off-topic-tech

[twin dew]

On that K10 Phenom family, the RAM is always accessed in that 128/144bit.
But in unganged mode, you get two successive 64/72bit memory lines to make the whole one.
Which seems to have been the same in Bulldozer-Excavator too.

128 bits without ECC, 144bits, spread into those ECC symbols with ECC

Using 144 bits, to store 128 bits.
In symbols that are 4 bits, 8 bits or 16 bits long.

Or was it that each symbol stores 4, 8 or 16 bits?

[wanton orchid]

So in x16 killchip still works

[twin dew]

Didn't look properly at those symbol tables.

[wanton orchid]

for 2x64/72

[twin dew]

Yes, higher that one is, more likely is that it works, but more likely is that other types of multibit errors slip through or cannot be fixed.

[wanton orchid]

2x8 = x16 per chip

[twin dew]

And yes, that x16 works for 2x64/72 with x8 chips to get chip loss capability.

Like that K10 documentation showed.

[wanton orchid]

lost bits are likely to be chip specific and close lines in real-life anyway
even without chip issue

[tribal kraken]

Will see if the AliExpress Icemancooler direct die block beats EKWB one. I bet it does

For Intel 12-14th gen

[wanton orchid]

anyway all this means whatever the ecc setting it will always fetch 128bit

which is logical and not a problem

my worry originally was it would fetch 128bytes

[twin dew]

Unless it was changed in some Zen, after Excavator.

[wanton orchid]

but it's unlikely

16bit code ecc is already strong

[twin dew]

Ah, this kind of thing.
Never heard RAM ECC working on anything more than single address.

[night girder]

Interesting article: tweakers.net/nieuws/230706/amd-extra-3d-v-cache-zou-9950x3d-amper-sneller-maar-wel-duurder-maken.html

Interview with Donny Woligroski from AMD.

AMD: extra 3D V-Cache would make 9950X3D barely faster, but more expensive

Translate the rest yourself 😄 Cba.

[twin dew]

Yeah, like talked few days ago.
Instead of single silicon chip with CCD on backside, on V-cache you have 4 chips of silicon.
Structural
Thinned CCD
Thinned V-cache
Structural.
All "glued" together.

[night girder]

It's an interview.

more QA. Not really technical.

[twin dew]

Yeah, but just as WHY V-cache CCD is much more expensive than just normal CCD.

[night girder]

"If you draw a line through a graph with the performance differences between the two processors, it is straight for a very long time, but at the end it suddenly goes down steeply due to games in which things go wrong now. We analyze them and try to solve the cause. The latest version of our 3D V-Cache driver contained a number of important improvements and we will also come up with an improved driver with the upcoming release."

[night girder]

eh seems like half the story.

[twin dew]

Significantly more expensive to make V-cache CCD than normal one.
And like that interview says, doesn't really give any performance boosts, as the code that benefits from that cache really doesn't like the inter-CCD latency.

To spread it out to two CCDs.

So currently any code that can benefit from that cache is better to be limited to running on just 8 cores of single CCD.

And not letting it spill to the second CCD, even if that too had V-cache.

Maybe AMD could make small run of dual-V-cache CCD 9975X3Ds or like for reviewing?

So GN etc. could do that kind of tests themselves and publish third party results.

[night girder]

"With more cache you don't solve the fact that the latencies are higher between cores in different CCDs, and many games are sensitive to that too."

Donny sounds down to earth.

[wanton orchid]

I mean 64bx16x2 = 256bytes

64b bus 2 channels 16 burst length

a single ganged read

[twin dew]

Yeah, understood.
But never even heard that RAM ECC would work on multiple addresses.
And you can anyways do a half-bursts too already.
And on DDR4, full burst wasn't 16 but 8 IIRC.

[wanton orchid]

it is 16 half is 8 I checked

ddr5 increased to 32 iirc

[twin dew]

No, DDR5 is BL16, BC8, with optional 32 long burst support too.

[wanton orchid]

what do you mean by "multiple adresses"

[twin dew]

And DDR4 is BL8, BC4.

[twin dew]

You select one address, you read, you get it and next X addresses out.

But you can select any address, not just multiple of that burst length.

[wanton orchid]

you mean it would have both shorter and longer burst ?

[twin dew]

So ECC couldn't work on multiple addresses as you would then have to do reads always on specific boundaries.

Seems DDR4 also has that second longer 16 long burst support.

[wanton orchid]

but isn't it what 2x64 mode is doing basically ?

[twin dew]

No, in that 2x64 mode you on Burst Length of 8, you actually just get 4 addresses worth of stuff out, as each takes two 64bit slots.

[wanton orchid]

16 byte boundary

[twin dew]

Basically real address is 2x logical address and 2x+1 logical address.

Or other way around, brain not braining.

[wanton orchid]

I mean it's more than that because 64b is 8byte already
so you mean real address is already 16 byte boundary

[twin dew]

Unganged: Two subsequent 64 bit RAM addresses on same stick, containing one 128 bit data chunk.
Ganged: Two 64bit RAM addresses on different sticks, containing one 128 bit data chunk.

The address shown before memory controller doesn't anymore have much to do with the address given to the RAM chips for the actual address.

[wanton orchid]

you are contradicting yourself
if ganged mode is 1 burst per rank
then ranks have 1 burst boundary i.e 8 bytes

[wanton orchid]

this still does not respond to ecc boundary problem
as it still is tied to memory controller

I mean
if ecc is 16byte what would prevent it from being 128

[twin dew]

No?
Addresses X, X+1, X+2 ... X+8.
BL of 8.
You can ask read starting from X, ending at X+7.
Or ask read starting from X+1, ending at X+8.
Where each address is 128 bits of actual data.

In theory.

[wanton orchid]

you said "because it can burst where it wants"
then processed to say that in unganged mode it can only burst on even columns

[twin dew]

But actuality is now 2x64bits on same stick.
So you actually ask for X, then get X to X+3, but the ram provides 8x 64 bits, you get 4x128 bits.

[twin dew]

Yeah, because the memory controller does translations, it hasn't been straight 1:1 for long time.

[wanton orchid]

and so it never responded to what I said

because nothing prevents it from doing 128byte boundary translation

[twin dew]

Sorry, brain not braining, I might have misunderstood.

[twin dew]

Because that would be very inefficient?
And isn't done?

[wanton orchid]

I mean there even exist "fake" ecc for mobile devices

it basically use part of the burst length for ecc symbols

[twin dew]

Inline ECC is just using more "slots" to hold the data and ECC in same physical DRAM device.

But brain is not working today to really try to talk about this it seems.

[wanton orchid]

iirc it's like 448/512

[twin dew]

But there isn't any limitations to only access memory starting from modulo burst length spots.
You can select any address.

[wanton orchid]

I understood that
but nothing prevent it physically from doing so

[twin dew]

Yes, there isn't current limit.
There isn't anything preventing memory controller from being made to do that, except the performance loss with current compiled code.

[wanton orchid]

all I'm saying is I couldn't be sure it was efficient at higher symbol sizes

because it's manual settings so "why not"
but now I get it
it works at logic_address>>4 boundary

[languid gulch]

think i'm going to have to redesign my desk

[twin dew]

Just seems to change how exactly that 128 bits of data is split and encoded into that 144 bits of physical storage.

[twin dew]

Would be really nice to know how it has changed with DDR5, with 2x 32 bits of data per stick, and either 2x 4 bit, 2x 8bit, or 1x 8bit ECC...
Where that last one is just weird.
Where both channels share one 8bit ECC chip or something like that?
That originally wasn't going to be included AFAIK?

[night girder]

Question, what defines a modern UI?

[willow pike]

not this

[night girder]

What about this?

Curved corners, flex grids and gradients it seems.

[wanton orchid]

ddr5 have wider ecc to compensate right ?

like is it 16 bit per physical channel or something like that (per 2x logic channel)

[twin dew]

Cannot know, the system side implementation details aren't public.
32 bits without ECC per subchannel, 36 or 40 bits per subchannel with ECC (4 or 8 bits extra)

So that 4 bits per 32 bits would be same ratio as with older ones, 72 bits per DIMM, and that better one is 80 bits per DIMM.

And DDR5 dies internally actually uses 136 bits to store 128 bits, for that on-die ECC.

Too much hidden...

Behind paywalls and NDAs etc.

[wanton orchid]

this feels like nda for having the seller offering in the first place

scammy

"All reads and writes must begin at a column address which is a multiple of 8" ahah
they are going with the way I was originally expecting

[twin dew]

Ok.

[wanton orchid]

8*32 = 256 bits minimum read boundary/size ?

(8 burst is chopped 16)

so with ECC lines this means minimum 256+32 reads

so same ratio and no need to do fancy interleaving

it is minimum 256+32 reads/writes PER channel

ngl this sounds much easier

you can easily cut that in 2x 128+16
for half latency on small reads

[wanton orchid]

@twin dew wait sorry for ambiguity, I'm talking about ddr5 spec there

[twin dew]

Yeah, DDR5 seems to have lot of extra complexity and latency from that on-die ECC being done on that once per 128 stored bits level.
As that is per chip.
So whatever you ask of the chip, it always has to do everything in 128/136 bit chunks.

[wanton orchid]

that means 16 bit chips would make sense

it is kind of matrix ECC I like this

chip ECC the rows
imc ECC the columns

[tribal kraken]

Oh well. Not a simple MB/CPU/RAM swap. Looks like full loop flush

[twin dew]

Thats bad.
Dissolved and oxidized copper.

Not enough anticorrosive additives in the water, and something made of copper/brass in the loop has corroded internally.

So most likely radiator.

[tribal kraken]

Yes that block has been in 24/7 use from 2018. First time opening it now. The loop is as old and had used radiators already in 2018. Just been changing the water every 2 year or so.

Impossible to get it all out, just flus with high flow tap water. But maybe scrap the radiators when next pump dies. It can be tomorrow or 5 years

That block had a lot flow restriction and did not cool 12900k proberly anymore for cinebench all core, 100C thermal throttle. Satisfactory server no problem at ~60C still

The coolant is clear EK cryofuel

[twin dew]

Just remember to use more anticorrosives in next fill.
Probably lot more as some will react with the left over copper gunk in radiator.

[tribal kraken]

Yeah that loop is simple to replace coolant.

[edgy hazel]

😰

disassemble the loop. don't torture it 😦

[tribal kraken]

Hand tight everything is enough yeah but Leatherman for opening just is Soo much efficient

And it can cut zip ties, open large Philips head screws that Ifix kit can't etc

And the knife on original Leatherman is sharp enough to do new clean cuts on soft tubing

And it's no use if I don't keep it always with me

[soft bloom]

how bad is this? definitely not new ssd

[twin dew]

That is HDD, not SSD?

[soft bloom]

most definitely ssd

[twin dew]

Ok, something weird.
It has both HDD only, and SSD only attributes...

[soft bloom]

well, maybe because it's so old? i think it already crossed 10 years of operation

[twin dew]

3 years on.
With 8925 power cycles.

