Review MSI MPG X570 Gaming Plus Review: Affordable Basics

[Crashman]

I especially like the part about how people shouldn't put 3900x's in an x570 motherboard.

I'm glad you liked that. Now read the rest of my comments above. I don't think there's enough power for a mild overclock when using the big processor.

 

[mossberg]

Where in the article does it say that?

I watched the video, and saw the response when Steve wanted a private conversation, with the editor of the article. It was a petty, if not arrogant, response.

 

[Crashman]

I watched the video, and saw the response when Steve wanted a private conversation, with the editor of the article. It was a petty, if not arrogant, response.

This is a review response thread for content that's actually in the review.

As for other people's results, maybe they were testing with old firmware. If you look at the first run of X570 reviews you'll find that there was a pretty big drop in power a few weeks after launch. I don't know. I can't do anything more than guess about data I didn't generate myself.

 

[freezed1]

Why not just retest with a 3900x / 3950x with avx load and post your findings? I am using a 200$ x570 board (gigabyte x570 ultra WiFi) and I went from 3600 to 3950x, so there are people like me who upgrade CPUs on 200$ motherboards.
Thanks

 

[Ncogneto]

You probably did not know this, but you CAN read a different opinion and just move on. I respect yours, but the point is totally missing.

So, just move on, say nothing, and watch people buy a garbage motherboard?

 

[Ncogneto]

Dude... MSI Unify is the fixed version that came out after the backlash. That mobo has no VRM problems.

I know that, it is why I bought it, where did I say it was a bad Motherboard. Interestingly enough, the Unify was MSI's answer to the crap boards they made before it. The Unify is a great board.

 

[Zigmars Zilgalvis]

So, just move on, say nothing, and watch people buy a garbage motherboard?

We are on the same page - guy called me a troll for saying the board is bad, so that was my replay to that. Nobody should buy or recommend this board.

 

[sarchasm]

"Most buyers in the sub-$200 motherboard market will never upgrade to the Ryzen 9 3950X, and few will even jump up to the 3900X, and hardly any buyers would expect boards in this price class to overclock those processors. When you’re mostly after the basics in an X570 board, so you can spend more on the processor or other components, the MPG X570 Gaming Plus is easy to recommend."

I can't come up with any logic to justify this statement. If you are just planning on running a Ryzen 3700X forever you should be recommending a B450. The only reason to have a X570 would be if you want a platform for future upgrades or are dropping one of the higher tier level chips in now.

I have a Ryzen 3900X running in a "B450 AORUS PRO WIFI". If I am going to drop $200 on a motherboard it had better be able to handle the high end chips of the current generation. Quite honestly a lot of the reason to go AMD has always been the easy upgrade path cause you know Intel is going to force you into building a whole new system.

 

[Orumus]

Tom's Hardware used to be a great place to go for feedback. We tested this CPU up to 230W via overclocked stress tests. The 3950X TDP is 105W. Reading instead of skimming could be helpful..tsk tsk

Is this a joke? So you OC'd a 3700x to above the stock TDP of a 3950x ...and? Did it not occur to you that a 3950x would be even higher wattage then a mid-range 65w CPU when OC'd? In one part of this article you say that people would buy a sub $200 board to save money for a better CPU then later you say few will get that better CPU on a sub $200 board. $160 isn't cheap man. If someone only wants "the basics" they go with a B450 board or less. The fact you think that your testing disproves anything is hilarious and sad. You use different testing variables and seem to think it disproves something.
Your assumptions are wrong and out of touch. People OC for performance like you said.... especially when they don't have the money to upgrade to a better part. Think about this. In a year or so when these CPUs are much cheaper but "better" CPUs are out, the people who bought these sub $200 boards with a 3700x may upgrade to that now much cheaper higher end CPU and they will probably want to OC it because they can't afford the faster R9 49XX. That is just one very possible scenario where someone buying a so called "cheap" sub $200 board might get a 39XX CPU and want to OC it.

