 10:12:30
|
And it's not on here yet https://haveibeenpwned.com/Passwords
|
|
|
|
Demiurge
|
2024-12-02 02:18:35
|
I recommend you use strong randomly generated passwords you store in a keepass file.
|
|
2024-12-02 02:18:55
|
Keepass2xc is great
|
|
2024-12-02 02:25:33
|
Helps a lot when storing and organizing all your login info. Just keep a lot of backups. And backup your decryption keys too obviously.
|
|
2024-12-02 02:26:07
|
Since losing a decryption key is just as bad as losing the encrypted file
|
|
|
|
DZgas Ж
|
|
_wb_
I guess you could apply an image-sized DCT and print the coefficients as an image...
|
|
2024-12-02 06:57:37
|
Well, that's an interesting idea, I'll give it a try. try doing dct analog
|
|
2024-12-02 06:58:06
|
|
|
2024-12-02 09:17:04
|
so 1 bit line DCT pattern encode-decode
|
|
2024-12-02 09:19:42
|
one-dimensional DCT scanline
|
|
|
LMP88959
|
2024-12-02 09:52:50
|
Wavelet might be better here
|
|
|
|
DZgas Ж
|
2024-12-02 09:58:36
|
<@723252970657022053> maybe https://discord.com/channels/794206087879852103/806898911091753051/1311826971831046154
|
|
|
|
LMP88959
|
2024-12-02 10:00:11
|
Oh yeah regular wavelets break with shifting
|
|
2024-12-02 10:01:38
|
How about dithering to 1 bit
|
|
2024-12-02 10:02:02
|
And then making a filter to approximate the inverse of the dither
|
|
|
|
DZgas Ж
|
|
LMP88959
How about dithering to 1 bit
|
|
2024-12-02 10:02:34
|
for the input image?
|
|
|
|
LMP88959
|
|
|
DZgas Ж
|
2024-12-02 10:02:55
|
There's no need for that
|
|
|
|
LMP88959
|
2024-12-02 10:03:24
|
I thought you want to convert a greyscale image to black and white and be able to reverse it or something
|
|
|
|
DZgas Ж
|
2024-12-02 10:05:27
|
I rewrote the STRESS algorithm from GIMP sources to python, and now I can do strong smart normalization. And for the conversion algorithm, I do pixel difference comparisons. not PSNR/SSIM. This is the best and very slow way, without compromise. But since I'm rounding the entire output to 1 bit, it's all making a lot of noise.
|
|
|
LMP88959
I thought you want to convert a greyscale image to black and white and be able to reverse it or something
|
|
2024-12-02 10:07:23
|
Well, yes, but in fact this is an impossible task for algorithms like wavelet. but DCT has a wave basis, which, with the exact location of the scan from the paper, can be decoded normally. The wavelet would need Pixel (dot) perfect
|
|
|
|
LMP88959
|
2024-12-02 10:09:45
|
Yeah…
|
|
|
|
DZgas Ж
|
2024-12-02 10:26:13
|
|
|
2024-12-03 12:22:28
|
|
|
2024-12-03 12:25:41
|
without translation to 1 bit, it looks more normal. But I made mistakes somewhere, because the image should be identical to the input one. do not pay attention to the dark-white lines, this is the normalization of the lines...some mirror reflection artifact...
|
|
2024-12-03 12:26:23
|
waiting for jpeg xl 256x256 dct <:jxl:1300131149867126814>
|
|
|
|
jonnyawsom3
|
2024-12-03 03:39:59
|
Edging NTFS
|
|
|
|
DZgas Ж
|
2024-12-04 11:37:35
|
<@853026420792360980> Listen, let's make up, you know there are a lot of people with different opinions, none of which is true, but you're not stupid enough to I didn't care
Let's go for something in common so as not to accumulate passive negativity, name the top 3 of your favorite games ❤️🩹
|
|
|
|
jonnyawsom3
|
2024-12-04 11:50:47
|
Wow, an apology on the internet. I'm honestly impressed
|
|
|
|
DZgas Ж
|
|
Wow, an apology on the internet. I'm honestly impressed
|
|
2024-12-05 03:45:49
|
Unfortunately, it is not mutual
|
|
|
|
Cacodemon345
|
2024-12-05 02:11:24
|
https://www.phoronix.com/news/RISC-V-PowerVR-Driver-Support
|
|
|
|
Quackdoc
|
|
Cacodemon345
https://www.phoronix.com/news/RISC-V-PowerVR-Driver-Support
|
|
2024-12-05 02:24:54
|
so exciting
|
|
|
|
DZgas Ж
|
2024-12-06 04:23:01
|
Oh great. I was able to find two Google vulnerabilities and index 1.3 million YouTube videos by gacha life 🎀 title, description, preview.