But most likely has bad SATA cable.

From that UDMA CRC Error Rate

Crucial/Micron uses that Attribute 202 for "Percentage Lifetime Remaining"

So you have 21% of rated writes remaining.

Or possibly that normalized 79%

And most drives don't actually die when that reaches 0, but just keep going until actual failures start to happen and those 171, 172 and 196 start to go up.

Some Intel drives for example did contain intentional firmware feature where they would switch to read only mode when the nominal writes were full.

[soft bloom]

if that's total count then totally possible that's remnants from older sata2 cable... remembering simthing like that long ago

[twin dew]

Don't know, would need to monitor and see if it ever goes down or not.

[soft bloom]

i need to make decision of expanding partition on 980 samsung for some more stuff
or move it all on this ssd in hopes it won't just fail despite quite active writes and reads

[twin dew]

All 4066 spare sectors are still unused.

No part of the actual flash has hit failure point for any of the cells.

[soft bloom]

His chinese brother-at-arms (used for mirroring for some time) is 'very informative'
but works

[twin dew]

That seems to have basically broken SMART reporting for most of the values.

[soft bloom]

and yonger brother used for windows doesn't even have attr for unused sectors? oh

[soft bloom]

sadly

[soft bloom]

today i learned i can tell make to use multiple processes, so that it will compile Python faster. that saved sooo much time

[short marlin]

@opaque grove

I don't think I have a picture with the new watercooled GPU

[opaque grove]

fak it's big

[short marlin]

Told you

[twin dew]

Why didn't you just attach actual image?
But bad phone screenshot of one?

[short marlin]

Cuz it was in my cloud.
And it doesn't save the pictures in an easy to find spot on my phone

[soft bloom]

it's all about framing

[twin dew]

Then next time, at least turn the phone into right orientation for the picture?

[short marlin]

Doesn't make the picture that much better

[short marlin]

@obtuse geyser

[obtuse geyser]

That card is bent and screaming.
Not much space in that case, the airflow may be choked. If you remove both top fans, glue front half of top grill shut, remove back fan and open up the space under gpu, maybe it improves the flow.
But at least decent psu inside 👍 .

[obtuse geyser]

Given power cycle count and total runtime, I would replace it. Copy data off, get new one, before it fails miserably. Can chuck this one into junk pc for doing groceries.

[winged valley]

Noctua fans on a rad is crazy

[obtuse geyser]

My current airflow setup - testing shown that fans on top and back did basically nothing, so I'm down to two very large at front running at minimum rpm most of the time (they ramp up to 60% when mb temp sensors go up)
CPU fan is stopped until 60-70C then slowly ramps up , but unless running all cores the cpu is effectivelly passive cooled 😂
Sapphire did very good job on their cooler, so all the card needs is a gentle airflow to flush the hot air outside.
The PC is optimized to be noiseless on desktop/light load, and surprisingly silent under full load too.

PS. You may want to check if your noctua fans are pressure optimized or flow optimized. Mine are from arctic and the front ones are flow optimized, so they dont work well with pushing air through say radiators, while cpu one is the "Pressure" variant.

[short marlin]

It was perfectly fine.
Upgraded to a water cooled GPU few months after that picture.

[obtuse geyser]

does watercooling gpu help much?

[pure karma]

with weight reduction

[edgy hazel]

depends on the gpu lol

[edgy hazel]

never had a full copper waterblock on your gpu huh?

[obtuse geyser]

My previous gpu was hanging on zip tie, that was cheaper way to reduce xd

[pure karma]

thats basically the air cooler's now so yes

[obtuse geyser]

arent you limited by gpu designed board power tho? Mine even on max limit will not go into dangerous temp territory, so adding more cooling is not sensible.

[pure karma]

yea power limiter makes basically any gpu not worth overclocking

altho there was a shift to the voltage limiter being the problematic one at some point

[obtuse geyser]

given that gpu cost as much as rest of pc, I would be worried to mess with factory cooling. no warranty and any damage means you loose couple hundred.

[edgy hazel]

it's really not hard

just tighten everything properly

[short marlin]

And I did my fair amount of testing aswell. And that was the best setup
Although they were all pretty much close due to the radiator being so big.
Only outsider was the panel removed with 2-3° C more.

[edgy hazel]

are you cooling your PC with crude oil?

[short marlin]

Regarding temps. They were a lot better. And my GPU was a lot quiter.
But I didn't chose the best cooler in that regard, it was AMD Founders edition.

[obtuse geyser]

if removing side panel improves temps, your cooling is doing something wrong

[short marlin]

Nahh, that's DP Ultra

[short marlin]

It made the temp worse

[obtuse geyser]

what are your thoughts on delidding?
My cpu is basically limited by IHS transfer rate

[pure karma]

a expensive idea

[obtuse geyser]

isnt it basically the same as gpu waterblock?

[pure karma]

once you get the ihs off sorta yea

just getting the IHS off without killing the thing is the challenge

[short marlin]

Depends on the CPU, some are easier, some are jarfer

[pure karma]

a delid is defedently nto something i would do considering my track record

heck i literally nuked a power supply earlier

[short marlin]

Delidding is pretty cheap
But the IHS pretty much protects the silicon inside.
Without it it's vulnerable.

[twin dew]

With right tools, delidding is pretty easy.
For AM5 it doesn't even need special tools.
In turn Arrow Lake needs very special tools.

[obtuse geyser]

Not that vulnerable. Old athlons had bare die with rubber stands. If you install it right, all works ok, and has good thermal transfer.
but my final take on deliding was to just buy fastest cpu I could fit into my mb.

[twin dew]

But as said, it is very easy to crack the dies without IHS when mounting cooling.

[twin dew]

Lot of them died to cracked dies.
If you didn't get third party protection shim.

[obtuse geyser]

I do not have long enough track record to argue that. But the bare die was factory default, so manufacturer considered it durable enough.

[twin dew]

And started to use IHS in next gen.

Because of the failure rate.

P2 and early P3 used integrated heatsink.
Later P3:s were bare.
P4 switched to IHS.

Early Athlons used integrated heatsinks.
Later Athlons were bare.
Athlon 64:s switched to IHS.

And before P2, CPUs had the die under cover on pin side of the substrate.

Basically only Socket A for AMD and Socket 370 for Intel used bare died CPUs.

[willow pike]

athlon xp was also bare

[twin dew]

Switched from the old CPU "backside" dies to Slot type CPUs with integrated cooling, to single generation of sockets with bare dies on topside, and then to new sockets with IHS on CPUs

[willow pike]

it was absolutely not durable

but it was cheaper

[twin dew]

Same Socket A.
Just branding change from Thunderbird to Palomino.

Same base arch, just branding change on revision.

[willow pike]

yeah more cache, higher clocks, and the 2000+ etc model numbers to match p4 ghz

[short marlin]

Back when Intel used shitty thermal paste you already saw a nice decrease on temps just by delidding and adding LM and using the original IHS

[willow pike]

they were doing that for a long ass time

[short marlin]

Aftermarket IHS or planing decreased the temperature further

[willow pike]

i think they only shifted in skylake

which skylake? I don't recall

[twin dew]

Even just delidding and even using same paste, or something better.
As just removing the glue gunk would allow that IHS to fit better to the actual die.

[willow pike]

(coffee kaby etc are all skylake go fuck yourself)

[tribal kraken]

Testing out the Icemancooler direct die block. Default clocks 13900k seems to be tamed

[willow pike]

i would say "hurr enable MCE hurr" for the memes but that might actually kill it

[tribal kraken]

TBH ambient here is only about 15C but still impressive for AliExpress piece

Now I need sauna and beer but after that maybe some test more

[short marlin]

I think 7th gen was the last gen, after that they used solder like AMD

[willow pike]

forced to as they continued to cram more volts into that poor 14nm die

[short marlin]

Time to oc the beast

[willow pike]

i could never oc a raptor lake

undervolting might not be a bad idea though

[tribal kraken]

It was 5,6GHz all core with the EK direct die but I was never happy with contact and cooling perfromance with it

I think this could go 5,7 as thermals look good but time for PUBG with the AMD system 😄

[tough owl]

who knew in 2025 laptops could get erect

[languid gulch]

yea but are they floppy or rock solid

[alpine cape]

is 80 c normal cpu temp while playing games

[stray badger]

Laptop: yes

Desktop: fine, but a touch high

[languid gulch]

i'd be wondering if there was some kind of program being run in the background, or if the plastic's still on the heat sink

[twin dew]

while playing games

[languid gulch]

right but i mean on top of that

[twin dew]

Intel CPUs have run incredible hot for last few generations.
And AM5 CPUs also run hot.
Unless you have very overpowered CPU cooler.

Both have been set up expecting that the cooler will be the limiting factor, and to boost until they hit that temperature limit.

[soft bloom]

I set up cooling so it's a bit loud, but disperses heat quickly
which makes it roar even from single threaded load

but at least that's 'free' heating in winter

half an hour of single thread being 100% and room becames noticably warmer

[edgy hazel]

*Absolutely fine for any x3D before 9000 series

[wanton orchid]

80 if fine for zen3/5000 too when boosted all core high power eating workloads

[twin dew]

But that 95C is the natural place under load for lot of CPUs today when loaded, not even full multicore load.

[night girder]

https://www.youtube.com/watch?v=-fnLD6QvDGU

[visual tree]

That is the most finnish thing I have read

I was actually thinking of buying an outdoor finnish sauna + harvia wood stove

[night girder]

This makes sense.

Because hiding, is the same as closing.

[vagrant portal]

When you want to reopen the window again, like a settings window, it could waste less CPU resources for the trade of memory resources. Hiding here would probably not deallocate the objects whereas closing would. But yeah it's a design decision and a tradeoff I guess. Has it's advantages and disadvantages.

[visual tree]

💀

Not sure if this is a mistake or someone is clearly insane

[night girder]

Then call it hiding.

These are methods provided froma library.

They call a method close, when investigating it's calling hide() method. Which just setSWhowing(false).

Which is not closing. Closing != hiding.

[vagrant portal]

You are completely right that it is something different. Maybe the library has issues with re-initializing a window a second time because of some object mismanagement? It might be a workaround for problems that I don't know of. Like some "singleton" hack. I guess the open function checks whether the window has already been built? No good practice, yes, but well...

[tribal kraken]

Just Asus MCE 5,5GHz all core, 295W of power and temps stay still under 70°C

[pure karma]

Intentional probably

[mental oriole]

I would say no. 😄

[verbal raft]

[visual tree]

The seller probably knows there is a very low demand for that product and only orders it from a supplier when a customer makes an order while using the "greater fool" selling strategy

Because out of 200 people, 199 will laugh at the price and refuse to buy the product and at least 1 fool will buy it

[night girder]

The life of coders ... trust nothing.

[verbal raft]

https://www.youtube.com/watch?v=8mufF-YQzXM

[pure karma]

Cpu bit at the end was the cherry on top

[verbal raft]

5fps interpolated to 430

[willow pike]

DLSS9 preview from the future

[vagrant portal]

DLSS9 be like: "You can purchase a license to this base image and your GPU will hallucinate the rest of the game dependent on your controller input."

