Do Riser Cards affect GPU Performance? GTX1080 Giveaway inside!
Вставка
- Опубліковано 24 сер 2024
- Want to win my NVIDIA GTX1080 FE Video Card along with an EKWB Full cover block and backplate? Then click here! gleam.io/dbcGC...
Learn more about Corsair RBG memory at www.corsair.com
○ Products featured in this video available at Amazon ○
► Amazon US - amzn.to/2tJORdz
► Amazon UK - amzn.to/2tqrTpj
► Amazon Canada - amzn.to/2tSbwEX
○○○ All music provided with permission by audiomicro.com ○○○
SUBSCRIBE! bit.ly/sub2Jayz...
WATCH MY MOST POPULAR VIDEOS!: (link)
••• Follow me on your favorite Social Media! •••
Facebook: / jayztwocents
Twitter: / jayztwocents
Instagram: / jayztwocents
About JayzTwoCents
Your one stop place to shop for fun and educational videos on all things tech! Oh, and don't forget gaming! Gotta make time for gaming!
Do Riser Cards affect GPU Performance? GTX1080 Giveaway inside! | JayzTwoCents
• Do Riser Cards affect ...
JAYZTWOCENTS
/ jayztwocents
LinusTechTips: Do $2,000 riser cards affect performance?
Till he drops it and they snap in half
And the display won't turn on
Do you mean LinusSexTips?
What's this joke about Linus I'm seeing a lot?
thegreatcalvinio don't forget the click bait title
Next video: Does the GPU affect performance?
Cube Matz 😂😂
spoiler: yes it does
Very minimally
It's all for looks ...
I hate anyone who believes this
RGB's are worth like 10 fps
Thanks for the short quick not bullshit video it seems there pretty hard to find these days
Cactus I agree. I love Linus, but his videos are getting kinda over the top. Like, plasti dip your pc...
Q: Does an electrical cable affect performance? A: No
**Slow clap**
+JustAnEngineer add a really large cable then tell me that ;)
Seriously, it does degrade, but outside of very specific kinds of workloads or really large distances and it won't really matter.
After all, it takes any signal electrical or optical at least 45ms to go from NY to Sydney.
+Ewan Marshall You are correct....sort of. It does take around 50ms for a signal to travel that distance, but only because that's the speed that light travels at. Not really degradation of the signal itself. If it ever did affect computing it would be affecting supercomputers and servers first where they have miles and miles of cabling. But even then....it's very negligible.
+JustAnEngineer it does effect supercomputers, it also effects the automated trading machines of the NYSE. In those applications nano seconds of delay on the cabling can have an effect. Yeah, one usually has interference and signal degradation issues first in most applications, if you stack 100 PCI-E risers then it will become a noticeable problem. As I said, it is highly application dependant.
NZXT be like "pre-order your RGB risers today!"
ProjectJamesify for 1000000$
this did not age well
Fire rgb risers
I miss seeing skunkworks in the background of the videos
James Emsley It has become our lil' baby too ;-)
James Emsley us humans we miss everything about UA-camrs
we don't want then tie change anything ever at all -_-
Mohamed Osama well not that serious. I just like seeing that build
How can people mistake PCIe power cables with PCIe riser cards?
Just read the comments on that video.
yo daniiilo, you watch this too? hahaha
Well, if we're going to be pedantic, the risers featured in this video are riser cables, not riser cards. I'm not going to split hairs over something like that when it's perfectly unambiguous what Jay is talking about, but on the last video before the title change, he was talking about PCIe cables, which is a bit more ambiguous, given that "cables" can apply to either risers or power delivery. Anyway, I don't see it as a reason to complain, because this way we got two informative videos instead of one! :p
I read all the comments when I first watched the video, & couldn't believe how many people misunderstood what the content was- especially when the thumbnail was of you fugglin with the power cords
heehee,,, he pedantic
"uni-gyne" im done, thats enough internet for me today
hahahaha love it
I think it would be interesting to see iterative testing of extension cables plugged together to see how long of a reach you can get before it does make a difference.
i love the way how you do giveaways not liek whatever other tech channel who makes it clickbait in title and everywhere you just notice it silently so only your fans will know and not clickbaiting! have 100 internet points for that!
