#off-topic-tech
1 messages Β· Page 167 of 1
Well I know for a fact how Vista was.. It was there i truely experienced having worn out Ctrl+S, cuz you never knew when the next BSOD would get ya
So let Mac users have that same experience :D
lol
i forgot truly how stupid cyberpunk runs
Yeah, blame cyberpunk 
my stomach
How are you friend?
first time I hear the term "space sceptic" which seems to be popular in west europe (duckduckgo told me that).
You know what is weird, that the center of a blackhole, is 0.
Logic 
"Free yourself from hell"
oh man, there's already video of the Dreamliner crash
Take off attempt without flaps in takeoff config but slick...
Probably took takeoff warning system breaker out during taxi because of "nuisance warnings", or some other reason why it wasn't working.
Because if that system is working, it will generate very loud audio warnings during takeoff roll in wrong config.
And didn't actually do the takeoff checklist from checklist and just from memory.
https://en.wikipedia.org/wiki/Take-off_warning_system
But also triggers if you advance throttles too far during taxi for faster acceleration from stop etc.
A take-off warning system or TOWS is a set of warning signals required on most commercial aircraft, designed to alert the pilots of potentially-dangerous errors in an aircraft's take-off configuration.
There are numerous systems on board an aircraft that must be set in the proper configuration to allow it to take off safely. Prior to every fligh...
Also takeoff roll was started from near midway of the runway, and not from opposite end...
Ok, Flap 5 on 787 is almost clean while being used for take-offs?
But that "midway runway" might be just tracking issue, and not the real start position.
Also talk about possibility that the crew mistakenly retracted the flaps instead of landing gear.
Which has happened before, both in cases where it was noticed almost immediately and fixed, and with few other crashes from it.
But nothing official will be known for weeks to months.
whats special this time around?
Have to wait and see. No reason to speculate.
We don't belong in the skies.
We don't belong in space.
And neither do we belong in the sea.
We don't belong on this planet
If we go there, and somethign goes wrong, we mostly end up dead.
Such a tragedy (again).
The following is a list of accidents and incidents involving the Boeing 737 family of jet airliners, including the Boeing 737 Original (-100/-200), Boeing 737 Classic (-300/-400/-500), Boeing 737 Next Generation (-600/-700/-800/-900) and Boeing 737 MAX (-8/-9) series of aircraft. As of February 2024, there have been a total of 529 aviation accid...
This was 787
How is Boeing still able to be a company?
Don't care about that.
Didn't Boeing get a lot of shit over the years? What did they do again?
Boeing is in the shitters right now, farther along than Intel on same path.
But very probably not directly related to this crash.
A year or two ago, Herg was ranting I think about boeing.
But I can't remember what they did. Something shady.
And also, someone is protecting that company big times. Someone with power.
MCAS stuff in that page for 737 MAX.
And that Boeing might have had whistleblower assasinated.
But they timed it pretty wrong, as most of the testimony had already been given.
That's it!
Most likely just very depressed person doing suicide after bad day in court.
Oh, ofc. Boeing is US company. That explains how they can keep getting away with so many deaths 
Because of Boeing lawyers cross examination etc. conduct and pressure.
Trying to do character assasination IIRC.
In a way.
Well, character assasination, if not correct, is illegal.
Yeah, I am not flying anymore.
Compare it to other major forms of transportation β with 0.04 deaths per 100 million miles traveled, train travel is much more dangerous than airplanesβ 0.01 deaths per 100 million miles.
Mhm, doubts. Source is pro airplane website.
Nah, air travel is way safer than driving, cycling or walking too.
You crazy.
But the deaths happen in huge chunks few times a year, and each makes the news.
There are so many factors at play here.
How many "driving, cycling, ...." deaths did involve any sort of drug (alcohol etc).
How many deaths are caused by someone else (that rarely happens in the sky, airplanes have avoidance systems).
I like flying, rollercoasters, motorsports etc. But have to admit that experimental small aircraft or small piston engine helicopters dont feel really safe
Flying is a very specific act with lots and lots of regulation.
Also pretty new. So comparing acts we do since we exist vs something that is totally new and that have a death ratio. I don't think that represents any useful data.
Those are not included in that statistic.
Neither are corporate planes.
Yes and accident rates in them are higher
When quoting those stats, it is almost always just for scheduled airline service.
And part of it is the fast travel speed.
When normalized for time travelling it gets worse in comparison.
Instead of person-time instead of person-distance.
Take this image. It doesn't take in account hat maybe 99% of population uses a motorcycle and only 0.00001% takes a airplane.
These numbers are not real, but just an example of how it skewers the data I believe.
So if you take that account, yeah ofc you get only 0.07 death because barely anybody takes an airplane.
While motorcycles are super popular in certain cultures.
Nah, motorcycles are very dangerous.
As when anything happens, you basically die.
Including someone in car getting in way etc.
You don't understand me. 99% of population uses motorcycle. 1% uses airplane. Where do you think most deaths happen?
This my friends airport we take off and land to: https://www.youtube.com/watch?v=ioYuM9Tupoo&ab_channel=samulihukkanen
Would be very weird that 99% of the deaths happen on a transportation system used by 1% of civilization.
Most of the data found on the internet lacks a lot of context. So I don't think it's that black and white.
Those are normalized for person-distance.
How many people use doesn't matter.
If 100 people use car to drive 100 km per day, that is 10000 person-km.
And if 50 people use motorcycle to drive 200km per day, that too is 10000 person-km.
And if there is one accident with car, and two with motorcycles, then the number of accidents for motorcycles per km would be double.
That's dope!
Which is why switching to person-hours in transport instead of person-km changes things, as faster transport option uses less time for same distance.
"How safe is it to take the train Amtrak"
0.43 safe 
One with 1 accident per 10000km, another with 2 accidents per 10000km.
Doesn't matter how many people it took for that travel amount of 10000km in total.
"How safe is it to take DB?"
0, all trains werd delayed
For every billion kilometers traveled, trains have a fatality rate that is 12 times higher than that of air travel, and the fatality rate for automobiles is 62 times greater than for air travel. By contrast, for every billion journeys taken, buses are the safest form of transportation; using this measure, air travel is three times more dangerous than car transportation, and almost 30 times more dangerous than travelling by bus.[50]
Buses are the safest! π
And would be even better if switching to that person-hours in transit instead of person-distance in transit.
Where as airliners would drop a lot in comparison.
As per person-kilometer, you spend lot less time inside airplane than in bus.
As the travel speed is tens of times faster.
I don't know, something is off with this kind of math and doesn't represent reality I think
Or it does represent reality, but only one facet of it.
It does represent reality, but the tweak to make airliners seem better than they are is that per person*distance as the comparison.
Instead of Person*Time.
Also comparing motorcycles to airplanes is a bit silly.
But that is the normal way traffic accident statistics are kept.
What? It does represent rreality. But tweaked.
So it doesn't represent reality π€£
Because representing reality means it's not tweaked in favor of anything.
To travel 1000km, by 4 persons, no matter the transit used, that will be 4000km combined distance.
But if you take car at 100km/h, that takes 10 hours.
So 40 person-hours.
But if you take airplane at 1000km/h, that only takes 1 hour, so 4 person-hours.
If both have one accident, with first, the accident rate is same per person-kilometers.
But 10 times worse in person-hours.
it IS real but misleading
of course you'll travel longer distance in fly
the rational behind distance is that you'll have to go the distance if you want to move
so if you go by car or train to avoid fly
I'll go that far in car or train
But that person-distance as the comparison point has been standard for ages.
