News Which PC Boots Up and Shuts Down Faster: AMD or Intel?

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Sep 11, 2020
3
0
10
These Ubuntu boot times hurt my eyes, consider compiling your own kernel specific to the machine you use.

Also I don't know how you obtain these boot times on Ubuntu but on any modern Linux distribution that uses systemd as their init system, you can use systemd-analyze to query the boot time, which will tell you how long the system spent in firmware.
 

nofanneeded

Respectable
Sep 29, 2019
1,541
251
2,090
We might as well throw every cross-CPU comparison out the window then simply because they didn't use the same motherboard.


Do you have any proof of this? Just because it's Intel on Intel doesn't mean much. I may as well claim they should've use an NVIDIA video card because AMD systems can recognize and configure a Radeon card faster.

1- Booting time includes loading drivers for each hardware , this will increase booting time if the drivers take longer , thats why when you compare booting times , you need identical onboard hardware. and this has nothing to do with CPU speed comparisons .. this is about booting time only. Benchmarks will not be affected much when the system is running.

2- Intel on intel does not mean much ? really ? The whole system is labeled optane ready , I wouldn't be surprised that it is optimized to boot faster .. as for your GPU comparison , well yes if you want to test booting time , it is better to use neutral card .
 
Last edited:
Sep 11, 2020
4
5
15
The CPU does not make any difference, its the speed of your storage device. If your using a hardrive then your windows will boot up slower and be slower in general. Use an SSD or an M.2 NVme SSD for a faster Windows Experience.
Wrong
 
D

Deleted member 2783327

Guest
I have an MSI Creator X299 and MSI X299 XPower gaming AC and ASRock X299 Taichi's. Boot drives are Samsung 970 Evo Plus or 970 Pro's. They take 42-48 seconds to boot. Just under 30 seconds before the dots start spinning, then about 15-20 seconds from there. And I've disabled RAM test and set fast boot on to get them that low :(

The 970 Evo Plus consistently outperforms the Pro in pretty much everything, including boot up.

But this is apparently normal for HEDT systems. I would so love to have sub 20 second boot up times :)
 
  • Like
Reactions: nofanneeded

rdmetz

Distinguished
Sep 9, 2015
10
4
18,515
Do you have anything worthwhile to say?

Do you?



He's pointing out an obvious bias from a ton of people having a very hard time accepting their decision led to them not having the absolute best at everything setup.



The amount of reaching being done here is next level and the fact is ryzen isn't "better at everything" they just don't want to hear it.


So many angry AMD fanboys, it's not a surprise to anyone who has owned a CPU from both sides of the market that intel boots faster. I upgraded from a 2-3 second boot on an old intel chip, to a 14-second boot on a new AMD one.
That's if the thing even boots at all.

My friends with ryzen builds (1800x 2700x and 3700x) have all had problems getting their systems to post the first 2 seeing this issue continue on even to today.

It's just random luck whether the next boot will actually happen or not.

But it's led them to the same findings as most of the ryzen defenders here in these comments.

Leave it on as much as you can to avoid dealing with it. Lol


My own thoughts on this is that so many want to try and make excuses and explain away the advantage while others who accept it just say it doesn't matter since they don't do it very often.

I hear that from people who went out and bought a cpu for their gaming pc that actually games at lower resolution WHILE claiming its because it's faster in other areas (areas they NEVER actually do anything with).

I guarantee if you ran a 6 month usage analysis on a huge portion of these amd believers you'd find that they only used their systems in ways where if they had Intel they would have had higher performance with never losing out anywhere else... Because they never do anything else with it. Just gaming.

How their fragile egos would crumble and the excuses would flow but it's true the competition can be 10x better at every other task but if your usage is 0% of those other tasks then your advantage is 0%.

Meanwhile you're DEFINITELY throwing actual useful performance in something most are doing 90% if the time or more with their setups. With advantages up to 20-30% in the ACTUAL tasks you're doing daily that literally just throwing away money.

Maybe everyone shouldn't jump off the bridge just because everyone else is doing it too. (they might have had a reason and a parachute but you just look screwed)
 
Last edited:
  • Like
Reactions: carvedinside
Do you?