no limit on requests. no protection from bots, no paying for api access, fuck free <:Google:806629068803932281>
|
|
|
DZgas Ж
<:This:805404376658739230>
I wrote all this in order to introduce the case of a problem that I cannot solve. maybe someone here knows how to solve it.
all my search work is solely the random finding of videos, despite the fact that, of course, I have saved all the popular channels, this is just a drop in the ocean. and all my work was saving random videos that "the Yandex crawler" found even without using the api.
The Problem: how can I find on YouTube all the existing videos containing the name gacha life, and having more than 100 thousand views.
if just think purely logically, this problem is solved very simply. — all the titles of all the videos are Write in Google databases, the number of views is just too. and Everyone has access to these databases, you just need to enter a word in the search bar. but here's the problem I'm facing — despite the fact that all videos can are sorted by the number of views, (the function is available on YouTube search). ....
But
I can't even get 0.01% of the videos because the youtube API **literally** limits the number of videos received to 50 pages of 10 videos (in api spec docs)
|
|
2024-12-06 04:26:04
|
a year ago
|
|
2024-12-06 04:31:03
|
if only scrolling millions... realize how unimaginably huge YouTube servers are. It's just unthinkable.
|
|
2024-12-06 04:33:11
|
to place it all on screen, need to use a size of 1 pixel
|
|
2024-12-06 04:38:30
|
and unfortunately, if it had been an avif or jpeg xl, my computer would have died. but webp. damn it, it can actually display 200 thousand webp on a single browser page. it's just... it just works. and it doesn't consume much memory.
and since the choice is from jpeg and webp. of course webp is a much better choice <:VP8:973222137202610226>
|
|
|
|
spider-mario
|
|
|
lonjil
|
|
|
diskorduser
|
2024-12-08 02:09:25
|
Hello <@226977230121598977> do you know any free russian email providers? I need it to create an account and post few threads on Russian forums.
|
|
|
|
DZgas Ж
|
|
diskorduser
Hello <@226977230121598977> do you know any free russian email providers? I need it to create an account and post few threads on Russian forums.
|
|
2024-12-08 02:19:03
|
I do not use ru email now. it exists, but as a tool it is dying in russia.
but if you are asking out of nescience, then the most popular Russian email is mail.ru and yandex
|
|
|
|
diskorduser
|
2024-12-08 02:19:41
|
Oh
|
|
2024-12-08 02:21:10
|
Do you have an account on sysadmins.ru?
|
|
|
|
DZgas Ж
|
|
diskorduser
Do you have an account on sysadmins.ru?
|
|
2024-12-08 02:21:52
|
No. I have no idea what it is. I'm a system administrator, but this is the first time I've heard about it
|
|
|
|
diskorduser
|
|
|
DZgas Ж
|
2024-12-08 02:22:16
|
Maybe I'm just from the younger generation
|
|
2024-12-08 02:22:27
|
<:H264_AVC:805854162079842314> 🎀
|
|
|
|
diskorduser
|
2024-12-08 02:24:59
|
I need to get some ||cisco|| firmwares from that site to unbrick a device. It is difficult to get those files directly from|| cisco|| 😦
|
|
|
|
DZgas Ж
|
|
diskorduser
I need to get some ||cisco|| firmwares from that site to unbrick a device. It is difficult to get those files directly from|| cisco|| 😦
|
|
2024-12-08 02:26:32
|
fuck cisco 👍
|
|
|
|
diskorduser
|
|
|
DZgas Ж
|
2024-12-08 02:28:15
|
Don't look at the fact that our study focused on Cisco. No one has explained to me why we use an iPhone and not an Android.
|
|
2024-12-08 02:29:05
|
having come to a real enterprise. What did I see? No. They don't use it. education and shit.