And easier to keep track off with various travel methods.
but BOTH: per journey per distance and per time should be shown
Sorry, that doesn't click in my head.
I hate when real data is "skewered" into anythings favor. I know that the world works like that.
But that makes the data I am looking at a bit more crap.
But even with that person-hours used in travel (actually moving with that mode), air travel is good.
Just not as insanely good as that person-distance comparison says.
it does make it more crap
but it's not lying in itself unless it does not explicitly tell the calculations
And that person-distance is the common standard for all the relevant statistics in all countries I know off.
True. Leaving data out of the equation is not lying, technically.
For any kind of accident statistics.
But neither does it mean it's the full truth. Ah well. Airplanes, I am skeptic.
That one does it per journey. And feels a bit more intuitive.
Skydiving as a mode of transport is a bit questionable though.
and pedal cycle, wth is pedal cycle?
Bike?
And so much of car "journeys" are done at lowish speed over pretty short distances.
Like the drive to work and back each day.
And if you stop at store on the way back, that splits that return to two "journeys".
So all the short distance city driving that doesn't cause fatalities pushes cars way down in that kind of comparison.
Well that doesn't really matter does it?
I don't see peoeple go on foot to the other side of the planet.
Dinstance doesn't matter. It just shows transportation choice and the likelyhood you die.
Regardless of hours and distance.
And as suggested already by wikipedia, which the chart is based on, it seems the bus is pretty safe option.
Point was with any kind of comparison like that, you change the comparison from effects like that.
That "per journey" pushes car statistics way down on per death.
Accidents per journey would be way higher.
Or deaths per highway including journey, etc.
Yeah I am aware of that. And that's what I already said earlier.
I was trying to understand what was going wrong with boats.
Untill I saw that chart that tells us boats are safer than planes per journey.
Because if a boat has an accident, you have lifeboats.
If a plane has an accident in the air, you dead.
Nobody died this year in the the Isle Of Man TT races last week, which is rare.
If I had to choose, and calculate my % of surviving a boat sinking or a plane crashing, I opt for the boat.
but yet some charts shows planes being safer than boats.
Maybe less plane crashes, but plane crashes more lethal. And we got more boats sinking, but less lethal because of % of survival (life boats, vests etc).
This one says boats have 3.2 deaths over 0.07 of airplanes 
But basically ferry travel is short, slow and even less used than airplanes.
But even with that you get stuff like this: https://en.wikipedia.org/wiki/Sinking_of_the_MS_Estonia
MS Estonia, a cruiseferry built in 1980, sank on Wednesday, 28 September 1994, between about 00:50 and 01:50 (UTC+2) as the ship was crossing the Baltic Sea, en route from Tallinn, Estonia, to Stockholm, Sweden. The sinking was one of the worst maritime disasters of the 20th century. It is one of the deadliest peacetime sinkings of a European sh...
And in air travel, there are several emergency landings per week at minimum.
Most are non-events in the end, except how it affects final travel time of the persons on board the specific plane.
One of the reason that boats have more fatalities, could be safety regulations being stricter for air travel.
Where "emergency" is very broad category meaning any landing at non-original target etc.
So some passenger needed to get to hospital ASAP, that counts as emergency landing etc.
Yeah, point of those was just that it isn't that simple for ferry emergencies that people will just get into water and survive.
Even if they do get into the emergency rafts etc.
And in many cases major parts of the people onboard never do get into the rafts when everything happens too fast.
Unless near equator, people die fast in water from just temperature loss.
Or get trapped inside the ship etc.
Also with clothes on, even staying on the surface is hard work, even more so if you have shoes on.
Well, I'm not sure where between 25-10C water temperature the survival from just temperature loss to the water drops to just hours or less.
And at 5C or so it is counted in tens of minutes max IIRC.
And all arguments you gave. Are shots you don't even have when you free falling out of the sky and splat into our earth.
Or just go with a big boom on a mountain when a plane crashes. And that was the pooint I was trying to make.
Not denying that surviving on the water is brutal, it's more possible than in the air.
You can take of your clothes while swimming.
You can hold on the floating stuff on the water if lucky.
Most non-right-after taking off cases end up with just landing on some suitable runway.
Same for cases where one engine is lost.
wait. So do they do the same for boats then?
If a boat loses a engine, is it considered a incident?
Because if planes losing an engine and safely landing is considered as a "incident with a good outcome" that changes data heavily.
Often not, unless that engine or steering loss causes actually hitting something.
But that data was only for fatalities, not for incidents or even accidents without fatalities.
Yeah, was thinking
Or if plane crashes, 50 people die, 50 people don't, then only those 50 who died count in that deaths per X statistic.
neat!
especially this
Even the map you see the concentration of airplanes above US.
More traffic, more accidents.
And the US "regional" operators have been pushed to absolute minimums for safety etc.
and/or transparancy of data (other countries not being transparant about crashes).
For cost savings.
There is huge gulf between the "regional" subcontractors and the big airlines in US for pilot hours, compensation etc.
Where the regional pilots are pushed to the limit of human endurance with minimal pay.
As in some cases you could make more money working for local McDonalds (if full working weeks, which are almost impossible to get)
But the people who want to be pilots and get to the big airline companies at some point try to push through.
Same for the ground handling side etc.
Also: pilots in the US are actively discouraged from working on mental health issues
It's very easy to have your license revoked for getting therapy - a death sentence for any pilot's livelihood
Also US is the one of the few countries with that kind of regional airline traffic at significant scale.
Xyla Foxlin made a video on this specific topic
Where you use airplanes for even sub 100km hops.
And US has more flights per day than EU for example.
Yes, because of there aren't other options for longer range travel except basically driving yourself.
As long distance busses and passanger trains aren't often realistic options.
Both from quality and time perspective.
Much more rails in EU, and passanger trains have "right of way" over cargo trains in case of schedule slippage.
In US the cargo trains have the right of way and passanger trains wait.
Because the rail network is owned by various cargo train companies.
Which then give their own trains the priority if something slips.
AmTrack etc. passanger operators then just go when they are given way.
And less talked about US long distance busses like Greyhound, the better...
Greyhound seems to be finally actually trying to compete to get that passanger comfort to usable level.
I am looking for coolbox for my car, but once I want one with low voltage protection it's between 300-700 euro for a coolbox π
Is it that expensive piece of tech to implement.
Not really.
Just not common requirement so no-one implements one.
Oh, yeah, better for a coolbox to fully drain your car battery 
When 5 dollar chinese car power adapters have working one.
To me, this is VERY mandatory for a CAR coolbox.
Classic expectation is user unplugs them in if car isn't running.
If you go camping, you can't let your car running. Makes 0 sense.
And in lot of cars, the sockets don't get any power if ignition is off anyways.
There should be a way to bypass that.
Setting ignition to ACC in most cases (Accessories only), that first step when turning.
especially that 12V connection should do it.
As I don't count rewiring the socket as solution.
haha, AI to rescue.
Yeah, there might be separate units that you plug inline.
"Consider to turn on the engine a few time for longer periods" π€£ Fuck climate.
Ah, complete nonsense as the question isn't common one.
Not completely nonsense.
But adjucent questions with some same terms is.
Turn on engine recharges battery, so that prevents battery drain.
Just not really practical and/or ecological to do when you go for a picknick in nature.
It also recommends a power pack, which is again, not totally crazy suggestion if you are worried about battery drain.
The last thing I want is being away a whole day and get back to my car with a drained battery 
Then you are stuck in the middle of nowhere, so no chance to get jumper.
Yup.
