The chipset doesn't run hot and the fan generally doesn't come on unless you are hitting it really hard with most MB. I have never noticed it on my gigabyte x570 master. People need to calm their breeches over the damn fan. Also you don't have to buy a $700 MB. There have been x570 boards as low as $150. It is not something unique to AMD. Intel also has stupid expensive MB.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
X570 Chipset Fans...
- Thread starter Aix.
- Start date
Oh true, there are boards that are in that price range on both sides, no issue there. But any board that requires a fan in that location is a hard no from me. If it didn't run hot, it wouldn't need a fan. I'm pretty sure the engineers did the math.
I repasted my chipset fan on my MSI Gaming Edge and it doesn't run anymore, nor does it exceed 60c. 75c is when the fan kicks on. Never does anymore since a repaste.
The fan is a none issue. Most people wouldn't even notice if it randomly died I bet.
Good to know. I was considering that if I bought one of these boards doing exactly that. 15W shouldn't really require a fan imho, but evidently they thought it does. Must have been in the specs for it to appear on literally every board but one.
Now if only Ryzen was winning the single thread war I might be willing to go that extra mile, but that's another topic. I can wait an extra minute for a render. Right now the actual benefit of PCIE4 is exactly zero (in consumer space) that I know of and I don't see it as anything special in the next three to five years. Maybe we see some graphics cards or drives that can actually use these speeds in a meaningful way but right now that's not the case for graphics and drives of this nature drop out after cache is exhausted that I know of. I'm not sure why these scenarios would run hot?
What I do know is that they don't pay extra for fans instead of heatsinks without a reason, especially when they had to have known this would have a negative impression on any system builder old enough to remember fans on chipsets.
Now if only Ryzen was winning the single thread war I might be willing to go that extra mile, but that's another topic. I can wait an extra minute for a render. Right now the actual benefit of PCIE4 is exactly zero (in consumer space) that I know of and I don't see it as anything special in the next three to five years. Maybe we see some graphics cards or drives that can actually use these speeds in a meaningful way but right now that's not the case for graphics and drives of this nature drop out after cache is exhausted that I know of. I'm not sure why these scenarios would run hot?
What I do know is that they don't pay extra for fans instead of heatsinks without a reason, especially when they had to have known this would have a negative impression on any system builder old enough to remember fans on chipsets.
If you max out the IO on the chipset with all PCI-e gen 4 SSDs and GPUs then the fan may be required. That's the only guess I could give. Most of these fans are designed like laptop fans and are pretty durable for longevity as long as you keep it clean from dust.
That's really all I could think of as well. Some of them are saying how durable the Delta fan they installed is, 60,000 hours and all that. Don't they know what images come to mind when you say "delta fan"? Delta isn't a name I think of as being quiet, but at least I can think of that as reliable.
The actual 1st x16 slot and one of the nvme has a cpu link, as well as 4 usb 10gbps. If I understand that correctly that's not even running through that part of the board.
That leaves everything else running over the chipset into the cpu over a pcie 4.0 x4 link. I suppose some calculation indicated it could overheat, but that's kind of the point where you say it's not a good design... and that's generally obvious by the presence of a fan. It really shouldn't require active cooling.
There are ways to deal with this, of course, but I generally prefer designs where I don't have to consider these issues. If I'm missing something here point it out.
The actual 1st x16 slot and one of the nvme has a cpu link, as well as 4 usb 10gbps. If I understand that correctly that's not even running through that part of the board.
That leaves everything else running over the chipset into the cpu over a pcie 4.0 x4 link. I suppose some calculation indicated it could overheat, but that's kind of the point where you say it's not a good design... and that's generally obvious by the presence of a fan. It really shouldn't require active cooling.
There are ways to deal with this, of course, but I generally prefer designs where I don't have to consider these issues. If I'm missing something here point it out.
You are making mountains out of ant hills.
Denpepe
2[H]4U
- Joined
- Oct 26, 2015
- Messages
- 2,269
CPU's and graphics cards use fans are they also badly designed then?
GotNoRice
[H]F Junkie
- Joined
- Jul 11, 2001
- Messages
- 11,994
What you're missing is that these chipsets are not designed with a typical usage scenario in mind, they are designed with a worst-case scenario in mind. They are designed for the guy living in the desert who builds a computer with no case fans and then leaves his computer on all day long with the AC turned off. I'm talking about ambient temperatures in the high 80's, if not higher. In your typical home/work environment, in more forgiving climates and/or anyone who uses air conditioning when it gets hot, the fan is a complete non-issue.
