That brings back nightmares about removing IDE cables. You couldn't pull the cable itself because it could damage it, which meant pinching both sides. When you finally got it loose, your hand would slingshot out and you could cut yourself on the sharp edges that used to be in cases.
@@qwertykeyboard5901 Either you're too young to have dealt with it, or you had some of the nicer IDE cables that had attachments for pulling on to avoid injury.
I like that, Gabrocki, but I like the video of machine gun even better! LOL, a gun spitting out numbers! He did a great job editing that! But oh yeah, Red Power Ranger, you have a good point there!
3:33 explaining clock recovery encoding protocols like 8/10 might be relevant for future videos, as it helps viewers understand how high speed communication work without a dedicated shared system clock between components. Also useful for understanding why 1Gbit=100Mbyte. Other relevant topic: Could be interesting to also cover the actual pins on a cable, e.g. USB, Ethernet, SATA, and wtf they actually do, and cover differential signaling and how it reduces interference problems.
Dude! You have one hell of a UA-cam channel. That energy is exciting. I appreciate what you offer in learning- fun graphics and easy to understand. Keep up the great work.
Sorry, but saying "USB got one data line in each direction" is false. USB 2.0 and lower only has D+ and D- as signal lines, but this is a differential pair. Meaning, that it basically carries the same information at the same direction at each time (half-duplex), but the signals are the inverse of each other to block outside interference. From USB 3.0, the cable got two more pairs for each direction (SuperSpeed tx/rx), but the old pair is still there to make your statement false and to confuse everyone.
This video had a few good points but was marred, and it could confuse people. Had you simply mentioned _"parallel to serial compression"_ (like USB) requiring (a chipset) doing processing of it, it would have cleared up the "magic" mystery of why and how people managed to get from the data rates of parallel to those of serial. The pay-off and trade-off would start to dawn on them because it isn't "magic". You could then have simply "name dropped" Fourier mathematics, and then people could at least look it up in a book or online. There was a reason parallel existed (high data rates and also acting as the first GPIO), and yet form your video people could walk away thinking serial should have only ever been the way forward and that somehow there was no hurdle to get the benefits of parallel into serial. You don't even have to "explain" the complicated stuff. Just "mention" the word and then people can have a look. The notion of pins being prone to damage is *not* an argument against parallel when comparing to serial. Not only can you wire to (and thereby adapt) the shape of something else like an rj11 or rj45 connector for a parallel (so the size and pins can be smaller), but also a serial (e.g. a great big RS232 null modem cable) can be about the same size as a parallel printer cable.
I needed ALL OF THIS ENTERTAINMENT. Reminds me of how engaged I was watching Bill Nye the Science guy as a kid. He made me fall in love with science and you make learning boring info easy.
Fun fact: Australia's first computer, CSIRAC, used a serial bus for moving data around internally. This was back in the late 1940s, when 2000 valves and 1000 IPS was impressive.
Future Video suggestion: Subwoofer types... the subwoofer must be on the floor or on the table/desk? Why not under the desk or why not on the table? Does the subwoofer speaker design matters? and why? Can the vibration damage the computer components, ex: hdd, ram, etc.?
2:56 im always bending my internal usb 3.0 connector and no other connectors lol? I plan on 3d printing some mounts to hold it better once its in because when it sometimes gets pulled or when bent out the way it comes off wonky and bends pins
I've used a similar example of the rate of fire of a shot gun and its dispersion of the shells loads to a automatic fire weapon, to explain some of the differences to people who don't really get the differences so easily.
Fun fact: You can get a PCIe x4, 8, 16, etc card working in an x1 slot by either sawing the extra pins off the card or cutting open the PCIe connector on the motherboard. I had to do this once, to get a video card to work in a PC that only had 1 x1 slot remaining. (The x16 slot was occupied by an HBA.) It worked reasonably well, although the drop in bandwidth/performance was apparent even at the Windows desktop.
well 12 years ago was in 2005 and USB was basically already the standard and SATA was gaining popularity over IDE. That and with IDE there was always a chance that the pins would bend if you didn't pull the cable out exactly at 90° and it was a huge chore to bend them back. Which, although it usually doesn't require IT assistance, is still a pain in the ass.
But remember that most companies have old hardware and it takes time to upgrade everything, we still have some Compaqs with P2, P3 and P4 runing(those fuckers wont break) they are used in deposits and cargo bays to show some stuff, so is not necesary to upgrade them for now
you dont usually remove parallel ports as often. thats why most of the external parallel ports have screws and screwholes and the internal ones are, well internal and not plug and play. if you were to plug an IEEE 1284 connector into a port multiple times a day it would break in no time. but usually you don't actually remove or insert a plug a whole lot.
This brings back memories from when PC parts were more expensive, and repairing parts was more common rather than disposing of them, and delicately straightening those bent pins on drives, cables, and motherboards. Speaking of pins, I still have a large jar full of gold CPU pins that I used to extract from junked PCs. I should smelt that down into a gold bar sometime.
A lot of serial communication stuff is clock and data ( i2c , i2s, spi) and usb is NOT receive and transmit, but Data+ and DATA- , one wire is inverse of the other... same for other things.
You didn't mention the speed of chipset meaning there was a limit to baud that each chip could interpret the data in serial. It was cheaper to put them in parallel and feed the more expensive CPU that way than to buy a chip fast enough to talk in serial at the same speed. It wasn't until these chipsets became more powerful and less expensive that serial was able to take over. It also helped that we were reaching limits with parallel speeds anyway due to the aforementioned crosstalk, etc.
USB does not have one data line in each direction. USB uses differential data signaling, this means that one data line can be mapped to 2 physical wires, one being D- and the other D+, so when one wire goes high voltage, the other goes low voltage. Differential signaling allow serial communications to be faster because they are able to send data with lower voltage swings without being affected by ambient noise. As for the clock it goes embedded in each data signal and is reconstructed by the receiver.
Parallel will catch up and out perform serial eventually. Sure crosstalk is a problem with ribbon cable but twisted pair advances along with proper shielding can offer more bandwidth along with TDMA style data synchronization and other transport level data timing matching technology.
thenoobcannon Well, I suppose a machine-gun is just a gun that operates by using mechanisms for reloading, ejecting, and whatnot instead of needing the input of a human every step of the way. And automatic is a general term I suppose, so he's not that incorrect. So long as it functions by itself without someone having to baby it every step of the way I suppose it's automatic, and if it does that with mechanisms, it's a machine. I dunno, though.
Great video! :) I would find it very interesting if you could do a Techquickie on computer component power consumption, particularly that of CPUs and GPUs and what constitutes it. What does the electricity actually do inside the components - what is it needed for? Also, the GTX 580 from around 7 years back and a modern GTX 1080 Ti both have similar power consumption yet the difference in performance is, of course, absolutely massive. Similar trends can be seen on CPUs as well. What has allowed these huge advancements in power efficiency and can we expect this trend to continue still as time passes on - is there still much that can be done to improve power efficiency? Where are the physical limits as to how low power consumption could be dropped in the future for a given level of performance (a certain amount of FLOPS)? Also, how are CPUs used on mobile devices and laptops able to achieve such low power consumption ratings despite still being fairly powerful? It would be very interesting to learn about these things! All the best to the whole team and thank you for making these videos. :)
It should be noted that there are dual port SAS drives, which have multiple channels, whereas SATA only has one, so that's another part that kind of blurs the lines a little bit, like PCI
Hey Linus, since you are talking about all those different ports, I was thinking about you making a video about firewire. It was said that it was faster than a usb 2.0 back in the day.
I thought crosstalk was between a receiver and transmitter (side by side) wires. Cause the transmitter has much more signal in the beginning of the wire and after some distance (that the receiver's signal is not faided that much) had electrical interference with eachother.
i definitely don't miss the days when you had to pin the master/slave etc and when you had to type in all the volume information for a hard drive into the bios, Plug and Play all the way.
I’m assuming this means that each wire in a USB cable (for example) handles one set of information bit by bit while other wires can handle other sets of info simultaneously. Parallel connections have every wire containing one bit of each byte so they’re effectively working together all the time whereas the wires in a serial connection operate independently
So wait if these are so technically different how do ide to sata adapters works? Or does it use a dram chip to store the data and then convert it from serial to parallel?
I thought crosstalk was between a receiver and transmitter (side by side) wire. Cause the transmitter has much more signal in the beginning of the wire and after some distance (that the receiver's signal is not faided that much) had electrical interference
There are definitely some limitations to standards like USB. Some industrial applications actually do work find parallel connection more reliable if they need a specific thing to be refreshed as quickly as possible. A serial bus really doesn't lend itself quite as well to real-time applications, and a couple of unusual examples would be USB keyboards not having NKRO, and the fact that an SNES controller can be more responsive on the original hardware than a USB knockoff can be on a PC. USB in particular has the limitation of relying on the OS polling a certain number of times per second, where some parallel standards can use straight up hardware interrupts, so the device can tell the computer about state changes as soon as they happen rather than waiting for the next poll. I'm kind of glad the PS/2 ports are still around.
USB keyboards not supporting NKRO is a common myth. USB mice are also not any less fast or reliable than PS/2 mice. USB has matured and evolved enough at this point that the advantages PS/2 once had are pretty much moot.
I have a whole bucket load of what appears to be DB 37 Serial data storage cartridges (predecessor to USB thumb drive by the looks of it, to me at least) and I'm trying to figure out more about them. I have these because my Grandfather custom built a few barrel organs in his life time, usually barrel organs are analog instruments which use rolls of perforated paper to produce music, my Granps being the absolute 1980s-1990s tech wizard he was, completely rebuilt the guts of the organs, turning them digital. They read MIDI files off of these extremely obscure, dated 37 pin cartridges (I think serial DB 37), the cardriges themselves are totally black plastic, about 3x2x0.75 inches with right angle edges and corners, and absolutely no branding or text on the device other than the hand written labels of what songs are on each cartridge. Anyway, I inherited all this stuff and would love to figure out more about it all, I'd LOVE to be able to hook up some of these cartridges to modern hardware and see if I can write any custom music to be played on the organ. I'm an actual musician, which is why I inherited all of this, so it would be a blast to give some new life to my Grandad's old project.
That brings back nightmares about removing IDE cables. You couldn't pull the cable itself because it could damage it, which meant pinching both sides. When you finally got it loose, your hand would slingshot out and you could cut yourself on the sharp edges that used to be in cases.
Not to mention all the dust that gets trapped with those wide ribbon cables. I've seen some real nasty stuff build up in there.
Am I the only one who has never had this issue?
@@qwertykeyboard5901 Either you're too young to have dealt with it, or you had some of the nicer IDE cables that had attachments for pulling on to avoid injury.
3:32 I too install my GPU while my house is on fire.
Son:DaD THE HOUSE IS ON FIRE!
Dad: Hold on i'm building a PC
Dad takes his own sweet time, checking all of the components, debugging, while everyone around him panics, running around like mad.
Suddenly, dad sees an error. He grows suspicious.
"this is fine"
He gets engulfed in flames, and hes like, "hold on, i'm playing the witcher 3"
My sister heard you talking and said "Is that bob the tomato from veggie tales?"
He does kinda sound like bob from veggie tales actually...
LOL! It's close, but I think Linus' voice is a bit too high and squeaky to be Bob.
.....i never fucking thought about that....OH DAMN HE DOES?!!!!
Sean Ramey #
Hey, this comment was in the latest "mean comments" video on LTT
You guys should do a collab with LinusTechTips
Hah
Mr Tom Waffles [COD Mapper] Wat they already are
or Taras kul
That's the joke..
Mr Tom Waffles [COD Mapper] that would be lit bruh, why hasn't that happened yet?
I loved the machine gun/shotgun analogy. Genius!
Should've been machine gun vs line firing
he forgot about 19th century multi-gun/barrell... weapons (pre-mg's)
I like that, Gabrocki, but I like the video of machine gun even better! LOL, a gun spitting out numbers! He did a great job editing that!
But oh yeah, Red Power Ranger, you have a good point there!
That was for the Americans in the crowd who until then had no idea what was going on.
gabrocki literary was
about to comment that like a second before he said
3:33 explaining clock recovery encoding protocols like 8/10 might be relevant for future videos, as it helps viewers understand how high speed communication work without a dedicated shared system clock between components. Also useful for understanding why 1Gbit=100Mbyte.
Other relevant topic: Could be interesting to also cover the actual pins on a cable, e.g. USB, Ethernet, SATA, and wtf they actually do, and cover differential signaling and how it reduces interference problems.
Serial the parallel killer.
LOL
Manmeet Singh how can one kill in parallel anyway
@@togwam Maybe dual wielding?
@@qetzyl9911 not very parallel
concise, well thought out, and well written. Keep up the good work, whoever wrote this episode and did the animations!
let's all eat cereal at the same time so we can be parallel.
and no, it doesn't all have to be the same cereal :)
Brolivia Wilde I got the cinnamon toast crunch
Ba tum tss
Haha, Brolivia!
@Williammc10 CoaCoa Crispies :)
Did he drop a Threadripper already?
jayz two cents dropped one instead
Mahmood Mohanad Ltt replied “dropping stuff is our thing@
2 days ago
seriously an awesome video man! studying right now for my embedded miscroprocessor systems design class right now and you are motivating me to study!
Dude! You have one hell of a UA-cam channel. That energy is exciting. I appreciate what you offer in learning- fun graphics and easy to understand. Keep up the great work.
Sorry, but saying "USB got one data line in each direction" is false.
USB 2.0 and lower only has D+ and D- as signal lines, but this is a differential pair. Meaning, that it basically carries the same information at the same direction at each time (half-duplex), but the signals are the inverse of each other to block outside interference.
From USB 3.0, the cable got two more pairs for each direction (SuperSpeed tx/rx), but the old pair is still there to make your statement false and to confuse everyone.
the difference is that cereal is a food you silly goose.
I didn't watch the video but if lyonous makes a joke about cereal I'm suing.
He did at the end🙂
I hope leonardus has good lawyers
Why not watch the video, Eprepp?
3:32 tfw you forge a pc in a dying star
This video had a few good points but was marred, and it could confuse people. Had you simply mentioned _"parallel to serial compression"_ (like USB) requiring (a chipset) doing processing of it, it would have cleared up the "magic" mystery of why and how people managed to get from the data rates of parallel to those of serial. The pay-off and trade-off would start to dawn on them because it isn't "magic". You could then have simply "name dropped" Fourier mathematics, and then people could at least look it up in a book or online. There was a reason parallel existed (high data rates and also acting as the first GPIO), and yet form your video people could walk away thinking serial should have only ever been the way forward and that somehow there was no hurdle to get the benefits of parallel into serial. You don't even have to "explain" the complicated stuff. Just "mention" the word and then people can have a look.
The notion of pins being prone to damage is *not* an argument against parallel when comparing to serial. Not only can you wire to (and thereby adapt) the shape of something else like an rj11 or rj45 connector for a parallel (so the size and pins can be smaller), but also a serial (e.g. a great big RS232 null modem cable) can be about the same size as a parallel printer cable.
Thanks very much! Been trying to wrap my head around this for a while and I am glad I found your video!
The only Linus content that is worth watching anymore
Shoving that GPU in sideways at 3:30 gave me goose bumps
I needed ALL OF THIS ENTERTAINMENT. Reminds me of how engaged I was watching Bill Nye the Science guy as a kid. He made me fall in love with science and you make learning boring info easy.
"I wouldn't bother trying to eat your bowl of frosted flakes one at a time." ..... Challenge accepted!
Linus thanks for helping me learn lots of stuff daily. Hailing from Nigeria as me and my buddy watch you all the time
3:26 "Isn't THAT parallel?" 😅
You took a funny example with the soldiers. Made my evening today.