|
|
|
|
diskorduser
|
2024-12-08 02:30:04
|
I deal with few cisco devices. I accidentally erased system partition of a wireless access point. now it is not booting
|
|
|
|
DZgas Ж
|
|
diskorduser
I need to get some ||cisco|| firmwares from that site to unbrick a device. It is difficult to get those files directly from|| cisco|| 😦
|
|
2024-12-08 02:30:25
|
Are you sure it only exists there? write the firmware names version and other information numbers dates and letters
|
|
2024-12-10 01:27:19
|
<@723252970657022053> <@608345430656417867> literally
|
|
|
|
jonnyawsom3
|
2024-12-10 02:57:55
|
JXL when
|
|
|
|
juliobbv
|
2024-12-10 03:49:04
|
codec2 when
|
|
|
|
Fox Wizard
|
2024-12-10 10:04:32
|
That fun moment when your ISP promised to connect 1 device with ethernet cable for free, but they hire a third party that doesn't know about it. So you end up connecting your phone to Wi-Fi and to your PC with USB which is somehow still faster, more stable and has a lower latency than my previous subscription <:KekDog:884736660376535040>
|
|
2024-12-10 10:06:34
|
Old subscription with ethernet cable: https://www.speedtest.net/result/17089691457
New subscription (which is 3/5 the price) with phone hotspot connected to PC with USB: https://www.speedtest.net/result/17111585323
|
|
|
|
jonnyawsom3
|
2024-12-10 10:14:03
|
Then there's me and my £35/m contract https://www.speedtest.net/result/17114530864
|
|
|
|
Fox Wizard
|
2024-12-10 10:15:34
|
That's more than what I currently pay...
|
|
2024-12-10 10:16:47
|
I currently pay €25 a month for 10 months, but after that it'll go up for €40 a month
|
|
|
|
CrushedAsian255
|
2024-12-10 10:17:32
|
Australian here: 90 AUD for 150/40
|
|
2024-12-10 10:17:41
|
(54 EUR)
|
|
|
|
spider-mario
|
2024-12-10 10:24:36
|
https://infosec.exchange/@wdormann/113625346544970814
oops
|
|
|
|
Fox Wizard
|
|
CrushedAsian255
Australian here: 90 AUD for 150/40
|
|
2024-12-10 10:27:50
|
For that price I can get 2gbps up and down XD
|
|
|
|
CrushedAsian255
|
2024-12-10 10:28:24
|
Apparently NBN Australian fundamentally doesn’t support symmetric
|
|
2024-12-10 10:28:30
|
best you can get is 1000/40
|
|
2024-12-10 10:28:41
|
Some places manage 1000/80
|
|
|
|
Fox Wizard
|
2024-12-10 10:28:57
|
Rip, that's the main reason I didn't choose for certain ISPs
|
|
2024-12-10 10:29:15
|
Many support upload speeds of 10% the download speed for some reason
|
|
2024-12-10 10:30:07
|
And lol. I can get 8gbps up and down for €80 for a month which I'll definitely do if I ever host a LAN party XD
|
|
|
|
Quackdoc
|
|
spider-mario
https://infosec.exchange/@wdormann/113625346544970814
oops
|
|
2024-12-10 10:31:35
|
this is so dumb xD
|
|
|
|
Oleksii Matiash
|
|
|
jonnyawsom3
|
|
|
Oleksii Matiash
|
2024-12-10 03:19:53
|
And 1 Gbps for 12$, but I don't need it
|
|
|
|
diskorduser
|
2024-12-10 05:08:55
|
My $3.18/ month unlimited 5G plan
|
|
|
|
Demiurge
|
2024-12-11 03:50:14
|
Where in the USA do you get a $3 plan
|
|
|
|
diskorduser
|
|
Demiurge
Where in the USA do you get a $3 plan
|
|
2024-12-11 04:01:43
|
I'm not in the USA. Just converting my country currency to dollar equivalent
|
|
|
|
yoochan
|
2024-12-11 04:07:36
|
east europe ?
|
|
|
|
Oleksii Matiash
|
|
yoochan
east europe ?