[edgy hazel]

Tbh pretty good for 5 to 430

[verbal raft]

stalker 2 official discord 👀

[stray badger]

who tf be voting for other?

[twin dew]

AMD had expected competition to 9800X3D.

[winged valley]

Datacenters

NVidia and IBM have CPU's

There's also Apple M-series, Qualcomm, RISC stuff

[stray badger]

Who tf is playing Stalker 2 on a POWER mainframe?

[winged valley]

etc

[winged valley]

Pretty sure mainframes arent a thing anymore

[stray badger]

Unfortunately, you are incorrect

[winged valley]

Yeah its power server now

I get no mainframe related results

[stray badger]

z/OS

[winged valley]

Huh I guess they are still a thing then

[stray badger]

Fun fact: about 50% of production code is still cobol

[verbal raft]

😂

[verbal raft]

https://tenor.com/view/intc-stock-intc-coffin-intel-coffin-gif-14403125517727857092

[north rock]

Hi guys, I hope this is somewhat the right thread to ask this question.
I have 60-80 fps with Ultra settings but when i activate Lumen I get 30-40 fps.
Can I upgrade CPU or GPU to be able to play with Lumen at 60-80 fps or do I need a total upgrade of my PC?

16GB RAM
i7-9700KF
RTX 2600 Super

[verbal raft]

i think its more likely to be the 2060 holding you back in this case

but i would say that the CPU is also on the edge of being able to achive 60 W lumen

the RAM is fine though

[north rock]

Thank you very much, then I don't think it's worth it to upgrade both economically. I might be able to get my hands on another 2060 used, for ~150EUR. I have no idea if SLI is still a thing and if Satisfactory uses it tho.

[twin dew]

No, SLI hasn't been a thing since DX12 came along.

[verbal raft]

SLI is basicly dead at this point......

[twin dew]

Will not work in any DX12 game.
And Nvidia hasn't been adding support for any new DX11 games for years at this point.

[north rock]

I have not been in the PC building buissenes for a loooong time haha

[twin dew]

And removed HW support for it several generations ago.

[north rock]

The things I learn 🙂

Thanks for info

[twin dew]

Dead since early 2020.

[verbal raft]

the 3090 was the last IIRC and even that had NVlink instead

though i think the entire 20 series has NVlink

[twin dew]

2000-series already had NVLink.

[verbal raft]

but as baldur said it wont work

[twin dew]

But Nvidia stopped adding those new SLI game profiles in 2020.

[dire igloo]

Last time I heard someone seriously consider dual GPU for gaming was dual R7 370

[north rock]

But if I would consider a new GPU, Is it the ones for 5-600 EUR I need to look at? Like RTX 5070?

[verbal raft]

i think the 3060/4060/RX 7600/7700 would be enough

also : the 5070 isnt out yet and no one other than NVIDIA knows what it performs like

for all we know it could effectivly be a renamed 4070 super

[verbal raft]

but we have RTX 50 and RDNA 4 right around the corner so i suggest waiting

[north rock]

Oh okay, so 400-500EUR. For eventuelly smooth fps and pretty lights ingame

[verbal raft]

(if you can)

[north rock]

Yeah, I think so. Absolutely no rush. The game runs perfect without Lumen

on Ultra

[verbal raft]

you could also use upscaling (DLSS)/FrameGen (from FSR) as a stop gap to be able to use lumen

[north rock]

I have Upscaling DLSS in settings and tried the "Upscaling Preset" options with lumen but still didnt get enough fps to be smooth with lumen. Or do you mean something else?

[verbal raft]

what presets have you tried with it?

(quality/balanced/performance)

[north rock]

all of them, yeah

[verbal raft]

oh

then it might actually be a CPU issue

[north rock]

it works great fps wise with "performance"

but looks like shit

shit is maybe the wrong word, but it doesnt look as good as i want

[verbal raft]

oh so it ran well, but didnt look good

is that what you are saying?

[north rock]

yeah, kind of. with performance used and high lumen. It's kind of smooth, around 60fps

but looks very grainy/blurry

[verbal raft]

so i was also suggesting FG that might look a bit better than DLSS

[north rock]

I dont know what that is, sorry

[verbal raft]

but for that this game requires that you use FSR AA

[verbal raft]

ok, no problem

[dire igloo]

So you want better visuals with usable fps?

[dire igloo]

FrameGen, Nvidia tech that hallucinates extra frames

And DLSS is upscaling tech equivalent to FSR and XeSS

[verbal raft]

and AMD tech too

[dire igloo]

afaik AMD's equivalent is called Fluid Motion Frames (FMF)

[verbal raft]

here is where you *should be able to turn it on

[dire igloo]

FrameGen is specifically Nvidia

[north rock]

something like that yes, I want to use lumen for the lighting and I want good visuals and good fps so its smooth witout drops/stutter

[verbal raft]

that is the driver level one : the in game one ( the one that is usable) is called FSR FG

[cyan crescent]

My windows 11 installation is stuck. Can someone help?

[dire igloo]

What resolution are you playing at?

[verbal raft]

so what it does is it takes 2 rendered frames and renders one in between this gives you more smoothnes but negatively affects responsiveness and can also introduce artifacts

[north rock]

only 1920x1080. I enabled FSR instead of DLSS, around same fps. Activaded FG. Dropped around 10-15 fps

[dire igloo]

Use dlss with your 2060

[verbal raft]

so it dropped you to 10-15 fps?

with FG?

[dire igloo]

Or did it drop by 10-15 fps?

[north rock]

no dropped from 60 to 50 after clicking in fg

[verbal raft]

in the menu right?

also : FG doesnt like the menu's / UI

[pure karma]

rtx 2000 dosent suport framgen you boomers

[north rock]

no in game in my factory, just now when not activated 55fps, when activated without moving camera i get 42fps

[pure karma]

its only 4070ti or higher

atleast oficially

[verbal raft]

ooff yeah IDK whats going on...thats strange

[north rock]

That's something i was thinking about. it says AMD

[pure karma]

works a bit on the 4050 for whatever reasson

[verbal raft]

it does support FSR FG

[verbal raft]

and NVIDIA doesnt specify the 4070 TIE as far as i know

[north rock]

I think I'll just go no lumen for now 🙂

[pure karma]

ahh yea might be any rtx 4000 gpu now

[verbal raft]

maybe a restart would help with that

[pure karma]

but i know when i originally got my 4070ti most games had it specified that framegen is only available for the 4070ti or higher

[verbal raft]

maybe back then there was no 4060 / 4070

🤔

[pure karma]

defedently was by mid 23

[verbal raft]

and the TI was the *lowest end model

[dire igloo]

Do you mean Nvidia's driver level feature or the ingame setting?

[verbal raft]

oh well maybe not by then

[verbal raft]

i think he forgor about FSR FG

bc thats what we were talkin about

[dire igloo]

I think he missed that we're talking about the ingame setting (FSR FG) and not Nvidia's product

[dire igloo]

Yeah, definitely the right course of action

[pure karma]

framegen's framegen its all the same blury garbage at the end of the day anyway

[verbal raft]

to me FG has **never **made the game look more blurry

other than AFMF 2 but we dont talk about that

[north rock]

550hrs before I found the setting so I think I'll survive haha

[verbal raft]

oof

IDK when it released that was the first thing i tried

back on my VEGA 3 that is

[north rock]

haha

thats me when i found the setting

[verbal raft]

do you know how blazing fast the VEGA 3 is?

[north rock]

no idea

[verbal raft]

so according to techpowerup my current GPU is 2200% faster

also : my card is about as fast as yours

[north rock]

but better cpu? or how are your settings? Ultra? High lumen?

[verbal raft]

i usually have evrything maxed out W lumen

1080 FSR AA 100% ultra preset lumen set to high

factory benchmark

47 avg
31 1%
27 0.1%

and i have an R5 5600G

which is faster

i ran some benchmarks at some point.... here are all the results

and the foliage heavy area is the red forest

and the factory is this one

better shot of the factory

and the benchmark pass also involved flying from the factory to this fuel power plant (granted it wasnt finished / in this state back then)

[north rock]

Nice one!

[verbal raft]

thanks!

got a few hours in😄

[north rock]

Im 170hrs in to this 1.0 save, and I thought that was much lol

[verbal raft]

well it takes time to decorate stuff with lumen in mind

[cyan crescent]

Why did they have to make w11? Its sucks

[verbal raft]

i guess im not alone in that sentiment

🙂

fuck WIN-DOWES 11

[cyan crescent]

There is literally almost no change in terms of software support as i was able to install w11 drivers on w10

Everything on w11 couldve been an optional update to w10 if it didnt have to do with improved security

[verbal raft]

the only difference between WIN 10 and win-dowes 11 is + 90% bloatware

[cyan crescent]

As soon as i get a replacement 4tb ssd for my laptop, im reinstalling w10

My old 4th drive borked itself and i have to put in the original 1tb drive which has w11

[verbal raft]

well for me and how i designed my factories...thats not an option (1st lumen off 2nd lumen on)

[wanton orchid]

win11 is basically patched windows 10

more bigger numbers better

[cyan crescent]

Didnt intel tell Microsoft to do w11?

[pure karma]

Youve got better performance than me im down to 10fps

[verbal raft]

in 1%'s ?

or 0.1%

?

[twin dew]

Probably all the time.
But I would somehow expect that be from too high VRAM usage.

[verbal raft]

🤔

i think @pure karma has a 4070

i have a 6600 8 gig

[twin dew]

UE5 does stupid stuff when it hits specific VRAM usage limits.
Tanking FPS to certain level that depends on RAM and PCIe speeds.

In 5-20 FPS range, depending on computer.

[verbal raft]

i know

[twin dew]

But depends what his stat unit says.

[verbal raft]

but i think my PC is slower than MagicZ's in every way

it would be pretty weird for MagicZ to run out of VRAM with 12gig's

[twin dew]

Impossible is possible with any PC that MagicZ touches.

[verbal raft]

ahh OK can i get the lore of that?😀

or is that classified?

[twin dew]

Seems to have some weird skill to break anything.
But I should have been in bed 2 hours ago.

[verbal raft]

oof

well have a good night

[mental oriole]

Why I'm going linux next.

[tough owl]

https://tenor.com/view/penguin-linux-tux-tux-linux-gif-341950137175367051

[broken stratus]

Tux Roll

[uneven bronze]

damn... got me beat. I got a Dell Dimension 8200 running. Has an old Office 2003 install with some archived emails I keep handy. I know where the bodies are buried.