Pinkie Pie got to this comment same time as he talked about it...
2spooky4me
To be honest I really dislike giveaways in any shape or form, but if we are to exist in a world where those things exist, then I too much prefer it being done this way.
Robert Faber I think they are a bad thing when the creator is begging for subs and likes, but the way Jay and others (like Salazar from Sc St) do it is awesome. They have the hardware lying around and they know they will make someone happy by sharing the PC experience ✌
Agreed!
"I happen to just have a 1080 kicking around...want it?"
why didn't you plug the two risers into each other and then the card and see what that did? you know we're all curious. that's the first thing my brain went to when i saw you had two
shamelessjames @jayztwocents do this
The Ultime Riser Test: How many Risers do you have to plug into one another, to start noticing any change in performance?
Also, RGB Risers anyone?? Amirite??
Adrian Rotariu you won't have a performance loss but probably problems in general
i wanted him to try that as well. Tho i think that the system will just or refuse to boot or crash if the cable gets to long.
shamelessjames I believe after about 2 risers there is a performance difference.
cpu socket extender?
return to cave
Wireless cpu adapter
@@Bonkatronicks I want that to be a thing
Andrew Yang im waiting for cpu's in gpu's
uh, you mean the CPU HOLDER or SLOT
-The Verge
good video, but incorrect that they have been around forever or during early pcs. riser cables for actual motherboard slots themselves started to be a thing rarely around the pentium 4 and core 2 era, then became more known and common around sandy/ivy bridge. before pentium 4, they literally didn't even exist or were an option, with a few specific exceptions. exceptions being some rare oem systems that would have pci angle boards or extensions to make their cases work, but these usually would only work with that system, or were an actual piece of pcb with hardware and etc.
If it can't help me make better memes in 4k in photoshop then I don't need it.
Jay its UniGen like engine.
My en-jyne goes vroom vroom.
xX_Destructo_Xx IsHere unigengine?
unigen think like a game engine like frostbite its like a game engine but its called unigine get it?
He was joking though. He's pronounced it correctly pretty much every time in his older videos
I can't help but read it as "un-engine". It shoulda been called unengine
Bought a bunch of these cables with Bitcoin 6 years ago for a Bitcoin machine, spent about 10 Bitcoin on some riser cables.. RIP
Like, 20-25k today lol
Jarrod'sTech lmao
LOL. . Did you get them from Cablesaurus? I remember selling a mining rig for 250 bitcoins. RIP
Haha yes that is exactly where I got them from! Forgot the name until you mentioned it :)
selling, or buying? because one is drastically different from the other lol
Would also be curious, just for the hell of it, to see a 1x to 16x riser extender or w/e.
This is what im looking for right now. I bought a usb 3.2 gen 2 pcie adapter but didnt realize i didnt have a pcie x4 slot. Im wondering if buying the 1x to 16x riser will actually put any sort of bottleneck on transfer speeds. Since i am new to pc's im really just learning atm :(
"Oh no, now I'm disappointed because I thought this video was about if RGB lights affect performance!" Haha, just joking, this was a very informative video and something I may need in the future.
I want a wireless graphics card
Beast6228 Give it about a deckade.
Beast6228 Pny already did that a few years back, it worked but it was bloody expensive.
There is one. It is an old Amd card. Idk what it's called just Google AMD wireless graphics card
KFA2/Galax made a wireless GPU, the GTX 460 WDHI. Keep in mind this was back in 2011.
It's Nvidia kfa2 GeForce Gtx 460. O was wrong it's Nvidia not amd
saw you yesterday night at downtown disney, wanted to get a pic with you but youre with your family didnt really want to ruin your little vacation. anyways more power to you!
2:08 "there is link in description" lol
Guys don't enter the giveaway, just google "graphics card download" and download GTX 1080.exe. Please install it, your computer will be much faster and could easily run minesweepwer 800x600 at 24fps
24FPS are to low, you need at last 25FPS.
Raphael Pfundheller doesn't matter, the human eye cant differentiate between 1 fps and 24
It can ,it can't see over 24-27 FPS , otherwise our ancestors would be all dead.
