#๐โtech
1 messages ยท Page 62 of 1
chances are theres an xoc bios that runs on your card
you already are in the overclocking server
if you have a bios switch theres no risk even
well
probably
not worlds best lel
its just very good
also heres some food for thought
potential 7990
top 10 for people who dont own 2 3090tis
its also not even in the top 100 if you include 3090s
if i could be a filthy cheater i would lel
i have a 1kw bios loaded on my 3rd bios position
dont use it tho even for oc
step 1 buy a kingpin or hof oclab
step 2 flick the switch
but wasnt a kingpin or hof ocl
you need both :/
sure but good luck booting
the core wont bootstrap an unsigned vbios
nah
you can flash it on most 3090tis
if youre talking about the strix leaked bios
no i mean actually modded
there is a leaked signed 900w bios for 3090tis
ye
and youd use this not the kp bc the kp is only supposed to work with 1 voltage controller
yeah the hof and kp only work on those cards as they are very custom under the hood
well not that custom lel
but the bioses are locked to 1 specific voltage controller of the 3 or 4 that are allowed
if you can run 4750 on your card you absolutely have a golden 3090ti lel
esp on air
a full 1500mhz over the current record on ln2
No it doesnt
Even if it tried the GPU is physically incapable of putting more than 12 volts on the core
It probably doesn't tho
The voltage controller probably only supports 1.8v
Might be a 2.5v option tho
nice
also i forget who it was, i think @vagrant pagoda
the gddr on the 3090 uses 3w at idle and 100w at full load
the 3080 with less than half the memory chips uses alot less than 10 watts
yea i cant match that
my water itself is at 28 lmao
its not
lel
Is AC software any better these days?
what
the software isnt present on my computer atm i think
But always hated the software
i never used the software once i set it up
You need the software for the remote to work completely so I always had it
the quadro doesnt have a remote
the point is you dont need the software on the quadro once its set up
it has alot of features tho that all seem to work as expected
and once renamed in the software it seems to save to the hardware
yea thats a bit low lel
Sponsor: Montech Sky One Lite on Amazon https://geni.us/sux2
In this hands-on with the AMD Radeon RX 7900 XTX, we mostly talk about the potential usefulness of the thermistor embedded in the reference design right-most fan and how it could be used as not just a GPU feature, but a case layout utility for end users. We'd love to see more cards ado...
new amd gpus look promising, even for people who record stuff, gotta wait for real world tests
Can't wait for Musk to wake up and realize he's breaking Cali employment laws with his mass downsizing of staff overnight 
@latent mural
i came here to see if theres any drawback using the overclocking mode tachyon whatever its called
is it related to polling rate at all?
and this calculation are done in cpu or kbd
can u use without software in background?

<@&375234916776017933> is having the software running an requirement for tachyon? sry tag
i did ctrl+f search before tag
In the future please use our support forum. ๐
#1019755933959733258
But no - As of right now nothing the keyboard does requires the Wootility app running. It's all in firmware. You only need Wootility to configure things.
maybe there is a forwarding service you could use?
yes i will, i use proxy, i live in brazil
but is bit sad that its a preorder process
Shop Micro Centerโs Great Selection of Computer Parts: https://micro.center/7re
Check out Micro Centerโs Top Deals: https://micro.center/kso
PDQ.com: Start your FREE trial now! At https://lmg.gg/PDQOct
AMD just announced their RDNA3 GPUs and dang.. the RX 7900XT and RX 7900 XTX look amazing for gamers.
Discuss on the forum: https://linustecht...
Better than I thought tbh
Eh... let's wait and see what the reviews say.
Though it looks like ray tracing performance is gonna suck again
@latent mural I looked into some more details
Yeah it wonโt be Nvidia level but still better than lavalace
and my NVME is actually a WD SSD
Ehh better than ampere I meant
I looked into comparing
WD the best one that they have the WD black SN850
is a lower cost then the NVME you talked about
Thats just from a website source with in depth about the device you told me about comparing others
with prices etc
The SK Hynix P41 offers amazing performance thanks to a new PCI-Express 4.0 compatible controller. In our Hynix Platinum P41 review we find that this drive is faster than Samsung 980 Pro and WD Black SN850. Thanks to an energy efficient design, there is no thermal throttling, even without heatsink.
Well the ray tracing from what they claim is gonna be as much better as normal performance.
So same % performance loss enabling ray tracing as RDNA2 which makes it not an option if you want to use RT
ehh no because it's stil less of a penalty and your base fps is higher
30fps less at 50 hurts more than 20fps less at 100
but we'll have to wait for some actual numbers
Itโs gonna be more like 20 less at 50 and 40 less at 100 comparison from their slides.
But yeah, without actual numbers hard to say.
After looking at the pic they showed and say the parts are corsair
I can most definitely say its this
for the memory
about 64GB of ram
so 2 orders of that
do you need more than 32gb?
@ashen spindle depending on the projects for modaling blender work and animations for video production for content
Yes
mainly the pc is for everyone's use
so video animation and such like that for youtube content better memory is better
makes sense then, you just will most likely not see it run at 6000Mbps though
Idk why they got corsir
most likely cause of RGB lighting
Since I did want to style my pc
to fit the theme of a galaxy
true
Idk man pretty much all the stuff is finshed
its not prebuilt
its a custom build
This wasn't assumabled before sold
this was paid by me first
before being assumabled
meaning its not prebuilt
Plus it's going to be colored and designed
Unlike prebuilds that doesn't have designs that you want on it like
your own organization themed galaxy on the case
why I said custom build
gonna get some rest now
The first known case of native 16-pin to 16-pin 12VHPWR cable (not adapter) melting?
https://t.co/eVhw0x1JFC
Now that is interesting
Also since it seems to have better internal construction than the nvidia one
One 12VPHWR is about as good as two 8 pins with good quality cables. Or probably 3 with smaller gauge cables.
oh and it handily beats the m1 ultra btw, 5.3 Tb/s of bandwidth if more than twice of the m1 interconnect ๐
What does?
5.3 and 53 is a small difference 
Yeah it is faster than that. Though itโs cache so not entirely the same thing.
yeah it makes it even more impressive because it has to travel more lol
also jfc amd is able to make these cards at such a lower prices if you compare 4090 size vs n31
Sponsor: Montech Sky One Lite on Amazon https://geni.us/sux2
In this hands-on with the AMD Radeon RX 7900 XTX, we mostly talk about the potential usefulness of the thermistor embedded in the reference design right-most fan and how it could be used as not just a GPU feature, but a case layout utility for end users. We'd love to see more cards ado...
ok, i wont click the bait.. what?