 

[Crashman]

We are on the same page - guy called me a troll for saying the board is bad, so that was my replay to that. Nobody should buy or recommend this board.

Are you the one who said "At the very least, Tom's is saying that MSI must have gotten its own spec wrong"?
But did anyone say that? We think everyone who buys a board should expect at least some flexibility in adjusting CPU settings. Given the capacity of the voltage regulator, we don't perceive that flexibility existing for the 3950X when used in this board.

 

[tamalero]

Are you the one who said "At the very least, Tom's is saying that MSI must have gotten its own spec wrong"?
But did anyone say that? We think everyone who buys a board should expect at least some flexibility in adjusting CPU settings. Given the capacity of the voltage regulator, we don't perceive that flexibility existing for the 3950X when used in this board.

Then why would you give this board the Editor's Choice then?
Why a board that barely can handle a 3950X is given a Editor's choice on top of the other issues in the article?
Why are you instead of correcting, double down on the flaws of the articles?
Why not dispel any nansayers by using a 3900X or a 3950X? I mean, put your money where your mouth is.

Like others said, this article stinks as bad as the infamous "JUST BUY IT!" from a year or so ago.
Its like a very hard attempt to justify something by twisting normal methods of testing.
Cramming a box in a circle hole.

 

[Crashman]

"Most buyers in the sub-$200 motherboard market will never upgrade to the Ryzen 9 3950X, and few will even jump up to the 3900X, and hardly any buyers would expect boards in this price class to overclock those processors. When you’re mostly after the basics in an X570 board, so you can spend more on the processor or other components, the MPG X570 Gaming Plus is easy to recommend."

I can't come up with any logic to justify this statement. If you are just planning on running a Ryzen 3700X forever you should be recommending a B450. The only reason to have a X570 would be if you want a platform for future upgrades or are dropping one of the higher tier level chips in now.

Right now the impetus is PCIe 4.0. It can make a big difference in storage performance. By the time we're done arguing over that, the B550 will be available.

And as much as we like to think that we're going to upgrade the CPU before we find a must-have newer board, I've been following this market long enough to say that most of us won't. If recent history is any indication, we'll be handing this systems down to someone else instead.

 

[svan71]

horrible review that ignores the fact the board vrm is sub par. Review a flagship chipset with a flagship cpu that is on the manufactures website as supported. You reviewed it open bench with a 65w cpu why not a 95 or a 105w which is the one causing down clocking and over heating.

 

[Crashman]

Then why would you give this board the Editor's Choice then?

At its price, it's basically a PCIe 4.0-supporting alternative to cheap B450 models.

 

[TechRevenge]

I bought this board. Your response smacks of being a very poor troll since I already said that. That's a rookie troll. You've got to pump up that troll.

The reason our recommendations are limited is that AMD processors don't often run at their rated frequency and we don't want to be stuck with "but I'm not overclocking, I'm just trying to get my processor up to its rated frequency" excuses on a board that obviously doesn't have enough power to overclock those CPUs.

You are an Editor. I was a Super Mod for AT, a Global Mod for MSI Forum, and an editor years ago on a site that no longer in existence. I was a teenager when I started and your level of professionalism is beyond head scratching here. You should never call people trolls, let alone the editor of Techspot which he is a veteran of the golden days of PC enthusiast hardware. I don't even know who you are to be honest, and really do not care because of your lack of respect for others.

View: https://www.youtube.com/watch?v=UZWUOldsxXQ
Fast forward to 16 minutes and you are using the same context to Steve as you are to members of this forum.

I really hope someone reaches out to the parent company of this site (used to be purch) and sets it straight because something is not right here.

You could have fixed your flawed review, instead, you resort to attacking others. Classy...

 

[BigMack70]

--- Tests using 65W TDP CPU
--- Claims this disproves multiple other reviewers who got bad results with a 105W TDP CPU
--- Pretends nobody uses a higher end CPU with these boards, because reasons
--- Calls anyone questioning this a liar and a troll

Nice.jpg

What a piece of incompetent garbage.