He's pointing out an obvious bias from a ton of people having a very hard time accepting their decision led to them not having the absolute best at everything setup.



The amount of reaching being done here is next level and the fact is ryzen isn't "better at everything" they just don't want to hear it.



That's if the thing even boots at all.

My friends with ryzen builds (1800x 2700x and 3700x) have all had problems getting their systems to post the first 2 seeing this issue continue on even to today.

It's just random luck whether the next boot will actually happen or not.

But it's led them to the same findings as most of the ryzen defenders here in these comments.

Leave it on as much as you can to avoid dealing with it. Lol


My own thoughts on this is that so many want to try and make excuses and explain away the advantage while others who accept it just say it doesn't matter since they don't do it very often.

I hear that from people who went out and bought a cpu for their gaming pc that actually games at lower resolution WHILE claiming its because it's faster in other areas (areas they NEVER actually do anything with).

I guarantee if you ran a 6 month usage analysis on a huge portion of these amd believers you'd find that they only used their systems in ways where if they had Intel they would have had higher performance with never losing out anywhere else... Because they never do anything else with it. Just gaming.

How their fragile egos would crumble and the excuses would flow but it's true the competition can be 10x better at every other task but if your usage is 0% of those other tasks then your advantage is 0%.

Meanwhile you're DEFINITELY throwing actual useful performance in something most are doing 90% if the time or more with their setups. With advantages up to 20-30% in the ACTUAL tasks you're doing daily that literally just throwing away money.

Maybe everyone shouldn't jump off the bridge just because everyone else is doing it too. (they might have had a reason and a parachute but you just look screwed)

Honestly?

There are 3 people spouting fanboyism into this thread, of which you're the main contributor.

The simple fact is 90% of us ACTUALLY don't care at all about a 5 or 6 second boot time difference.

Fanboyism has nothing to do with it.
 
  • Like
Reactions: King_V

nings

Commendable
Mar 5, 2018
14
2
1,515
Again a fake news from Toms Hardware. You are computer specialist or not ?

Windows XP have finished his booting process when reach the desktop. Vista too but when all customers complaint Microsoft about VERY SLOW boot time create a new trick for speed up desktop appear without start most services and other windows slow boot process but the system need 2 more minutes (on HDD) for complete the boot process. Try to launch a VM on a computer with HDD on the startup folder of Windows 7/10 that's can take a pretty high amount of time. You do the same on Windows XP or Linux your VM boot very fast even if the boot process can take 10 more seconds. I prefer have a computer 100% resources available than Windows 7/10 who take 99% of the resources available one time on the desktop and catch computer newbie who are happy to view their desktop quickly but just need to wait more for can begin working at a correct speed.

Have you try a linux distribution without systemd because that's a little the same than windows add a long boot delay for my own experience. Come back on a Devuan without systemd give a lot better boot process but even without that I'm 100% sure than systemd have finished his boot process and free up 100% available resources when Windows 7/10 take a long time but a HDD show you that really better than a "multi-tasks" sata or nvme drive.

Please review your own news before posts some people have some knowledge more and more disappointed about Tom's Hardware. And your war with Intel/AMD where you try to strike AMD but AMD have already win and take months after months market share to Intel since desktop Ryzen 3xxxx. Check AMD/Intel sales and Nasdaq stock price !
Intel have a biggest market share and produce more than just CPU but the price is a lot lower than AMD.
Instead to try to destroy AMD invest on AMD when produce good CPU and enjoy your 2.5x reward (stock price from summer 2019 ~30$ to januar 2020 60$ and now still at ~80$). I let look your Intel company from 2019 to now.
 

nings

Commendable
Mar 5, 2018
14
2
1,515
And sorry for the boot time may be true but it's two different chipset and two different BIOS from only one company. You need more manufacturer for say one chipset is faster than another.

Boot time has two phases :
  1. POST push on the power button until boot loader
  2. OS Starting from bootloader until the computer let user use 100% of the computer resources so CPU at ~0% and SSD at ~0% too.