Thermistor
Apparently it measures air intake temperature somehow
And you can fuck with it to make cool custom air presets
For fans
basically you can have case temp sensors without having to buy some expensive things
so you know if you set up a new fan if it was the wrong way if it increased case temps etc.
or measure intake air temp? nice
anybody know what to do for this error on ventura update?
tried couple of times but the same error always, has enough space too
with the red texts?
Right
I would definitely buy these over the 7900XTX 
Why not spend like 600 euros extra
Support me on Patreon:
https://www.patreon.com/der8auer
Save 10% on your iFixit purchase: DER8AUER10
https://eustore.ifixit.com/der8auer
Pro Tech Toolkit: http://bit.ly/2JOFD8f
Find my products at Caseking:
Delid Die Mate 2: htt...
Valve seem to be working on Peer to Peer transfer between Steam Clients.... FUCK YEAH
Wireless transfers of games between Deck and computer will be possible then
Peer to peer?
Torrent style?
That would be amazing for download speeds
I hope it's only for LAN though
Id imagine so, even just purely for licensing and DRM purposes
stemdeck to desktop distribution would be nice
Steam has DRm built in, so peer to peer is fine
Loads of games use P2P for updates.
I don't ๐ฆ
I am kind of happy with the XTX performance
It's about similar or just a bit slower
At what, half price? 
Huh apparently the results are skewed on TPU
Regardless, early December will be interesting
I hope EU pricing is not shit
you can already make the conversion for price
it will be the same as always, conversion and like idk 20bucks ton top
- AiB prices are bit higher
I am glad tho that AMD didn't pretend to compete with the 4090
But it really seems like Nvidia got it spot on, I thought the 4090 was cut down too much but full n31 doesn't even seem to match this cut down ad102...
And that's not to mention the raytracing not keeping up with their raster improvements
Now I want to know how RX 7950 X3D will look
In processors it makes a big deal
In GPUs?
me after you dont realise not everyone has 1000 euros let alone 1900
Overclocked 7900xtx
*most likely
Not certainly
ok, based on the data we got assuming there is not some fuckery ๐ it wont do much
and?
wh omegalul
Who cares?
The lad that said that "AMD RT perf will go over 2x"
Which... it didn't, it seems
so leaker with very spotty track record removed account
Shocked I say
so.. who cares
I don't know why people give so much time to rumours, it makes no difference.
Just stop giving it airtime, people make money from just lying. Why waste your time on it.
Tell that to people who still make headlines after Moore's law is dead posts a video
Greymon at least seemed somewhat reliable
you're the one who brings those up the most here lmfao
Got MLID guy is just so irritatingly bad.
I have more credibility saying RDNA4 will be 40x faster than rdna3.
Source: trust me bro
Who cares, i hate that i have to buy Nvidia for their Cuda cores to train neural networks and shit
Maybe it's an age thing but in the last few years it's seemed stupid to even give these the time of day
It's all the same bs sources, greymon was just more competent at removing the unbelievable nonsense
Both are using the same "sources"
I wish AMD sells like crazy and performs very well
I don't think GPU market is cooled enough to not sell good
All I hope is the drivers are solid
Yeah if you use CUDA you are kinda stuck there, although AMD made some brief mention of AI improvements but I doubt it will be enough to move over.
They are really consumer boards = gaming
Commercial boards = compute
These days anyway
Did they change their logo? I thought it was a Chinese knock-off but that might just be me haha
It totally wasn't me who convinced Tony to get that specific model
It's been like that for a little while, I think?
I can't say I like how emo it looks though
Imo it looks shrek not emo 
I still like the brand logo of my mic the most
Optimised for critical listening though, hopefully you can criticise everything you hear now ๐
They make good av recievers and amps.
Wait, wait
Marantz do make headphones?
They made headphones back in 70s
But they still make them
Marantz Professionals MPH-1 headphones feature powerful 40mm drivers and a comfortable on-ear design, perfectly suited for both long hours in the studio monitoring in-field recordings or listening to music. With their authoritative 1528kHz frequency response, the MPH-1 delivers the detail and acc...
They look similar to Shures
I wonder if they come from a similar factory
Oh fuck off MLID
They don't really do headphones anymore though. I don't think they did well
Today we analyze marketing stuff
And talk about a product that's 2 years out in the future
So he will copy and paste the graphs people on Reddit have done with TPU chats with AMD claims.
Solid work.
I wonder what he'll be saying about RDNA4
you know if you just stop mentioning him, it will mean less exposure for him, what you are doing is how he can make this into a job lmao
That it's faster than RDNA3. Nothing new
The real question is why do you read his Twitter
Ah I see, I'm glad I don't use Twitter as I generally avoid seeing these pointless people haha
I don't even have a twitter account
but there are some cool people on there
So it doesn't hurt to check out twitter like once every few days
It's like worse than Facebook these days, there are some useful people on it just a shame there are so many real idiots on it as well !
well i like em already
The only reason I ever turn on facebook is to check out the marketplace
A lot of peeps tend to flip their stuff for cheap
Solid, how are they?
What did you have before?
He said Astros A50s
well i did have some hd sixhundos before
but they are gone in the wind
(i just lost em when moving somehow)
Try to get Dag to sell you his HD800 for cheap 
and my astro a50 was just a standin for comfort cause wireless at home while gaming
i dont game much anymore tho
but for travel headphones these shure will be what i wanted
sound nice
fit well
and cable
I liked my a50s gen 3 or whatever. I have recently gone full circle again though.
Used to always buy "decent" headphones for my budget and then went wireless and enjoyed it for a few years it was a great change.
I then questioned it and then did an impulse buy of a secondhand topping a90 and a schiit bifrost 2 to go with some hifi ananda headphones.
Can't go back now haha
i have the gen3 and for gaming and wireless they are awesome
nice fit and theres some really really nice sounding eq profiles out there
i just mainly wanna get back some good headphones for listening to music and stuff due to the decline in my gaming time
The best feature I miss is discord volume control, it was great to be able to change the volume from game and discord chats when things change rather than faffing with windows volume slider or the game
Wireless > wired since the hd220
listening to laptop speakers when a dt770 lies next to me
Yeah no
Valuation is imaginary money
Have you tried the hdr220s?