|
|
2024-12-11 05:43:20
|
I'm here
|
|
|
|
Quackdoc
|
2024-12-11 10:20:11
|
so Pop Os' cosmic ecosystem is wild, cosmic panel/dock uses a sort of nested wayland support for their applets, which means you can do some really dumb stuff like this
|
|
2024-12-11 10:20:57
|
functioning xeyes in the panel lol
|
|
|
|
DZgas Ж
|
|
|
CrushedAsian255
|
|
DZgas Ж
lol
|
|
2024-12-13 10:42:02
|
https://www.blackmagicdesign.com/products/davinciresolve
|
|
|
|
DZgas Ж
|
|
CrushedAsian255
https://www.blackmagicdesign.com/products/davinciresolve
|
|
2024-12-13 10:43:06
|
This shit is lagging on my computer just because it's running. One of the most unoptimized programs for working with video
|
|
2024-12-13 10:44:18
|
I really tried everything new that exists. And unfortunately there is nothing better. Maximum functionality with a minimal interface. The same as offering me to switch from Cinema4d to Blender
|
|
|
|
Quackdoc
|
|
CrushedAsian255
https://www.blackmagicdesign.com/products/davinciresolve
|
|
2024-12-13 10:44:36
|
olive [av1_chad](https://cdn.discordapp.com/emojis/862625638238257183.webp?size=48&name=av1_chad)
|
|
|
|
DZgas Ж
|
|
CrushedAsian255
https://www.blackmagicdesign.com/products/davinciresolve
|
|
2024-12-13 10:45:01
|
If it needs to be cut very slightly, I use Avidemux. This is enough in 99% of cases
|
|
|
CrushedAsian255
https://www.blackmagicdesign.com/products/davinciresolve
|
|
2024-12-13 10:51:41
|
anyway, for my current project. No program will help. I had to program the frames rendering in python. And I opened **Sony **Vegas just to synchronize the sound with the video. and nothing else.
|
|
|
DZgas Ж
if only scrolling millions... realize how unimaginably huge YouTube servers are. It's just unthinkable.
|
|
2024-12-13 10:55:49
|
I processed 170 gigapixels (170,000 megapixels) compressing into WEBP.
Having created a "Google map" like, so tomorrow I will share this project site
|
|
2024-12-13 10:58:17
|
that it is technically possible to encode jxl in independent blocks of 256 pixels. And also decode in parts with block 256. This is implemented at the standard level, but not implemented at the encoder or decoder level.
But jpeg xl with a size of 300kk by 300kk pixels size 6 gigabytes, as one picture file, is technically possible
but unfortunately in 2025 I still have to have 2 million files of tiles for the map. Alas, it's a pity.
|
|
2024-12-13 10:58:27
|
<:JXL:805850130203934781>
|
|
2024-12-14 10:44:18
|
webp do brrrbrbrrrbr
|
|
|
DNFrozen
|
2024-12-15 02:07:24
|
https://cdn.discordapp.com/attachments/452572560987521034/1317816632525193297/2024-11-19_23-40-08_0.mp4?ex=67600fe1&is=675ebe61&hm=794c3b6e3f0258ef271981c49f58a1c0c45f3baf7fe016f44bb6d5924c1d0514&
|
|
|
|
DZgas Ж
|
|
DZgas Ж
I processed 170 gigapixels (170,000 megapixels) compressing into WEBP.
Having created a "Google map" like, so tomorrow I will share this project site
|
|
2024-12-15 09:16:52
|
yep the map site
https://gachamap.robdot.su/
|
|
|
|
Demiurge
|
2024-12-17 05:44:58
|
Yes, the <:JXL:805850130203934781> bitstream standard is extremely impressive but software must be written that takes advantage of the possibilities and features.
|
|
2024-12-17 05:46:06
|
That's the downside of (next-gen) alien technology from the future: it takes forever for it to be used to the fullest and for the power to be unleashed
|
|
|
|
_wb_
|
2024-12-17 09:43:58
|
Our main focus so far has been to make sure that decoders are complete. We don't want to repeat what happened with JPEG where de facto only a relatively small subset of the standard became widely deployed.
Making better use in encoders of what is available in the standard would of course also be nice but it's less urgent, in a way.
|
|
|
|
Oleksii Matiash
|
2024-12-17 06:47:39
|
It took exactly 2.5 years for Adobe to fix the issue I reported. Blazing fast
|
|
2024-12-18 12:38:15
|
`Introducing Ultra HDR, a new format in Android 14 that lets users capture stunningly realistic photos with incredible dynamic range.`
🤦♂️
https://android-developers.googleblog.com/2024/12/whats-new-in-camerax-140-and-jetpack-compose-support.html
|
|
|
|
yoochan
|
2024-12-18 03:33:50
|
I'm trying to cleanup scans that, in the worst cases, can look like this. The goal is to end up with a clean, low color, image (ie. b&w or a few levels of grey). I started with something inadequate, then implemented the otsu threshold.