 

[svan71]

At its price, it's basically a PCIe 4.0-supporting alternative to cheap B450 models.

yes and "at this price" there are boards with VRM that don't overheat.

 

[Crashman]

yes and "at this price" there are boards with VRM that don't overheat.

I went through all the bother to order a board in part to confirm overheating, used an overclock to bump the CPU power consumption up by well over 50W, and was disappointed by a maximum load temperature of 90 °C after an hour of Prime95 small-FFTs.

Small-FFts is the same CPU load that I've been using to deny other people's claims of a 100% stable overclock at substantially higher frequencies, so you can imagine how I felt when, even with the overclock, I couldn't get the voltage regulator overly hot. That is the perspective from which this was written.

You may have noticed a reference to a July article that showed higher CPU temperatures: Those correspond to the higher power readings we found with early firmware. Since everyone has updated their firmware, those old numbers are only relevant within that firmware's timeframe.

 

[mossberg]

Right now the impetus is PCIe 4.0. It can make a big difference in storage performance. By the time we're done arguing over that, the B550 will be available.

And as much as we like to think that we're going to upgrade the CPU before we find a must-have newer board, I've been following this market long enough to say that most of us won't. If recent history is any indication, we'll be handing this systems down to someone else instead.

The whole AM4 platform is based on the fact it is upgradeable, to a better CPU, in the future. 1st gen Ryzen users, doing a drop in upgrade, to a 3rd gen, is more common than you may think. I fully intend to drop in a 3rd gen cpu, into my board, when 4th gen is released, and the fire sale of 3rd gen starts. Previous to Ryzen, yea hand me down systems probably were more common. Intel requiring a new socket and/or chipset, needlessly, for a new release will cause that.

PCI-E 4.0 is a feature 99.9% of people have no need of, nor would they notice the difference, in everyday use. Most cannot tell a difference between SATA, an entry level NVME, like the 660p, or even the very good 970evo. They would be better served by a B450 tomahawk max, and using the savings towards a faster GPU, or a higher capacity SSD.

People that can use, or need, such storage speed, are most likely professionals looking at a far more expensive setup, than a low end x570 board, that cannot properly handle anything beyond a 3700x.

 

[Crashman]

The whole AM4 platform is based on the fact it is upgradeable, to a better CPU, in the future. 1st gen Ryzen users, doing a drop in upgrade, to a 3rd gen, is more common than you may think. I fully intend to drop in a 3rd gen cpu, into my board, when 4th gen is released, and the fire sale of 3rd gen starts. Previous to Ryzen, yea hand me down systems probably were more common. Intel requiring a new socket and/or chipset, needlessly, for a new release will cause that.

PCI-E 4.0 is a feature 99.9% of people have no need of, nor would they notice the difference, in everyday use. Most cannot tell a difference between SATA, an entry level NVME, like the 660p, or even the very good 970evo. They would be better served by a B450 tomahawk max, and using the savings towards a faster GPU, or a higher capacity SSD.

People that can use, or need, such storage speed, are most likely professionals looking at a far more expensive setup, than a low end x570 board, that cannot properly handle anything beyond a 3700x.

I might be a little overly enthused about the ever-increasing performance data I'm seeing from PCIe 4.0 storage.
So I take it you want me to procure a 3950X and see what it takes to burn the board? No big loss if the board dies, but I need to know that it's worth the effort to get the CPU. Maybe a 3900X instead?

 

[mossberg]

A 3900x, at the very least, would make a lot more sense, for board testing, than a 65w 3700x. I doubt any would argue the board can handle a 3700x, decently enough, but there are arguably better options, at this general price point. This board really doesn't deserve a 4 star rating, or editors choice.

 

[InvalidError]

At its price, it's basically a PCIe 4.0-supporting alternative to cheap B450 models.

Except that there is a handful of better alternatives within a few dollars of the same price. Makes it hard for people to swallow an "easy recommendation" for a board that even the manufacturer has admitted to being sub-par.

While the B550 would make a whole lot more sense for people looking for more budget-conscious boards and don't intend to upgrade from something like a 3600X, the chipset isn't out in the wild yet.

 

[Taterrr]

Original Response

I bought this board. Your response smacks of being a very poor troll since I already said that. That's a rookie troll. You've got to pump up that troll.

The reason our recommendations are limited is that AMD processors don't often run at their rated frequency and we don't want to be stuck with "but I'm not overclocking, I'm just trying to get my processor up to its rated frequency" excuses on a board that obviously doesn't have enough power to overclock those CPUs.

Edited Response

The reason our recommendations are limited is that AMD processors don't often run at their rated frequency and we don't want to be stuck with "but I'm not overclocking, I'm just trying to get my processor up to its rated frequency" excuses on a board that obviously doesn't have enough power to overclock those CPUs.

This explains so much. I couldn't believe that you were being so stubborn in your complete lack of accepting ALL the tests performed outside of yours, but this initial comment opened my eyes. You were already convinced no one was correct other than you. For the love of God, even the MANUFACTURER of the board acknowledged the issue and replicated it. YOU are literally the only person who has managed to not replicate the overheating results. So...MSI are liars and trolls as well, correct? At least according to your logic...

I can't say for certain what underlying issue caused one of the worst tech reviews I've ever seen, but I can say, without a doubt, that any company wanting to maintain their reputation (or recover it at this point) needs to have anyone acting like this as far away from the public and/or customers as possible. The highly flawed test methods as I read the article were a big red flag, but the terrible attitude and complete lack of professionalism is something that makes me, and I am certain thousands of others, lose all trust in such a website.

We come to websites like TH for trustworthy information before we spend our hard-earned money. This article and your subsequent responses have completely removed any trust I had in TH.

I hope you learn from this, but I sure won't be around to find out.

Edit: I need to point out that I did just register my account today, or at least this linked account. I have been visiting TH for tech advice/info without registering or under an old account that I can't remember for, I would estimate, 10 years. The GPU Hierarchy page is bookmarked in my Google account. Currently trying to decide if it's worth it to research that article for flaws or just go ahead and remove it.

 

[Ry_Trapp0]

Just posting up to say that I will no longer be visiting this site for any reason. The forums are awesome, but there are other options - however, it's pretty clear that the testing methodologies are... 'amateur' at best, which brings into question any review results. I mean, I don't know what the hell needs to be said when THE MANUFACTURER THEMSELVES REPLICATED THE POOR RESULTS; and, yet, ONLY TomsHardware is "correct", while literally everyone else, INCLUDING THE MANUFACTURER, is wrong?

...yea...

...OK...

I've always checked both TomsHardware and AnandTech for reviews and for referencing hardware - but I guess I'm going to have more time on my hands by reducing this to just AnandTech now.
Would be nice to see a followup to this total cluster... not getting my hopes up though...

Also, you're completely, utterly, positively out of touch with your readers and the market if you REALLY think that people are buying expensive X570 boards just so they can run PCIe 4.0 NVMEs that literally cost as much as their budget CPU that they'll never upgrade in their X570 board - I mean, WHAT?!
Linus & co just did a "blind taste test" where they had their editors try different systems and guess which was a SATA SSD, an NVME 3.0, and an NVME 4.0, in gaming and video editing - they all guessed that the system with the SATA SSD was the one with the NVME PCIe 4.0, which was obviously completely and utterly wrong. Yet, I'm supposed to believe that there's just SO many budget buyers that will ONLY buy a 65 watt or lower CPU - for their X570 board that they're going to put THE most expensive SSD format into, for NO real world performance gain??? You've gotta be kidding me right now, this is a complete joke...

 

[LordConrad]

Considering many other hardware sites have found this board to be plagued by overheating VRMS, I find your testing of this motherboard to be highly suspect. I don't find this article to be at all trustworthy.