No, and not interested either
I know you haven't because basically no one did because they were too expensive
Wireless being better than wired? Depends entirely on use case but generally wireless is a compromise, quality is less but portability and lack of wire is better for a lot of people
Not the Sennheiser digital wireless systems
Cable never bothered me, when they sound as good as my K702s then I will consider wireless not shit anymore
Full lossless digital wireless with solid amplification
The hdr220 are likely solidly better than the k702
Yeah no
Considering that's the consensus on hd600 and the hdr220 are literally identical to hd600s
Besides being heavier and better pads
How does that work then? I guess price wise is a thing as well.
It must have some latency still so music ok but video and games?
Although I did see someone selling Ananda-BTs for 500$ recently
I should take a look if they are still available
Why would it have latency?
Wireless? It has latency on transmission
Why?
Because physics?
Why?
yeah about the same latency as you get on a cable lol
This
cables are also a distance and not instant teleportation
wireless != bluetooth
last time I had BT headphones, the sound never got sent to them in games and sometimes when it did, it was laggy and barely worked..
That's just bluetooth
It's really not good
I guess if you have your own contemporary wireless standard it might be better
if you understand those graphs thats the one I have and that's like 12 year old stuff
Bluetooth has served me well. I had EarStudio ES100 MK2 for like a year and a half before they broke. The plastic casing could not handle my use. Now i think two years on Qudelix 5k
I'm tall so running a cable to my phone is awkward, so i go with a portable bluetooth dac/amp combo so i can run the cable to that
and have my phone in my pocket
But most non Bluetooth wireless headphones have inherit latency, it's less but it is present.
Like the a50s are wireless and have 15ms as one of the fastest wireless ones to my knowledge (with compromises for it).
If you have normal wired it's essentially 0ms no?
I take back what I said in general though, I will leave it as a question for high framerate gaming though as a potential issue. Be great for music though.
Still don't see how it's better than wired on quality as you aren't limited by amp and DAC being built in so you can make a better loop overall.
Again came to compromise as I said, portability for quality, it's a much less difference on that specific one but it's still a compromise
that idk, I just know that I dont notice any form of delay in csgo at 175hz with them or some cable headphones
That's only the newer ones that literally use Bluetooth under the bood
latency does not matter that much when listening to music ๐คท
granted I only compared with shitty hyperx cloud something and idk how good they are
The analog chain is shorter, and the entire loop is exactly known, the amps can be tuned to provide exactly the desired signal
As long as you aren't throwing away data, it's the same as building a dac into the headset, and you have the advantage of not having any electric interference, only emi which is alot easier to charqcterize
the only disadvantage is if you have an originally analog signal you have to convert it to digital first, which is why quin's headphones exist where the wireless signal is fully analog
there because theres no dac in the headphones, they have actually speed of light latency
the hdr220 are the latency of the dac, which is the same as an external dac would be
dont even remember how I ended up with these, dont think someone recommended them and I just tried them because it seemed good
so I got kinda lucky getting a good pair lmao
Makes sense that the DAC aspect is the main reason in the context of non Bluetooth ones like you mentioned.
I assume the reason the likes of rtings show 0ms for wired is because there is an additional hop from audio port to DAC then headphones that is adding the difference as well.
no, because there are headphones that do analog to digital conversion and do internal processing
they show 0ms because they cant (more likely just dont care) about measuring under .1ms
They do test it but maybe not in the decimals.
So it be the fact the headphones tend to bypass/turn off their additional audio processing when wired in instead of wireless. Like you can with av receivers, turn off signal processing and just straight pass through reducing latency as it just hands off.
Yeah makes sense that way thanks.
I wonder why no real headphones did lossless audio wireless like the sennhiser for pc then, seems like a missed opportunity.
Audeze Penrose audio system is pretty bad but the headphones are meant to be good. Always a compromise on bandwidth when it seems like sennhiser already done it.
Maybe it's a sad patent thing or I miss a logical reason
One of the side gens of the Astro series did have it
It's called kleer
Well, hopefully, RDNA4 will bring the RT throughput 2X-4X per compute unit. With the next generation in 2024, expectedly, NVIDIA will essentially create a revolution within the industry, by enabling fully RT/Path-Traced gaming experiences at native 4K60
man the amount of insanity some people have, nvidia beats rdna3 by a bit and now they think that shows how nvidia can suddenly just 10x the performance next gen??
Who?
nobody important
Is it the usual?
Makes me wonder how much quantum stuff will make things interesting once its past the baby steps
Uhm do you mean for personal computing? If so not within our lifetime I suspect
RX 7900 โup toโ 3 GHz, but donโt expect such clocks from AIBs PCWorld spill the beans on potential Radeon RX 7900 frequency potential.ย The rumors about very high clocks on Navi 31 GPUs were wrong, but is there any chance we will see a high clock on custom RX 7900 graphics cards? According to [โฆ]
That's like.. 500mhz extra?
Dunno if it'll give much more perf
500mhz is nothing to sneeze at
If it were to reach that number
Which without exotic cooling feels unlikely to me
Hmm
remember they have two different clocks now
frontend and shader clock
so could be for either one which would have different effects and scales differently
so might do 2% or 20%, likely on the lower side
It's possible that AMD decided not to clock it within an inch of its life again.
They get called out for poor power efficiency for cards like Vega 64 but that was in part due to them clocking it so high trying to match/beat Nvidia halo cards at the expense of efficiency.
It's why Vega on mobile and lower power devices was excellent, even the desktop cards when clocked down to their peak efficiency point were decent.
They get called out for doing that and then also get called out for not matching a much more expensive card so lose lose.
Gives partners room to do overclocked models though and a potential 7970XTX 3ghz edition haha
(Of course this could all be a lie and it just really is peak clocks for most working die yields)
Honestly now that electricity prices are an issue in europe I actually care how much my gpu consumes
Yeah it is a little more important this generation I think
Before it was more like: "oh this'll keep me warm in the winter"
honestly if I were AMD I would wait for the 4080 release then give first party perf/W numbers
Yeah i think it'll be more like 4080 vs 7900
When is the 4080 out? Wasn't that this month, I assume we will see soon.
It's probably why AMD didn't compare it as they can only go off Nvidia numbers at the moment
nov 16
What's the price of the 4080?
so if they beat it, show numbers that beat the 4080, then show how much more efficent it is vs 4090, so you hide that you lose in raw performance but still make valid claims
1200
So amd stays winning on price
At Finland it's priced from 1800 - 2000โฌ
Atm.
its not priced at anything yet since its not released yet
placeholder prices can be just that, placeholder
so we can just compare MSRP
I've found a 4080 TUF for 1509โฌ and it's release date is 16.11
Not long then, I think AMD has taken a bit out of Nvidia sails with rdna3 but only for the 4080.