[pure karma]

Cpu… every core pinned at 100%

Vram is close tho

[pure karma]

Everything is possible

i had 18GB of vram ussage not too long ago in fact i sent a picsure in here as is tradition at this point

[pure karma]

Go trough the history of this channel its about 2 years of colateral damage and counting

[edgy hazel]

MagicZ has the insane ability to break everything by just looking at it and then coming in here to ask for help

[pure karma]

I literally broke a psu yesterday

[night girder]

things stop working like magic

[pure karma]

Stated on the hush about it but yea

[edgy hazel]

We have forbidden MagicZ to ever touch any overclocking tool ever, because if he did, the nearest nuclear powerplant would have a meltdown

[pure karma]

at least i know what happened this time

the cpu turboed to over 3x its advertised power draw and the psu said "im tired boss"

[wanton orchid]

they do everything they can until they have a problem

then they talk about it here

[tough owl]

He’s got inverse Midas touch

[willow pike]

magicZ was the first owner of a 14900k

[cyan crescent]

Does a synology nas allow me to access it from any device? Phone, tablet, pc?

[twilit blaze]

From what i can see online it has to be a wired connection

[cyan crescent]

As in the nas has to be wired to the router?

[twilit blaze]

Anything connected to the NAS has to have some form of wired connection

That could be wrong

Ok no sorry

Are you connecting it to a router or a switch

[pure karma]

Unironically would probably be too scared to degrade under my posesion

[jagged snow]

It's just a nas, at the end of the day

So yes, you can

[stray badger]

Very

[stray badger]

Yes

[twin dew]

Which says you need to DDU your GPU drivers and reinstall.

[tribal kraken]

Proshop has listed few 5080 serie GPUs with EU prices, think FE cards will sell out fast

[soft bloom]

You know the stories that it takes millions of years for planet to move continents, split them, form mountains, valley, etc?
Well, this year started with signs that we will observe how it all happens rapidly.
But will we even notice the signs and save people who are in immediate danger?
Yep, this is not aligning with what mainstream scientists where tells us, but... just observe.

[dire igloo]

Hitchens's Razor would like to have a word with you

[languid gulch]

i have no idea wtf that's about

all i know is that i'm conditioned to be instantly suspicious whenever the term "mainstream" pops up, because of how politicized it is

[soft bloom]

how exactly does that apply?

[soft bloom]

fair
but science is not place for politics, right?
mainstream science wasn't fond of quantum physics when it was just starting to accumulate evidence
I don't mean anything bad with it, just a way to say that some ideas are more used than others

[soft bloom]

Ethiopia, for example

[languid gulch]

the great rift valley ain't gonna rip in half tomorrow

[soft bloom]

write that down and revisit in 5 years

[languid gulch]

i'd love to see where you're seeing predictions of it

[soft bloom]

also i do hope that nothing will happen tomorrow. and day after that, and so on

[languid gulch]

it might happen, but all evidence thus far is that tectonic movements happen on geologic time scales, not....days.

sure, there are things that happen quickly

but not plate boundary reshapes

and again, i'd love to see where those claims are coming from

[edgy hazel]

Pangea Gang

Imagine the train network 🤤

[dire igloo]

"mainstream science" always implies science that isn't majority agreed on - which is way too often exploited by shady salespeople, pseudo-medics or conspiracy theorists.

Using that term the way you did has an implied meaning which is directly linked to politics and pseudoscience. If you want to talk about unconventional science, feel free to do so - but maybe don't use loaded terminology that makes you look like a conspiracy theorist.

Also, quantum physics was heavily debated, not commonly disagreed to. And the primary reason for that was that it introduced a whole new model of how the world works - a model that needed verification.
Your claim isn't a new model tho, it's a prediction. And that prediction has to achieve two things: make realistic assumptions and work within the current best models.
If it needs a new model to work, said new model needs to be verified to be accurate.

[dire igloo]

it applies in the most basic way there is:
Provide evidence for what you're yapping about or it'll get dismissed

[edgy hazel]

https://tenor.com/view/tempting-fate-sts-saving-throw-show-gina-devivo-pocketgina-gif-15108344

[dire igloo]

Imagine that same train network if it was operated by DB

[edgy hazel]

Oh no

Oh nonono

[soft bloom]

Wait, but isn't mainstream by definition "what majority agrees on"?

[dire igloo]

Exactly why it's bogus terminology

[night girder]

The term "mainstream" refers to the ideas, attitudes, or activities that are considered normal, conventional, or widely accepted within a particular society or culture. It often describes trends, beliefs, or practices that are prevalent and dominate the general public's preferences, as opposed to those that are niche or alternative. In various contexts, such as media, music, or fashion, "mainstream" can indicate what is popular and widely consumed by the majority.

[dire igloo]

Scientific consensus is scientific consensus.
"Mainstream" implies opinion and allows for alternatives without critical questioning.

A song can be mainstream because its popularity is entirely based on how much people like it.
A scientific theory can not be mainstream because in science there's either proven correct, proven incorrect or unproven - or a percentage scale for accuracy

[dire igloo]

Using "mainstream" (a term linked to popularity) in the context of natural science (a field dominated by accuracy) is bad faith

[dire igloo]

Hm, pretty bad explanation, I'll try smth different

[night girder]

"mainstream implies opinions" is a bit of a stretch

[dire igloo]

Basically: using the term "mainstream" in this context is wrong because it inherently means that disagreement is a perfectly valid position to have.

You can disagree with scientific consensus, but only in one of two ways:

• you provide enough evidence to your position that the scientific consensus has to be re-evaluated (as it happened with heliocentrism and quantum physics)
• you're just straight up wrong

[soft bloom]

Then I meant *consensus"

[dire igloo]

Then the question becomes:
Is what you are saying actually contradictory to the current scientific consensus or just contradictory to what is commonly presented to the public (for example by middle school teachers telling children a simplified version of the world's physics)

[languid gulch]

also, still failing to specify what exactly isn't being accepted by "mainstream science", with sources, & why

[dire igloo]

"mainstream science wants you to believe that XYZ but actually it's ABC" is bad faith reasoning - it's incredibly loaded terminology and discredits the actual opinion.
Pair that with the fearmongering that is "but will we even notice the signs and save people who are in immediate danger?" and you just provided the prime example of what not to do when you want to actually have an informed scientific discussion

It's not worded to convince a scientist, it's worded to manipulate a commoner.
I recommend that you watch this video, it does a great job of explaining why rhetoric like this is bad and harmful to scientific research:
https://youtu.be/Pc2psN0PFTk

[dire igloo]

Now, back to the original message.
You mentioned "signs that we will observe how it all happens rapidly".
What signs? How fast is "rapidly"? Who's saying this? Where did this come from? Is any of this backed by evidence?

[languid gulch]

has some real "yellowstone's about to erupt" vibes to it

[night girder]

you haven't seen those documentaries?

[languid gulch]

oh i have

but also the margin of prediction & error is give or take 80k years

[dire igloo]

That's another thing: what accuracy does the simulation have?

[languid gulch]

exactly

[night girder]

especially the 4th one is ia charm.

[dire igloo]

There's always a variance in results when it comes to making predictions.
That's why it's important to know what assumptions were made and how moderate the simulation is run

[languid gulch]

that channel & its IP should be banned for that kind of disinfo

[wanton orchid]

how many cm is in 5000km ?

[night girder]

YT is just a bad source of information, unless you really trust the content creator.

[languid gulch]

yup

[night girder]

or you are familiar with the subject at hand. So you can use your own head.

but besides that, YT is full of crap.

[dire igloo]

YouTube is a platform where anyone can publish anything.
It's as accurate of a source as Google is - it's not a source, it's a means of obtaining sources

[night girder]

Well, I don't agree that it's not a source.

[wanton orchid]

for most people this is trap
as they say they are familiar with it as soon as they watched an other video about it

[languid gulch]

it's a source of something 🤣

[night girder]

https://www.youtube.com/@StarTalk Neil Degrass Tyson is still a ascientist.

[wanton orchid]

instead of actually having real life experiences with it

[night girder]

Like I said; there is good content creators on there. But it's the minority.

It's not black and white.

[dire igloo]

That's not YouTube publishing this content tho, is it?

[languid gulch]

the good ones will link their sources for you to go & check stuff yourself

[dire igloo]

It's someone else using YouTube to publish their content

[night girder]

You literal said this: "YouTube is a platform where **anyone **can publish anything."

[languid gulch]

but also, still kinda waiting on what the claim is, & where it's coming from 🤣

[night girder]

So ofc it's not YT publishing it. It's still on their platform, which was the discussion.

[dire igloo]

You're judging scientific literature not by its authors or the accuracy of its content but the publisher

[night girder]

Yeah, about that. Kurzgestag is hated by some in this channel. But they always link their source material in a "professional" way.

The same people criticizing Kurzgestag never even bothered to look at source (scientific papers).

[languid gulch]

they also publish stuff about how & why they're wrong when they're wrong, which is more than you get from disinfo channels

[night girder]

But you are right, some do post the source. There is just 99% 💩 out there.

[night girder]

Indeed! I like them.

[dire igloo]

You said YouTube is a bad source of information - and I'm disagreeing on the semantics.
YouTube is a bad platform to use when trying to find information. It is however not a source - the source is whoever created the content.

[wanton orchid]

bait

Wien you do a search prompt

YouTube is source

[dire igloo]

YouTube is as much of a source as Google

[wanton orchid]

exactly

[night girder]

@languid gulch another good one is: SciShow.

[wanton orchid]

and that's why you should need to avoid centralized search prompt

[languid gulch]

i'd call it an epi-source, it's a resource you use to get other sources from

[dire igloo]

They provide you with access to the content, but you'd never go around and say "Source: YouTube" unless it's something that the company YouTube published

[night girder]

Something like wikipedia? Which isn't a source either.

[languid gulch]

<-awesome socks member

[night girder]

According to the semantics. Which why I don't discuss semantics in this channel because it's so freaking dumb.

[dire igloo]

Wikipedia is a source, they're just not a primary source.

[night girder]

🤦‍♀️

see?

[languid gulch]

exactly. it itself has plenty of content, and can be a source of information itself, but to have information on claims verified, you need to have links to the sources on those

[night girder]

Mind tricks to make the narrative fit.

[wanton orchid]

People still figuring out that it's not "friends" who populate your Twitter feeds

[night girder]

No, FW just said that is a source.

[dire igloo]

Wikipedia publishes the information and is in charge of curating its contents

[night girder]

So YT isn't a source according to them.

Wikipedia is according to them. 🤷‍♂️

While wiki links 100000 other sources, so it's not a source. When applying their own logic.

[dire igloo]

They crowd sourced it but with a moderation system in place

[languid gulch]

that's why we need a 3rd term for it. epi-source.

a source that can be made of sources, but isn't necessarily a source itself based on what you're looking for

[night girder]

It's always borderline philosophical points being made about semantics, it's stupid and leads to nowhere.

[wanton orchid]

btw these issues exist even for professional scientists

[night girder]

Nothing is a source if you really want to dig into it.

Because every information is probably build on other information, good luck tracing the real source.

[languid gulch]

usually 🤣 idk, the discussion itself can shape how people think about these things, which isn't necessarily a bad thing

[night girder]

Anyway, to me, YT can be a source for information. Especially the discussion between scientists (which isn't documented anywehere else).