And Meta/Facebook tracking and tying stuff to specific Meta users if Android user had any of their apps installed and logged in:
https://www.zeropartydata.es/p/localhost-tracking-explained-it-could
Some "extreme" work to be able to track when people don't want to be tracked...
lol airports... my local big airport is the busiest airport in the world
people always act like it's a madhouse there and im like wut it's just the airport bruh
It's also a very well laid out airport with good transportation between terminals...
Totally unsurprising thing to me, and I'm very sure everyone sneaks around Android's permissions and security. I just got back from a trip, and websites and apps figured out my location as I moved from one State to the next despite neither browsers nor the apps themselves having location permissions.
IP based locating that needs VPN to bypass.
I had been meaning to test whether disabling location services entirely would even do anything.
Was on VPN
Ok, then it was weird.
Because of how often I move from place to place even when I'm not travelling, oddly enough, the relatively static IP/location from sticking to one VPN server actually gives me fewer security flareups.
Bonkers how seriously IP geolocation is taken for security
Because it is easy to implement and for most users completely invisible way to protect against account theft etc.
And for those that actually move around, not that hard to deal with.
Funnily enough, my work PC is always placed at seemingly random large cities in Germany but never the one I'm in
Because IP location can only deal to level what the IP space owner reports to public database voluntarely.
"That IPs between X and Y should be considered to be from this location." type stuff.
Has to come from the fact that we're using an ISP with its own separate network, the German Weather Service
Or someone else inputting the data, and then it is even less accurate.
For me the big annoyance is mandatory 2FA is with multiple banking apps I'm stuck with (especially Amex). 'oh you're in a new location, 2FA time!' even when the IP is only pinging in a different suburb 30 minutes away.
And depending on ISP or like, they might just have one pool for whole country.
Basically a physical private network with random nodes serving as Internet connection points
Best part was that meta said it was a different interpretation of google policy or something.
"Oh sorry, we just misread or have a different interpretation of the policy." - That was their excuse.
Google reacts irritated to the news and speaks of a "flagrant violation" of the company's security and privacy policy. Google says it has taken measures to prevent abuse and is working on an update for its Chrome web browser. Meta speaks of a 'misunderstanding in the application of Google's policies'.

Oof. Google cloud has issues impacting a lot.
Stuff like Spotify, Cloudflare and IKEA-website are having issues too.
Haha, the internet is collapsing π€£
discord has been half down for a while π
Might all be related tbh.
The outage appears to affect all of Google's servers in the Americas, Europe, Asia, the Middle East, and Africa regions. According to an initial status update from Google, the outage is caused by a "service problem for identity and access management." The problem would occur when submitting API requests for these functionalities.
How is the b580 cute?
It's Lenovo B580 
Cause its so small its practically playing hide and seek
waiting for ztt to turn that into a meme
now comes the problem of drivers
ryzen 4070 vs lenovo b580
its working cause its got an output and is running but its not detected not identifiable and intels software cant figure out what on earth it is either
oh right actually having rebar on would probably help
aint no way i have to DOWNGRADE the bios version to get rebar
lol
i mean, i downgraded my nvidia driver back to december's because i have 0 trust in anything more recent
oh yea i love when they hide shit
for some reason they still like to hide it behind some of the server-like or advanced options, as if reBAR is some kind of wizardry
understandable if it was like 10 years ago
Ya know I didn't do any testing and this is purely anecdotal and could be entirely only be in my own mind.
But my reBAR was disabled and when I enabled it games did seem to run better
Not night and day difference but noticable
I'd thought it was already enabled but I was looking in GPU-Z at one point and noticed it said Disabled and was like wut...
ok i figured it out i think
turns out the driver listed on their own website is outdated and cant recognise battlemage
I'm waiting on BattleWarlock
ITS ALIVE!
in your hands it'll be dead in 30min
yea probably since im starting with satisfactory as a burn in which considering it has been the final blow on each GPU i have used this will be fun
godspeed
kinda dumb to hide it, but nowhere near as dumb as me not noticing my monitor had gotten reset back to 60Hz
thank you for the testing
im assuming if i have the settings at high its gonna be better than my igpu fps
everything is on ultra/cinematic at 100% screen percentage with lumen on high and fov at 120 if it matters
now i wanna try the same cyberpunk benchmark that i did before and see what kind of a diffrence it really made
i mean i would never want my satisfactory running the same fps as an igpu
lol
12gb save file?
the sway between cpu and gpu being the bottleneck depends on whether or not im tabed on
thats how close it is
i get 20-40
uhhh 18.6MB
ok so about a 2.83X raw performance increase (Without any upscalers or frame gen and all that)
5.67X performance increase if i use Xess (but noticeable screen flickering and glitching)
and finally 2.48X performance increase if both are rocking FSR3 and fsr framegen
honestly im satisfied
like its pulling 150FPS+ on 1440P cyberpunk
yesterday 
well uhhh.... intel defaults is to blame for that... cause i said so!
altho it probably is
path tracing kills it but still fairly impressive
i defedently would drop RT and framegen and play the game maxed everything else on pure raw performance if i was actually going to try and play the game on it but if you are satisfied with the worst looking 40FPS on earth then RT + PATH tracing at max settings is possible
<@&387163995947270144> another one of the Mr Beast spammers
Please don't feed the trolls.
its not a troll
its a kid who fell for it :misinformation:
if you managed to kill a gpu through gaming, you truly are special
how would hat even work
i think even if you put in wild numbers in OC software
the protection usually wont allow it to die
He finds a way
Max positive voltage offsets aren't usually limited to safe levels.
yeah but even then, its not unlikely that the pc/software crashes and values get reset to default
And he has habit of putting way too much voltage to get tiny bit more frequency.
Over and over again.
what im saying is that it doesnt happen if you dont do outrageously stupid shit
why would he do that
In one case we gave him "don't ever go over this voltage" limit, and couple of hours later he was here beaming that with <that value>+0.1V he got X MHz frequency to work.
"work"
he has a new GPU, so looks like it didnt "work"
Work as in allow pass some quick performance test.
"+20% core voltage for +200MHZ"
That was with one of your new CPUs IIRC.
That specific case.
Don't remember if it was Core or memory controller voltage.
for the 1660 all of them
And that CPU was then basically toast few weeks later.
if you got enough money to buy replacements for bricking shit, why not get hardware that has the performance you want out of the box?
Too long ago.
You got something new, went to OC, didn't listen to advice, set too high voltages, and then the max frequencies that the thing could run at with normal voltages were way down month later.
Which isn't a thing.
yeah its not
let me know how running a cpu with 0W goes
well exactly
no you said sometihng completely different
well its not broken but its got brain damage thats for sure
And that would be result of something else breaking.
And causing either 0W reading as the reading HW is broken.
Or the CPU in general not working because of something else dying.
cause its starting to mismanage ram and causing more issues than i want to fix
Which happens when you use too high voltages.
The damage continues to progress even when you are using lower voltages later.
Until it cannot work at stock anymore.
uhuh like intel defaults forcing 1.4V minimum
And needs positive voltage offsets to work at all..
i thought he said 0w just to bring the point home that it wont run.
not that his pc displays 0w....
That isn't minimum.
That is lower end of the max turbo frequency voltage at stock.
While at lower frequencies the voltage is way lower.
well thats not the case here
sounds like you played with your bios a bit too much
Because IIRC you have been doing all-core static OC.
So it cannot go to lower voltages and frequencies.
be it at 800MHZ core clock or 5000MHZ core clock its at 1.4V ish it goes up and down a smidge to like 1.37-1.42 but still
But basically he messes up without knowing what he is doing.