On rasterization it looks like it will be 10 - 20% behind 4090 but is 600 dollars cheaper.
Ray tracing is obviously the big one still, looks like it's near 3090ti which is ok but in context if you use most raytracing games then you have to consider that.
The 4090 16gb is much less than the 4090, it had what like 60% less shaders or something so it's already cut down significantly from it and is still $200 more than AMD.
RDNA3 sits comfortably in between 4080 and 4090 it looks like
Acording to the website
and 1500โฌ is pretty much msrp, but could still change till the release on the 16th
Yeah and with AMD revealing their prices might make Nvidia change things up
Probably not but
yeh
tbh they could afford it, they are so much cheaper to make than a 4090 lol
could be half the BOM depending on some numbers, but I assume someone with more info will make a video about that soon that would be interresting
Yep we'll see
The 4090 has 68% more CUDA cores than the 4080 16gb...it's incredible as it's also the same clockspeed so there is definitely a big drop in performance and won't be made up by clock speed.
It's definitely miles ahead of the 4080 but less than the 4090
Pretty crazy since 4090s could have been gotten for under 2000โฌ
Can't imagine Nvidia keeping the price when it's $200 more than AMD.
Makes sense why they pulled the 12gb 4060 tier card haha
Nvidia can do Nvidia thing and sell things at prices I'd deem too high
People will still buy it
Just because "it's the best available" in an absolute sense
Intel arc better value than nvidia GPUs 
idk amd has a lot more capacity now and nvidia doesnt want to lose marketshare
intel is losing to price cuts on old gen parts lmfao
Yeah but the prices might change when they're closer to release ๐
Maybe we'll see. I think they'll keep a premium price on their halo product
Wdym?
I mean Nvidia is the apple of GPUs, they have correctly spent their time marketing and making it know they are kings on every aspect of it so the basic consumer will just pick it because the halo card won.
They incorrectly assume because the top tier card beats another top tier card that ALL their cards must be good
sure but 4080 costing 200 more for less isn't a premium they can afford really, if it loses to the 7900xtx, or 300 if it also ties the XT
Yeah I'm only talking about the 4090
It helps that they actually manage the performance crown, they didn't sit on their hands unlike Intel Vs AMD in the cpu market
I would tbh argue Intel is more like apple than nvidia
Should have made that clear
none of these companies are like apple tho
like actually
nvidia natively supports 3rd parties
Marketing, first party and proprietary standards to leverage market position.
Sounds like apple to me in a way ๐
and a huge pile of stuff is available on github too
The 4090 prices are from 1999,90 - 2799,90โฌ and there's 2 KFA2 4090s in stock at 2599,90โฌ
nvidias marketing actually includes methodologies that could theoretically produce the numbers they say
I mean Nvidia business practices over the decade have been to strongarm their position and not really "open" standards.
I don't blame them as they are a business, AMD has to do the opposite to get technology adopted but still it's not ideal
apples marketing is all about the feeling
if the only requirement to be an apple is to not have your entire stack be open source, then we have a lot of apple companies suddenly
this
Yep, but Nvidia successfully marketed raytracing as a must have to feel good with GPUs even when they couldn't achieve usable framerates on their first gen.
https://www.kfa2.com/kfa2/graphics-card/40-series/4090-series/geforce-rtx-4090-sg.html Very thick cards ๐
NVIDIA Ada Lovelace Streaming Multi
but
they literally could?
laughs in RTX2060
just because you didnt like the framerate target of 1080p30 on a 2060 and 1440p60 on a 2080ti, doesnt mean it wasnt fully useable
the difference is apple slide are showing the basic m1 outperform a 3090 while using only 15w lmfao, nvidia might use dlss and all but it makes some sense at least
yap
Apple so actually produce good hardware though, and market it very well to think people need X and it's the most important thing since they were born.
I feel like Nvidia manages the same things these days
do they?
or do they produce good software?
the answer is neither
they produce synergistic software and hardware that makes "both" look good
Yes? They do both good software and hardware.
You can't deny it really, there are designs I don't agree with but technically their chips are beyond the competition and the phones surrounding hardware is well made.
are they?
So a good product?
Software and hardware are both very important, sum of its parts and all that.
product yes
but hardware and software individually? no way
their hardware and software are independently pretty
but they are not independently great
they are low functionality on their own
Nah I think you are mistaken here.
Their mobile chips are significantly ahead of Qualcomm and Samsung designs, hardware is better fundamentally.
ah but its not
its significantly less capable
but
Care to explain?
what it is capable of its significantly better optimized for
it just doesnt have general accelerators
it has accelerators for exactly the tasks it needs
But you want your phone to do phone things, why does it matter that it can't do something outside of its remit?
because i for example need my phone to output video over its charging connector?
the iphone has exactly 1 video clockgen
the iphones also have exactly the camera isp ports for the number of cameras the flagship of that gen has
the video codec support on the iphone is also relatively limited
compared to the other vendors vpus
I don't see how that's a problem?
It doesn't make the hardware bad, it is great hardware for what it needs to do.
It's faster in browser rendering, video recording, gaming.
Can you not output video??
like you may not care, but its only because ios doesnt/cant use these capabilities that it doesnt matter
the hardware itself is specialized
Still tho in games iPhone while get higher framerates
the product overall is good
not doom
basically it's easier to make A better if you dont have to use any chip area for B because you dont implement B
not quake
Yes it's optimised for its use case and the general use case. This makes it better at the tasks it's doing than a general Qualcomm Snapdragon SoC.
because you arent allowed to run doom and quake
Cmon man u know damn well I'm talking about regular ass mobile games that run on both android and ios
I'm not arguing about the walled garden and restrictions they impose for more sales as I hate apples business practices.
It's why I can't use apple, I only have to use it when building and shipping apps on iOS at work as I'm forced to ๐
yes but there are plenty of games that outright dont run on ios...
Yeah it's a shame
apple locks the software down so you cant expose the weaknesses of the hardware
which is fantastic for a product
but the soc itself isnt magically more capable
No they lock it so you are forced into their services. it's fantastic for sales ๐
They still make great hardware.
What you are arguing is not my point.
You may as well be saying my microwave can't be a good microwave because it doesn't have an oven function built in. But others do.
I don't see why other games wouldn't run well on iPhone as well as long as they are ported natively
yep its a win win win at least for apple
you said their chips are beyond the competition
that is incorrect
I said they make great hardware and you said no.
I then also said they are ahead of Qualcomm which they are.
yes
no
products are not just hardware
They also do get acces to the newest node every time tho
And the hardware is also great?
the hardware is good enough
This is what I don't get your argument for saying no. It does what it sets out to do in a better way, it's good it's competent.