I would use it for my thesis if I had ato write one.

[wanton orchid]

and that's mainly most Youtube content is bullshit

[night girder]

don't disagree with that. But I said; it's not black and white.

[dire igloo]

A scientific paper also links hundreds of sources, it's still a source.
A metastudy doesn't do any actual research but gathers, prepares, arranges and presents existing research in a new fashion.
Both of these are source.

The question by which I decide what is or isn't a source is simply this: "Who is in charge of curating the content?" (in regards to its accuracy, not legal compliance)

[languid gulch]

if the youtuber is the originator of the information, it counts as a source. they can choose to link to other information, or just have the video be the publication

[wanton orchid]

because it just get some other online shit and render it a video

[night girder]

I am not going to advocate that ALL content is bad.

Maybe there is 1-2% trustworthy content out there. I don't know.

[languid gulch]

content needs context

[wanton orchid]

I said most

[dire igloo]

YouTube and Google are as much of a source as a library

[night girder]

Yeah.

[dire igloo]

"Episource" still implies it to be a source.
It's a platform, simple as that

[night girder]

#off-topic-tech message

What I said, is still a source.

[languid gulch]

god i hate platform as a term 🤣 reminds me of app

[dire igloo]

Would you quote YouTube?
Or would you quote the creator and their content piece?

[night girder]

Since a discussion between scientists can lead to new insights you cannot read anywhere else. And if that's recording, and that video is on YT, it's a source.

You can 100% link a YT video as a source. It's the same as referencing a nature documentary on VHS.

[wanton orchid]

not true
when you looking in library you mostly have a specific reference in mind
Google can be used like that too but it's not the primary use anymore and it's getting even more complicated lately

[dire igloo]

That makes YouTube a shitty platform for a lack of search functionality, doesn't mean it ain't a platform

[night girder]

The internet is the library. Google is the person sitting at the desk when you enter the library and you say; "I need this xy, where cna I find it?"

[dire igloo]

So... VHS is a source too?

Source for me is something that actually curates the information.

[wanton orchid]

it's completely up to YouTube to serve you content
and it's well known Youtube does do whatever IT wants
unless you input very specific search prompt
and even then now (hello shadow bans etc )

People still not getting how dystopian this is, just because it is happening slow enough

[dire igloo]

You can absolutely use a YT video as a source, I never disagreed with that.

I'm saying that "Source: YouTube" is not how scientific research works.
"Source: interview with XYZ published on YT" is a different thing - it's YouTube acting as a platform for the information, but not the source of it

[night girder]

Maybe just use your 🧠 ? If the video is made by MisterBeast, probably not? If the video is made by Stephen Hawkins, probably yes?

Just dig into it, check out the material. If it holds up, I would use it as source. Especially if it's someone trustworty, if I have to reference to somethign Stephen Hawkins said in a video, I would use it as source.

[wanton orchid]

that's exactly all been said already

• some things can be good sources ON YouTube
• Youtube is a bad source

[dire igloo]

Simple question then:
Does YouTube curate the contents of the videos?
Last time I checked, they don't. Beyond legal compliance, the contents of a scishow or kurzgesagt video aren't decided by the company youtube

[night girder]

Ofc not. Same as you would put in your thesis: Source: library

that's not how to reference to source material 😄

[wanton orchid]

💀

how naïve are you really ?

[night girder]

They said they going to be more strict with clickbait content a few weeks ago.

but "saying" != "actions".

[wanton orchid]

YouTube literally bans Videos broadly as soon as it does not comply with its direction
and heavily filter spreading of the videos otherwise

creators are only doing the heavy lifting for YouTube

[night girder]

honestly, I whished there was a better scientific platform for videos. Something like wikipedia.

or just a alternative to YT. It's a monopoly.

Why does nobody (and I mean higher ups) dare to say, it's a monopoly.

[wanton orchid]

People are being made hostage by Google and still try not to understand it

[night girder]

Crazy, or do I live under a rock? Is there another video platform, as mighty as YT? That offers the same?

Netflix has a different goal. Instagram too. And the rest aint so big.

Vimeo?

Btw; blog.google/intl/en-in/products/platforms/strengthening-enforcement-against-egregious-clickbait-on-youtube/

[dire igloo]

Just to clarify:
"YouTube" is a company that provides a video hosting platform under the same name.

The contents on said platform are moderated to be legally compliant but other than their own content, YouTube doesn't have any direct influence over the contents.

[dire igloo]

In this context, you're not talking about YouTube.
You're talking about creators that use YouTube.

Mister Beast could put his content on Pornhub or vimeo, wouldn't change a thing about the scientific accuracy of the content.

You're not evaluating YouTube as a source, you're evaluating content creators as a source. YouTube is just their platform of choice

[night girder]

Fireworker .... you know that aint right. And I know you know.

[wanton orchid]

THEY removed the accuracy and relevance of search results
then THEY pushed people to watch whatever is thrown at them
then THEY complained "it only show most viewed videos so we need to put random things here with a black bkx sophisticated algorithms"
then THEY decided to directly forcefully manage what people are watching
then THEY are complaining "it's people don't know searching so we must provide other ways"

[night girder]

So, you don't reference a book but the writer of a book? 🤦‍♀️

[wanton orchid]

it's all the same assholes creating the problems then "solving it"

[night girder]

Librrary -> book -> writer.

Youtube -> video -> creator.

I agree with SharklienX on this.

YT does have influence over their content.

The magical algorithms. And content creators know this. So they change their content, to the algorithm in the hopes to get famous and some money.

[dire igloo]

I 100% agree with Number 1, but number 2 is just straight up wrong unless you're talking about YT as the publisher (example: their annual business report).

YouTube is a platform.
It's not a source.
You can't quote YouTube. You can only quote a piece of media on YouTube.

YouTube is a shitty platform for scientific research due to the amount of bad content and lack of search functionality.

[wanton orchid]

you don't know how to read ?

[dire igloo]

That's literally publication bias and totally a thing in scientific literature

[night girder]

What is this discussion?

[wanton orchid]

"whatever is thrown at them"

[night girder]

Wikipedia is not a platform. It's a company.

Google is not a search engine.

It's a company.

What kind of logic is this?

[dire igloo]

Fucking hell

[wanton orchid]

how is that not relevant "because Youtube does not create its own videos" (and btw they did)

[dire igloo]

Did you miss the part about the "under the same name"?

[dire igloo]

It's pretty fucking obvious that we're not talking about the company - I just wanted to make it clear that YouTube IS a source when talking about the company

Their platform however is NOT a source

[night girder]

https://tenor.com/view/what-the-wtf-gif-25758871

[dire igloo]

This

[dire igloo]

And this

[wanton orchid]

this

[dire igloo]

Ignore this

[night girder]

afaik, they still do?

[wanton orchid]

I haven't checked that
last time I knew they were reconsidering the scope of it

[night girder]

https://www.youtube.com/@YouTube/videos

That's the official Youtube channel.

my source is wikipedia.

Oh fuck. Wikipedia isn't a source.

[wanton orchid]

I'm contemplating the irony of "you-tube" own channel

[dire igloo]

YouTube does create their own videos - that's where the company uses their own platform and that is when they actually are a source (equivalent to the business report example).

However, the platform as a whole is not a source, it's a platform.
The content there is curated by the users of the platform, not the company that runs it (tho there is monetary incentive).

If I told you that I got a piece of information from YouTube, you'd request the actual source - the video.

[night girder]

me too 🤣

and it's trustworthiness.

what is all this crap on the channel?

Looks all like ads?

[dire igloo]

That's bad teaching.
Teachers always said Wikipedia isn't a source when in reality it absolutely is.
It just doesn't qualify as a good source for them because they fear user manipulation interfering with accuracy (Croatia, Alan MacMasters)

In that sense, Wikipedia is a bad source but a source nonetheless

[dire igloo]

And as I explained above, this is the company YouTube acting as a source

[night girder]

I am just applying your logic 🤷‍♂️

[dire igloo]

You're applying your misunderstood version of my logic

[night girder]

If I say "YT can be used as a source" and your argument is that it can't unless you reference annual reports from YT, then the same can be said about wikipedia.

[dire igloo]

Wikipedia is a source because the platform itself is in charge of curating its contents

[night girder]

Most people would understand me when I said: "YT can be used as a source".

[wanton orchid]

the content is curated by both
and again your argument would only work if

1. search results would be strictly defined
2. search would be the main asset to get to videos (with external links too)
   the only way to achieve that is :
3. get to Videos only from other referencing mechanisms (forums discord communities blogs etc)
4. subscribe to your content creator
   which is how reliable channels almost exclusively works now, and is getting retaliation from YouTube algorithms and policies

[dire igloo]

It's crowdsourced through editors and a moderation hierarchy, but the responsibility of content curation is with Wikipedia

[wanton orchid]

YouTube is much closer to Wikipedia now than you think
: random guy publish
: "platform" decide how where when if it is published on their "platform" (American you remember)

Wikipedia works more like a platform for articles than YouTube as a platform for videos rn

[dire igloo]

Just to clarify:
my argument is that the website of YouTube (not the company, not the creators) is not a source, it's a platform.

And I don't see how 1 or 2 would be of any importance here - those are quality metrics by which to judge a platform, not defining characteristics of a platform.

Good platform: Google scholar
Bad platform: YouTube

Good source: Coffeezilla documentary
Bad source: Jake Paul vlog

[edgy hazel]

Ah I love #off-topic-tech ❤️

[night girder]

YT is tech

[edgy hazel]

https://www.bleepingcomputer.com/news/microsoft/microsoft-to-force-install-new-outlook-on-windows-10-pcs-in-february/

[wanton orchid]

YouTube home pages works as well as an asshole next month tv series show recommendation personal blog
so every recommendation personal blogs are platforms

YouTube got you by changing slowly

[night girder]

is it clickbait?

wasn't W10 EOL?

[twin dew]

In next fall.

[dire igloo]

That's primarily because the moderation structure isn't scaling as well as it should be.

Let's take a simpler route:
Who do you contact when you want to correct a piece of inaccurate information?
On YouTube: you have to reach out to the creator, the platform can't do shit unless the creator was in violation of any laws.
On Wikipedia: the wikipedia moderation team

[wanton orchid]

there ARE definitions for things being a platform or not
and it managing all the way up to specific content without user being too much involved means it's content
same as tv channels

[night girder]

You can't block the installation. But you can remove it atleast. Which is kinda weird lol.

Microsoft being microsoft

[twin dew]

And that "Outlook (New)" has been autoinstalling for a year already.

To replace older Mail.

[night girder]

oh your right 🤣 I have it on my system.

time to remove it.

[wanton orchid]

I don't know in what heaven country idea you live into
but youtube does A FREAKING LOT more than what laws ask for
and even sometimes does nothing at all even though laws are involved

[night girder]

It's seems only a reference or link is installed?

[twin dew]

Again pure enshittification of UI compared to the old one in utility.