On my x570 Aorus Ultra, there are 3 selectable fan profiles, and even the most aggressive still allows the chipset fan to completely turn off at low temperatures. And keep in mind, just because a chipset doesn't have a fan doesn't mean it runs cool - it just means it's probably throttling instead whenever it gets hot. If you prefer that route, then just select the "Silent" fan profile and let it roast.
You can see that in all the fan profiles, the fan doesn't kick up to anywhere near full speed until the PCH temp starts approaching 90C. If your PCH is getting that hot, you have bigger issues.
"Silent":
"Balance":
"Performance":
Last edited:
Addressing this reply specifically, I think you know that's not a direct comparison. These parts are expected to be actively cooled and have much more robust solutions for doing so.
Denpepe
2[H]4U
- Joined
- Oct 26, 2015
- Messages
- 2,269
Yet when not stressed certain (high end) graphics cards fans do not spin up. I know it's not a direct comparison, but it's better to have an option for cooling and not needing it then not having it and needing it.
If marketing is to be believed most chipsets are designed to be subjected to these conditions as well, without fans. I'm not sure how much throttling occurs under full load or heated conditions, I've never tested this, and I've never seen such reports. I'm not saying that information doesn't exist, obviously it does, somewhere. What is a direct comparison in the most obvious sense is chipsets running the same CPU, or even chipsets running different CPU's on x86. In B550, where the layout is more traditional and the 4x link to the cpu is pcie3.0 x4, this solution doesn't require active cooling.
So something changed, and that change led to the decision to actively cool the chipset. I don't pretend to be an expert on these matters, but I know to ask questions about what I don't understand. What it appears like to me is that too much was put over the chipset for passive cooling to be a reliable solution, and your fan graph at the most aggressive profile shows fan on at what, 40C? I can't imagine there's any throttling there, and at what point this, or another chipset throttles is unknown to me.
I can put a spare tire on a car and it will drive, that doesn't mean I want to do that all the time. And yes, you could turn down the fan profile and let it cook, but if that was how it was designed to operate (and plenty of systems do just that, but not in consumer space) then the fan wouldn't be there in the first place. Somewhere along the line some engineer said if you do it this way and it gets too hot performance will suffer, it will fail, or both. The point being that without specific understanding of what exactly happens when it gets too hot and under what use cases this occurs one cannot really look at the design and trust that it is solid without its cooling solution. I'm not trying to bash what otherwise seems to be a good product, but I would be far more comfortable if the chipset on the board wasn't prone to failure without active cooling under any scenario in which it is designed to operate.
I don't think it should be needed on a chipset, and if it is, it shouldn't be located where the GPU is going to go over an active cooling solution. The whole setup there speaks of bad design and a point of failure. There are a ton of boards out there that don't require this, just not for x570.
wait 2 months for Zen 3 boards...
I'm thinking you're not wrong here. Whatever comes next for motherboards on AMD should be really good.
GotNoRice
[H]F Junkie
- Joined
- Jul 11, 2001
- Messages
- 11,994
Throttling is actually very common these days in many different cases. For example, many SSD controllers will throttle at higher temperatures. It's one of the reason why so many motherboards these days include heatsinks for the M.2 slots. Those SSD controllers are still designed to work just fine without cooling, but obviously cooling is preferable over throttling if given the option.
It's a common theme in many areas of computing that higher data transfer rates cause increased temperatures in the chips that facilitate that data transfer. That's true with everything from SSD controllers, Network controllers, CPU's, and yes, chipsets. If PCIe 4.0 is the reason why x570 potentially requires occasional active cooling, then a takeaway from that is that in most cases, when you're not pushing speeds above PCIe 3.0, the chipset probably won't get very hot, and the fan will spend more time turned off. But again, AMD has to plan for a worst case scenario - someone pushing the PCIe 4.0 bus to it's limit in a hot environment.
PCIe 4.0, obviously. What chipset does PCIe 4.0 with passive cooling? None?
There is certainly no throttling at 40C. The "Performance" fan profile is not the default fan profile. If you purposefully go into the bios and select the most aggressive fan profile, then obviously you are willing to potentially live with a bit of noise for the sake of keeping temps cooler. Really no different than any other device with an adjustable fan profile. As far as what temperature it throttles, I don't know either. One thing all three graphs have in common though, is that above 85C all bets are off and the chipset needs the FAN at that point. So the throttle temp is probably in the 90-100C range, which is actually pretty typical.