I'll hate on apple all day but you are really being pedantic here on the actual point ages ago xD
you are thinking of the product
basically, you are saying their hardware is ahead of qualcom, but it's more that their score in certain tests are higher becuase they are specialized for that
this
so if you say they have the higher score on benchmark X, sure, that does not come as a result of hardware better in general
Render a webpage faster on phone X than Y, it's good hardware at the tasks it does.
yep
yes, you are correct, for the task, not as hardware vs hardware comparison
But the actual generalised tests are faster, it may have features lopped off it but it does the tasks you need
my android renders webpages 2-3 seconds faster than my iphone does
and it takes the iphone 2-3 seconds
Android is OS ,not the hardware though ๐
ok my s21u
What browser is that with as Samsung allow others.
But again, hardware is more than just the chip. The enter phone is hardware
Because chrome is the IE of the modern day and ignores standards? ๐
yes and its not the fault of the a14, its explicitly an intentional ios optimization
i use chrome on the iphone too
basically if you could translate the hardware architecture into a score, apple's score would be lower because while they have a bigger number in one test, they lack/ignore other parts
so it does it's part well, but if you say you look at the hardware standalone, it's not ahead
the actual reason is that my s21 keeps the webpage loaded in the background so that when i click on chrome its already open
the iphone on the other hand yeets it nearly as soon as you open another app
I didn't know iPhones did that lmfao
basically compare it to the intel gpus, not great as all around, but in certain tasks it smokes the competition, that's how apple made their chips, by removing everything that would show up as not winning
it depends on how much ram the app uses
I did originally say apple makes good hardware generally and was told no. That covers compute chips and the actually phone which is all "hardware" which is why the point was being picked apart wrongly.
"technically their chips are beyond the competition"
As a person who's browser always shows the tab number as ๐ this is interesting lol
each tab is separate in chrome
so on both my s21 and iphone the background tabs get yeeted
True but it seems to still have some sort of preview snapshot saved
It's much better that way anyway
I mean this is apple being a walled garden and its own OS choices.
Your Google Chrome on iOS is using WebKit rendering not blink that Google uses.
This is forced by apple to keep control of it all.
Apple force things to be saved when switched task instead of running generally for power optimisation
but again its more that they make specialized hardware with software that doesnt use the weaknesses of the hardware so the PRODUCTS are great
Specialised hardware can be good hardware no???
and also i dont dispute that the hardware is good
the hardware is "good" its just not "great" or "the best"
the products
Ok I said they do both good software and hardware and you kept saying do they? And said no haha
the hardware itself definetly isnt the best
If people do not use the weaknesses of the hardware it basically does not exist
the first time you said good was 2 minutes ago
before that all you said was great and the best
So for the right person apple hardware is the best
apple products
dont say hardware
4:48 mate. "Yes? They do both good software and hardware.
You can't deny it really, there are designs I don't agree with but technically their chips are beyond the competition and the phones surrounding hardware is well made."
Ok I'll stop now it's pointless I don't know why I waste effort on a silly point overall let's get back to other tech talk
These 2 are inseparable in that context
so just say products
Semantics
yes
Tomato tomato
If I get you a phone it's technically hardware and software it doesn't have to called a product
oh I just got my email I send to myself exactly one year ago about what the rumours said back then
funny how well it was timed with the amd launch
not perfect accuracy but hey, thank fuck the 3k pricepoint didnt happen lmfao
Lol
Ada Lovelace and rDNA 3 cure world hunger and cancer?? That be the best rumours haha
spot on besides navi 33 tbh
Think you should start a YouTube channel with that, yearly "leaks" haha
nvidia is not above 450, but it was more about amd having better power draw numbers so it being 355 instead makes up for there being an >
actually i didnt see that 3k
why did you think that?
idk I dont remember where I got leaks one year ago from lol
The 450 we'll give u as it's not that far off
but that was with performance more being like 2.7 rather than 1.7
Well if mining didn't die along with the energy crisis I'm pretty sure that was at least a possibility albeit tiny
I think once we get AiB models and see how much they get from memory oc and such we can see if a refresh on top seems realistic or not
Why would it be pain? It is the flagship. Maybe a 7970 next year
im coming around to 3 link 384b total bus between 2 dies but id much prefer 2 link 512b dual die
Idk what else amd would have they already used the most expensive letter twice
nvidia has another 20ish% perf available above the 4090
ehhhh
Is that Nvidia figures? Haha
no thats math
12% of the cores are disabled but they wont scale to 12% more perf
12% more sms and 9% more memory bandwidth
along with a siiiignficant amount of cache
It's a massive die they probably don't have enough usable full dies to use yet.
It also won't be cheaper than the 4090 it will just be a TI with small gains
Why would it be cheaper?
its aloooot larger than 3090 to ti
it wont be if theres nothing to compete with
yeah but it would have to reduce clocks to keep more cores up even if power draw goes up
the 1080ti was an absolute monster thanks to nvidia misjudging how good vega would be
Like why would Nvidia ever sell a higher bin for less
1080ti again
I mean that's not them misjudging AMD that's just them executing a great design plan in the years it takes
yes
so if amd doesnt have anything over the 7900xtx then that would be pain
Eh 4090 can be expensive for all I care
no the point is a 1080ti could've been a thing with like 5% less cores active for the same price, ofc not a completely different design
be better if it wasnt
I want reasonable video cards to be affordable
be better if the 4090 was affordable
Yeah but to beat it they just enabled what was possible and sold them with a nice price tag premium.
If the design wasn't good for a fully working die then it wouldn't exist. didn't matter than Vega was off the mark for it to be possible
no it wasnt fully enabled lol
I miss Hawaii days
the 1080ti was 11/12ths of a full core lel
almost exactly
was like 93% or so
Eh ๐คทโโ๏ธ that thing is a powerhog would be nice if humanity as a whole could reduce our usage instead of increasing it
they were already selling a full core at that point
I meant is a full die wasn't good it wouldn't be possible, not that it was a fully functional die, poor wording on my part.
As in the maximum die was a good design, they had that much headroom to enable with working dies.
there are always full dies laying around
nvidia still makes bank from them at low yields
Of course, that's why they charge a fortune for them.
And the poor ones get binned down to us peasants ๐
nvidia could sell the 4090 at 900 and it would just be less research budget
Eh the peasants get completely different dies I think
Could being the strong word haha, they won't as it's a halo product and they own the market
yes
which is why i hope theres a product above the 7900xtx
I member when top tier cards were ยฃ350, those were the days!