[wanton orchid]

it's not driven AT ALL by laws

[night girder]

8.00 KB seems very efficient for a new mail app 😉

[wanton orchid]

you have to remove it with system prompts
or also from the iso before installation

and then it's actually on request

(oh btw that should be how it works to be eu compliant, but who cares )

[edgy hazel]

"Microsoft has also announced that beginning with the 2025 spring update all users will have to agree to eat baked shit at least twice a week to make sure their device is working according to the newest security guidelines layed out by the NSA.
Failing to do so will result in an insecure device so Microsoft has recommend to start migrating to this new diet as soon as possible."

[dire igloo]

You're judging YouTube's quality as a platform.
You're saying that YouTube is a BAD platform. That I agree with - and the primary reason is that yt is tailored for entertainment, not scientific accuracy. Hence why it works how it works.

Let's add two more examples:
Not a platform - the library of babel
Not a source - blank paper

[wanton orchid]

distraction , not entertainment anymore

[dire igloo]

The company YouTube does not have direct influence over the content on the platform aside from legal compliance (copyright, NetzDG).
The company does have an indirect influence through monetization practices which (just like other scientific literature) causes publication bias

[wanton orchid]

no
not monetization
you stuck in 2016 buddy
it's been at least 6 years it actually unreference videos

it has no tie at all to its user as platform
it can and is completely disregarding "platform" features

[dire igloo]

The way the algorithm works is for YouTube to get the most ad money. They get the most ad money by disallowing illegal content, hiding unpopular content and promoting popular content.
In that sense I guess you're right - it's not about entertainment, it's about profit and popularity

[willow pike]

oh heck

[dire igloo]

Wdym?

[wanton orchid]

YouTube sign you for not controlling anything even indirectly on the referencing process
such that it can not ever reliably be used as an actual platform

[willow pike]

that new outlook mandates all of your emails be uploaded to the "Microsoft cloud", and you cannot opt out

[wanton orchid]

i.e not a platform

you could only arg it's platformoid at this point

[dire igloo]

Again, I'm talking "platform" in the very basic sense that it's a technological framework for publication.
Similar to GitHub or Reddit.

There is content moderation and there is content promotion alongside monetization.
None of that changes the fact that YouTube is a platform, it just makes it a shitty one in regards to scientific accuracy

[wanton orchid]

and I'm talking in basic sense that it exactly is NOT a framework for publication anymore

[dire igloo]

I can't follow your logic, can you provide examples or simplify it?

[dire igloo]

So I am unable to publish content on YouTube?

[wanton orchid]

all the example of requirement for it to be a platform you disregarded

why would it be different this time

[dire igloo]

How is search a requirement?

And especially: how is a well functioning search a requirement?

[wanton orchid]

yes
exactly
you are able to give your content to YouTube (almost literally btw)
for them to decide if they publish it and how

[dire igloo]

It's not about whether or not it's good at its job.
It's about what job it has

[wanton orchid]

because it provide what makes it a platform : permitting wished access to your content

[wanton orchid]

and again it does not have this job anymore
read the contracts

[soft bloom]

5000 000 times 100
500kk cm 🙂
0.5Gm

[dire igloo]

And what would stop them from publishing said content?
As in: make it generally inaccessible even if directly linked to. Not just immensely hard to find or hardly any public availability

[soft bloom]

Bruh

The typo

[dire igloo]

Okay, so YouTube used to be a platform back in the "broadcast yourself" days when content publication was solely tied to legal compliance, is that what you're saying?
Then what are they now if not a platform?

[wanton orchid]

"what would stop them" who is "them" here ?
YouTube ? they are deciding publication not you which is my point
You ? Youtube is doing whatever

[dire igloo]

Exactly, how does YouTube decide whether or not they put my content on their platform?

[wanton orchid]

personally with their own direction

are you just about to witness its closed logic driven or what ?

[dire igloo]

Do you have an example of how this goes beyond legal compliance?

[dire igloo]

It absolutely is not, but from all I've gathered, the only reasons that YouTube disallow publication are legal compliance (mainly copyright)

[wanton orchid]

you say suicide or Hitler and your video basically goes offline
no laws mandate that

you seem completely feeded by survival bias

[dire igloo]

They allow disincentivize publication of "critical" content through moderation, but they're not actually blocking said content from the platform

[night girder]

So I did some checking, and it's not installed on my system.

[wanton orchid]

"but I've seen videos on YouTube so.."

[wanton orchid]

they are blocking it
you almost can't search for specific videos anymore anyway

[dire igloo]

"basically" is the key phrase.
It remains online. You can link it. It's technically public.
However, nobody will ever be able to find it and you won't ever see a cent of ad revenue.

That's still being a platform tho imo, just a super shitty one when it comes to content availability

[wanton orchid]

having the video link works
it's not a platform behavior
it's a data hosting behavior

but happy you completely prove my point that people completely forgot what a platform is

[dire igloo]

That's not what I meant tho.
Is the video actually gone? No.
You can still access it through the link.

It's just that the search functionality is fucked

[wanton orchid]

it's not platform anymore

in platform there is plat

you can think of it as circulating area

you cut circulation it's not platform

it's not "on" the platform at least

it's just data hosted

and Youtube does not like being only used for hosting

So it's starting predation on these 'channels'

[dire igloo]

in the sense of source vs platform, it absolutely is a platform tho

And all of what they're doing just makes it super super unviable for scientific research and publication

Because you risk your content being hidden (not removed) and can't find certain content even if you wanted to

[dire igloo]

*unless the company uses its website to publish content, in that scenario, YouTube (the company) would become a source - however, the website still remains non-source

[soft bloom]

Well thankfully it didn't erupt. Maybe it even won't.
Still shady practice to remove sensors from public access and convincing public by stating that "everything is normal" while the evidence that checks their own "voclano about to erupt" list (usgs).

[dire igloo]

Still shady practice to make outrageous claims like these with zero evidence to them

[languid gulch]

uhhh yea, would love to see some sources on that

[dire igloo]

Actually, it's super simple to explain:
The technological foundation absolutely is a platform.
The company and its business decisions make it poorly moderated which makes it a horrible platform for the sake of scientific research and publication.

[soft bloom]

But basically any idea that is emerging or didn't get enough evidence is not mainstream, as opposed to popular (by number of papers, funding, evidence etc).
Quantum mechanics wasn't popular. Climate change wasn't popular. Washing hands wasn't popular at some point in time.
I don't claim which model is right or wrong, better describes observations or not.
Mainstream was just about being widely adopted

[languid gulch]

lemme see the evidence of reliable information being removed

[edgy hazel]

The phrase "mainstream science" is tainted by conspiracy theorist. So if you don't want to look like one, don't use it. End of the story.

[soft bloom]

Yet there's a problem of merging classical mechanics and quantum
They have different fields yet people still take their finding to headlines because they discovered some new way of how matter behaves on the boundary of those two.
Does it make both unproven or otherwise? Just asking what it is in your understanding.

[soft bloom]

Eh, both

[languid gulch]

still waiting on some sources as to when and how ethiopia is going to be ripped in half in the next year

[dire igloo]

The only reason one would use the term "mainstream science" is to imply validity for a claim that contradicts consensus.

[languid gulch]

also the whole "Climate change wasn't popular" was oil companies knowingly spreading disinformation for profit. they knew in 1954 according to court-discovered documentation.

[night girder]

plottwist.

there is no such thing as mainstream science.

there is science. and pseudoscience afaik.

[soft bloom]

That shaping of continents, moving of tectonic plates has to take millions of years (aka geologic timescale)

[languid gulch]

still waiting on a link to a journal or something

describing exactly what you're trying to claim

[soft bloom]

Fair.
Also sad

[soft bloom]

Well quantum mechanics was pseudoscience to Einstein afair

[languid gulch]

it's trying to come up with the best explanation for what's observed with the information that's available.

[soft bloom]

They - oil companies or scientific world?

[languid gulch]

oil companies

the courts subpoena'd the documentation from internal records at the oil companies. it's not like PNAS was hiding it

www.bbc.com/news/science-environment-64241994

[soft bloom]

Ok but that doesn't mena it was known to majority. Not even talking about being accepted.
I am saying that while what you aya is probably true, it doesn't denyy claim that it wasn't mainstream.
By definition, there's always period when idea isn't widely supported adopted and developed.

[languid gulch]

you keep using mainstream as if it means something

as if there's a Mainstream Society that dictates all things

it was well known by the science community, and is in textbooks at the time, that carbon dioxide has a specific heat capacity. apply that to our atmosphere, it's easy to conclude that putting more of it into the atmosphere would warm it up over time. now, publish that knowledge at the same time the most profitable industry in history, known for going into areas and killing people to get at "their product", and see how far you'd get.

[dire igloo]

Both are proven true within their context and false within another.
Newtonian mechanics fail near lightspeed where relativity comes into play. Quantum physics collapses on the macroscopic scale.
That's the accuracy range that I mentioned - knowing when a model is or isn't accurate and to what degree it is accurate.

"The coin lands heads" is a roughly 50% accurate description of a coinflip.
It's neither true nor false - that would be the case if you made a definitive statement ("the coin always lands heads") or if you made a more accurate model ("disregarding the edge, a fair coin will land heads 50% of the time and tails in all other cases")

[languid gulch]

i'd also love to see the polling data you've somehow managed to come across on when exactly different ideas crossed the 50% threshold of acceptance by.......the public? scientists in general? the ones only in that field? what exactly?

[dire igloo]

Btw, referencing times of scientific disagreement/debate or conspiracy theories that turned out to be true is another way of implicitly validating invalid claims.
It's manipulation by rhetoric.

[languid gulch]

also failing to define a term like "mainstream", then using it in different contexts, only leads to a failure of understanding

[dire igloo]

Pseudoscience is anything presenting as proven fact that isn't founded in the scientific method.

"This method cures cancer" can be proven scientifically through clinical trials or it's stated pseudoscientifically through anecdotes and handwavy reasoning but without any actual scientific evidence to back it up.

Theories from pseudoscience can technically turn out to be scientifically true, but that's a matter of chance not science.

[languid gulch]

most scientific debates end when a specific question recieves a specific answer. "The Great Debate" of the 1920s was "what are the weird dusty blob things in the sky that aren't stars or planets? One side said it was gas clouds close to us, the other side said they were other galaxies not too long after our own Milky Way was confirmed.

both sides were right

[dire igloo]

"science" isn't some elitist conglomerate, it's a process.
A scientist is any person operating by that process.
It's designed to be deductive and produce the most accurate understanding of the world - through conjecture, experiment, specification and reproducibility by falsification and verification

[night girder]

theorem + tests = science

Just like how the double slit helped in establishing quantum physics.

Which was an idea of Albert Einstein 🤣

[dire igloo]

The biggest scientific debates in history came when the current model failed to describe certain aspects of reality ("what are those sky smudges?" or "why is light speed a limit?") or when a new model was conjectured that needed proof (quantum mechanics, heliocentrism)

[night girder]

So, that's the funny part about science, Albert was a critical thinker, so he challenged quantum physics. But he also helped in making it better by asking the right questions.