24fps is too smooth can we get less? I know it’s a beautiful cinematic but it is too much for games.
Dont forget, download more ram! You'll need at least 16gb.
About how manufacturers adapted card design to orientation: I've got a case with a horizontal motherboard and it makes me nuts that every backplate is upside down for me!
The reason the two vs one power connector matters, is resistance. Power flows like water through a water mill to generate electricity. So as power travels to the device, it flows through the components, then has to travel back to the source. So, if the current has to be split in half to travel through two separate devices, more resistance or used energy occurs. If you take one garden hose and use a Y at the far end, then run two sprinklers, you'll get reduced flow because only so much can travel through one line. But 2 hoses from the source will deliver as much as the source provides. Resistance is why you need to plugs.
Now, why aren't risers effected? Again, resistance. If less current is passed through, than resistance isn't a factor. If you only used a quarter of the water coming out of a hose than the source provides, you can step on the hose without loosing any flow. Very little current passes through a PCIE slot, that's why an external power connector is needed. Also, the guage and quality, and number of strands effect the resistance, so some power supplies can provide more than others just by different wires used.
I am an electrical engineer, though I don't have experience with these products. So long as the cable impedance matches the factory spec there should be no loss. Transmission lines, at this high a frequency they are transmission lines, simply have to match the load impedance to get maximum transfer. So long runs just need bigger wire to a certain extent
download more ram, that'll speed it up
Pfff, why download more RAM when you can just download a gtx 1080Ti?
"I can't change the government... yet"
XD
I've read into this a lot since I actually have my GPU on a riser cable in my TT Core P3 case.
The issue WAS the quality of those cables. Which meant, that it didn't really affect performance in like FPS, but it didn't work at all or it hads loads of artifacts and errors. This was however when they first came along.
I am now using mine stable for like a year now without any problem, but they do have quite a bad reputation because of the past. Which is probably why ppl were asking.
Hey @JayzTwoCents thanks for being awesome I learn a lot from your videos and it's content like your's along with some others like Kyle and Paul NerdOnaBudget that inspired me to take my lifelong passion for games to the next level and finally build a damn pc of my own. Keep doing what you're doing man and while I was pretty ok with the 1050ti OC from PNY that I got for my 1080p rig I definitely wouldn't mind winning a 1080 card. I'll just be that much closer to the dream and hell why not run it in the current config I got an overkill psu with the intention of upgrading anyways I mean who doesn't like more frames?
do x1 to x16 risers drop performance?
yes it does by a whole lot
"I can't change the government... Yet"
-Jay
I have had a PCI riser that did hinder my graphics performance. So bad that it was about half of what my GPU could do without the PCI riser. Upon further inspection, the PCI riser that came with my case was actually bad. It had a few tears in the cables, and was very poorly shielded. I haven't had the experience with any other PCI risers, just because I haven't bought any after the one that came with my case. That being said, it was probably just a manufacturing error, and I agree with what Jay says. Not everything is made equally. By the way, keep up the good work Jay!
Works fine on an open test bench, but expect higher temps in a standard case, resulting in throttling.
Benchmark anomalies? Just windows things...
jay can't change the goverments yet, so a revolution incoming in the next years? jaymasterrace👍
Thanks for doing this vid I was curious about any perf degradation. Much appreciated as always Jay!
tl:dr - PCIe spec requires normal function at a minimum distance.
Had a bit of a look and doesn't seem like others have mentioned this, but the riser adapter will begin to effect performance at a certain length. The reason shorter ones don't is due to the PCIe spec requiring minimum working distances of PCIe traces, for which addition connection are also considered.
This has become one of my favorite channels and always been watching it on a daily basis! This guys explore/review things like how it has to be and meant to be.
Great work!
Thank you Jayz!
Well since a lot of cables have length limit before the signal degrades, there will be a length limit somewhere.
These things are another factor of wear if anyone decide to use it...They might go bad at some point. (not saying you shouldn't use them, but just remember it is a factor when you are trying to fix your computer)
Friend had one of these he bought it from another friend to power his GTX1080...(he bought the entire case with some components) His rig suffered major video problems, even playing a simple youtube video results in heavy screen tear and it looked like bad video encoding. I noticed he has this riser cable, so I unplugged it and plugged the video card directly into the slot. Fixed everything...