8 years ago 290x launched haha how time flies.
I think you're hoping for too much I'd say there is probably like a 7950 refresh at some point
Not that much more
speaking of no competition, nvidia had many gtx 480s with 512 cores available for a potential gtx 485 or 490, and they just never released it bc amd had nothing before the new node
i absolutely agree
doesnt mean i want the 7990 any less...
Would that be to just buy a cheaper 4090 when they cut the price?
AMD won't be releasing a higher card for a long time to be honest.
Only with bigger stock clocks maybe
i want a cheaper 4090ti
the 4090 isnt really capable of 4k240 and its too expensive to buy and not be capable of driving my monitor
its not far off tho which is why i think a 4090ti would work
I see, I like how AMD is just there to try and keep Nvidia prices lower but not to get the sale.
It happens every generation xD (no hate, buy what's best it's capitalism )
I wonder how Nvidia and rtg r&d Budgets compare
if amd releases a 7990 tho id buy it in an instant
I have a sneaking suspicion rtg is far lower
its not even close
AMD is like a hobo compared to Nvidia budget
alot of nvidias budget does go straight into ai research to be fair
Considering that I'd say they are doing pretty good
hard to say since rtg is now split into doing rdna and cdna, nvidia also splits stuff though, it's hard to thinkg amd has more
Nvidia 5.2 billion dollars
AMD 3.9 billion dollars ( was 2.8 year prior)
That covers CPU and GPU....
amd as a whole might have more than nvidia puts in gpu
but rtg vs nv gpu nv gpu is alot more munz
It was only like 1.5 billion a couple of years ago though, they finally got some money around to spend significantly more
the other thing tho is that microsoft's and sony's research is on their books but helps amd
and the money that goes into those partnerships are huuuuuuuge
getting pensando and xilinx also wasnt cheap for amd the last years
Intel spending 17 billion on r&d also puts it into perspective the combined fronts it's competing against
I feel like many people thought amds ati acquisition to be a blunder ending up with amds departure from the foundary business but in hindsight looks like it worked out well
gotta remember that intel rnd also includes silicon fab dev which is hyper omega expensive
idk that shit has too many layers to see if ATI acquisition was the right choice or not
absolutely agreed there
Oh yeah for sure it's not cheap, good point for context
AMD buying ati was wrong I think but right at the same time xD
if you throw tsmc rnd in there it probably dwarfs intel
only worked because saudis stepped in to pump cash into GF, they could only make chiplets because GF gave up on 7nm and allowed them to use TSMC if they keep using GF wafers for other stuff
just so much stuff that at any point could've gone either way
4 billion this year apparently
that surprises me
So you could call Intel's budget 10 billion maybe as a rough estimate which is still massive
think tsmc lowered rnd for later half of this year
Certainly a lot of factors but I don't think amd would have had better luck with their own fabs
with n3 crashing and burning im not sure its the time for that...
they didnt lel
oh N3 is already gone, N3E is a replacement
their fabs were money sinks
Ye that's what I'm saying
n3 will not see a product anymore pretty likely
yes but n3e isnt final yet is it?
and these new nodes are all about product specific design rules
i still find it crazy that nvidia made two different designs for h100, old method and new one using ai, and the ai placed transistors ended up with a 20% smaller die
The problem is AMD CPUs at the time were excellent, on merit they were destroying Intel.
Unfortunately Intel bought them out of the market with silly anticompetitive payments to system builders like dell (and servers) to avoid using AMD which killed their sales and their money.
They had to then survive on a shoestring budget and the graphics division was successful for this but couldn't carry on indefinitely.
Eventually the swing with Zen meant they could finally invest more in GPUs and the fruits of that is only really just being shown finally
Doing designs on a shoestring budget against companies significantly bigger and bought the market is hard haha
It's why Vega was actually not bad in context, it was scalable on the market and done on a tiny budget.
It wasn't amazing but it did its job in compute and in games
no it didnt lel
its job was compete with the preeminent 1080ti
which it wasnt capable of
it was supposed to be 16gb and a cheaper titan
Not really, it was for server work and sold plenty in server and render farms.
as evidenced by that being the product it released
that wasnt the intention nor what koduri wanted it to be
It's interesting idk if it was in my seminar or if I read it somewhere but it seems we should be trying to understand these ai algorithms more instead of just letting them go to town
yes
It's why he left he wanted more money for r&d that wasn't possible at the time
knowledge debt from using ai to make black box magic will have a bad end
agreed
and the technical lead at the company investing the most money into agi saying "we shouldnt worry about the safety until we have a system at least as smart as a toddler" is highly concerning to me
We will all be dead by then as bigger problems aren't being dealt with haha
i dont think we will all be dead by 2040
idk if putin and xi got the same memo as you got 
ok but i think the entiiire world becoming a nuclear hellscape is under a 50% chance
I mean global warming isn't stopping, that will kill off a few people by that time and the rest will be killed by growing divide between rich and poor then nuclear winters maybe ๐
global warming only kills poor people (which is why nothing is being done about it)
Who cares about sunny having his revolution against the new machines while we are fighting amongst ourselves for the last scraps of land
At first
When they can't get people to flip their burgers anymore they will notice
no, anyone who is still rich will eventually become poor thanks to the warming, then they arent rich any more ๐ ๐
anyways type c on rdna3 is awesome
since they are working closely with valve I wonder if they know valve tries virtual link again
i hope that type c becomes common on gpus so that we can get 1 cable monitors
or if thats just for displayport basically
i hope they do usb 4 basically
only skipping out on pcie tunnelling
Can someone tell me how long is the type-c cable in the box ?
anyone have experience with both unraid and truenas? and which you prefer? Unraid seems to be the more noob friendly choice, but lower performance.
define performance
if you are network limited, it doesnt matter
I want to do 2.5 gbe at least
1.8m iirc
and unraid will perform better when 1 drive dies
hmmmmm
unraid can also add single drives at a time
yeah truenas need a whole "block"
unraid also works fine with shingled drives which are getting harder to avoid now
infinitely longer on smr drives lel
yikes
Fyi unRAID dose ZFS
You can setup a ZFS pool outside of the unRAID array
I have the Unraid array as my bulk drives for backups
yea but then theres no point in paying for unraid
And use ZFS as the actual used pool
Very nice gui lol
hmm
Also with even sata cache drives doing 2.5g isn't hard
Hell good mech drives will saturate 2.5
yea just need a teeny bit of ram
Yep
Last one I built was using a Dell prebuilt tower. 4770 and 16gb with 4 12 TB and 2 1tb sata ssds as cache mirrored
It had no issues saturating 10gig
Using ZFS for the cache
With more ran it would do better
hows the speed of zfs with 1 parity vs raid 5? assuming 4 drives for both?