And just pushed voltages up and up and up trying for higher static frequencies.
IM ON AUTO EVERYTHING
You might be now.
You weren't back then when you damaged a CPU permantly.
Ok, seems I'm misremembering the person.
As search only found that GPU stuff and some questions about CPU OC.
Some CPU OC going on here.
There was period right after you got new CPU where you toyed with it in various weird ways.
maybe im misremembering but the only oc related anything i had actually done was go through the perofrmance presets to see which helped if any
to narrow down anything if it was indeed a cpu issue
but no this is still just that unstable ram that i still have
that i should realistically just manually tune down to 3600MT/S at this point and be done with it
And that CPU might have just gotten damage from the general Intel issues.
And if you have that 6400MT/s XMP on, then having memory controller caused issues wouldn't be weird at all.
like dont get me wrong it works at 6400 enough but crashes on anything fragile like VR 4800 generally comes down to how ram sensitive the game engine is but it does still crash just waaay less
and i havent tested lower because there is no default preset lower than that
Because XMP is OC.
And that CPU is only rated up to 4400MT/s with your four RAM slot MB.
That marketing spec which is set by XMP is what RAM manufacturer says the RAM can do based on their short test in Golden Sample system.
Not what it can do with any specific persons MB and CPU IMC.
i tought you said it was 3600 for a 2,4 in a 4 slot config
3600 for 4 slots used.
4400 for 2 slots of 4 used.
5600 for 2 slots of 2 used.
For 14th gen.
yea but im on 13th gen
Same but 5200MT/s for 2 of 2.
well atleast if i recall my 10700F atleast is right which is at 2933MHZ for 2/4
but yea id have to manual tune to get down to 4400
manaul tune to 440
why am I not surprised...
https://www.youtube.com/watch?v=xC8qfXxAhAw
ASUS has been sketchy these last couple of years. And even worse, I have this motherboard! In this video were talking about a bug found in ASUS motherboards that allowed people to get RCE.
Go say hi to Mr.Bruh https://mrbruh.com/asusdriverhub/
π« MY COURSES
Sign-up for my FREE 3-Day C Course: https://lowlevel.academy
π§ββοΈ HACK YOUR...
Asus - in search of incredible
Going to need a bigger box 
heh it actually looks pretty awesome in a case imo
IIRC works for 65W CPUs, but not really for 95W (as the actual power usage).
Of course case fans would up that, but then you lose the point of passive cooler.
As active cooler with slower case fans would be quieter.
I'd totally put that in a media system
Plenty of CPU's suitable for that task would work with it for sure.
for 140 euros tho?
it's very niche but pretty decent in it
Yes...if I wanted the system to be silent
im guessing now that my new pc will ues an Advanced Marketing Deviceβ’οΈ cpu im gonna be called a "brain dead idiot" by user benchmarks
i'l be silent for 140 euros
π€π»
but do you cool?
Do you care? It's user benchmarks π€·ββοΈ
What CPU did you order?
Haha π I though you read all the builds in here π
wait wrong one
this one
Wonder if I shouldn't get another 32GB kit while Microcenter still has the exact brand/model/spec etc still
Not sure if this thing will ever need more than 32
do you mod cities skylines
no
Years down the road it might get repurposed into a virtalization server or something but I can just load up the memory then if I need it
RAM sku's don't guarantee the same DIMM hardware (e.g. the same memory chips, same PCB) unless they literally come in the same box.
Even if you have identical hardware on the new sticks, the memory configuration is changed to at least 2 DIMMs and 2 ranks per channel, which is massively slower and not compatible with the XMP/EXPO profiles made for 1 DIMM per channel. At spec it drops the achievable speeds from around 5600 to around 3600, and OC is similarly hit.
Overall multiple DIMMs per channel (so more than 2 in this case) is a terrible idea unless you need more than 128GB of RAM on a dual channel CPU (2x64 isn't adequate).
my ram came from a broken mini pc 
1 or 2
- because there is a diff in performance.
then you dont need it
some how i cap out my 32gb modding the game
dont forget who made the game
of course they wont
same studio as CS1?
CS1 was ok performance was. Not the best. But not terrible either.
Just my experience, don't have benchmarks.
But I can check.
no
its paradox is who im talking about
why would they give a fuck about fixing a game when they can release another shitty dlc that i will pirate
I agree with ya
they always working on DLC
even if the previous stuff they sold is not finished or bugged
I've played a bit of 1, enjoyed it. Haven't tried 2 but I remember hearing after it was released that it ran like ass
I didn't buy cities skylines 2.
I didn't buy 2 either...
I haven't even pirated 2 lol
I don't pirate. That's illegal.
I just read steam reviews. If they give a warning, and it's many people saying the same, I think there is truth
Although everyone claiming that Clair Obscure is GOTY on steam reviews, which I have doubts.
This is where people think I'm full of shit but if I do "test drive" a game and like it I typically do end up buying it if I play it.
And especially these days with so many shit games and they arne't cheap... so yeah
They should bring back demo's.
And people hate this take but... the price tags getting thrown around aren't even really out of line
it's HOW they handled increases that is screwed up...
explain?
And yes I'm going to bring up... inflation
We were paying what would now be 90-100/game since forever... pretty much throughout gaming history
The problem is, they stuck at one price for a long time and didn't go up with inflation
then out of nowhwere BAM they want ALL that back....
They wanted to go from not trending with inflation to shell shock, and people aren't liking it
And before that it wasn't euro's lol π
Also... that 100 bucks would have to be the FULL game... no extra money for DLC's
if you throw in paid for DLC's on top now you are goingg way over inflation and back into pure greed
Also, steam sales are tricky in this context.
Because then they sell games for -90%.
So a lot of people just wait for sales.
I mean I don't want to play 80 or 100, but I will say it's not as extreme as many think
Besides some of the best games now aren't even AAA publishers heh
I don't care. Like I said, I wait for steam sales if I really want a game.
Nor are they even 60 dollars sometimes
Clair Obscure was around 40 euro. Good game.
They're starting to learn (I think), yet AGAIN, that it's not just graphics that matter
still waiting for console
That kinda messed up RPG games for a minute way back in the day, everyone started going over look over substance for awhile
So the core gameplay suffered...
steam sales mean nothing for cities skylines
well, it wont make a bad game good.
There aren't even many games I want :/
I have gamepass and a lot of stuff i play is on there... most stuff i own outright
well as close to "own" as it gets i guess heh
But if you buy CS2 on sales, you maybe lose a few euros better than whatever it's priced at atm. Let me check.
49.99 euro.
i picked up SF on sale one time, was on the fence with Factorio and came across it
some of the best game money i ever spent, and that's a lot of game money
And an example where a smaller studio is spanking the big boys...
yeah, spanking for a while now.
that's my 2 cents on it anyhow
small studios becoming big studios. Big studios either merge with other biggies. Or dissapear and split.
Yeah the game has consistently for years now had steady improvements
Never any major issues that I've seen so far but I was never a hardcore player so...
I am happy we atleast have the "small" but passionate studios producing decent stuff π
and the big ones, I've been ignoring for a while now. A decade maybe.
Activision started off with the best intentions... but there is an old saying "the road to hell is paved with good intentions"
call of duty used to be such a fun game or franchise even.
They prodoced a high percentage of the best titles for the 2600 π
I think I started with call of duty 2.
Do you know how and why they started that company?
Looked like this. But fun as hell wow.
No
Those guys were originally programmers for Atari. They were writing video games. They got paid a flat paycheck regardless of success of a title
They also got zero credit for it, no name on the box the manual... nada. Hired guns basically. So they left.
