Updates will always be problematic

Discussion Hub General Discussion
21

Ok, let’s take a bit of a dive into that premise.

AMD, Intel, Nvidia. You are likely correct. Almost every PC out there will be built using these primary components.

Two Intel or AMD CPU’s with the same number on the box are going to be darn close. If you are O/C’ing that CPU you will find there are differences in the tolerances from one to the next but overall, negligible difference.

This is about as far as I am willing to go with the “no 2 PC’s” premise.

Now lets look at GPU’s. Pick a number. Either a AMD or Nvidia chip. Now pop into Amazon and shop for that chip. How many different board manufacturers are there? What is the price range? Remember to filter just that one chip.

I haven’t shopped components since last August when I built my MSFS system. I first decided on my CPU. I went with an i5 8400. Then I looked for MB’s that supported that CPU. Hundreds of choices from nearly as many manufacturers. Boards from $150 to $750. The number of possible options were soul destroying. Built in WiFi or no, 1 NVME or 4, 2 RAM slots or 4, Optical audio or no… on and on. I settled on a board maker I trust with the options I wanted. I ignored the price point.

Now we need to pick a graphics card. Same issue here. Hundreds of choices for the same GPU. Then we start shopping memory. Wow! the price ranges for 32GB of memory. Power supplies, same.

After selecting my components and placing the order, I decided, just for fun, to see how cheap I could have gone. Selecting the same base. i5 8400 / 2060s /32GB. I could have purchased all my components for just over $1000. The order I had just placed, $3150.

No way that the $1000 no name brand hardware would have been as reliable and trouble free. I cruise along nicely in MSFS with 0 CTD. I do hi end 4K video processing/editing. No issues.

Between case, cooling, PSU, CPU, GPU, Mem, MB, even quality of internal cabling and routing, the number of possible combinations, price ranges, hardware quality and reliability is off the chart.

Now that you have a system built, let’s set it up. Are the MB drivers proprietary? Are you using the default intel drivers that Windows wants to install? Does your Wovelot card use their own driver or have you installed the Generic NVidia driver?

So you have your PC up and running and you install MSFS. Now you start plugging in head trackers, control boxes, joysticks, throttle quadrants, pedals, headsets, VR. Each with there own proprietary drivers and configuration software. You start running into issues because you can’t get your headset software past your virus protection.

Yup. There are only really 3-4 primary players in the computer world. The end result though, once that "FLY NOW" button lights up, is that no 2 PC’s are ever the same.

@blueline308 @Grumpy3355 @IroncladBoat941

2 Likes
22

And how many of them are making their own silicon?

None of the OEM’s are.
They are using Nvidia or AMD silicon which they slap onto a PCB which doesn’t really deviate much from the reference design.
None of them change the way that the silicon works and yes, all of them will work with the GPU manufacturers own drivers. Indeed downloading and installing the latest driver is usually recommended by the OEM.

The same thing is happening with motherboards.

And nothing that any of them do with motherboards or graphics cards in any way changes the way a CPU or a GPU work and how they process instructions or pass data to and from memory.

Because graphics card and motherboard manufacturers don’t define that.
AMD, Intel and Nvidia do.

Even Dell, that has a ■■■■■■ habit of making motherboards with proprietary power connectors for their proprietary power supplies don’t change the way the CPU works. They just mess up your potential upgrade path.

Just as laptop manufacturers working within a small custom form factor may create a specific motherboard for a specific model of laptop - that motherboard doesn’t change the way the CPU processes instructions.

23

So, you maintain that there are no quality or performance differences between a hi-end hi performance motherboard and a cheap generic motherboard? As long as it has the same socket, I won’t notice any difference? That $139 LGA1200 board will perform the same as the $699 ROG Maximus XII Hero board?

Every user has the same level of knowledge and as such knows the best way to match hardware and configure BIOS for best performance and stability? Every off the shelf $2500 system is configured the same and all the preloaded software is carefully selected to ensure no conflicts or resource hogging?

Circuit board design and manufacture can make a huge difference in the quality and speed of the signals being passed. One bad solder joint on a cheap board can cause huge signal degradation. Just a small temperature change on the board can result in corrupt data.

In the most simplistic overview you are not wrong. On paper, reference standards are supposed to insure compatibility. When viewed from the pure electronics realities even two LEDs from the same box will perform differently. When the variabilities of manufacturing standards and price point targets during design are taken into account, do you really believe all components are created equal?