The fan doesn't turn on all the time. In most enthusiast systems, with decent case airflow, the fan will be off most of the time. So just like a spare tire, it's not something you use all the time, but it's nice to have when you need it.
Again, a chipset has to be designed with the worst-case scenario in mind. The AMD engineers have to design the chipset without knowing the exact environment it will operate in. They don't know what case the motherboard will be installed in. They don't know what kind of airflow that case will have. They don't know what the ambient temperatures will be. If they didn't plan for a worst case scenario, they could easily end up in a situation like with the old nForce chipsets that got a reputation for cooking themselves to death in cases with poor airflow.
Thankfully we now have people who have been using x570 boards for upwards of a year now and have plenty of empirical data that shows how the fan operates. It doesn't run most of the time in a typical enthusiast case. Just because a motherboard has a chipset fan doesn't mean it's a repeat of what we saw on your typical VIA athlon-era motherboard ~20 years ago, where they typically used cheap fans that weren't even thermally controlled and just ran full blast 24/7 until they failed.
If your usage scenario doesn't correspond with the worst-case scenario that they had to design for, then why does it bother you so much?
Throttling is actually very common these days in many different cases. For example, many SSD controllers will throttle at higher temperatures. It's one of the reason why so many motherboards these days include heatsinks for the M.2 slots. Those SSD controllers are still designed to work just fine without cooling, but obviously cooling is preferable over throttling if given the option.
That's one point of view, mine happens to be that it's preferable in this case that a chipset operate without the need for active cooling, as active cooling will fail at some point, whereas passive cooling, properly designed, has been proven to last the life of the system, which in many cases is beyond it's relevant lifespan.
It's a common theme in many areas of computing that higher data transfer rates cause increased temperatures in the chips that facilitate that data transfer. That's true with everything from SSD controllers, Network controllers, CPU's, and yes, chipsets. If PCIe 4.0 is the reason why x570 potentially requires occasional active cooling, then a takeaway from that is that in most cases, when you're not pushing speeds above PCIe 3.0, the chipset probably won't get very hot, and the fan will spend more time turned off. But again, AMD has to plan for a worst case scenario - someone pushing the PCIe 4.0 bus to it's limit in a hot environment.
Agreed, but this is true of all systems, and even on B550 running the same CPU this isn't needed, or isn't present.
PCIe 4.0, obviously. What chipset does PCIe 4.0 with passive cooling? None?
As discussed earlier, the only board that does this is $700. It's possible, at some absurd price point, but I would like to think it could be done at a much lower price.
There is certainly no throttling at 40C. The "Performance" fan profile is not the default fan profile. If you purposefully go into the bios and select the most aggressive fan profile, then obviously you are willing to potentially live with a bit of noise for the sake of keeping temps cooler. Really no different than any other device with an adjustable fan profile. As far as what temperature it throttles, I don't know either. One thing all three graphs have in common though, is that above 85C all bets are off and the chipset needs the FAN at that point. So the throttle temp is probably in the 90-100C range, which is actually pretty typical.
Agreed, what concerns me is that there is some design consideration here that suggests this could be enough of a common issue to require an active cooling solution. I'm not a ... wait for it ... fan of that.
The fan doesn't turn on all the time. In most enthusiast systems, with decent case airflow, the fan will be off most of the time. So just like a spare tire, it's not something you use all the time, but it's nice to have when you need it.
Yes, well, you don't preinstall a spare tire on the actual hub, you put it in the trunk or under the car.
Again, a chipset has to be designed with the worst-case scenario in mind. The AMD engineers have to design the chipset without knowing the exact environment it will operate in. They don't know what case the motherboard will be installed in. They don't know what kind of airflow that case will have. They don't know what the ambient temperatures will be. If they didn't plan for a worst case scenario, they could easily end up in a situation like with the old nForce chipsets that got a reputation for cooking themselves to death in cases with poor airflow.
Ah, now here's the rub, and the mountain of it for me. I remember these designs and the same old arguments of "it's fine, no big deal" and blah blah blah. It wasn't then, and it likely isn't now, which was what I was getting at when I said not in the last ten years and not now. It is exactly the small things that have kept me on Intel for so long. I'd like to give AMD a chance, especially if I believe they have a good product. I tend to look for a bulletproof design, and this isn't it. It may be fine if you are less demanding than I am, but poor people can't buy twice. If I had a bigger budget I'd just buy both systems and let it ride and find out.
Thankfully we now have people who have been using x570 boards for upwards of a year now and have plenty of empirical data that shows how the fan operates. It doesn't run most of the time in a typical enthusiast case. Just because a motherboard has a chipset fan doesn't mean it's a repeat of what we saw on your typical VIA athlon-era motherboard ~20 years ago, where they typically used cheap fans that weren't even thermally controlled and just ran full blast 24/7 until they failed.
And yet, here we are cooling a chipset with a fan.
I can keep it cool with a variety of solutions. I can also buy something that has no need of that. That's kind of exactly the point, why would I buy something that is designed to need active cooling when other solutions don't?
If your usage scenario doesn't correspond with the worst-case scenario that they had to design for, then why does it bother you so much?
I think I summed this up earlier, but just to say it again, I would be far more comfortable if the chipset on the board wasn't prone to failure without active cooling under any scenario in which it is designed to operate. I have never cared for any system designed in this way when alternatives exist. My use case today may not be the same in a year or two, especially with my current plans.
mvmiller12
[H]ard|Gawd
- Joined
- Aug 7, 2011
- Messages
- 1,519
Honestly, I just worked around the entire issue by using my 3900X CPU on a Crosshair VI Extreme (x370 chipset) board. It is, IMHO, probably the single best overall AM4 board Asus has yet produced.
Stock + PBO CPU performance is the same on this board as on the latest and greatest Crosshair VIII (x570) boards, and I am running my DDR4-3600 RAM at 3600 without issue. F-Clock is set to 1800, also without issue. The only things I am missing are PCIe 4.0 support and compatibility with the last AM4 (DDR4) CPUs that will be produced (Ryzen 4000 series). PCIe 4.0 is useless to me - it does nothing for graphics cards, and PCIe 4.0 SSDs are STOOOOOOPID expensive for a performance uplift that will only be seen in benchmarks for the vast majority of users (myself included). As to the lack of Ryzen 4000 support? Well...
This motherboard has hosted a Ryzen 1700, 2700X, and a 3900X. It will eventually host a 3950X because I can actually make use of those cores and threads. That is a LOT of upgradability on a single board, and so I don't really feel gypped on the lack of upcoming support for Ryzen 4000. Honestly, will just skip that generation in favor of the new DDR5 boards and CPUs that will follow it. I'm sure the 3900X, much less the 3950X I plan to get next year, will "get me by" until then.
Stock + PBO CPU performance is the same on this board as on the latest and greatest Crosshair VIII (x570) boards, and I am running my DDR4-3600 RAM at 3600 without issue. F-Clock is set to 1800, also without issue. The only things I am missing are PCIe 4.0 support and compatibility with the last AM4 (DDR4) CPUs that will be produced (Ryzen 4000 series). PCIe 4.0 is useless to me - it does nothing for graphics cards, and PCIe 4.0 SSDs are STOOOOOOPID expensive for a performance uplift that will only be seen in benchmarks for the vast majority of users (myself included). As to the lack of Ryzen 4000 support? Well...
This motherboard has hosted a Ryzen 1700, 2700X, and a 3900X. It will eventually host a 3950X because I can actually make use of those cores and threads. That is a LOT of upgradability on a single board, and so I don't really feel gypped on the lack of upcoming support for Ryzen 4000. Honestly, will just skip that generation in favor of the new DDR5 boards and CPUs that will follow it. I'm sure the 3900X, much less the 3950X I plan to get next year, will "get me by" until then.
sirmonkey1985
[H]ard|DCer of the Month - July 2010
- Joined
- Sep 13, 2008
- Messages
- 22,414
considering he could of saved him self the last 2 days of needlessly typing useless "i saw it on reddit so it must be right" crap and just looked at the first page of this thread, he would of known why the fan was there in the first place. not worth the time or energy with that guy since his opinion is the only opinion that matters to him.
Jamie Marsala
Limp Gawd
- Joined
- Mar 9, 2016
- Messages
- 316
Zen 3 will be out this fall hopefully and the X570 and B550 are the ones so far for it. Zen 4 will bring a new chipset.
I was a bit worried about the fan at first, but on my board (a rather cheap ASRock x570 Phantom Gaming 4) the fan only ramps up when first powering up the system before the splash screen displays. Then it slows down and is practically inaudible and I've never heard it speed up while using the PC. Of course fan failure is always a possibility, particularly with these small fans, but we'll cross that bridge in the future if necessary.