Jump to content

Does Bottlenecking matter?

greenhunt2003

Guys, I used this bottleneck calculator: http://pc-builds.com/calculator/

 

I have tested my Ryzen 5 3400G with a GTX 1050 Ti and it produces around 30% of the bottleneck.

 

Ok, I don't even care about "bottlenecking". You wanna know why?

No. 1 - Just because bottlenecking means that the CPU limits the performance of the GPU, doesn't mean that they don't work together at all. The CPU handles and deals with the instructions, while the GPU renders video.

No. 2 - The powerful GPU can still run well with a weak CPU, but less than full expectations of performance.

No. 3 - The word "bottleneck" is just Cassandra.

 

Bottlenecking is almost like comparing the size of an elephant to a mouse. It is when one or more components limit the performance of a particular component.

 

Goddamn dude, I have rights to put whatever components I want. I can even install an RTX 2080 Ti to a Celeron, Pentium or i3 machine.

Gaming and Video Editing PC: HP MO1-F0021na

Processor: AMD Ryzen 5 3400G @3.7GHz

RAM: 8GB 2666MHz DDR4

Storage:

KBG30ZMV256G KIOXIA 256GB Solid State Drive (Boot Drive)

STD1000DM003-1SB102 1TB 7200rpm Hard Drive (Secondary Drive)

Graphics card: AMD Radeon RX Vega 11 (Integrated)

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, mscrafter2003 said:

Guys, I used this bottleneck calculator: http://pc-builds.com/calculator/

and that's your mistake... these calculators base their info on a single scenario or even just out of thin air

 

1 minute ago, mscrafter2003 said:

No. 1 - Just because bottlenecking means that the CPU limits the performance of the GPU, doesn't mean that they don't work together at all. The CPU handles and deals with the instructions, while the GPU renders video.

yes, but will be held back if one is slower with something, as the other will have to wait around, which is technically speaking the "bottleneck", but those will always occur, depend on workload and everything.

 

2 minutes ago, mscrafter2003 said:

I have rights to put whatever components I want. I can even install an RTX 2080 Ti to a Celeron, Pentium or i3 machine.

yes, but I'd probably call you stupid if you were to actually create a bottleneck this big

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, LukeSavenije said:

yes, but I'd probably call you stupid if you were to actually create a bottleneck this big

I'm just joking man.

4 minutes ago, LukeSavenije said:

 

yes, but will be held back if one is slower with something, as the other will have to wait around, which is technically speaking the "bottleneck", but those will always occur, depend on workload and everything.

 

which is why that you won't get maximum performance from a GPU because a weak CPU is the limiting factor.

6 minutes ago, LukeSavenije said:

and that's your mistake... these calculators base their info on a single scenario or even just out of thin air

tbh, I wouldn't recommend this to anyone because it will just spread fear and stress.

Gaming and Video Editing PC: HP MO1-F0021na

Processor: AMD Ryzen 5 3400G @3.7GHz

RAM: 8GB 2666MHz DDR4

Storage:

KBG30ZMV256G KIOXIA 256GB Solid State Drive (Boot Drive)

STD1000DM003-1SB102 1TB 7200rpm Hard Drive (Secondary Drive)

Graphics card: AMD Radeon RX Vega 11 (Integrated)

Link to comment
Share on other sites

Link to post
Share on other sites

12 minutes ago, mscrafter2003 said:

Guys, I used this bottleneck calculator: http://pc-builds.com/calculator/

I have tested my Ryzen 5 3400G with a GTX 1050 Ti and it produces around 30% of the bottleneck.

On average, depending on what you do. Something this website only has a little disclaimer for.

15 minutes ago, mscrafter2003 said:

Ok, I don't even care about "bottlenecking". You wanna know why?

No. 1 - Just because bottlenecking means that the CPU limits the performance of the GPU, doesn't mean that they don't work together at all. The CPU handles and deals with the instructions, while the GPU renders video.

No. 2 - The powerful GPU can still run well with a weak CPU, but less than full expectations of performance.

No. 3 - The word "bottleneck" is just Cassandra.

1. yeah, so if something is a bottleneck depends on what you do

2. of course, but what you said is a bottleneck. Something not running as fast as it could, because another component is holding it back. So I don't see this as an argument on why one should not care about bottlenecking.

3. I fail to see how 'a Cassandra' is relevant here. You said yourself with bottlenecking not necessarily something bad would happen. 

18 minutes ago, mscrafter2003 said:

Goddamn dude, I have rights to put whatever components I want. I can even install an RTX 2080 Ti to a Celeron, Pentium or i3 machine.

This sounds a lot like a 'I can do whatever the heck I want as a sovereign citizen of planet Earth!'-kind of thing.