So you need to let the Windows boot loader pop and press enter when you need to select the system until windows launch task manager on resources usage and wait than Windows finish is boot process and free all resources for user. On Linux it's the same boot loader show in all case.
And now you can give a true number with the real Windows boot time only CPU/chipset dependent without POST time who is mainly caused by the manufacturer. Same for linux where you can see all time gain to have a system 100% usable.

So please be computer specialist and write good news !

Thank you for all newbie people who not know really good all software/hardware trick but yes you can continue to give BIOS/UEFI trick for speedup POST process.
 
Again a fake news from Toms Hardware. You are computer specialist or not ?

Windows XP have finished his booting process when reach the desktop. Vista too but when all customers complaint Microsoft about VERY SLOW boot time create a new trick for speed up desktop appear without start most services and other windows slow boot process but the system need 2 more minutes (on HDD) for complete the boot process. Try to launch a VM on a computer with HDD on the startup folder of Windows 7/10 that's can take a pretty high amount of time. You do the same on Windows XP or Linux your VM boot very fast even if the boot process can take 10 more seconds. I prefer have a computer 100% resources available than Windows 7/10 who take 99% of the resources available one time on the desktop and catch computer newbie who are happy to view their desktop quickly but just need to wait more for can begin working at a correct speed.

Have you try a linux distribution without systemd because that's a little the same than windows add a long boot delay for my own experience. Come back on a Devuan without systemd give a lot better boot process but even without that I'm 100% sure than systemd have finished his boot process and free up 100% available resources when Windows 7/10 take a long time but a HDD show you that really better than a "multi-tasks" sata or nvme drive.

Please review your own news before posts some people have some knowledge more and more disappointed about Tom's Hardware. And your war with Intel/AMD where you try to strike AMD but AMD have already win and take months after months market share to Intel since desktop Ryzen 3xxxx. Check AMD/Intel sales and Nasdaq stock price !
Intel have a biggest market share and produce more than just CPU but the price is a lot lower than AMD.
Instead to try to destroy AMD invest on AMD when produce good CPU and enjoy your 2.5x reward (stock price from summer 2019 ~30$ to januar 2020 60$ and now still at ~80$). I let look your Intel company from 2019 to now.

You're adding fuel to the fire when it comes to fanboyism posts.

The article wasn't biased at all, it's people's views that are biased when see see it as detrimental towards AMD, which it isn't purposefully.

I'll repeat it again '90+% of users do not care about a 6 second boot time/shutdown difference'

It has total irrelevance to actual system performance once inside the OS.
 
  • Like
Reactions: King_V

st379

Distinguished
Aug 24, 2013
169
69
18,660
First I want to say that I don't like articles where Amd loses. (Sorry, had to troll all the fanboys who went full crazy on a boot up articles.)

My pc 90% of the time is in sleep mode. Only doing restarts on windows update.
But if you decided to do this article I think you should know that different motherboards vendors and chipsets gives different boot up time.
If you could at least add Asrock vs Gigabyte vs Asus vs Msi it would be nice otherwise this article is pretty pointless.
If you decided to dive into this topic at least don't give one motherboard vendor and one chipset. It is a little bit unprofessional in my opinion.

Before all the fanboys from Intel will go full crazy on me I want to calm you down and say that I am sure that Intel will win. Just want to see Msi vs Asrock vs Gigabyte vs Asus.
 
Last edited:
So many angry AMD fanboys, it's not a surprise to anyone who has owned a CPU from both sides of the market that intel boots faster. I upgraded from a 2-3 second boot on an old intel chip, to a 14-second boot on a new AMD one.

Yeah AMD takes an eternity to POST compared to Intel. The Windows boot time is almost the same but the time between switching on the power and the POST sound is atleast 5x higher than Intel. It matters to me a lot since I shutdown my PC whenever I am not using it. I was wondering why weren't any sites doing Time to POST benchmarks. You can just do a google search and see how many people are complaining
 

deksman

Distinguished
Aug 29, 2011
233
19
18,685
It looks like AGESA MOBO firmware from AMD's side takes a bit longer to initialize than Intel, otherwise I believe the actual boot up times will be almost 1:1.