[dire igloo]

Included in the testing phase: documentation, falsification and peer review

[night girder]

Another critical thinker about quantum physics was Schrödinger ofc.

[dire igloo]

Exactly - Einstein searched for the holes because he wanted them filled, not because he wanted to rip apart the fabric

[dire igloo]

Higgs is actually a good example too - proposed the Higgs Boson while he was still attending uni courses and got dismissed by his professors because of his lack in expertise.
He designed a way of verifying its existence - albeit very costly and quite unreliable (giant ass particle accelerator).
Decades later, he was proven right and received the Nobel prize

Ramanujan is another great example of a brilliant mind being blocked not by the scientific method but academic elitists - a slum kid from India that only had a book with mathematical formulas was really frowned upon by the majority English academic elite.

[soft bloom]

Easy to conclude != Easy to prove or according to observations though

Also logic isn't in just adding particles that have heat capacity...
Anyway, good you recognize that there are nuances to even publishing findings

[dire igloo]

Another one: Marie Curie. Just read Einstein's letter to her about not giving a fuck about gossip magazines trying to pull her into the mud.
And of course Katherine Johnson who not only had to combat sexism but also racism, leading to one of the space race's most important mathematicians to be hid away in a basement

[dire igloo]

As my compsci prof so beautifully said: "A proof is just one scientist trying to convince another scientist of their theory"

Any proof will have to involve logical conclusion sooner or later. The question is, how much wiggle room there is:
Are contradictory conclusions possible? Are there incorrect or inaccurate assumptions? etc

[dire igloo]

Nothing sad about it. "Mainstream" is a media concept relating to popularity. If you wanna keep media and popularity out of the discussion (for example when trying to discover truth), there's no other way than to avoid the term.

Anyways, back to the actual topic of discussion:
Go provide evidence for the claims you made.

Link a source. Provide the data. Give us the info we need to judge the truth of what you said

[night girder]

Yeah know, I have a problem.

I need to find a front-end technology stack I want to use for a little project 🤣

[edgy hazel]

Tell me what u want, tell me what you need

[night girder]

Swing, JavaFX, all feel too old. Angular etc, you need to have browser installed.

I was thinking React.

but ... that's chromium based.

I want to write UI with minimum css and back-end with something else. The main thing is, I want to ship it as application. So it starts in it's own instance and not in the browser.

[pure karma]

I have no clue what that means so im just going to ignore it like everything else

The game runs

Thats good enough

[twin dew]

Uninstalling Nvidia GPU drivers with DDU, and then reinstalling them.
#1328095572845924434 message

One way Nvidia drivers break down for this game is to start taking 100% of CPU.

When the driver install is broken.

[night girder]

Ah fuck not react.

Electron!

[verbal raft]

well i know what im watching nexthttps://www.youtube.com/watch?v=RhvdF-6syeA

[mental oriole]

I don't care about it...

[night girder]

It being DLSS or Nvidia? 😄

[night girder]

So herg, 👍 👎 in using Electron framework?

when I look into it, it's maintainer is openJS foundation.

[mental oriole]

Dlss in general

Just fix your fucking game.

[edgy hazel]

I mean it's still chromium

[tough owl]

👎 electron bloated

[night girder]

I am so not used to javascript 🙈

but got it working.

[tough owl]

There’s a GitHub tool that will turn any website into an electron app

[night girder]

it was suprisingly easy to setup.

[rustic panther]

Is this something you do for your own fun?

If yes, use whatever the fuck you think is fun to play with or easiest for you, if not, go do some proper requirements research and go from there

[soft bloom]

Fun
I am just now realizing how much evidence there actually is...
But problem is not in aggregating it, but in viewing without assumption of "random event" and "fluctuations" but as signs (avoiding tautology for evidence) of single process.
It's not like they are hidden, secret or hard to find. Scientists are aware. But most operate divided from each other and under false assumptions.

[dire igloo]

how much evidence there actually is
I've seen exactly zero

It's not like they are hidden, secret or hard to find
for all intents and purposes yes, yes they are.

you are 100% regurgitating the bullshit rhetoric of the misinformation pipeline - emotionally loaded language, distrust in authority, oversimplification of scientific processes, anonymization, claiming a massive amount of evidence but sharing none, talking around the matter instead of the actual science

[night girder]

What is the discussion about, earthquakes? Tectonic activity? Volcanoes?

[dire igloo]

allegedly, tectonic movement is multiple orders of magnitude faster than what geological timescale would suggest
#off-topic-tech message

however, there is zero reason for any of us to believe that to be true

[languid gulch]

yea, still waiting for a link to something, anything

[dire igloo]

oh holup, I got a nasty idea:
go into an online space that prides itself in scientific enlightenment, then provide a somewhat plausibly backed theory as a challenge for the "consensus" - but make up a consensus that isn't true (but still realistic) to see how long it takes them to pick up

[soft bloom]

scientific enlightenment
the what?

[dire igloo]

people who think they're intellectually superior because of their belief in the scientific method - which there is some merit to it (rather that way than some rigid misguided ideology) but it can become too much when the blind spots of science are dismissed because they show an imperfection

in German, the word is Aufklärung - there also was a time in human history after medieval times where religious dogma was questioned and replaced by the scientific method

[soft bloom]

Oh I would like to revisit this message in like 5 years.
This must be good.

Sometimes i think that i should bet money on my implausible statements.
Except people in future will think it was immoral, or I will actually loose some money

But rn I remember the bet I wanted to make with my boss at old workplace.
Ez money

Just run from the boss who uses "it will be a problem of my children" as excuse to not handle problem on horizon

But yeah, probably should make any statements while remaining anonymous right?
Where does it even come from - making statements that are not in line with consensus, may cause fear, and don't have immediately provided evidence?
Definitely something shady

Still, I named 2 examples that show what happens and where.
The "lots of evidence" is in seismic data, data from volcanoes, including changes in composition and temperature of magma.
I don't think there's a comprehensive analytical paper yet, but you can definitely find individual case studies.
There's GeologyHub as good source of cases, but he doesn't see what I am talking about (at least doesn't talk about it) (actually i got it recommended from the same guy who Fireworker mentioned)
The unending fires are also linked to the same process.
That's not all of it... But should be a good start.
Anyway, conversation started as curiosity "will we notice and [take action, or not]?"

Btw,
how many news of "earth is splitting into two (but over millions of earth)" did you saw lately?
Doesn't that remind of "earth will heat 3 degrees (but in 100 years)"?

[past agate]

Linus hitting the bigtime huh? /s
https://x.com/FallonTonight/status/1879039226487984526

The Tonight Show (@FallonTonight) on X

.@LinusTech showcases some of the new tech unveiled at @ces 2025 to Jimmy & @sanbenito! #BadBunnyOnFallon #FallonTonight

Twitter · Jan 14, 2025, 5:33 AM

[dire igloo]

Ding ding ding, welcome to the misinformation pipeline.
A seemingly innocent topic presented in a super bullshit misinformative matter that leads to the same logic being applied to other fields of science - and in this case, seismic activity leading to climate change.

This is why the rhetoric you display is so dangerous and manipulative.
Climate science isn't geology. Climate scientists aren't geologists. Yet you applied the same rhetoric to both fields

You absolutely NEED to watch this video
https://youtu.be/Pc2psN0PFTk

[dire igloo]

Is this "lots of evidence" in the room with us right now?

You say we can definitely find them, how about YOU find them because it's YOUR job to prove YOUR statements?

How are wildfires related to seismic activity?
What data is there? How was it obtained? Were the things that got ignored?

That's why Hitchens's Razor applies: prove your claims or they'll get dismissed.
I don't need to prove you wrong and I especially don't need to provide proof that you're correct.
Prove yourself right or stop rambling nonsense.

[edgy hazel]

I like the way you do your own thing and aren't constrained by logic or information

[dire igloo]

"I named 2 examples that show what happens and where"

1. That's called anecdotal evidence and unless it's used to disprove the inexistence of something, the scientific value is rather small.
2. You didn't provide any information on it, didn't link anywhere for further reading, gave zero background info on why it's not supposed to be like this.

Hell, you explicitly mentioned immediate danger for the population.
There is NOTHING in what you said that makes me even consider an immediate threat

[languid gulch]

i'm expecting yet another "just wait, you'll see"

[dire igloo]

Ah yes.
A: "The world is doomed"
B: "Where? How? Huh? What? Why? Please tell me more."
A: (unspecific rambling) "you'll see"
B: "anything more specific?"
A: "DOOOOOOOOOMEEED!"

some xkcd style exaggeration

[dire igloo]

Also, to answer your question: none. I have not seen a single piece of news on that.

[soft bloom]

I think you are missing the point...
I don't want to proove anything, but to encourage investigation.

Recent video from Primeagen is a good example of why having a "proof" can be a bad thing (because it may not be full picture but peopel will interpret it without critical thinking).
Everyone here is capable of using internet to find factual data, as well as discussions.

[languid gulch]

a link would be useful

[twin dew]

So sprouting any kind of shit is ok as long as it is framed as being "encouraging investigation"?

[languid gulch]

known tactic for spreading disinformation

[soft bloom]

I was thinking multiple time of what I could link, but hoenstly, I don't think I can anything...

[languid gulch]

then all your claims are dismissed

what can be asserted without evidence can be dismissed without evidence

[twin dew]

That was the point.

[tribal kraken]

Aquasuite watercooling software feature, remote monitoring the 13900K "server" at home running prime95 small fft torture test: aquasuite.aquacomputer.de/9d5b93cf-319d-49df-a3f5-2ee730b2ae4d best way to spend time at work 😄

[languid gulch]

oh i know, just saying it out loud so everyone knows

[soft bloom]

and what would be a point of such misinformation?

[languid gulch]

fearmongering and spreading doubt about science

[soft bloom]

which acomplishes?..

[languid gulch]

"don't trust them, trust me"

[soft bloom]

still, what's the goal?

[languid gulch]

k, done with this. you're clearly choosing to continue the misinformation tactics and i'm done feeding it

[twin dew]

Most common target is just to get people to not trust anything.

[edgy hazel]

"Don't get the Vax. Vax is bad by big pharma. Supported by government. Government bad. Now government increasing taxes (for the rich). Remember how government bad? So tax increase bad aswell. Vote against."

Pretty easy flowchart for most conspiracies

[twin dew]

https://en.wikipedia.org/wiki/Fear,_uncertainty,_and_doubt

[tribal kraken]

When in doubt, flat out

[dire igloo]

"See the craze about the earth falling apart? Remember the craze about earth boiling to death? All just ridiculous fearmongering nonsense, don't let that fool you. Resist against it. Resist against restrictions imposed onto you. Fight against legislation made in the name of 'climate change'. Fight the people who demand such dangerous nonsense."

Watch the video I sent you. Please. I beg you.

[dire igloo]

I think it's you who is missing the point.
You are spreading misinformation. You are participating in harmful rhetoric. You are damaging the discussion.
If everyone is capable of using the Internet to find data, how come you aren't?