Sure you can do that, for a lot of purposes, a more balanced setup makes more sense.

 

There will be scenarios where this setup is logical (like for compute tasks, which you only do on your videocard; like F@H or Cryptomining, if that was still relevant on videocards), but in most scenarios, it would have been better to shuffle your money and put a bit more into the CPU and less into the GPU.

 

I think that is the issue with people screaming 'bottleneck!'. It's a husk of its former meaning. Especially websites like the one you linked do not give the proper information on why something is a bottleneck.

Instead it just recommends the user to get a different CPU or GPU. It does not take into account many people will use their computer for different purposes.

One PC/CPU+GPU combination could be a huge GPU bottleneck in one situation, while in another situation it can be a big CPU bottleneck.

"We're all in this together, might as well be friends" Tom, Toonami.

 

mini eLiXiVy: my open source 65% mechanical PCB, a build log, PCB anatomy and discussing open source licenses: https://linustechtips.com/topic/1366493-elixivy-a-65-mechanical-keyboard-build-log-pcb-anatomy-and-how-i-open-sourced-this-project/

 

mini_cardboard: a 4% keyboard build log and how keyboards workhttps://linustechtips.com/topic/1328547-mini_cardboard-a-4-keyboard-build-log-and-how-keyboards-work/

Link to comment
Share on other sites

Link to post
Share on other sites

23 minutes ago, mscrafter2003 said:

Goddamn dude, I have rights to put whatever components I want. I can even install an RTX 2080 Ti to a Celeron, Pentium or i3 machine.

looks at Pentium 4 motherboard, 1x PCIe gen 1.0

 

looks at RTX Titan

 

I know what I must do

Main: AMD Ryzen 7 5800X3D, Nvidia GTX 1080 Ti, 16 GB 4400 MHz DDR4 Fedora 38 x86_64

Secondary: AMD Ryzen 5 5600G, 16 GB 2667 MHz DDR4, Fedora 38 x86_64

Server: AMD Athlon PRO 3125GE, 32 GB 2667 MHz DDR4 ECC, TrueNAS Core 13.0-U5.1

Home Laptop: Intel Core i5-L16G7, 8 GB 4267 MHz LPDDR4x, Windows 11 Home 22H2 x86_64

Work Laptop: Intel Core i7-10510U, NVIDIA Quadro P520, 8 GB 2667 MHz DDR4, Windows 10 Pro 22H2 x86_64

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, mscrafter2003 said:

Guys, I used this bottleneck calculator: http://pc-builds.com/calculator/

 

I have tested my Ryzen 5 3400G with a GTX 1050 Ti and it produces around 30% of the bottleneck.

 

Ok, I don't even care about "bottlenecking". You wanna know why?

No. 1 - Just because bottlenecking means that the CPU limits the performance of the GPU, doesn't mean that they don't work together at all. The CPU handles and deals with the instructions, while the GPU renders video.

No. 2 - The powerful GPU can still run well with a weak CPU, but less than full expectations of performance.

No. 3 - The word "bottleneck" is just Cassandra.

 

Bottlenecking is almost like comparing the size of an elephant to a mouse. It is when one or more components limit the performance of a particular component.

 

Goddamn dude, I have rights to put whatever components I want. I can even install an RTX 2080 Ti to a Celeron, Pentium or i3 machine.

Agreed.

 

It all depends on the workloads (if using a 2080ti for machine learning, who cares it runs on a i3).

 

Plus, even when gaming bottlenecks are relative. My CPU (old i5 w/o hyperthreads) was bottlenocking my GPU (2060) by some noticable amount. Bad combination? Nope. Cause I still had a significant boost in fps from that upgrade and later when I went from 1080p to 1440p things balanced out nicely.

 

Next up is upgrade to 3900x likely. Will the 2060 be underpowered then?  According to bottleneck calc,  yup. In reality, "nope" since the exfra speed/threads are for development/running of specialist software and not gaming. Ill then still play the same games for leisure that I do now, and the 2060 already pushes enough frames to my monitors refresh limit.

 

I9900k with 2080TI can form a huge bottleneck (if where talking mining where multiple gpus would fair a lot better).

 

Fit for purpose, thats where its at. A bottleneck only surfaces in specific workloads, even not all games are equally balanced in cpu/gpu load so any system will have some margin of it.

 

That all being said, I really do need that future 3900x to keep my cpu usage below 1% while moving the mouse at 144hz to click minesweeper (which is grossly overdue as far as its rtx enabled update goes, Its 2020 for crying out loud, we should have reflecting mines by now...)

 

Link to comment
Share on other sites

Link to post
Share on other sites

No, it is you experience using it that matters, not a bottleneck.