If that's the case, it shouldn't be surprising when you consider the fact that AMD has multiple CPU generations written into the said AGESA... whereas Intel does not.
 
If that's the case, it shouldn't be surprising when you consider the fact that AMD has multiple CPU generations written into the said AGESA... whereas Intel does not.
If the agesa has to go through every entry in its list to figure out which CPU is in the socket then that's the worst design ever.
Or is there something else there, how is supporting multiple CPU generations supposed to make post slower?
 

seanwebster

Contributing Writer
Editor
Aug 30, 2018
191
68
10,690
1- First , You cant compare two different motherboards in booting times , because hardware wise these two motherboard are not 1:1 identical .

2- Second , Using Optane is wrong for intel is more optimized to recognize Optane SSD faster on booting .. you should test this again using something neutral like Samsung NVME 970 pro , or 970 evo plus.

1 - I did. You can't use an Intel chip in an AMD motherboard and vice versa...They are both very similar boards overall. Both ASRock's Taichis. I also used an Asus ROG Crosshair VIII Hero Wifi. 6C/12TB CPUs - close enough. I left stock boost/turbo settings. I could probably see what locking both at the same GHz could do to change the results.

2 - My preliminary testing says otherwise. NAND flash can not match Optane's read performance at boot.

I agree with this. Most people (or all the people I know) do not use optane. The results might be slower with a samsung evo, but definitely interested in seeing those results just to know how much the optane is worth it for both intel and amd systems.

It's literally within a second or two difference. I just used the Intel Optane SSD 905P to remove my storage bottleneck.

These Ubuntu boot times hurt my eyes, consider compiling your own kernel specific to the machine you use.

Also I don't know how you obtain these boot times on Ubuntu but on any modern Linux distribution that uses systemd as their init system, you can use systemd-analyze to query the boot time, which will tell you how long the system spent in firmware.
Yeah, I used systemd some. I decided to just keep things simple tho. The point is to compare hardware. I just wanted to see how the chipset/CPU combo made a difference more than OS, so I left both OSes fairly stock. Utilizing Ubuntu just verified/supported the Windows results.

Times captured via camera/stopwatch - 20 iterations of each test on each system. Boot times are measured from powered off to desktop reveal. Reboots after 2 minutes idle time to desktop. Shutdowns after 2 minutes of idle time until the PSU switches off. The UEFI on the AMD system takes longer to load by a few seconds, hence most of the difference.

1- Booting time includes loading drivers for each hardware , this will increase booting time if the drivers take longer , thats why when you compare booting times , you need identical onboard hardware. and this has nothing to do with CPU speed comparisons .. this is about booting time only. Benchmarks will not be affected much when the system is running.

2- Intel on intel does not mean much ? really ? The whole system is labeled optane ready , I wouldn't be surprised that it is optimized to boot faster .. as for your GPU comparison , well yes if you want to test booting time , it is better to use neutral card .
Intel's Optane 905P boots faster than Gen4 SSDs to the desktop on both platforms. Its random read performance is 5-6x faster than most NAND flash-based SSDs.

I have an MSI Creator X299 and MSI X299 XPower gaming AC and ASRock X299 Taichi's. Boot drives are Samsung 970 Evo Plus or 970 Pro's. They take 42-48 seconds to boot. Just under 30 seconds before the dots start spinning, then about 15-20 seconds from there. And I've disabled RAM test and set fast boot on to get them that low :(

The 970 Evo Plus consistently outperforms the Pro in pretty much everything, including boot up.

But this is apparently normal for HEDT systems. I would so love to have sub 20 second boot up times :)
Unfortunately so. My old X99 + 12C/24T system took over 37 seconds to boot up after optimizing my UEFI settings lol.

Again a fake news from Toms Hardware. You are computer specialist or not ?