Raid 5 in whay
well maybe not raid 5, but default unraid using only 1 parity and 3 other drives
Raid-z1 is faster
ZFS dosent grow well
The way they implemented adding drives to a raid-z seems half assed
Also raid-z makes a bunch of drives one
unRAID dosent
They always separate, just protected
hmm
So generally the worst possible failure for unRAID is losing one disk of data
The other disk are completely uneffwcted
interesting
you are supposed to add and remove servers not drives lel
9gb overwolf kinda concerns me lmao
discord probably steaming a lot or moving files via it
seems normal to me
the discord without a logo is kinda spooky?
maybe an update changed some id and windows lost tracking?
Firefox really consumes a lot
I imagine YouTube for data capped peeps is probably a point of awareness
[web browser you use primarily] uses alot
It's mostly video streaming I suppose
lol
could be better discord or so maybe that changes the process
also have two discords, could be the streaming process?
Mine is definitely broken
it does
most vpns that arent windows native will add a virtual nic device
Discord 250gb+ not much?
streaming uses data faaaaaaast
i mean
if im streaming games and dont use as much as him how tf is he using discord so much
yeah, if you stream at 1080p or above at 60 fps with nitro that uses quite a lot
and if those streams are nitro quality that is an ammount you get quite fast
should be able to hit 250gb in less than 8 hours while sending video to a friend and receiving at the same time
for 6mbps its 12 hours of video
if its something ungodly like 4k60 then you are there even faster lol, assuming discord allows more bitrate
dont know what the discord bitrate limit is tbh
watchable quality for 1080p60 is at least 12 generally
which is like double of what twitch allows 
what
unless you are one of the really big streamers you are locked to like 6mps bitrate
oh was it 8, could be
I think it might be 6 for smaller ones
and more than 8 really only for a few, 30k viewers aint enough to get more 
oh maybe brazil internet is just too bad for ESL then 
but yeah gl finding one of the like 3 streams that do more than 8
big streamers probably
lirik doesn't have it, biggest german streamers with 30k viewers each don't have more
I remember there was on launch of Nioh for pc a channel that was somehow streaming at 20k bitrate or something like that, random japanese channel
man that was crispy
yea ive run test streams to youtube at 20m and they are sharp
i was testing my laptop out, and i managed to stream on battery on wifi for 2 hours on battery on my laptop
if only some big company that has good server infrastructure would buy twitch so they can afford more servers...
You can only upgrade Intel GPU drivers if you have an Intel CPU
Ffs Intel
Wait....
Now it works on any CPU?
This is confusing
who says it only works on intel cpus
It seems that is not the case anymore?
ntel told Ars Technica that it is possible for both Intel and AMD-based platforms to update Arc GPU firmware, and that Intel's Management Engine wasn't actually required for firmware updates.
first sentence aka they tried to use IME which might only work on intel cpus but is not needed
I wonder how these GPUs perform on Linux
Given their many problems on Windows currently
but it's all just buggy beta hard/software so wouldn't surprise me if that bug happens
just as bad
phoronix had linux benchmarks on launch
They only seem to work flawlessly on DX12 games
Any other DX version or Vulkan had stutters or shit performance
it's beta hardware that is for people who want to tinker and try new things
They are very tempting though....
Once they fix the drivers ofc
how are they tempting
both amd and nvidia have better performance at same or lower price
Do they?
I saw a ray tracing comparison and the Intel beats them by a lot
yeah if you want raytracing in some game that isn't using the nvidia RTX, since that only runs on nvidia and then you get decent RTRT performance but your base framerate is already lower than other options
@ashen spindlestill recommend? xD
keep in mind oled has a much higher powerdraw 
Wait wait ...
If a game says RTX, ray tracing only works on Nvidia GPUs
But if it says ray tracing, it'll work on both?
no because marketing mumbo jumbo just mixes it all up
most games have DXR now so it runs on all
but something made with nvidias own RTX plugin before DXR was put into dx12 or vulkan only works on their cards
Worth it tho
oh yeah for sure, but I also don't pay more for electricity than I did last year 
also you wont want to run that screen with a power limited gpu lmao
Interesting tidbits on RDNA3
AMD regarding competing against the 4090:
Radeon RX 7900 XTX] is designed to go against 4080 and we donโt have benchmarks numbers on 4080. Thatโs the primary reason why you didnt see any NVIDIA compares. $999 card is not a 4090 competitor, which costs 60% more, this is a 4080 competitor.
โ Frank Azor to PCWorld
AMD regarding FSR3 :
AMD FSR3 is not a reaction or a quick thing to DLSS, it is absolutely something we have been working on for a while. Why is it taking a little bit longer for it come out, that youโd probably hoped for? The key thing to remember about FSR is the FSR philosophy and FSR until now did not just work on RDNA2 or RDNA1 they work on other generations of AMD graphics cards. They also work on competitors graphics cards. It is exponentioally harder than if we just made it work on RDNA3. We really do want to work on more than just RDNA3.
โ Frank Azor to PCWorld
I wonder what FSR3 can do
Same thing dlss 3 does but 3 months later
So if we are to believe amd
Both Nvidia and amd decided to add extra frames to boost their algorythms at roughly the same time
Which is kinda weird
The tech has been in VR for 6 years
It's super easy to do it's just impossible to do well
Itโs like when fsr 2 was announced, people said itโs just amd pretending they have something to compete on paper with dlss2 and just started working on it, but then it was ready like 2 months later, you donโt make those things that fast lmao
Thatโs a really nice price ngl
Though 150% sRGB is weird
Kinda makes me wish I didnโt have a monitor already xD
the shure 840 is great love it
just needed a very slight touch in the treble but eq fixed that
otherwise super comfy and the travel bag is also superb. so is the cable
fidellity on these is heaps better than the kraken is used or the astro a50
I'd have been very dissapointed had it been worse than the kraken
ikr
this feels so much close to my sennheiser music experience and im happy
now ill have a happy trainride
Good thing you listened to me and not a bunch of people telling you to buy gaming headphones 
Glad you are enjoying them though, they are pretty affordable for the quality
they are literally what i wanted for a pair of travel headphones and with how rarely they will get used
ye idk gaming headphones seem to pretty universally be crap then again it's been a while since I had one
well depends
the hyperx cloud hedset I have laying around is not great imo but I got it for free, so good value 
my last set of gaming headphones were about 10 years ago so I guess I shouldn't judge
I know the hyperx are takstars which is ok ig
Ehh takstars are pretty mediocre
when you use an EQ in windows, does that override headphone virtualization settings?