 

if you get the experience you want at a price you can afford then it doesn’t matter.

i5 8600 - RX580 - Fractal Nano S - 1080p 144Hz

Link to comment
Share on other sites

Link to post
Share on other sites

Does Bottlenecking matter?

 

Uhm yeah, unless you don't care about money

PC Setup: 

HYTE Y60 White/Black + Custom ColdZero ventilation sidepanel

Intel Core i7-10700K + Corsair Hydro Series H100x

G.SKILL TridentZ RGB 32GB (F4-3600C16Q-32GTZR)

ASUS ROG STRIX RTX 3080Ti OC LC

ASUS ROG STRIX Z490-G GAMING (Wi-Fi)

Samsung EVO Plus 1TB

Samsung EVO Plus 1TB

Crucial MX500 2TB

Crucial MX300 1TB

Corsair HX1200i

 

Peripherals: 

Samsung Odyssey Neo G9 G95NC 57"

Samsung Odyssey Neo G7 32"

ASUS ROG Harpe Ace Aim Lab Edition Wireless

ASUS ROG Claymore II Wireless

ASUS ROG Sheath BLK LTD'

Corsair SP2500

Beyerdynamic DT 770 PRO X (Limited Editon) & Beyerdynamic TYGR 300R + FiiO K7 DAC/AMP

RØDE VideoMic II + Elgato WAVE Mic Arm

 

Racing SIM Setup: 

Sim-Lab GT1 EVO Sim Racing Cockpit + Sim-Lab GT1 EVO Single Screen holder

Svive Racing D1 Seat

Samsung Odyssey G9 49"

Simagic Alpha Mini

Simagic GT4 (Dual Clutch)

CSL Elite Pedals V2

Logitech K400 Plus

Link to comment
Share on other sites

Link to post
Share on other sites

I'm running a 2400G with a 1070 and I have yet to see any bottleneck at the moment.

Ryzen 5 3600 | Nvidia Quadro P400 | Gigabyte B550I Aorus Pro AX | Klevv Bolt XR 2x8GB DDR4-3600 CL18

Acer Chromebook C720 @ GalliumOS

Link to comment
Share on other sites

Link to post
Share on other sites

For the non-tech enthusiast a bottleneck isn't a big deal.  For a tech enthusiast who enjoys eeking out the performance of his machine/s - bottlenecking is a big deal.

 

Bottlenecking is also a big deal if your machine makes you income.  

 

Bottlenecking bad, and if not bad, than your use case doesn't care and that's okay too.

Workstation Laptop: Dell Precision 7540, Xeon E-2276M, 32gb DDR4, Quadro T2000 GPU, 4k display

Wifes Rig: ASRock B550m Riptide, Ryzen 5 5600X, Sapphire Nitro+ RX 6700 XT, 16gb (2x8) 3600mhz V-Color Skywalker RAM, ARESGAME AGS 850w PSU, 1tb WD Black SN750, 500gb Crucial m.2, DIYPC MA01-G case

My Rig: ASRock B450m Pro4, Ryzen 5 3600, ARESGAME River 5 CPU cooler, EVGA RTX 2060 KO, 16gb (2x8) 3600mhz TeamGroup T-Force RAM, ARESGAME AGV750w PSU, 1tb WD Black SN750 NVMe Win 10 boot drive, 3tb Hitachi 7200 RPM HDD, Fractal Design Focus G Mini custom painted.  

NVIDIA GeForce RTX 2060 video card benchmark result - AMD Ryzen 5 3600,ASRock B450M Pro4 (3dmark.com)

Daughter 1 Rig: ASrock B450 Pro4, Ryzen 7 1700 @ 4.2ghz all core 1.4vCore, AMD R9 Fury X w/ Swiftech KOMODO waterblock, Custom Loop 2x240mm + 1x120mm radiators in push/pull 16gb (2x8) Patriot Viper CL14 2666mhz RAM, Corsair HX850 PSU, 250gb Samsun 960 EVO NVMe Win 10 boot drive, 500gb Samsung 840 EVO SSD, 512GB TeamGroup MP30 M.2 SATA III SSD, SuperTalent 512gb SATA III SSD, CoolerMaster HAF XM Case. 

https://www.3dmark.com/3dm/37004594?

Daughter 2 Rig: ASUS B350-PRIME ATX, Ryzen 7 1700, Sapphire Nitro+ R9 Fury Tri-X, 16gb (2x8) 3200mhz V-Color Skywalker, ANTEC Earthwatts 750w PSU, MasterLiquid Lite 120 AIO cooler in Push/Pull config as rear exhaust, 250gb Samsung 850 Evo SSD, Patriot Burst 240gb SSD, Cougar MX330-X Case

 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×