1 Like
24

Thank you. I am trying this now

25

"Yup. There are only really 3-4 primary players in the computer world. The end result though, once that “FLY NOW" button lights up, is that no 2 PC’s are ever the same.”

That is exactly what I am trying to say. As you said in your post, the choises when you build your own PC and the components that are available is eye watering. The only thing with people’s choices is… “How deep is your pocket and what are you prepared to spend”. :face_with_raised_eyebrow:

2 Likes
26

I’m an MCP, now retired. Much of my knowledge is now out of date (if you want to know anything about DOS, Windows 3/9x, Windows XP, Office XP/2000, IDE drives, configuring system and autoexec files etc, I’m your man), but much of it’s still relevant.

And I can say quite categorically that a PC is much more than the sum of its parts. I’ve lost count of how many times I’ve come across an organisation with many computers, all identical, that all had their own idiosyncrasies.

I remember one Windows XP machine, when it “forgot” it had a USB device connected (common with XP), would freeze and the only way to unfreeze it was to open Task Manager and stop Rundll.exe. All the (identical) others were fine if you simply removed the USB device and replaced it.

They’re like cats; all with their own personality.

2 Likes
27

Beta testers are doing exactly what they’re supposed to be. They find, track and log bugs. What Asobo do with them after is what you should really be asking.

There have been plenty of NDA-bending comments by beta testers both here and on other platforms that outright say that bugs that made it into production (sometimes serious, game breaking ones) had been reported and logged early on in the beta process, - some within hours of the beta going live - but never acted upon until the public backlash after the broken update was launched.

While it’s true that all PCs are unique, which introduces a huge wildcard into the mix, most of the problems we see with updates (outside of stability) are not related to that. It’s issues that are universal to most or all users on both PC and now Xbox as well - like broken mechanics, broken planes, broken subsystems, etc.

It doesn’t take a whole lot of forum browsing to find that problems that people seem to want to attribute to hardware differences, overclocking, community mods, etc with absolute certainty are also affecting Xbox users who all have the same hardware and have no access to community mods.

1 Like
28

no if u follow the right produres!

29

My main gripe is that they’ll break something that’s been working fine, and then not fix it for a couple of months.

Over the past year we’ve had to put up with flaps, trim, mixture / condition levers and bush trips break, controls randomly inverted, performance jumping all over the place, not to mention the constant CTD’s some unlucky users experience.

I can only imagine how frustrating it must be for beta testers to spot these issues and not have their voices heard.

30

Same for me

31

At least your username says it all :rofl:

32

Agreed. I feel for them as well.

I wouldn’t bother signing up for a beta knowing what I know now, as it seems like bug reports from testers are treated the same as bugs reported to Zendesk. They get logged and no action is taken until those reported bugs hit the live build and the ■■■■ hits the fan with the community. Then you get a bunch of people ragging on the beta testers for “not doing their job”.

My free time is extremely valuable to me. When I choose to give it up for some reason, I expect it to be respected. And I honestly don’t feel it would be the case beta testing MSFS. At least not based on what testers have reported on the beta cycles so far.

I feel for the testers who volunteer their time and “stable” builds to test and have their input ignored and then have to deal with community members questioning what the testers did.

1 Like
33

At the risk of starting something here, I have to clarify some of the beta test rumours. I participated in a couple flights. The reports that bugs reported during testing are logged and ignored, are somewhat exaggerated. Depending on what the bug is and when in the process it is reported there may not be time to implement a fix.

Most betas have a specific focus. In many cases the things that get reported are existing bugs that were not slated to be addressed. There seems to be a belief that during beta everything that can be fixed should be. If there has not been a branch set up then by the time a flight is instituted the bug fix list is pretty much carved in stone. The job of the testers is to confirm that what was planned to be fixed, is in fact, fixed.

I wish, just like every other user, that there was unlimited resources and that when we test a new update, every bug we find can be addressed before release. In generic terms, usually a beta is at least a month in the making, before it goes to public test. There may be a week before release. In some cases it is possible to pass out a repaired beta during testing but often the beta is closed a couple days prior to release to allow time for some last minute fixes and then compiling the branches. The number of man hours needed in a project of this scope forces a defined focus and, yes, some stuff just isn’t on the docket.

1 Like
34

I have had a great experience with SU 6. Good job Asobo!

1 Like
35

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.