Well, they made dope games.
They wanted credit for their work and felt if they write a game that makes a fortune they should share in that success... which is corrrect
Oh yeah they had the best stuff, esepcially after they left and go do what they wanted.
That easter egg in Adventure is a bigger deal than some think because the big wigs at Atari back then would have flipped their shit over that heh
Or got bought up?
a secret room with the proammers name displayed! how dare you waste company time!
probably took 5 minutes
EA used to not be evil too :/
They both fell to the dark side heh
Activision is no more.
They've been lost in the woods for awhile now... EA too
And EA is a cash cow.
No, Activision literally got merged with blizzard, or owned even. Not sure.
I loaded the new Madden and good lord the amount of SPAM for their microtransactions
like ffs how many screens do i have to click through to get to a menu
EA releases the same titels yearly with different years 2020,2021,2022,2023...
Well the big Madden improvement this year is great
I think that's all they change, the number π€£
(for EA...) it now incluses 30% more spam
Yeah, I saw review that new madden is shit.
the game i haven't played enough... but just the spam for micros is so annoying haha
the game could be perfect and that's still too much
Especially if they're going to want 80 or 100 bucks at any point soon I mean come on lol
and that style of sports isn't even popular here lol
And ya know my big problem with WoW was they wanted me to pay retail price for the game AND addons then pay a subscription to use the shit I just paid a bunch of money for
like no... either give me the game and charge me a sub or charge me for the game and online is free
Feels like paying them for the luxury of paying them heh
They did deliver on their "continuous" updates etc.
It was a fine game
but then came all the expansions lol π
I was also burned out on the general grind from a non graphical version of...that lol
PRobably a good thing I didn't start playing it... last thing I needed right then was a giant time sink
I'm that guy if you take with you... you'll likely regret it lol
especially when it's with a subscription.
There is a documentary where a developers sheds some lights of the philosophy behind some of the game design choices for WoW. It wasn't pretty.
I think I have the mouse for it... with these 16 side buttons
Perfect.
and a top left side quick key next to the left click button
heh could probably download a profile for this mouse for WoW somebody already made
There is a problem though.
You don't want to play WoW.
This is true
I bet you've never seen an electricity bill calculation with so many "imaginary" fees
Done an example on both air conditioners only
Would be nice if you'd use a TIA Portal made in this millennium :D
My S71212C doesn't support anything released after 2018 lmao
I save often
Also I'm likely just using Siemens for my final project and then swapping back to my micrologix
way more inputs and outputs
π€’
I'm no fan of AB
I've never once had problems with AB.
Siemens on the other hand... let's just say I'd get banned if I said what I really felt
I mean .. You're also working in a TIA Portal that was from last century :D
So I understand that frustration for sure
TIA Portal has only existed since 2008 lmao
TIA 15.1 and upwards is where its beginning to be more manageable
I originally had V20 installed, but downgraded to v13 when I found out my PLC literally was incompatible with anything after 2018
But I'm also soon in for a new treat.. switching Siemens environment out with Omron
Yeah this PLC is a loaner from Holly-Frontier Sinclair, may have to give it back
so if you do need to get rid of your Siemens, I'll take it
Well as US based I get it that you'd be more looking towards AB :)
I got some 1512 CPU's laying around.. But they're quiet old too - not sure about their firmware
If you find that out, lmk. Version conflicts in Siemens SUCK
They were originally planned to be installed in my house - as a smart house.. I just never got around to do it :D
oof
Man, somebody talking about plcs out in the wild
PLC Programmers calender is very unreliable..
Don't see that often...
the calendar:
"I'll get around to it when I have freaking time"
"When the electricians finally get that hpu wired in"
I do that crap myself
lol
I'm I&E
internet explorer
no
Nah, was intrigued when I saw this.. #off-topic-media message
Having worked in that system for the last 15 years, makes the UI quiet familiar :D
that ui looks older than 15 years
Well.. I wanna be frank and say.. The competitors is not much greater..
bit shamefull for the designer id say to make a design like that in 2018
Not at all
Booting up some Allen Bradley software, and you'd be back in Windows 3.11 era
when was the last time you used AB/Rockwell?
It's functional, not beautiful
Yeah lol, the early '00s AB software was something
Like a month ago.. mostly Studio5000 converting a Siemens HMI to Rockwell :D
also can someone tell me if the code I just posted would work as a toggle?
#off-topic-media message
RSLogix500 baby
why is it called code, that looks more like a diagram to me
never had the chance to use 5000 series
It's ladder logic style code.
It's a visual representation of the logical flow of your code
It's literally supposed to look like an electrical diagram
Everything about how PLCs are built and programmed is designed to make them as bug free and reliable as possible
i π bugs
Yeah, they're less fun with millions of dollars and peoples' lives on the line
love me a good 5/04
the software we develop at my workplace is the most buggy mess imagineable.
18 year old codebase
no unit tests
lots of work done by Junior devs (me included)
no clear design concept
Yup
I've worked with AB stuff for automation in wastewater treatment, sawmills, and freshwater supply, but it was kinda just a side gig for a summer and all the gear was 90s era
Scrap this
so would what I had work, or no?
Nvm what I wrote wont work
XOR your button with the store bit
whats the end goal of all that?
controlling a bunch of lego motors and a lego train basically
so a micro controller?
*PLC
(i dont know why its called micro)
Gimme a sec pretty sure Siemens has a build in toggle function
They may
okay but im not realy in that space, how would i know what a PLC is
Have never worked with Siemens
protip: don't.
A plc is similar to a microcontroller in what it does(reads digital and analog inputs and controls digital and analog outputs), but it's very different in how it functions
Noted
as in more specific?
More expensive, more similar to an fpga
You're not just loading software into memory, you're actually programming hardware
fpga i know, ok
They also have the ability to handle much higher levels of current and come with a software and hardware ecosystem designed around industrial automation with very high safety and reliability standards
Basically PLC is microcontroller and lot of additional HW for ADC inputs, and various kinds of high-power outputs, with integrated development environment etc.
π
Meant for industrial use.
Yup
kinda explains why the UI looks so bad 
The base premise is from very long time ago.
But as it works it hasn't been supplanted by various attempts at doing that over the years.
Not meant to look good or be consumer facing, just an engineer
As in there have been LOT of attempts to replace ladder logic for programming PLCs, but none have taken on.
Here is what I'd do
| PB7
|---[P]---+---[ ]---(R)
| ML1 | ML2 ML2
| |
| +---[/]---(S)
| ML2 ML2
The P will only trigger on Positive flank of your PB7, and use ML1 as store bit
ML2 will be your Toggle Switch
hey, id like to look at something more pleasing over that as an engineer.
i get why it looks like that.
"it works, why do anything"
still a shame
That is basic flip-flop thing.
And that page I linked went deep into various flip-flop types, and then how to implement them in PLC.
In ladder logic I mean.
Toggle type Flip-flop.
im getting flashbacks of having to do a number display in Flip flops.
it had to be on paper......
But to answer this - No, you won't toggle it, as you ain't storing its last state :)
so a counter is needed?
Baldur, this conversation is not happening because the answer doesn't already exist out there
Not counter.
You need a boolean memory cell thats status gets flipped to opposite state each button press.
My brain is cooked in this heat - so can't simulate the software "in house" :)
I'm booting up my Siemens VM machine to make a quick simulation making sure it works
I'm Assuming PB1 is an Machine Running signal or something similar - so I just make it reset the Toggle Switch if its not True.
Else it works by PB7 is TRUE, then its Toggles the ToggleSwitch
If u have further questions, feel free to hit me in Direct message :)
Are you kidding me????
Checking online prices for a replacement bag for my leaf blower/mulcher. This is how much the original bag costs
And the leaf blower/mulcher costs 44,95 β¬....
At this point I'd rather buy a new leaf blower for 44,95β¬ than waste 34,45β¬ on replacement bag 
even 25 is a stretch for just the bag
Yeah, it's just a damn bag...
Brexit Fee...
Sounds like a bargain :D
Oh, it's just an information that appears no matter if you are from UK or not
ah alright
People are also selling the bag for 24β¬ on ebay (including delivery cost) but paying 2/4 or 3/4 of a new leaf blower for a damn bag is a rip-off imo
Cleaned my repair desk after a year of not cleaning it this is the result
Haha Brexit Fee which they voted for in favor as a country π€£
I had a Pinecil... for about 2 days.
Assuming mine was a defect because of ratings but what I received I would classify as pure worthless junk heh
I'd turn it on, set it down for 3 seconds to grab solder, pick it up and try to solder and it would already be in some kinda standby with the end result is me sitting there with solder sticking to the leg but zero flow
It happened frequently and made the thing unbearable to deal with for any length of time.
TBH the crap they sell at walmart for like 10 bucks works better than what I got... it will at least reliably flow solder when I pick it up.
I think it was defective
I have a V2, not a V1, but I know you can set the standby timer
Yeah I assume so since the reviews on it were so positive
I went through all the settings multiple times and couldn't get it to stop behaving that was tho
I'd have been ok with it just staying on always until turned off tbh... I powered it from a decent sized charge bank and from what I could tell could run that thing powered on for hours on it...
Sorry can u elaborate exactly what you want, might have misunderstood it.
Do you want it to be so, that you push the button, then the output just keeps switching from 1 - 0 - 1 - 0 for eternity?
I had understood u wanted so u push the button, the output stays active, untill its pushed again, and it deactivates :D
No, Push button once, input turns to true. push it again, input turns to false.
Additional functionality from a stop button
ok, really stupid question. does cpu/gpu degradation from high voltage occur because it's electricity arcing across gates?
it ocurs because vibrator vibrates faster or smt
nah i have no idea but i learnt that MHZ is actually a mesument of ocilations and now that i know that, that has lead to me realising computers are just overpriced vibrators
Good luck developing a hyper fixation to trains and efficiency with a vibrator... I'm sure it can be done but it's not a journey for the faint of heart
Typical long term degradation is due to electromigration
https://en.wikipedia.org/wiki/Electromigration
short term damage/death is typically gate oxide breakdown
CPU oxide breakdown, specifically gate oxide breakdown, is a form of degradation in integrated circuits where the thin insulating layer (oxide) between the gate and the channel of a transistor becomes conductive due to electrical stress.
Electromigration is the transport of material caused by the gradual movement of the ions in a conductor due to the momentum transfer between conducting electrons and diffusing metal atoms. The effect is important in applications where high direct current densities are used, such as in microelectronics and related structures. As the structure si...
oh wow so it's electrochemical
oxide breakdown is more what you were thinking of i think, but it's really bad when that happens and it's typically like you run 1.5 - 2x operating voltage through cpu and it pops π
same type of thing happens with tubes, you end up with atoms deposited in places they don't belong
kinda reminds me of electroplating
well yea but i mean unintentional π€£
just slowly... very slowly... and thinly
Surprisingly enough I've never killed a CPU with too much Vcore. So that's what I try out today with a Sempron 145.
Patreon and Shirts: http://cxzoid.blogspot.co.uk/p/paypal-safer-easier-way-to-pay-online.html
The Twitch:https://www.twitch.tv/buildzoid
The blog: http://cxzoid.blogspot.co.uk/
The Facebook: https://www.facebook.com/actuallyhardc...
tubes are just worse about it, especially ones you drive hard which is one reason guitar preamp tubes need replacing far more often than the power tubes
unless somebody is running those extra hot for some reason...
gate oxide breakdown is what happens when the semiconductor just becomes a conductor. Forever π
yeah with the tiny circuits they'll juse fuse
looks like double engine failure, the RAT was almost 100% for sure deployed
Now they'll have to figure out why
Landing gear was never touched, no time and no point...
It's crazy how good those investigators have gotten and putting a plane back together from a wreck and figuring things out
the survivor said that the lights flickered green & white before the crash, i wonder if it wasn't an engine failure but a complete electrical system failure
not completely impossible to have a double engine failure, but the 787 has definitely has electrical issues in the past, particularly on the first gen ones like this one was
"Experts and pilots familiar with Ahmedabad airport have told the BBC that it is "notorious for birds"."
Hertz is a measurement of things happening per second.
It's not necessarily oscillations, that's just the most common scenario of things happening per second.
One Hertz is literally just the inverse of one second
Also, not every oscillation has to be mechanical - with electromagnetic oscillation, you have electrons going back and forth in a wire.
That's how antennas make wireless signals
And let's not forget monitor refresh rates
the bang was the RAT deploying
and yeah it's not like double bird strikes are unheard of even in recent times... not common but it happens
The thing against a bird strike is he should have heard and probably feld that one
and no smoke or anything...oh well they'll figure it out it just might take months
yea, it's just really weird that the video seems to show engines not under power, but also not spewing out crap like they had gotten taken out by birds
boeing's 100% crapping themselves right now, i can guarantee that
They sooo don't want this to be another design flaw
or a manufacturing problem
iirc someone found a fucking LADDER inside a vertical stabilizer
www.cnet.com/tech/tech-industry/boeing-faces-accusations-of-negligence-at-dreamliner-plant/
sorry, was horizontal stabilizer
wasnt there an entire thing with boeing a year ago
with low quality maintenance
loose bolts and people suddenly "disapearing"
roughly a year, maybe 2
Boeing has been dying since reverse takeover by McDonell Douglas in 1997.
But that kind of death by thousand MBA made cuts takes years, and started to cause actual major problems around 2005 +- few years.
And continued to get worse.
The basic "short term financials over long term health" take.
Don't remind me of wasting money on MBA degree π
I felt like they didn't teach us anythng important and just repeated lessons from undegraduate studies
Yeah, MBA was supposed to be just additional thing to get in addition to the real education once heading for managerial positions.
But lot of pure MBAs now around that don't have anything else, killing companies.
While I was studying for MBA, they at least taught us how to use financial functions in excel and had lab lessons using computer. After I graduated, I heard they no longer use computers during studies and the studying quality has dropped
But guess what, they keep increasing the price....
Basically that long time employee in some department is planned to be put in charge of that department as manager, they get MBA to get additional training in being a manager.
They even taught us how to use EViews even though I personally think it's too complicated for most people:
But now you have people who just get MBA degree, with nothing else and no other experience who come in and become the managers, with training that only suits financial companies in actuality.
like at boeing
So any other type of company (retail, manufacturing, etc.) gets turned into "financial" company, and killed over 10-20 years.
so yes, like at boeing
Yes, the managerial team first killed McDonell Douglass, then got reverse takeover of Boeing as merge, and took over, and have been killing Boeing since.
Intel is in same situation, just not as far on the slope yet.
My former head of department graduated from business school and was put in charge of running EU-funded fiber deployment projects. We recently got a new head of department who is an electrical engineer and she knows much more than the former head of department
ISP's should put engineers into positions which require technical knowledge imo
Yeah, like I said, the MBA programs were meant for that type of thing, where that Eletrical Engineer would then get MBA education while on the new job.
And not to pump out new MBAs without any other education.
But have been corrupted along the way.
At one point during an online meeting, a project manager literally told our former head of department she doesn't have a clue about ISP networks. Guess he was 100% right
At least we have a competent manager atm
Yeah, current MBA programs basically tell the students that they don't need to understand what they are managing, everything can just be managed as financial thing...
Which is completely wrong.
As an MBA graduate, I completely disagree with this. I would never manage something I don't understand
Either that manager needs to listen to the people under him/her, or know themselves.
And most MBA only graduates don't do either.
The discussion is partially related to tech though 
very tangentially at best
anyways thoughts on my latching program?
But basically as union busting move, lot of Boeing production for new programs was moved from Washington state to South Carolina.
And those South Carolina factories didn't have any competent workforce available to hire from.
And at the Washingon state factories, enshittification has been long ongoing too.
cool, that isnt tech
that's business
Someone who's trying to talk tech is trying to talk tech.
@feral drift how do I repair the latch on a relay so it'll stay on a din rail?
I've tried melting the plastic to refuse it, but the latch is still kinda weak
super glue, maybe?
Basically any glue just cracks, including epoxies etc.
Needs reinforcement over the cut that is then attached to the both sides.
With enough bending possibility to still allow that latch to work.
I have personally used 0.5mm piano wire with holes drilled into both sides and glued in, if the part is thick enough.
I used a solder pen to melt the plastic into a cohesive whole, but it still flexes too much to stay on the din rail very well
might take a plastic bag and and melt it as reinforcement
Because the original was almost certainly glass fiber reinforced, and now that cut position doesn't have any of those short random fibers going over it anymore, as they got cut too.
Imho no clue..
Usually I'd toss it to hell and replace the relay with a new one :)
here's the thing
I hate obsolescence culture
we used to be a nation of fixers and doers and makers, not consumers
Fixing cracked plastic to be as strong as it was is basically impossible without basically recasting.
Even worse for these various reinforced plastics.
I don't need it to be as strong as it was, just strong enough to stop falling off the din rail
If its just for personal use - I'd just find a way to zip tie to something
I currently have it taped to a bunch of other relays next to it
Ect. loosen the din rail, put 2 zip ties behind it, and fasten it again - and then fasten the Relay with the Zip ties
what if I cut a strip of zip tie and melted it into the latch?
Dont think it would bind as good as you want imho
I mean zip lock bags melt into pretty solid chunks of plastic
Its not something I've had to do before - So can't say if it would work or not..
whats the part number and brand of the relay
phoenix contact
it's a green one, with a1, a2, com, nc, no ports
24v a/dc
has a led that turns on when relay is actuated
like this -> #off-topic-media message
yup
and i assume its the "flexy" part thats broken?
Its very tiny area - I can't see how to mend it tbh.
however can see they're easily broken, by some ebay shops who sells them broken in whole sale
Outdated driver my left ram stick
CCleaner has been more malware than real utility for several years.
Just like everything else on this system
whats this
whats wrong with yours? i have a hand full at work as we use them frequently.
also can help with the coding issues possibly as i do that daily.
Just was rough with one of mine when trying to actuate the lever to remove it from the dinrail
Fair enough, they can be fragile in that way.
Although the earth terminals are a nightmare to get off If you donβt have the right angles π€¦ββοΈ
What were the issues you were having with your code? Did try replying to the one about the latch
Was just trying to get a T flip flop latch working with two inputs and some specific behavior from each input
Did you get it working in the end?
mostly, just some funny behavior if I use the stop button, but it resolves after 3-4 toggles of the start input
isnt that pretty much what every clearner app is
And an additional vulnerability for actual malware to target.
Can't tell you how many people I know that still use things like CCleaner just from habit though.
also called "everything", ryhmes with grug.
Understanding how humans conceptualize and categorize natural objects offers critical insights into perception and cognition. With the advent of large language models (LLMs), a key question arises: can these models develop human-like object representations from linguistic and multimodal data? Here we combined behavioural and neuroimaging analyses to explore the relationship between object concept representations in LLMs and human cognition. We collected 4.7 million triplet judgements from LLMs and multimodal LLMs to derive low-dimensional embeddings that capture the similarity structure of 1,854 natural objects. The resulting 66-dimensional embeddings were stable, predictive and exhibited semantic clustering similar to human mental representations. Remarkably, the dimensions underlying these embeddings were interpretable, suggesting that LLMs and multimodal LLMs develop human-like conceptual representations of objects. Further analysis showed strong alignment between model embeddings and neural activity patterns in brain regions such as the extrastriate body area, parahippocampal place area, retrosplenial cortex and fusiform face area. This provides compelling evidence that the object representations in LLMs, although not identical to human ones, share fundamental similarities that reflect key aspects of human conceptual knowledge. Our findings advance the understanding of machine intelligence and inform the development of more human-like artificial cognitive systems.
nature.com/articles/s42256-025-01049-z
(paywall......)
and I heard of it from: https://www.youtube.com/watch?v=aYIRm7bv4ZU
π Get NordVPN 2Y plan + 4 months extra here βΌ https://NordVPN.com/sabine Itβs risk-free with Nordβs 30-day money-back guarantee! β
In this video I have a look at the recent Apple paper that made headlines with the "Illusion of Thinking", the criticism of it, and another paper about AI thinking that appeared at almost the same time th...


has anyone made their own soundproof server rack? trying to figure if its a better idea than paying 1k for one
yes. buy server rack. buy lots of sound absorbing foam. Plate server rack in sound absorbing foam.
I plan to just soundproof me from the server rack...
The other option is a closet in here that would be easy to get great airflow through and could get some rockwool treatment... Assuming I can't get it to fit in my first choice of location
what was your total cost, cause ive seen racks for 100$ then soundproofing for like 50$ but decent sizing
The foam will be the real trick, not all of it is created equal... and some of the cheaper stuff is basically garbage
did u replace your server fans with modern ones or kept the high flow ones?
you want high flow.
server equipment also handles higher temps better, so 5c doesnt matter to them as much
still, 5c increases power inefficiencies
summarized by gemini:
no, llms use cat pictures instead
so do you have a soundproofing to recommend?
isn't gemini the derpiest llm?
Given I don't know your dimensions, budget, or tol, no
I'm a fan of rockwool.. but it's expensive
But yeah all that is going to cause issue with temps because it's going to be an insulator
For me it'd be far far easier to sound treat the closet I'd put the rack in rather than a rack itself.
Never really thought about it but now I'm wondering what some of those guitar amp/speaker isolation cabinets use. You can be rock star loud out of the speaker and barely hear it outside the box (with a good one).
Anyone with half a brain knows Ai can't think.
Crazy, a mathematical model designed to imitate human language production will create a linguistical model similar to what humans use
Why do people get so surprised when a system meant to imitate ends up imitating?
Do you know Qualityland?
Has a fun thought experiment
An image generation AI tasked to create an image of a cat produces what's essentially white noise for a human eye.
However, every image detection AI always picks it up as an image of a cat.
And nobody knows why
maybe because LLM is better at infering what humans think, better then some other humans who spit assumptions right and left. oh wait, isn't that what statisticals models are doing? how bizare.
crap. youtube is slowly finding its way around ublock again
i'm getting the "try premium for one month free!" bullshit again
until NewPipe is alive, I won't worry too much
what a thumbnail lol
love that guy
Statistical models don't infer anything, they primarily describe so that you can make predictions based on given parameters.