When the image contains no highlights and shadows, otsu works... not perfect but well enough. However, with high variations of luminosity and contrast over the page it fails miserably.
Would you have some good algorithms in your bag for this kind of issues ? Before I try again to reinvent the wheel ?
|
|
2024-12-18 03:36:49
|
(once cleaned, for uniformly lit pages, the results are amazing ! with a full page like this saved as just 38Ko of jxl, lossless, black and white)
|
|
|
|
spider-mario
|
|
yoochan
I'm trying to cleanup scans that, in the worst cases, can look like this. The goal is to end up with a clean, low color, image (ie. b&w or a few levels of grey). I started with something inadequate, then implemented the otsu threshold.
When the image contains no highlights and shadows, otsu works... not perfect but well enough. However, with high variations of luminosity and contrast over the page it fails miserably.
Would you have some good algorithms in your bag for this kind of issues ? Before I try again to reinvent the wheel ?
|
|
2024-12-18 03:44:09
|
something like this? https://en.wikipedia.org/wiki/Thresholding_(image_processing)#Global_vs_local_thresholding
|
|
2024-12-18 03:45:36
|
it seems OpenCV has it as “adaptive gaussian thresholding” https://docs.opencv.org/3.4/d7/d4d/tutorial_py_thresholding.html
|
|
|
|
yoochan
|
2024-12-18 03:51:08
|
indeed, local thresholding ! I'll have a look at opencv 🙂 thx
|
|
2024-12-18 04:01:21
|
the explanation of the algorithm is scarce 😄 but I have something to dig
|
|
|
|
Quackdoc
|
|
Oleksii Matiash
`Introducing Ultra HDR, a new format in Android 14 that lets users capture stunningly realistic photos with incredible dynamic range.`
🤦♂️
https://android-developers.googleblog.com/2024/12/whats-new-in-camerax-140-and-jetpack-compose-support.html
|
|
2024-12-18 04:30:19
|
ignoring the fact that they introduced ultrahdr as just hdr, cameraX really is a nice api to work with, a decent number of foss camera apps now use it so this is a nice win for them
|
|
2024-12-18 04:31:15
|
krokiet on android via termux, not usable yet tho
|
|
|
|
Oleksii Matiash
|
|
Quackdoc
ignoring the fact that they introduced ultrahdr as just hdr, cameraX really is a nice api to work with, a decent number of foss camera apps now use it so this is a nice win for them
|
|
2024-12-18 04:58:21
|
I know, I'm Android dev. My point was about ultrahdr
|
|
|
|
VcSaJen
|
|
Quackdoc
ignoring the fact that they introduced ultrahdr as just hdr, cameraX really is a nice api to work with, a decent number of foss camera apps now use it so this is a nice win for them
|
|
2024-12-19 12:59:46
|
I wish manufacturers actually enabled access to cameras in the API. Most of the time only 1 out of 3 back cameras work in non-system apps, other two are dead weight.
|
|
|
|
jonnyawsom3
|
2024-12-19 01:39:35
|
I've got a dedicated monochrome camera that I can't get RAWs from after a firmware update, and doesn't show at all in anything other than the built-in app
|
|
|
|
spider-mario
|
2024-12-19 07:05:35
|
https://www.tiktok.com/@le_danielle_/video/7434990254181977375
|
|
|
|
DZgas Ж
|
2024-12-24 02:02:16
|
fluffy expansion <:PepeGlasses:878298516965982308>
|
|
|
|
AccessViolation_
|
2024-12-24 12:51:19
|
reading the draft specification
|
|
|
|
Meow
|
|
DZgas Ж
fluffy expansion <:PepeGlasses:878298516965982308>
|
|
2024-12-24 02:11:05
|
Are they stored as .jxl?
|
|
|
|
DZgas Ж
|
|
Meow
Are they stored as .jxl?
|
|
2024-12-24 02:49:42
|
|
|
|
|
TheBigBadBoy - 𝙸𝚛
|
|
Meow
Are they stored as .jxl?