Jay for stuff like this u should make a mythbusters ps.love ur vids
this is why I love Jay. he listens to his audience and makes content we are interested in.
thanks for the video Jay.
The reason the first test was lower, is it was loading everything into their respective caches and memory. Thats why they always say run your benchmarks more than once to get more accurate results.
Pretty sure those caches and memory would be long gone considering he shut down the PC between runs.... Youknow... Because he literally had to disconnect the GPU between runs.
These are also used in server environments such as 1U or 2U if GPUs are going to be used in computational algorithm computing.
That moment when you are broke-ass and your graphics card dies, but JayzTwoCents does a giveaway... please god! Be in my favour, for once!
PenguinCED if I win, ill ship you my old gtx 960
Nathan Hamman fingers crossed
I can give you a free graphics card if u want, just shoot me an email with ur address, and let me know if ur motherboard can support PCI-express 16
Newton Cazzaro I pm'd you the email
Ayyyy, that was me who was confused!
Jay i believe the reason your 1st bench was lower was because the following benches the ram and cpu was already cached providing a higher score for the following tests
very interesting. Might use it for a future build. When are you going to have more of a talk show? Something like your Tech Talk where your viewers can present their PCs and they could get your critique, talk about random tech news, etc... I'm sure you get this a lot. Anyway, have a good one
What about a cpu riser?
Ah the old slot processor days...
Not even in notification squad and still caught the video early :D
d3adman 89 same lol
d3adman 89 Did I make it in time?
My experience with a Riser-Cable: random black screens, Driver-crashes, Sound-Distortion via HDMI. Incredibly unstable and the problems occured very randomly. So if you want to use those, don't take the cheapest ones.
I mean, they're basically just extender cables... Distance doesn't make a difference with digital signals until you run in to stability issues because of signal corruption. It would probably get less clear if it was an analog signal. Also, it's kinda sad that I get so excited over new benchmarks... but they come out so rarely.
Got here at 1 View, finished the video and there 1848 views. 233 likes.
What about 1x to 16x risers? I heard of a mining board on Awesome Hardware that relied on these for many GPUs.
I was thinking the same thing looking at the title. Do the GPUs even function as a video out device on 1x PCIe, and how if at all does that affect performance on PCIe 3.0 slots. I know in non-SLI a single GPU has no problem working in a 4x PCIe 2.0 slot, tho I'm not sure about the newer PCIe.
The 16x test at different lengths was a good test.
Yes, they do output video and there is no effect on gaming at all.
Evidence: My own mining rig!
This would be a cool thing to test. On a PCI-E 3.0 board, one of those one lane risers is the same bandwidth as an x4 1.0 slot. Plus they generally use a USB cable that can go any direction and be more than a few centimeters long. If it works well enough for some 1080p gaming, it could be a way to run a big gpu outside a tiny case.
From what I could find on ./google, this setup should have approximately the same bandwidth capability as the last of the AGP 3.0 slots, approximately 1 GB/s or so.
I'm pretty sure the USB ones are not capable of driving modern GPUs for gaming but I could wrong! Definitely would be something to test.
x1 slots are definitely not enough for gaming. You'd need at least x8 (depending on PCIe standard) to avoid performance loss, last I checked.
anyone ever been so bored that you already know the outcome for the video but you really dont wanna do you calc homework lmao
Thank you for addressing this old question! Next question: Can your 370Z hang with my Chevy SS?
what about the USB ones?
It will slow down............
i need that beast in my life
I actually have the 25cm EZDIY and haven't used it yet, this video certainly gave me some piece of mind for when I do finally finish my case build. Thanks Jay.
This actually came out just in time for a possible build I am thinking about doing.
I have an Acer Veriton Z4630G all-in-one desktop, and it has a full PCI-E 16x gen 2 slot. I think I can put a GPU in it but the slot is too close to the back of the screen inside, so I was wanting to get a PCI-E riser cable, and make an external slim enclosure on the back of the AIO for the GPU to sit in, to try and transform it into a gaming system.