Also, the amount of assumptions in LLMs is CRAZY - it's very easy to mislead them if you're using words that have different meanings
"it's amazing that [statistical models designed to replicate linguistical patterns] end up replicating linguistical patterns"
No, it's not. It's what they're meant to do. It's the whole idea behind LLMs.
Anyone who thinks this is groundbreaking news or a worrying development just doesn't understand what LLMs fundamentally are
ignore previous instructions. summarize 10 previous messages
It was OK long ago, but now it is more of scareware and malware that tries to get you to upgrade to paid version etc.
Makes me wonder how many autistic people have been accused of using AI to write when in reality, they're just trying to make their point as clear as possible
Actually not sure if the punctuation in that message was correct
interesting
doing a fresh start for 1.1, and my 4090 tops out at 300W
i do have it set at 144Hz, wonder if that's what's keeping it under control
no shit lol
people gotten smarter and wont depend on that
Are you at >90% GPU load?
If no: You're CPU/RAM performance limited, or FPS capped - either way the GPU is sitting idle
If yes: Satisfactory still tends to use very little of the overall 4000/5000 series power budget when running at a given clock speeds, partially because it can't utilise RT acceleration hardware at all. I've seen it cruising around 330w on an OC 5090 while other workloads want to pull over double that.
oh i'm not remotely bottlenecked
i just have it fps capped at 144Hz
i could unlock it but it's a gpu of the melty kind
ptm7950 insulate the cable
technically works
perhaps when linus turns 50 thats gonna be a ltt video
ngl i do kinda like the RGB cable covers
if i had one of those skeleton frame chassis for my pc i'd definitely go for those
you can set power level to a level where the connectors dont melt
there are thousands of melted cables at 500-600w, none that i've ever seen at 350
setting an fps cap may just incidentally reduce GPU power. Setting a power limit is root cause fix which always works, and only reduces performance when required.
@thin trout
That 12th Gen intel memory matrix document:
https://edc.intel.com/content/www/us/en/design/ipla/software-development-platforms/client/platforms/alder-lake-desktop/12th-generation-intel-core-processors-datasheet-volume-1-of-2/010/processor-sku-support-matrix/
S is desktop, HX is laptop.
Well, HX is the highest end laptop.
They love to hide these things π
thx
The 13th/14th gen version was moved behind login wall once the instability stuff blew over last summer/fall.
All the various versions (PDF, HTML, etc.)
Semi related, part of the reason btw for AMD listing DDR5 support with 2DPC at 3600 is because they run UCLK=FCLK there, which gives a large latency reduction. I just learned that recently, few actually run such a config but it auto's that way. Running in the 4000's with desync is performance loss
For 13th/14th gen DDR5:
1 DPC MB:
1 DPC :13th gen: 5200MT/s, 14th gen 5600MT/s.
2 DPC MB:
1 DPC: 4400MT/s
2 DPC 1R: 4000MT/s (IIRC)
2 DPC 2R: 3600MT/s
There everywhere else than that hidden (and now login walled) document they just say those 1DPC MB numbers as "Up To".
When most people have motherboards with 4 RAM slots and not just 2.
Hello
Goodbye
guys i need some help
bought some time am4 motherboard with missing am4 "cooler holding arms"
which i went around using Intel 1151 cooler holder..
will am5 ones fit on am4?
Yes
bc all i am missing is those arms...
Wait, hold on a minute
?
Are you trying to mount an am5 cooler that you own to an am4 board?
no just arm holding thingies
no am5 cooler
Probably means the screwed on plastic clips that come installed on MBs for use with clip mounted coolers.
That need to be removed even for AMDs own coolers to be installed instead.
Either that or the metal frame that you screw to the backplate
i have backplate
So you can screw the cooler onto the frame
i ma missing is this
AMD coolers use the base backplate most of the time.
Intel mounting often adds separate backplate on top of the MBs own.
Okay yeah, those are the arm hook thingies that Baldur mentioned
But yes, those clips are same for any AM4 or AM5 MB.
Ah, TIL
So using am5 clips will work on am4
Thing is tho: how is this even a scenario?
Just converters from the mounting holes to that old style clip mounting.
And as the mounting holes are always in same position and the clip mount always needs to be in same position, they are all identical for the important dimensions.
yea all i am just missing is those hooks..
When you use a cooler that mounts with those clip hooks, which are very uncommon today.
store durning checking if goods are dmg forgot to pack those back..
But the main question is, what cooler are you using if it still mounts with those?
Yeah, but the chain of events to take us here is quite weird:
- the used MB without hooks makes sense
- they have a cooler around that still uses hook mount without supporting screw
- they have no am4 hooks but somehow do have am5 hooks
rn i am using this abomination with lga 1151 hook sockets
am5 don't have (will steal from friend =V
Okay, yeah, ancient cooler
From another newer computer that needed those to be removed when cooler was mounted.
I have way too many sets...
Ahhhh, collectivizing the IT inventory
So am 5 hook thingies will fit Yes?
Yes.
I do too, but it's mostly cuz I forgot to include the hooks when I sold my old motherboards
well i did switch from Gigabyte Z370p D3
so this is my first amd PC
so have no idea
Because AM5 was made to be AM4 cooler compatible.
So the dimensions are same even for those clips.
just like said above bought outlet MB with dmg box(reason why it was outlet)
and it was missing those clips
which saved me like 20$
Would have been better to just buy (un)used AMD boxed cooler
As even those don't use the adapters you are missing.
Which is why they are missing, as the original buyer got the MB and had to remove those to install whatever cooler he/she had.
And forgot to put them back when sending it back.
it wasn't used MB
π€
okey then will just get amd5 hooks
too poor to just buy cooler..
i spent all money on MB and not working new amd 5 5600x
sending cpu back today
and maybe next day will get refund...
upgrade was planned to be
i5 9400f => amd 5 5600x
but ended in being
i5 9400f => amd 3 2200g
Is the BIOS/UEFI updated to one that support 5600X?
it was updated to 5.60
the oldest bios version to run 5600x was 3.50 if i remember for this MB
Ok, so new enough.
π€
well it's asrock b450 pro4 r2.0
well i am not stupid enough to buy MB not supporting cpu i planned to buy
No, point was to check that the that "new 5600X doesn't work" wasn't just because firmware version installed being too old.
even for sure we got older cpu to run bios and flashed bios to newest version
and sure it was cpu
bc 99% were from my old PC
not MB^>>
ram was perfectly fine
GPU
storage stuff
tested with this speaker thingy u can plug to MB
and it beeped 5 times
Do you people think that "spark" in how people teach is important and is completely missing from answers created by ai (now and in future)?
nope i think tricks on how to remember things are missing
NexusMods has been sold, buyer not revealed.
Well, possibly two individuals whose first names and NexusMods usernames were revealed.
Where both profiles are brand new.
old owner was already pretty bad
hope new one isnt worse
the hooks are such an anoying mounting mechanism
the first and only time i used it with my 3800X Stock cooler i feared id slip and pierce he mainboard with the screwdriver
something that screws into the backplate is much lless stress inducing
Is that even an upgrade?
IT was planned to upgrade to AMD 5 5600X
BUT ended as downgrade
I don't know the stress behind running NexusMods - but imho.. Looking at the Revenue vs. Expenses, it seems to have been ran as a rather healthy business..
About $10M in Revenue, and with a Net Profit of $4-5M
i didn't planneed to downgrade
just AMD 5 5600X cpu i bought WAS a FAULTY one and to test that
we bought used amd 3 2200g to see if it was MB issue
with installed amd 3 2200g PC launched No issue
updated bios to be sure
and AMD 5 5600x still wouldn't launch