Windows XP have finished his booting process when reach the desktop. Vista too but when all customers complaint Microsoft about VERY SLOW boot time create a new trick for speed up desktop appear without start most services and other windows slow boot process but the system need 2 more minutes (on HDD) for complete the boot process. Try to launch a VM on a computer with HDD on the startup folder of Windows 7/10 that's can take a pretty high amount of time. You do the same on Windows XP or Linux your VM boot very fast even if the boot process can take 10 more seconds. I prefer have a computer 100% resources available than Windows 7/10 who take 99% of the resources available one time on the desktop and catch computer newbie who are happy to view their desktop quickly but just need to wait more for can begin working at a correct speed.

Windows XP, Vista and 7? They are all irrelevant in 2020. Why bother with an HDD ofr OS/VM usage? They are slow and flash is affordable.

When testing my boot speeds, the resource manager showed little to no activity after booting for minutes after booting. Again, I'm using one of the world's fastest SSDs for this. Not archaic tech that takes a few ms to respond to a request.

Have you try a linux distribution without systemd because that's a little the same than windows add a long boot delay for my own experience. Come back on a Devuan without systemd give a lot better boot process but even without that I'm 100% sure than systemd have finished his boot process and free up 100% available resources when Windows 7/10 take a long time but a HDD show you that really better than a "multi-tasks" sata or nvme drive.

Please review your own news before posts some people have some knowledge more and more disappointed about Tom's Hardware. And your war with Intel/AMD where you try to strike AMD but AMD have already win and take months after months market share to Intel since desktop Ryzen 3xxxx. Check AMD/Intel sales and Nasdaq stock price !

Intel have a biggest market share and produce more than just CPU but the price is a lot lower than AMD.
Instead to try to destroy AMD invest on AMD when produce good CPU and enjoy your 2.5x reward (stock price from summer 2019 ~30$ to januar 2020 60$ and now still at ~80$). I let look your Intel company from 2019 to now.
I simply chose Ubuntu as it is the most popular distribution. We could go more in-depth and test various distros if that is something of enough interest. This article is mainly about hardware. While Windows beats Ubuntu in boot speed, the focus was not Linux vs Windows. Ubuntu was just used as supporting evidence.

And sorry for the boot time may be true but it's two different chipset and two different BIOS from only one company. You need more manufacturer for say one chipset is faster than another.

Boot time has two phases :
  1. POST push on the power button until boot loader
  2. OS Starting from bootloader until the computer let user use 100% of the computer resources so CPU at ~0% and SSD at ~0% too.
So you need to let the Windows boot loader pop and press enter when you need to select the system until windows launch task manager on resources usage and wait than Windows finish is boot process and free all resources for user. On Linux it's the same boot loader show in all case.
And now you can give a true number with the real Windows boot time only CPU/chipset dependent without POST time who is mainly caused by the manufacturer. Same for linux where you can see all time gain to have a system 100% usable.

So please be computer specialist and write good news !

Thank you for all newbie people who not know really good all software/hardware trick but yes you can continue to give BIOS/UEFI trick for speedup POST process.

What you mention I have already done. I just haven't graphed all my data for this article. For the purpose of this article, it's more about overall times (the user experience) rather than breaking them down bit by bit like the one I linked to in the beginning. I've written articles about storage and OS boot processes multiple times in the past decade now. I do know what is what.

First I want to say that I don't like articles where Amd loses. (Sorry, had to troll all the fanboys who went full crazy on a boot up articles.)

My pc 90% of the time is in sleep mode. Only doing restarts on windows update.
But if you decided to do this article I think you should know that different motherboards vendors and chipsets gives different boot up time.
If you could at least add Asrock vs Gigabyte vs Asus vs Msi it would be nice otherwise this article is pretty pointless.
If you decided to dive into this topic at least don't give one motherboard vendor and one chipset. It is a little bit unprofessional in my opinion.

Before all the fanboys from Intel will go full crazy on me I want to calm you down and say that I am sure that Intel will win. Just want to see Msi vs Asrock vs Gigabyte vs Asus.

My Asus X570 ROG Crosshair VIII Hero Wifi shows similar boot results to the ASRock, actually a second slower. I doubt that a Gigabyte or MSI board will boot faster. I haven't experienced such with any of their boards I've used, but who knows 100% without testing, right?