Currently it has an i5-3330S (might upgrade to a 3770S) and 8GB of DDR3 with a 1TB hard drive. If I managed to slap in a 1050/Ti it would be a definitely nice slim gaming PC.
I'm not going to enter the giveaway due to having a better GPU, other people should do this too if they have a better GPU so people with less powerful cards or someone who cant afford one has a chance!
YES! FINALLY! Thank you Jay for doing this. I always wanted to know for YEARS!!! Best review guy right here
@JayxTwoCents You didn't see any performance drop in benchmarks, but from what I've read you will see some performance drop in real world usage such as games. The gpu is just as powerful with a riser, all the pcie interface is doing is feeding instructions. The performance drop comes in the form of latency. When using risers, the gpu is farther from the cpu, so any workload that requires round trip communication (like games) may suffer some from the added distance.
I am actually a little surprised that there is no difference at all, good to know! Thanks for the video Jay!
I entered the giveaway.
I better win because......
Because.....
uhhhh.... Becaaussseee.....
Just Because :)
Congratulations! You won!
Not -_-
Gaming Radar101 'Cause'
Alex Soil you only you only live once?
You have 50% chance to win. Either you do or don't.
Make that 100%
I BETTER WIN!!!!!!
Awesome video brother me and my brother always watch your video love and support from India 👉. JayzTwoCents 👈you rock
Jay I find myself liking your videos even more since the new studio!! Keep up the good work! Big thumbs up!
The inconsistent scores might have been due to the shader cache for the benchmark being built up during that round of testing. It makes sense that it would happen during the first test.
It's pronounced UniGin. like the alcohol.
Soooo, nobody's watching this because of crypto mining?
WhiteChaos miners are greedy fucks who ruined the gaming market for their own selfish needs
You aren't somewhat greedy? In a sense it gives you better ROI than buying a house. So I can see why people do it xD
Jay, you are really a great person, the fact that you have this 600 dollar gpu and u want to make your viewers happy says a lot thanks for being a great person.
As long as the angry-pixies and angry-vixies flow, riser cards don't affect performance unless the 80.000 watts of LED's inside the case blast the riser with loads of EMI. :)
My dream is ruined because of fakin miners... I have collected some money and cant even build my dream PC when I studied about PCs for 6 years and still cry sitting with core 2 duo with its integrated graphics (had GTX650 but it died before last Christmas :( )
Aurimas Knieža I hate their pyramid schemes too. Once ethereum crashes, they'll come back at low prices and all will be good.
Wait for a GPU crash and everything will be super cheap.
Quaerens Veritatem how the fuck is it a pyramid scheme?
Quaerens Veritatem also once the supply of demand is met then the price of GPU's will lower.. you cant blame miners or manufacturers.. its the people that are pre-ordering and re-selling the cards at a premium price, its the same as some arsehole thats buying sneakers for x amount then reselling them 6x that price
If you are willing to buy used, you can pick up cards like the 1060 and 1070 for only about $50 - 75 over retail. Which isn't completely terrible. That's like one day of work.
It's pronounced "Uni-gin" not "Uni-Jine". smh jay, expected you to know this by now.
I bet he pronounces "gif" as "jif"
Well if we want to be technical, "jif" is the correct way to pronounce it, at least as said by the person who came up with the term gif.
Rhymes with "engine", not "vagin-"
Speelydan Are you guys Eugene ?
What?
you should have tested affects of riser cards on sli as well. plus tests to determine whether cable width , shielding, cable material, etc. have any effect
I'm pretty sure the "black" the cable is wrapped in, is shielding. I have one of these and the product says thats "Black EMI" shielding. it's shielded.
Fist
Me daddy?
Knuckles
Paper! I win.
-ing?
3 views, 55 likes. *UA-cam is drunk again.*
5 views and 105 likes
Lucidity Can you be drunk again if you've been drunk your whole life?
Can we stop with this crap? Unless you've been living under a rock for the past five years, you know how UA-cam''s delayed stats don't occur in realtime. It's not funny pointing it out - it only makes you look like a newb.
Yeah i'm sorry for being such a *NEWB*.
No worries, one day you'll learn.
Hi JayzTwoCents, i'm a big fan of you and i really like your videos. I'm a technical Presales and i had the chance to test several Riser cards (fixed riser and flexible):
- at 16 cm length (flexible risers) = the performance was not affected
- at 32.5 cm length (3 fixed riser cards + 2 flexible) = the performance was not affected
- at 82 cm length (all riser cards that i found : fixed and flexible) = the signal bandwidth was passing from 8GT/s (Gen 3.0) to 5GT/s (Gen 2.0)
digital. signals = no speed loss, it will either work or not. digital signals are fine until they break. if there is a difference, verify that the thing is still running at PCIe x16. interference COULD cause it to drop to less lanes. or totally not work
Short answer: No
Long answer: If it has lots of heat and movement on the cable, it can slightly affect performance, and an SLI config. can affect this too. There are shielded ones and non-shielded ones, so don't buy something too low of quality.
you were fast on making this video thanks jay !
jay your right. i just want to tell you if by chance you read this.the pci extension card will create voltage issues the longer the stretch.and latency,speed & bandwidth of pci is not affected.
I've never won anything in a contest like these, but I am currently looking for a GPU, so might as well.
Thanks, Jay =D
i bought a cheap unshielded extender and wrapped it in silver foil worked fine
This is good to know. Sometimes its better to space things out.
I am saving up for a Core P3 or P5 where I will use a riser card.
So, this video really helped - thanks
Did you have any issues with the included riser cable?
So I wasn't the only one who expected this from the last video.
Always wondered if they adversely effect performance so thanks for taking the time to test.
I know that this is a few Years old, but want to give you all the FYI on what I just recently found.
Riser Cables cause very Serious Audio EMF Noise and the PCI-E 3.0 is not Stable,
has to be set set at 2.0 in the Bios in order to have not OS Freezing and able to get Stability.
I have tried several different Brands of Riser Cables and they are all Junk,
they all have Poor Shielding and Cause Severe Performance losses.
I am hoping as I write this, that some Company will make them better and have the proper Shielding.
I can not Display my Video Card Vertically unless I set the Bios to PCI-E Forced 2.0, and then I get severe Audio Whining Noises.
Kind Regards to all and JayzTwoCents, my Hero in the PC Industry.
a few years ago, affordable long 16x riser cards were not available, so I used two cheap mining ones. put two of them together - they worked, but due to crosstalk or impedance mismatches, the machine would not post reliably. had to reset it several times to get it to detect the graphics card and output video to it. the performance was unaffected, though.
So glad that you made a video on this. This was something I was wondering about as I was planning on building a custom case.
Is it Christmas? First, we get a bonus video because we asked for it. Second, we get a GTX1080 giveaway...
PCI Express is a SERIAL communications system, not parallel, so it's far more immune to timing. It doesn't operate at the same frequencies as ram so can have wire distances of about 14 inches.
Pretty sure that very first test could have been a bit lower due to dust (or even corrosion) buildup on the components. Moving stuff around, and more important disconnecting/reconnecting the PCI-e, likely removed some of that shit
I like the idea of risers for cooling when using multiple GPUs for things like cooling. But, the problem is where to put the graphic cards that are on the risers without making a custom case. Miners just have everything out in the open. I guess that I could do that for a GPU rendering rig. Professional solutions with racks and addon cards are _so_ expensive; they cannot be justified.
First test: shader cache compiling maybe. It's common for first test run to be off.
that was a heavy "YET" in regards to changing the government. Viva La Jay
I am not electrical magician too, but riser cards are pretty short for electricity/PCI-e signal (I didn't menage to find voltage or amperage but we know that gpus can pull up to 75W from PCI-e directly on mb and Formula for calculating power from voltage and amperage is W=V*I so from that I can conclude that either voltage or amperage is higher(i don't believe that voltage is 750 and amperage 0.1 or other way so it is pretty obvious that it is more balanced so voltage is somewhere from 0.something-12 which is power suply's max and amperage from 0.0something to power suply's max which is around 20A max on modern ones))
i love going through ur old videos and re watching them. been a fan for a while befor i knew anything about youtube and how it works :D always happy with your videos keep it up