You can get better stuff than takstars for similar price offering
Hey it's possible to be connected to two connections?
For example via my phone dare by cable and wireless internet of my.t box
Does it increased the speed of the network?
For my pc
Who
short answer no
I'm not sure what they mean with the phone part, but most likely no
Ah
Ok ty
no
well its possible but no speed increase
Oh ok
You CAN set up things like this, but it is NOT easy.
?
the only gain for normal consumers is a fallback connection if one dies
Definitely possible but I have no clue how to make it work reliably, back in the day I sometimes managed to get steam to download using both connections for example. Really helped with speeds!
Damn
Okay ty
basically only works with load balancing when single requests can be split to either connection and get routed to the same endpoint
lets say its not worth the trouble unless u need to start combining multiple gigabits into a single uplink so to say
But really what should be done is something called bonding but that I have no experience with
bonding is eh at best from my experience
is that the same thing? have the feeling that realtek setting gets ignored when I use an equalizer
I did say "ok I guess" which I wouldn't interpret as a big endorsement
pretty sure they both get applied seperately
android i think does the same
after installing the EQ I feel like the virtualization toggle no longer changes anything
basically wanted to give this a shot to see if I've been missing out on quality or not https://github.com/jaakkopasanen/AutoEq/tree/master/results/headphonecom/headphonecom_harman_over-ear_2018/Sennheiser RS 180
urgh maybe the AI overlords should take over asap if that means sounds settings just are perfect from the start on everything
how would auto eq work?
AutoEQ is a project for equalizing headphone frequency responses automatically, and it achieves this by parsing frequency response measurements and producing equalization settings which correct the headphone to a neutral sound.
basically maths to reach a target i assume
from my understanding yes
https://github.com/jaakkopasanen/AutoEq front page has more infos
from the picture also yes
and seems like you can also compute it yourself
wait but I'm guessing there is some variation when it comes to the same product basically tolerances and such#
so does it have refference curves for headphones and eq them to a target
I guess so
my issue is I have always used headphone virtualization because without it sound, especially conversation just felt wrong, but idk if a good EQ setting also fixes it or not, because now it's in the like an inbetween where I don't know if it sounds wrong or if I'm just used to years of "wrong" sound settings
hmm I think I used to have such a programm that made the headphones dead neutral don't remember what it was called tho
think @thick elk posted it a while ago here first and I just looked into it again recently
it definitely sounds better than without EQ and without virtualization, I just dont really have a reference to compare if it's better than no EQ with virtualization or not lmao
Support us on Patreon: https://www.patreon.com/hardwareunboxed
Join us on Floatplane: https://www.floatplane.com/channel/HardwareUnboxed
Buy relevant products from Amazon, Newegg and others below:
Intel Core i5-13600K:https://geni.us/lJgCa
Intel Core i7-13700K: https://geni.us/Gpuqh3
Intel Core i9-13900K: https://geni.us/AieGiL
AMD Ryzen 5 7600...
Good lord
The 5800X3D slaps
What if they re-released 5800X3D as 6800X3D and gave it DDR5 and AM5 support?
How much would that add to its performance
thats not how things work
How hard would it be to just re-release it with different pins to adjust to the new motherboards and add DDR5 support
"what if they did this hpyothetical thing that has insane amounts of variables that we dont know"
but also why bother doing that when you just make zen4 v cache
all the changes, validation, steppings etc. for having one specific procressor work on different socket and with new IMC?
so a new IO die for it?
I am afraid 7800X3D will be too expensive 
So I'd rather have the 5800X3D repackaged for same price as when it launched
you do realize a "6800x3d" would cost like quadruple what a zen4 x3d will be
How so
all the cost saving would be gone because you need to make validation, packaging, new memory controller, different socket for a single cpu sku
it's more than just changing the pins to make that work
To rerelease the 5800x3d you'd definitely need to change a lot of things to get it to work on am5 that'd also change the value propositions a lot
You can't just do ddr5 on am4
Weirdly according to HU findings, R5 7600X is a good value
I guess it's only shit in EU where those things cost absurd amounts
no it was bad on release because b series boards were not out yet so you had to use expensive boards only
- some older boards from alder lake are not price adjusted so cheaper in euro relatively to newer stuff
I see B650 is still shit
Cheapest ones
Oh no, there's this one
But that's a .eu shop so not local 
Cheapest one locally remains 204.10โฌ for B650M PG Riptide
And I really don't feel like going with ASrock
On other news
Musk does what his little brain thinks is a good choice 
eh twitter would't be much of a loss tbh
I feel bad for the unemployment levels
But maybe whoever they fire will get poached by more respectable companies
software engineers are very saught after
you better be running that ram with an nvidia card
hwo ever you turn it, it is an important platform for lots of peeps, and probably do not want to leak DMs. security is important
Join The Full Nerd gang as they talk about the latest PC hardware topics. In today's special episode we are joined by Frank Azor, Chief Architect of Gaming Solutions, to talk about the Radeon RX 7000 announcement. And of course we answer your questions live!
Check out the audio version of the podcast on iTunes, Spotify, Pocket Casts and more so...
you know im somewhat of a typescript programmer myself
it spreads entry into a new object
... is the spread operator and basically splits an areay into its single values or an object into its properties
o
entry is an object returned from a prisma db query which doesnt let me use the setting as an index so i recreate a standard js object to access the setting referenced more dynamically
and since the function safe guards te setting to a string array i can just do this
ud still never do this in a production environment and just make smarter functions with a switch or smth
Nixie tubes <3 <3
Today I'm admiring the largest Nixie Tubes I've ever seen on Millclock's new ZIN-70 clock.
LINKS https://millclock.com
AFFILIATED EBAY LINK: https://ebay.us/3yYNgX
Many many thanks to Millclock for supplying the clock shown in this video.
00:00 Background
02:23 Un-crating and set-up
06:34 Features
17:09 Size comparison vs ZIN-18
17:55 Case ...
I wish there was a way to cover the circuit board, cause this house is dusty AF
๐
I love how these look so much
I always post in the wrong channels ffs
"AMD has doubled the compute throughput by adding dual-issue SIMD to their design, but the core count has not been doubled. This is why the Stream Processor count is 6144, but effective single-precision compute performance has increased to 61 TFLOPS."