|
|
2024-12-24 03:17:01
|
no lol, they are stickers from this Discord server
|
|
2024-12-24 03:17:16
|
added by <@219525188818042881>
|
|
2024-12-24 03:18:12
|
they had to remove one to add JXL pancake <:KekDog:805390049033191445>
|
|
|
|
Fox Wizard
|
|
|
AccessViolation_
|
2024-12-25 02:53:24
|
cute fops <:BlobYay:806132268186861619>
|
|
|
|
Demiurge
|
2024-12-26 08:12:04
|
Happy Yule and first night of Hanukah
|
|
|
Nova Aurora
|
2024-12-26 08:17:09
|
<:BlobYay:806132268186861619> 🎄 🕎
|
|
|
|
VcSaJen
|
|
TheBigBadBoy - 𝙸𝚛
they had to remove one to add JXL pancake <:KekDog:805390049033191445>
|
|
2024-12-26 03:25:01
|
Five more boosts for more sticker slots
|
|
|
|
TheBigBadBoy - 𝙸𝚛
|
2024-12-26 10:01:35
|
not like I want more fox stickers <:KekDog:805390049033191445>
|
|
|
|
VcSaJen
|
2024-12-28 02:18:32
|
Is RustRover a full-blown IDE like Delphi, Visual Studio, etc., or it's a text editor with kinda-sorta IDE functionalities like VSCode, Vim, emacs, etc?
I heard reports that it lacks some basic functionality, but that was a while ago, maybe that changed in recent months?
|
|
|
|
CrushedAsian255
|
2024-12-28 02:28:24
|
its *meant* to be a full-blown IDE but not sure if it is or not
|
|
|
|
VcSaJen
|
2024-12-28 02:45:52
|
I used Elixir programming language plugin for IDEA, and while it helps, it's not to the level of full IDE. So I wonder if it's on that level
|
|
|
|
Quackdoc
|
2024-12-28 02:53:53
|
abandon IDEs, return to vi
|
|
|
|
VcSaJen
|
2024-12-28 02:57:46
|
Abandom raw XML editing for GUI, return to visual programming of GUI
|
|
|
|
spider-mario
|
2024-12-28 08:05:04
|
this but non-ironically
|
|
2024-12-28 08:05:28
|
(the image comparison tool and the flicker test tool in libjxl were made with the help of Qt Designer)
|
|
2024-12-28 08:31:28
|
|
|
|
|
Quackdoc
|
|
VcSaJen
Abandom raw XML editing for GUI, return to visual programming of GUI
|
|
2024-12-28 08:39:31
|
I miss stuff like that lol
|
|
|
|
spider-mario
|
2024-12-28 08:56:33
|
I mean, it still exists
|
|
2024-12-28 08:56:36
|
https://www.embarcadero.com/products/delphi/starter/free-download
|
|
2024-12-28 08:57:06
|
or, say, https://www.xojo.com/
|
|
2024-12-28 08:57:23
|
(formerly “REALbasic”)
|
|
|
|
VcSaJen
|
2024-12-28 10:15:14
|
Uno (paid cross-platform .NET GUI framework) recently announced some kind of visual form designer: https://platform.uno/blog/uno-platform-studio-featuring-hot-design/ . But corporate/marketing speak makes it hard to discern what it actually is.
|
|
2024-12-28 10:18:08
|
Looks like instead of traditional IDE designer, it's more similar to videogame level editor.
|
|
|
|
spider-mario
|
2024-12-28 10:59:06
|
Uno itself is not paid, it’s Apache-licenced https://platform.uno/pricing/
|
|
2024-12-28 10:59:46
|
don’t know about this “Hot Design” thing though
|
|
|
|
VcSaJen
|
2024-12-28 12:00:49
|
Yeah, pure framework part of Uno is free, plugins/tools are paid:
> The Uno Platform Studio Pro version will be paid, and it is going to offer Hot Design, Design-to-Code and other productivity boosters which supplement the core benefits provided by the core Uno Platform Open-Source project. While we haven’t decided on the price yet, our plan is to offer developer-friendly pricing like typical offerings in .NET space, be it 3rd party UI component providers or IDE plugins.
|
|
|
|
Posi832
|
2024-12-30 03:41:57
|
Comparing DirectX versions in Portal for fun
|
|