You say that proof can be a bad thing if it's incomplete but treated as complete?
See it this way: your current position is backed by zero proof. Zero data to even help make a concise point.
In fact, you haven't even made a concise point - there is nothing specific for you to prove because you haven't made a specific claim.
The most specific thing you mentioned was the existence of immediate danger. Danger of what type? How much time is there to prepare, how much to react? What would one need to protect oneself?

I would absolutely want to know more about this. Tectonics are a cool thing, geological activity is fascinating. I really enjoy learning more about the gigantic lump of rock we call our home planet. Even more so if there's immensely powerful stuff going on, things that deeply impact how we as a species interact with the planet.
If you shared ANYTHING on that topic, anything I could look into, anything to provide a deeper understanding of the topic: I would devour that shit.
I WANT to believe you, not because I want it to be true, but because I want there to be knowledge to obtain. Please stop regurgitating this wishy washy nonsense, please give me concise scientific information - even if it's a meta-analysis.

And to address why it's YOU that has to provide the proof:
Telling someone who isn't informed to research themselves has an immensely high chance of them finding misinformation

[soft bloom]

Finding misinformation is not the problem. These days it's pretty much expected.
But problem is stopping as soon as confirmation bias is satisfied. Not checking "what if it's wrong? How can i test that?" Is a problem. Stopping observations is a problem. Thinking that any worldview (model of it's part) is complete is a problem.

[soft bloom]

If everyone is capable of using the Internet to find data, how come you aren't?
I think you didn't understand, my bad.
I didn't say "i don't have a link(source)", I said that I can't provide one.
And I know that my udnerstanding of it is incomplete, so it seems like inviting to find more information is also a better course of action than just sharing what I know so far.

[soft bloom]

I will right after the vid that expands my udnerstanding of I claimed. (well, to be fair, conclusion is not mine)

[soft bloom]

haven't made a specific claim
I believe I can shorten it to this:

1. plate tectonics is not a valid model of Earth.
2. change of continental shape happens much faster
3. immediate danger for people in Africa - Ethiopia.

[twin dew]

Based on what?

Most importantly that last step, why would the specific areas people be in immediate danger?

[wanton orchid]

I mean it looks like flatearth "proof" to me for 2. point
"it looks like that when you look at horizon"

• but you are looking at an horizon in the first place
  why can't you do the actual calculations ? instead of repeating yourself ?
  because yes
  if you calculate horizon view, you can estimate a curve
  and in every directions
  also known as sphere shape

here :
yes tectonics moves
then why are you claiming "much faster" ?

do the calculations

[soft bloom]

how the heck do i balance learning more with this conversation?
I think its' better for me to stop engaging for some time...

[wanton orchid]

YOU claimed
you have to tell WHY

People can't and will not be in your head

[dire igloo]

1. Why not? Are there inaccuracies/limits (Newtonian physics) or is it fundamentally false (geocentrism)? Does it need readjustment or full replacement?
2. Says who? What makes them say that? How much faster? Why did we notice it just now and not much earlier?
3. Why those places? What danger? How much time to prepare? How much time to react? What are the expected damages? What are the proposed solutions?

[wanton orchid]

if you expect people to follow you AND not getting why (because you don't provide)
you are not only being in difficulties explaining yourself
but are being manipulative

[twin dew]

Throwing around extraordinary claims, with absolutely no backing.

At least none I have seen you say or link here.

[dire igloo]

My issues is: You're sharing nothing of value.
Anything you said could come from a group of experts that went "Oh shit! Oh fuck! Oh no!" or some college guy ripping the fattest bong hit ever while watching the movie 2012.
We don't know - and that's where you failed.

You're not inviting to find information. You're making extreme claims with zero backup.
The topic is interesting and definitely deserves to be discussed. But without scientific evidence and/or analysis, there cannot be an educated discussion.

[night girder]

2 days of this madness...

[dire igloo]

Simple:
Whatever resource you use to educate yourself, you share with us.
That way, we all have a common understanding of what we're talking about and we stop asking you to provide proof.

And if you are not yet educated on the manner in a matter that allows you to formulate an educated opinion:
Don't go around preaching about impending doom

And if you're concerned that we will rip apart those resources for being untrustworthy or exaggerating:
You probably shouldn't use them to educate yourself in the first place

[twin dew]

Somehow I expect this to be some weird interpretation from all the quake data he has been collecting and analyzing.

[soft bloom]

shhh, don't give me away

[night girder]

That's an assumption. Without saying the data, what can we say.

"Hey all, I have watched a bunch of data and according to me, the moon will hit the earth tomorrow!"

Chat: "Can we see that data?"

"No".

That's the whole discussion these past two days.

[soft bloom]

I deeply disagree with this because that would mean that whoever is under any sort of (for example, radical one) political/religious etc propaganda shouldn't share their finding from sources outside of it with people around because it will get discouraged in pretty much the same way.

[night girder]

Chat is annoyed for not seeing facts.

Kekw of the summer is annoyed nobody believes them.

And both talk around the issue 🤣

[soft bloom]

also let's forget I ever posted any of such analysis. I won't explain why, just pretend I never did anything like that. ok?

[wooden vessel]

Name one PhD who is writing research on this subject.

[night girder]

There isn't one. It's just public data on the internet.

Aggregated with who knows what code.

Spitting out a prediction or calculation we can't even seen.

Telling Kekw the tectonic plates are moving faster than scientists predicted.

That's what I get from the whole conversation.

And that's all an assumption, because I've seen 0 proof, or code, or anything.

[wooden vessel]

If there’s public data there will be papers since then they don’t even have to organize the experiment, just publish. And post docs love to publish

[soft bloom]

I never claimed such connection.
if anything, seismic data shows there's process that also influences plates. aka A causes B and C, but not B causes C (at least, i didn't conclude that)

[soft bloom]

that's where you are sadly mistaken...
geologists strongly believe that "nothing happens", and will die on that hill

[wooden vessel]

No they do not

[dire igloo]

In general, fair point.
In this context, eeeeeeh - unless you think that we are brainwashed fundamentalists that can't be reasoned with.

[wooden vessel]

I recall the earliest animations of earth formation and India moved at 10x normal speed to form the Himalayas and this remains settled science to today

[soft bloom]

haha, no. but... it's complicated. I definitely run out of time to discuss al lthe details.

[dire igloo]

Hence the "how much faster" question in this message:
10 times super slow is still slow

[soft bloom]

code? there's no code. what code? there's never been any code

[dire igloo]

Except you're not discussing anything, you're just throwing around hypotheses with zero proof to back them up

[soft bloom]

yeah, i see that. And i don't know how to fix that giving circumstances

[dire igloo]

You're essentially claiming that earth is flat and that you have proof and that proof can be found easily but the amount of actual proof you're submitting is exactly zero

[soft bloom]

at least i gave couple direction for search

[wooden vessel]

Dynamics of the abrupt change in Pacific Plate motion around 50 million years ago
van Hinsbergen, D. J. J., & others. (2019). Nature Geoscience.

NoThInG hAPPEnS

[soft bloom]

wait what?

earth is flat

[night girder]

I am assuming.

[soft bloom]

"on scale of human life"

[night girder]

You stated that. What is your evidence.

Don't start back paddling (peddling?) now.

If you don't collect the data yourself somehow (which was my assumption). Then where is paper?

[wooden vessel]

I’ll take what are earthquakes for $500 Alex

[soft bloom]

the most common argument I hear when there's discussion of increase in seismic activity is that "changes happen on geological timescale" and "we just got more & better sensors"

[wooden vessel]

The coastline in one part of Japan moved 300 ft out in one spot last year

[night girder]

It's just, in this chat, making bold statements without data to back it up. Will get called out. Happenens to me to.

Don't take it personal.

[soft bloom]

yeah i have no problem with that

[dire igloo]

Could you be right? Absolutely.
I'm no geologist, I got no idea whether or not you're right. So I'm very much open to discuss it.

Do I believe you? Definitely not, no.
Simply because the entire discussion has been tainted by you using FUD rhetoric, making ridiculous claims, nurturing the misinformation pipeline and providing zero actual scientific backing.

This is a prime example of Hitchens's Razor:
What can be asserted without evidence can be dismissed without evidence.
You asserted a theory and despite multiple requests for evidence you provided nothing to back it up.
Hence, your theory is dismissed.

You can re-introduce it for discussion once you can back it up with evidence.

[dire igloo]

It's an example of your rhetoric.
It explains why you fail so miserably to actually have an educated discussion.

[wooden vessel]

I just don’t even understand what you’re trying to assert. In our lifetime there have been tons of geological events of huge note. Mt st Helens. Mount Pinatubo. Montserrat. In my memory dozens of earthquakes. “Nothing happens”??

[edgy hazel]

Can we have CPUs back? Or GPUs? @pure karma do you want to overclock something and tell us what happens? Please anything but this

[night girder]

You giving MagicZ permission to overclock?

[pure karma]

i have been summoned

[night girder]

Go overclock MagicZ!

[dire igloo]

Okay, so, these are the possible scenarios:

1. You have evidence but for whatever reason can't share it (not accessible, classified, etc). Then say so. Tell us why you can't share the evidence.
2. There is evidence but you just suck at finding it. Tough luck I guess, there's nothing we can do to help you there. But don't go around loosely pointing us in random directions claiming that evidence is easy to find.
3. The theory is true but there isn't any evidence for it. That's not scientific research, that's just guessing randomly. Stop that. And stop acting like it's true if there isn't any evidence for it.
4. There is no evidence because there can't be because it's untrue. In that case congrats, you fell for misinformation. Quit the discussion, admit your mistake, learn from it and do better next time.
5. The theory itself is false but there is data suggesting it's true. Then there's an inaccuracy in the models or an unforeseen development. It needs to be researched further - and that requires proper data and analysis.

In either of these cases tho: your communication in this manner was utter dogshit. Not only did you fail to get your point across, you actually made the audience (us) distrust you.

[edgy hazel]

Please

Overclock your GPU

[night girder]

What happens if you overclock your PSU?

[pure karma]

it goes kaput

[night girder]

You sure? Did you try a good brand?

[pure karma]

problem...

[edgy hazel]

Can you even overclock a psu?

[dire igloo]

Maybe it's smth like continental drift that's happening on a scale of meters per day instead of millimeters per year

We don't know.
We can't know.
We don't have the data.

[night girder]

www.bloomberg.com./news/articles/2025-01-14/china-discusses-sale-of-tiktok-us-to-musk-as-one-possible-option

[night girder]

Holy fuck.

Wait what?

[dire igloo]

good product >>> "good" brand

[night girder]

Why would China consider to sell to Elon. They get money.

[pure karma]

dunno just the chinesse one that was in my alienware lol

[night girder]

But they lose a major product.

[edgy hazel]

@dire igloo go take a nap please 🙏

[dire igloo]

Aight, it's your taxpayer money

[edgy hazel]

Idc just make it stop

@soft bloom u 2. Go sleep