NVMe M.2 or Sata M.2

[Bassman]

I wish to relate my query specifically as related to MS2020 and specifically to a 2TB size storage. All the web information seem to concentrate on OS boot times and how long a program takes to start and there does not seem to be a big difference between a M.2 PCIe NVMe connection and a M.2 SATA connection once everything is started (subjective of course; 15 seconds longer would not bother me). So once all is up and started is there a major benefit in one connection over the other, again specifically as it would relate to running MS2020?

I further advise that I am looking at a Pioneer 2TB NVMe PCIe M.2 (AUD$370) a Crucial 2TB NVMe PCIe M.2 (AUD$408) or a WD Blue 2TB SATA M.2 (AUD$329). Cost is a factor because I then also have to persuade "the banker" to allow the additional purchase of MS2020. Pioneer have been around in various forms for many years, are they still a reliable brand because at AUD$370 it would appear to be a good compromise between connection and price.

[Jon Clarke]

The biggest difference between the SATA M2 and the NVMe M2 is the read write speed.

I have a 1TB NVMe M2 for my XP and MSFS. 

The read/write speed of the NVMe M2 is 3400MB/s

The SATA M2 will be just like a SSD which is 500MB/s

Loading times with NVMe M2 is way faster than SSD but once in sim there would not be any difference. My 1TB NVMe M2 cost me £120.

I already have 65GB of MSFS mods installed so loading time of the sim and the loading time of a flight start to become more important to me as the addons/mods increase.

[Nick Cooper]

Purely as a matter of interest, I bought an M2 to go with my latest hardware upgrade.

I did have a look at it and saw benchmark speeds of 500MB/s or so and thought nothing of it.

The motherboard manual is not the most detailed that I have ever read.

There is nothing on how to set the BIOS at all.

This morning, I reset it from SATA to NVMe and the same benchmark now reads

2434MB/s, so there clearly is big difference, though to be honest, MSFS does so many

other things when it starts that the loading time seems not to be drastically different.

Certainly, Jon's blistering broadband speed will be helping him along as well.

 

[Bassman]

As I am not too concerned about initial loading time, but I am concerned about cost, it seems to me that a M.2 Sata card is the way to go?

 

[Nick Cooper]

Here is a very informative article about M2 drives.

As far as I can see, a 2TB M2 can be not much more than a 2TB SSD.

Of course, the matter is complicated even more by different standards of M2

which of course means wildly varying prices for the same sized drive, much

like SSDs, in fact.

[Donka]

I have two NVME drives in my current system but my previous build had Sata m.2 drives. While the NVME drives benchmark a lot faster, the difference is game loading isn’t all that different. I think the biggest reason is the game engine is loading up lots of relatively small assets and these smaller file sizes don’t transfer as quickly as a single large file transferred when determining max speeds in a benchmark.

 

This article gives a similar insight...

https://techreport.com/blog/3467943/nvme-vs-ssd-test/

 

[Nick Cooper]

Hello,

that and not all of the time waited for a game to start is file transfer from a disk at all.

There are many things advocated by those who have bought them that as your article

points out, are only discernible to a benchmark.

Of course the market for them is full of those whose main aim is to out-benchmark

the next person.

Much like the relentless pursuit of frame rates in the flight simulator world, even though the frame rate

is not the measure of how well a simulator is working, nor how good it looks on the screen.

I chose my 1TB M2 because it cost near enough the same as a 1TB SSD and because I already

have eight drives and a DVD player in my case and it takes up no extra space at all, being mounted

under a pre-existing cover on the motherboard.

That was the main advantage to me.

[Bassman]

"Much like the relentless pursuit of frame rates in the flight simulator world, even though the frame rate

is not the measure of how well a simulator is working, nor how good it looks on the screen."

 

Absolutely love this from Nick Never a truer word typed. The best way I ever improved my fps was by turning off a fps counter.

[wolfko]

On 9/16/2020 at 10:42 AM, Bassman said:

 The best way I ever improved my fps was by turning off a fps counter.