If the manufacturers want to sponsor more testing on their platforms, that can always be arranged. I chose ASRock's X490 Taichi because it was the only reasonably priced USB 3.2 Gen 2x2 supporting motherboard out (I need for portable SSD reviews) and since I had an ASRock X570 Taichi already for internal SSD testing, I figured they were close enough to one another...it 's much better than comparing say an eATX motherboard with a bunch of optional controllers to a mini iTX or having different chipsets as well as different brands and saying its apples to apples, lol.
 
Last edited:

Theresa N

Admirable
Dec 10, 2019
88
18
7,165
I normally boot my AMD x570, 3900x system once a day and it boots up plenty fast for me. Shutting down is slowed down by one windows process that takes several seconds to shut down. I've never felt distressed by "slow" boot times.
 
Sep 12, 2020
1
0
10

st379

Distinguished
Aug 24, 2013
169
69
18,660
My Asus X570 ROG Crosshair VIII Hero Wifi shows similar boot results to the ASRock, actually a second slower. I doubt that a Gigabyte or MSI board will boot faster. I haven't experienced such with any of their boards I've used, but who knows 100% without testing, right?

If the manufacturers want to sponsor more testing on their platforms, that can always be arranged. I chose ASRock's X490 Taichi because it was the only reasonably priced USB 3.2 Gen 2x2 supporting motherboard out (I need for portable SSD reviews) and since I had an ASRock X570 Taichi already for internal SSD testing, I figured they were close enough to one another...it 's much better than comparing say an eATX motherboard with a bunch of optional controllers to a mini iTX or having different chipsets as well as different brands and saying its apples to apples, lol.
I don't want apples to apples. I would like to see x470, b460, x570, b550, mini itx, m-atx full atx, Gigabyte, Asus, Asrock, Msi......

I am sorry one vendor and one chipset it is not enough.

With my once a month reboot this topic does not concern me but if you decided to open this pandora box at least add a little bit variety in my opinion.

I just think that one sample is too little to publish an article and draw a conclusion.
I would never tackle this subject it was your choice so if you decided to go for it I would love to see more motherboards.

My opinion anyway. Thank you for the article and I hope you will update it for the one who do care about 10 seconds boot time.
 
Last edited:
Sep 12, 2020
2
0
10
What shill I write about today? shill I write an article that shill provide people with useful information, or shill I extol the virtues of a platform that, this year, shill be based on 6 year-old architecture that shill make it look better for my benefcators preferred platform? I know...I shill do the latter!
 
Last edited:

COLGeek

Cybernaut
Moderator
Since some of you seem to lack objectivity, I would like you to understand that TH is not paid by vendors to review products.

Attacking the authors serves no other purpose but to expose your own biases and insecurities.

Civility and respect toward others is required by all. Attack ideas, with facts and sources, don't attack people.
 
  • Like
Reactions: madmatt30
Sep 12, 2020
2
0
10
The fact that Intel isn't paying you for this article doesn't automatically mean that you you can't be biased. To me this article comes across as heavily biased towards Intel, especially given that you chose to use an Intel Optane SATA drive (providing an additional plug for Intel hardware, and hardly representative of what the average consumer would buy) rather than PCI 4.0 NVMe - which would have been limited to PCI 3.0 on the Intel platform. Also, let's not discount your convenient highlighting of 'fastest Windows 10 boot times we hit an amazing 4.9 seconds on an Intel platform' - even though that had nothing to do with these tests. But if it makes you happy, I'll change my previous statement.
 
Last edited:

nofanneeded

Respectable
Sep 29, 2019
1,541
251
2,090
2 - My preliminary testing says otherwise. NAND flash can not match Optane's read performance at boot.

We know Optane are faster in booting , but to test Intel or AMD is faster you should use neutral SSD on both and see which one is faster in booting , we are not benchmarking booting time here we are comparing which board boots faster. And anyways almost 90% of people dont use optane because they are very expansive per TB ., so this Comparison helps no one in reality ...
 
Last edited: