Which driver is newer, MB maker or AMD?

[Waldorf]

both seem to use same numbering, but the number on the AMD is lower than the one from MSI, while it was released later.

MSI is 6.02.22.053 from early march
AMD is 6.02.07.2300 from mid march

im getting ready (finally) for clean install, so i want to use the latest (version wise) from the beginning..

 

[Bill_Bright]

I always go with the motherboard. Same with GPU drivers; I go with the card maker and not the GPU.

I have to assume MSI knows what their board supports and they know the tweaks, if any, they made. AMD can only assume and therefore be all inclusive with CPUs and RAM, even if your board does not support all the CPUs and RAM the chipset might. The newer AMD driver may also include support for a recently added CPU that, unless you just bought a newly released CPU, I suspect you don't have.

Also, not sure it means anything but MSI says the driver is 63.99 MB in size. AMD says 63MB. I checked other AMD chipset drivers and this one, for example, says 65.4MB, suggesting they do use decimal points when called for.

So how do we explain the difference between 63 and 63.99MB? I don't know, but that would cause me to, again, go with MSI's.

 

[Waldorf]

yeah, just weird, as up until now, the msi one was released after amd put it out.

 

[Noci]

Quite a question and it seems there is no subtantiated answer to find with a quick Google search.
The discussion is not new, that is one of the first things I noticed when searching.

For example: https://www.google.com/search?q=msi...MQrQIoBHoECC8QBQ&biw=3440&bih=1305&dpr=1#ip=1

But like I said, no answer that put all the pieces into place popped up. What Bill_Bright says makes sense, but what I read on Reddit the tendency is more to go for AMD.
This can be the start of a long thread , sorry I can't help better.

Maybe contact MSI customer support for an explanation?

 

[Bill_Bright]

I am clearly biased and am about to pat myself on my own back to illustrate that. But frankly, I'd trust the advice given here at TPU long before I trust what I read on Reddit.

 

[c2DDragon]

This is from MSI. As you can see it has been modified on february 22nd so I would get the one from AMD.

Also :

The one from AMD is a bit bigger.

 

[Ruru]

I always go with the motherboard. Same with GPU drivers; I go with the card maker and not the GPU.

I go the total opposite direction, I download always the drivers from chip manufacturers' site.

 

[Bill_Bright]

This is from MSI. As you can see it has been modified on february 22nd so I would get the one from AMD.

Why? Why does the fact it was modified (by the board maker - which to me, is significant) February 22 suggest to you to download from AMD? I don't see the logic.

The one from AMD is a bit bigger.

So? That could easily suggest it has unnecessary bloat in it MSI does not use in that board. When it comes to drivers, bigger does not suggest better.

I go the total opposite direction, I download always the drivers from chip manufacturers' site.

Also why? AMD drivers will be generic, for all brand motherboards with that chipset. MSI is likely to be specific for MSI.

 

[c2DDragon]

Why? Why does the fact it was modified (by the board maker - which to me, is significant) February 22 suggest to you to download from AMD? I don't see the logic.

Because the one from AMD is newer ? OP wanted to know which of them is newer. It's in the title. I guess the one from AMD is newer.

So? That could easily suggest it has unnecessary bloat in it MSI does not use in that board. When it comes to drivers, bigger does not suggest better.

Why would you think so ?

 

[Waldorf]

To go a bit into detail:
while i prefer to use MB specific drivers, up until my (personal) switch to MSI, all other brands i used (ryzen),
were either not providing newer ones past initial ones, or at least were way behind, hence my use of the ones released by AMD.

and while bigger doesnt mean better, there is always a chance AMD did change/fix "something" (so to speak),
which is not (yet) included in the one from the board maker.

i can still wait for another 10-14 days and see what happens, but will probably go with the MSI one (for now).

@Bill_Bright
nothing the chipset driver "touches" when it comes to sw, is going to be tweaked by MSI,
as were not talking about hw being different/modified, vs what usually comes with any ryzen boards (e.g. different usb chip),
(and amd is still the maker of the cpu/chipset, not msi).

would be different if talking about stuff like fw/bios, which will definitely specific to brand/board and its capabilities.

just wish MSI would have release notes, and i could compare the separate drivers by themselves,
and see if there was any change compared to AMD.

edit:
should have tried that before posting, lol
unpacking the amd package, the chipset driver itself is from early feb, so the msi would still be newer, if just comparing that (single) driver.

 

[Bill_Bright]

AMD (and Intel) are essentially OEM makers. Neither make motherboards. Intel used to but stopped almost 10 years ago. Instead, they provide CPUs and chipsets to motherboard makers who then integrate them into their boards - just like the GPU makers and chip makers for RAM sticks.

While MSI, ASUS, Gigabyte, etc. may all use the same chipset, are you going to suggest those makers make identical boards that perform identically? Of course not.

So why do you not think they have haven't modified the chipset drivers to perform more efficiently and optimally for their specific boards? If I were MSI or ASUS, I sure would.

Why would you think so ?

I already explained why - twice! Please read through the thread.

were either not providing newer ones past initial ones, or at least were way behind, hence my use of the ones released by AMD.

I understand this but IMO, this is (most of the time) flawed logic. Why? Because drivers don't suddenly go bad and thus need updating just because a new update is out there.

I never update a driver unless the update (1) addresses a problem I am having (rare, except for very new hardware). (2) Addresses a newly discovered security issue (this is very rare). (3) Adds support to new hardware I am going to buy - this is the most common reason. That is, it adds support for newly released CPUs or RAM or something else only the most current motherboards "may" support.

In other words, I do my homework. You need to too. If the driver history/readme file does not say it fixes a problem I am having, there is no need to update.

 

[Waldorf]

example: when i have drivers released BEFORE the cpu (i5xxx) was released, i doubt its better optimized,
hence i used the amd ones.
and that's not even looking at things like improvements or fixes (e.g. security/scheduling/gaming etc),
which the latest amd included, but the one from GB (back then) didnt.
no one can tell me there wasnt anything done within 2y, as thats the time between the drivers that Gb released for the board i had (aorus ultra)...

its fine if you want to stick with drivers that work for you, i dont ,as i prefer to update chipset/gpu/audio drivers if possible,
and that has nothing to do with "homework" i have to do (or not).
its my personal preference, no matter who likes it (or not).

ignoring your arent even on ryzen.

 

[Ruru]

Why? Why does the fact it was modified (by the board maker - which to me, is significant) February 22 suggest to you to download from AMD? I don't see the logic.


So? That could easily suggest it has unnecessary bloat in it MSI does not use in that board. When it comes to drivers, bigger does not suggest better.


Also why? AMD drivers will be generic, for all brand motherboards with that chipset. MSI is likely to be specific for MSI.

AMD updates their downloads hella more often than an AIB

 

[Bill_Bright]

AMD updates their downloads hella more often than an AIB

Again, so? Does that mean it affects each and every user out there? NO!

Do you seriously think MSI, ASUS, Gigabyte etc. are stupid, lazy morons who could cares less that their users are left vulnerable to security risks? Left with poor performing motherboards or graphics cards?

If that is you, then I can't help you.

I say again, AMD and Intel are OEM makers. It is VERY common for motherboard and graphics card makers (and PSU makers and others) to tweak OEM products to enhance this feature or that feature - if for no other reason than to make their product stand out from their competition. If you don't believe that, then again, I cannot help you.

So hey! If you want to stick with the "generic" "one size fits all", OEM driver thinking, naively, that you have the latest therefore it must be better , then go for it. I'm not going to argue with you.

I have said my position and explained, with logic, why. The rest of you have said your position but failed to give any sound reason why - except, it is newer or bigger therefore it must be better!

Yeah right.

 

[c2DDragon]

I already explained why - twice! Please read through the thread.

I was asking why do you assume MSI would have removed some bloat ???
If you did your "so called" homework, you would have seen that ASRock is providing the same (here per example). You can find it on ROG forums too.

I renamed them for it to be clear.
But it's fine. Don't change anything.

---------

@Waldorf
I don't get why AMD prefers to provide the 6.02.07.2300 by the way.
If the 6.02.22.053 works fine, it's fine in my book. Also there is no trickery, the numbers just show it's newer despite the dates.

 

[Shihab]

On the "which is newer" topic:
I *think* the difference in dates listed on each website is due to testing. Those drivers aren't fresh off the compilers, of course. They have to be tested and vetted before publishing. The actual timestamp on the executable for the 6.02.22.053 (downloaded from a random MSI mobo support page) is Feb 22nd.
Compared to the same offering from Asrock (also a random mobo). Both have the same hash (i.e. exact same driver) and timestamp, and the Asrock offering was uploaded on March 26th.
Asrock was even nice enough to include a changelog, from which you can spot a few entries that are a couple sub-sub-versions above the ones in the package hosted on AMD's website.

tl;dr: MSI's version is more recent.

edit: Didn't load newer posts before posting this. Didn't see c2DDragon's comment. :|

 

[Bill_Bright]

I was asking why do you assume MSI would have removed some bloat ???

I am not assuming anything. What I know is the one on AMD (or Intel) are OEM drivers, essentially generic drivers for every motherboard ever make with that chipset.

If you like generic go for it.

But understand you have given not rationalization or logic why those are better than those provided the OP's motherboard maker.

I've given my opinion with my rationalization for it. I see no reason to argue further until you can give some sound reasoning why using generic is better.

If you don't trust your board maker, then I recommend buying a different brand next time.

Have a good day.

 

[Ruru]

Again, so? Does that mean it affects each and every user out there? NO!

Do you seriously think MSI, ASUS, Gigabyte etc. are stupid, lazy morons who could cares less that their users are left vulnerable to security risks? Left with poor performing motherboards or graphics cards?

If that is you, then I can't help you.

I say again, AMD and Intel are OEM makers. It is VERY common for motherboard and graphics card makers (and PSU makers and others) to tweak OEM products to enhance this feature or that feature - if for no other reason than to make their product stand out from their competition. If you don't believe that, then again, I cannot help you.

So hey! If you want to stick with the "generic" "one size fits all", OEM driver thinking, naively, that you have the latest therefore it must be better , then go for it. I'm not going to argue with you.

I have said my position and explained, with logic, why. The rest of you have said your position but failed to give any sound reason why - except, it is newer or bigger therefore it must be better!

Yeah right.

I'm not gonna even continue arguing as it seems to be totally pointless. Go ahead and use old drivers, not my problem.

 

[c2DDragon]

I am not assuming anything. What I know is the one on AMD (or Intel) are OEM drivers, essentially generic drivers for every motherboard ever make with that chipset.

If you like generic go for it.

But understand you have given not rationalization or logic why those are better than those provided the OP's motherboard maker.

I've given my opinion with my rationalization for it. I see no reason to argue further until you can give some sound reasoning why using generic is better.

If you don't trust your board maker, then I recommend buying a different brand next time.

Have a good day.

I can't beat belief.
You may have missed the part where I pointed that other motherboard makers provide this exact same driver which is 100% from AMD, so...generic but it's fine.
Have a good day too, thank you.

 

[Noci]

After the heated discussion above and giving it some thought;

Assuming both would work fine (and they should), I would go for the MSI drivers. They know as mentioned all the ins & outs of that specific MB and for all I know it could have to do with the specific RAM support on that MB due to an updated QVL. If AMD finds some serious exploit, it will update and the MB manufacteres will follow instantely.

Shoot me if I'm wrong or talking nonsense, just venting some ideas that make sense to me

like I said, can be the beginning of a loooong thread

 

[Shihab]

generic

There is an assumption here that board manufacturers can/do customize the chipset itself, which is, imo, a claim that needs proof.

What the oem can customize is already facilitated by the connectivity means the chipset exposes. That is, OEM-specific features are extraneous to the chipset, and whatever chip/controller they add gets its own separate driver and firmware. I do not see a reason why the oem would need to modify the hardware, much less AMD risking them disrupting its product stack. After all, AMD is not ARM, they don't just license an IP other OEMs can implement however they want, they sell the chips itself.

 

[Waldorf]

@Bill_Bright
lets go from a different view then:
statistically speaking, anything below +2000 times comparing, is not relevant, and only a guess.
so until that happened, or that you are involved in the actual "making" of drivers, you are making an educated guess,
and your "theory" of sticking with the what works and/or the board maker, has no less or more value,
than my theory of going with the latest (no matter who provides it).

and Nv is the best example for mine (being correct), as the discussion on new released drivers shows ppl getting improvements/fixes for things NOT listed
in the release note, meaning if ppl wouldnt have updated (to the generic Nv driver), they would have never actually gotten the "fix".

and thats on the "generic" drivers provided by Nvidia, not from the AIBs, that can put out cards way beyond what the FE cards from Nv can do,
no matter if its hw (amount of power phases/brand of vram/pcb layers) or things like higher clocks, which Nv doesnt know about.

e.g. the latest driver for my gpu coming from Gb is 3 years old.
guess what, the "generic" driver from Nv released this month has not only improved perf,
but provided many fixes for issues i had (one being micro stutter), and made (old) games work with settings that didnt work before,
unless i modified the games profile with Nv inspector (ambient occlusion), and most of them, NOT listed in the release notes,
as it was either fixed with a previous release a while back, or belonged to the "other improvements" part.

all completely going against "sticking" with (the least newest that) works/what the hw maker provides.

 

[chrcoluk]

Newer isnt always better, board vendors one would think they have done at least "some" testing with what is on their website so probably is a good idea at least as the starting point, and then only change the driver if you have a reason to like there is technical issues. Dont change it for the sake of it.

 

[natr0n]

for trouble free - board maker driver

for an adventure - newest driver

I like adventures though

 

[oxrufiioxo]

I've been updating direct from AMD since I've owned a Ryzen system almost 5 years now with zero issues. I doubt either option is bad though or will cause any issues.