TechQuickie, explaining the *LATEST* tech!
Fun fact: Australia's first computer, CSIRAC, used a serial bus for moving data around internally. This was back in the late 1940s, when 2000 valves and 1000 IPS was impressive.
Future Video suggestion:
Subwoofer types... the subwoofer must be on the floor or on the table/desk?
Why not under the desk or why not on the table?
Does the subwoofer speaker design matters? and why?
Can the vibration damage the computer components, ex: hdd, ram, etc.?
thanks Linus, this gonna help me for my Computer Science Exam!
DIFFERENCE BETWEEN PARALLEL AND SERIAL:-----
SIR,
YOU EXPLAINED WELL AND I FOUND THE VIDEO USEFUL TO NOTE.
THANKS
VATSA INDIA
Thank you so much Linus in helping me digest Serial I/O.
You are King Linus! Continue with the good work !
3:31 this pci-e video card was FORGED IN THE FLAMES OF HELL
What was burning the background of the video card install? Did someone try to overclock the Threadripper?
Maybe...or just Satan changing or installing his computer's GPU
i'm in love with this channel
great visuals to help. thanks for your content.
I was searching for this and couldn't find it, and i just found this video from very long ago from Linus himself!
Good video again well done Linus
2:56 im always bending my internal usb 3.0 connector and no other connectors lol? I plan on 3d printing some mounts to hold it better once its in because when it sometimes gets pulled or when bent out the way it comes off wonky and bends pins
and talking of bent pins at 3:07 it looks like the pins on that fan header is a bit wonky
I've used a similar example of the rate of fire of a shot gun and its dispersion of the shells loads to a automatic fire weapon, to explain some of the differences to people who don't really get the differences so easily.
Fun fact: You can get a PCIe x4, 8, 16, etc card working in an x1 slot by either sawing the extra pins off the card or cutting open the PCIe connector on the motherboard. I had to do this once, to get a video card to work in a PC that only had 1 x1 slot remaining. (The x16 slot was occupied by an HBA.) It worked reasonably well, although the drop in bandwidth/performance was apparent even at the Windows desktop.
In my 12 years in IT at a company i have seen more sata and USB ports broken than IDE, Paralels and COM ports
Dito, just 25y :)
well 12 years ago was in 2005 and USB was basically already the standard and SATA was gaining popularity over IDE. That and with IDE there was always a chance that the pins would bend if you didn't pull the cable out exactly at 90° and it was a huge chore to bend them back. Which, although it usually doesn't require IT assistance, is still a pain in the ass.
Same
But remember that most companies have old hardware and it takes time to upgrade everything, we still have some Compaqs with P2, P3 and P4 runing(those fuckers wont break) they are used in deposits and cargo bays to show some stuff, so is not necesary to upgrade them for now
you dont usually remove parallel ports as often. thats why most of the external parallel ports have screws and screwholes and the internal ones are, well internal and not plug and play. if you were to plug an IEEE 1284 connector into a port multiple times a day it would break in no time. but usually you don't actually remove or insert a plug a whole lot.
This brings back memories from when PC parts were more expensive, and repairing parts was more common rather than disposing of them, and delicately straightening those bent pins on drives, cables, and motherboards.
Speaking of pins, I still have a large jar full of gold CPU pins that I used to extract from junked PCs. I should smelt that down into a gold bar sometime.
A lot of serial communication stuff is clock and data ( i2c , i2s, spi) and usb is NOT receive and transmit, but Data+ and DATA- , one wire is inverse of the other... same for other things.
3:14 No matter how many people make jokes on Linus he ain't gonna stop those puns.
@3:31 "Hey, your room is on fire!" - "Just let me put my pc together first."
American way, use guns to explain stuff
I KNOW, THEY ARE CANADIANS
They are all british, kinda... xD
I thiuguht that's the Russians way XD
You didn't mention the speed of chipset meaning there was a limit to baud that each chip could interpret the data in serial. It was cheaper to put them in parallel and feed the more expensive CPU that way than to buy a chip fast enough to talk in serial at the same speed. It wasn't until these chipsets became more powerful and less expensive that serial was able to take over. It also helped that we were reaching limits with parallel speeds anyway due to the aforementioned crosstalk, etc.
Neat video, explains a lot, is very visual and easy to understand :)
Salary bonus for the guy who came up with machine gun vs. shotgun example :D
USB does not have one data line in each direction. USB uses differential data signaling, this means that one data line can be mapped to 2 physical wires, one being D- and the other D+, so when one wire goes high voltage, the other goes low voltage. Differential signaling allow serial communications to be faster because they are able to send data with lower voltage swings without being affected by ambient noise. As for the clock it goes embedded in each data signal and is reconstructed by the receiver.
If only you had uploaded this yesterday. I had an exam on this today :(
Parallel will catch up and out perform serial eventually. Sure crosstalk is a problem with ribbon cable but twisted pair advances along with proper shielding can offer more bandwidth along with TDMA style data synchronization and other transport level data timing matching technology.
The shotgun comparison was nice.
Make great great great sense to me! Thank you so much, dude
This was very well exolained
I'd love to see Linus try to build a DOS Gaming PC.
Those bear icons are awesome, give my compliments to the guy who has designed them :)
Cool description.
Thumbs up!
the Traffic and Machine gun/ shot gun analogy, blew my mind lol.
love that you're giving some props to the legacy stuff~ next you'll be assembling an IBM486 ;) (hey, a girl can dream)
3:30, so in parralel? (i get it's not the same kind of parralel, but it's still paralel)
1:03, found the "wait a minute, who are you" kid, all grown up
2:31 It is all about sending a message :D
Linus, could you explain the difference between Fiber connections (SX, LC, ST ; Maybe even direct dopper and SFP/+) and Multi Mode and Single Mode ?
That gun analogy is perhaps the most American way I have ever heard anyone explain serial and parallel.
What's crazy is that SCSI is STILL around. Granted, the switch from parallel to serial in 2004 probably played a big role in that.
If I could see Linus in person.
huge fan btw
from West Bengal
FIRST I WILL GIVE A BIG HUG,THEN I WANT AN AUTOGRAPH ON MY 1ST PC
Automatic machine gun? as opposed to one of those bolt action machine guns?
perhaps in contrast to the gattling gun or metal storm.
There's also the pump-action machine guns, he's just trying to be specific
Nate I can't tell if you're joking or not
Gattling guns are non-automatic machine guns
thenoobcannon
Well, I suppose a machine-gun is just a gun that operates by using mechanisms for reloading, ejecting, and whatnot instead of needing the input of a human every step of the way.
And automatic is a general term I suppose, so he's not that incorrect.
So long as it functions by itself without someone having to baby it every step of the way I suppose it's automatic, and if it does that with mechanisms, it's a machine.
I dunno, though.
Great video! :)
I would find it very interesting if you could do a Techquickie on computer component power consumption, particularly that of CPUs and GPUs and what constitutes it. What does the electricity actually do inside the components - what is it needed for? Also, the GTX 580 from around 7 years back and a modern GTX 1080 Ti both have similar power consumption yet the difference in performance is, of course, absolutely massive. Similar trends can be seen on CPUs as well. What has allowed these huge advancements in power efficiency and can we expect this trend to continue still as time passes on - is there still much that can be done to improve power efficiency? Where are the physical limits as to how low power consumption could be dropped in the future for a given level of performance (a certain amount of FLOPS)? Also, how are CPUs used on mobile devices and laptops able to achieve such low power consumption ratings despite still being fairly powerful?
It would be very interesting to learn about these things! All the best to the whole team and thank you for making these videos. :)
Tunnelbear is dooope! I use it on my Mac and iOS, works briliiantly most of the time. Thx Linus for suggesting such a well designed seamless app!
Even a noob can understand this. You are a hero :)
Was genuinely thinking this video would be about parallel and serial wiring in circuitry.
Good video man I like the machine gun explanation thank you
It should be noted that there are dual port SAS drives, which have multiple channels, whereas SATA only has one, so that's another part that kind of blurs the lines a little bit, like PCI
I loved that machine gun illustration.
Hey Linus, since you are talking about all those different ports, I was thinking about you making a video about firewire. It was said that it was faster than a usb 2.0 back in the day.
SO MUCH TO LEARN!!!
Fun fact, the S highlighted in SCSI at 0:55 does not stand for Serial.
I thought crosstalk was between a receiver and transmitter (side by side) wires. Cause the transmitter has much more signal in the beginning of the wire and after some distance (that the receiver's signal is not faided that much) had electrical interference with eachother.
Great explanation , thanks
Good video! Great info! Thanks Linus!
i definitely don't miss the days when you had to pin the master/slave etc and when you had to type in all the volume information for a hard drive into the bios, Plug and Play all the way.
wow with this funny guy i really remember the work
so why does usb 3 and 3.1 usb more than 4 pins? are there moultiple serial sonnections like pcie?
3:33 wheres that take from?
3:32 I too build my PCs in the depths of Hell
you took me back eons
I’m assuming this means that each wire in a USB cable (for example) handles one set of information bit by bit while other wires can handle other sets of info simultaneously. Parallel connections have every wire containing one bit of each byte so they’re effectively working together all the time whereas the wires in a serial connection operate independently
Please do a video on PCI vs other serial protocols like I2C
a video talking about ipconfig ports would be great
Why couldn't the parallel standard be improved to allow for the pins to transfer data independently like in PCIe?
I WAS EATING FROSTED FLAKES WHILE WATCHING THIS AND THAT PUN MADE ME SPIT IT OUT ON MY MONITOR XD
So wait if these are so technically different how do ide to sata adapters works? Or does it use a dram chip to store the data and then convert it from serial to parallel?
3:27 I am gonna Meme the Hell outta this.
Really great video
I thought crosstalk was between a receiver and transmitter (side by side) wire. Cause the transmitter has much more signal in the beginning of the wire and after some distance (that the receiver's signal is not faided that much) had electrical interference
You can be a serial but not a parallel killer..
haha liked the Shotgun/MG comparison ;)
There are definitely some limitations to standards like USB. Some industrial applications actually do work find parallel connection more reliable if they need a specific thing to be refreshed as quickly as possible. A serial bus really doesn't lend itself quite as well to real-time applications, and a couple of unusual examples would be USB keyboards not having NKRO, and the fact that an SNES controller can be more responsive on the original hardware than a USB knockoff can be on a PC. USB in particular has the limitation of relying on the OS polling a certain number of times per second, where some parallel standards can use straight up hardware interrupts, so the device can tell the computer about state changes as soon as they happen rather than waiting for the next poll. I'm kind of glad the PS/2 ports are still around.
USB keyboards not supporting NKRO is a common myth. USB mice are also not any less fast or reliable than PS/2 mice. USB has matured and evolved enough at this point that the advantages PS/2 once had are pretty much moot.
I have a whole bucket load of what appears to be DB 37 Serial data storage cartridges (predecessor to USB thumb drive by the looks of it, to me at least) and I'm trying to figure out more about them.
I have these because my Grandfather custom built a few barrel organs in his life time, usually barrel organs are analog instruments which use rolls of perforated paper to produce music, my Granps being the absolute 1980s-1990s tech wizard he was, completely rebuilt the guts of the organs, turning them digital. They read MIDI files off of these extremely obscure, dated 37 pin cartridges (I think serial DB 37), the cardriges themselves are totally black plastic, about 3x2x0.75 inches with right angle edges and corners, and absolutely no branding or text on the device other than the hand written labels of what songs are on each cartridge.
Anyway, I inherited all this stuff and would love to figure out more about it all, I'd LOVE to be able to hook up some of these cartridges to modern hardware and see if I can write any custom music to be played on the organ. I'm an actual musician, which is why I inherited all of this, so it would be a blast to give some new life to my Grandad's old project.
0:58 hella wrong. USB 3.0 has two seperate "lanes", so it can send and recieve at the same time.