Crippling updates and disappearing features

[eslader 1,774]

19 hours ago, Paraffin said:

Maybe it was done to allow VR to achieve decent frame rates?

I'm not understanding all these theories that they're reducing graphics for higher performance. Some of us were getting nearly flawless performance right from the start. MS has always historically taken the stand that performance issues should largely be solved by upgrading hardware. Even when that stand made no sense as the hardware capable of running FSX smoothly didn't exist until years after FSX was released. 

Now, clearly, the hardware does exist to run MSFS smoothly as it was at launch. I don't even have a bleeding edge computer and it was doing just fine. So there has to be another reason for it if they really are reducing graphics quality. And I don't buy the Xbox compatibility theories either, because we're a very long way from the days of the Atari 2600. Modern console systems are absolute beasts and frequently outstrip even robust PC builds.

 

[Maxis 1,832]

46 minutes ago, wthomas33065 said:

Of course it is.  That's why it is ludicrous to believe that they are intentionally crippling PC branch to "optimize for XBox".   Xbox system X hardware exceeds much of the hardware that we are currently using.   Current Xbox One hardware is 7 years old.  

If Asobo is doing this, then it is the first time in Xbox development that a PC title was intentionally crippled to shoehorn it to Xbox.

I believe the individuals who partake in this conjecture are probably not "gamers" and so they are simply using this as some sort of excuse.   I mean Xbox development has been blamed for everything from reduced visuals, to delay in bug fixes.  And that type of conjecture is simply not founded.  In many cases, ports are handled by completely different teams.  So to blame everything on Xbox simply is not founded unless there is some evidence to support the claim.  And the simple existence of reduced visuals or delayed bug fixes does not count as evidence that it is due to xbox integration.

 

wthomas and i dont agree on much but we do agree on this.. I have to see evidence of this. Cant speculate on it. Regardless i dont see much  difference between the hardware of MY pc and the Series X (Which i also have).

Xbox series X Zen 2 8 core  16 thread  CPU 3.6 or 3.8 GHz depending on configuration  / My Pc Zen 2 Ryzen 3700x 8 core 16 thread  both Zen 2 and both on the 7nm process

Series X Storage 1tb nvme  PCIE gen 4 / My pc 1tb nvme  PCIE gen 3 dedicated to the sim 

Current gen RDNA2 graphics with 16Gigs of DDR6 RAM (from what im reading this is probably equivalent of  a slimmed down 6800 or RTX2080 TI

My Pc just has an RTX 2080 super at 8 GB DDR6

So i really don't think Asobo is trying to shoehorn the sim to fit the hardware. The Xbox series X hardware is very much up to snuff if not even better in some respects. Also considering the Shell requires lower resources / processes are more optimized for gaming and resource requirements of a more general OS are dropped for a purposed built GUI for less overhead.

IF they did all this degradation to fit this in for Xbox series x then i would be genuinely surprised and disappointed because honestly the hardware is very capable.

Where i diverge from Wthomas is in Asobo/Microsoft 's QA and project management and how it affects the end user experience  for current PC users overall.

Edited December 10, 2020 by Maxis

[wthomas33065 994]

1 hour ago, Maxis said:

Where i diverge from Wthomas is in Asobo/Microsoft 's QA and project management and how it affects the end user experience  for current PC users overall.

We only disagree because we have only conjecture to base that opinion as well.  Neither of us have any insider information on what or how current QA and project management is affecting the product.   

We see the same patch issues, and new bugs being created, but I tend to think that Covid-19 and a biweekly update schedule more likely contributed to the low quality of testing that appeared to occur on the patches.  I also point to my own experience in testing.  More and more testing departments rely on scripting and not blackbox testing.  Test Scripts are great, but rarely test a product the way a real user does.  And unfortunately many test managers today emphasize testing against design documentation  (simply testing what development says works and keeping to that script), not testing on real user cases (test what users most likely will do when using the product.).

This doesn't excuse such mess ups as the livery change or the blank controller CTD which affected so many.   But I do believe the latest AP issues can be directly contributed to only testing the actual PID changes via scripting and not seeing how they affect real life performance.

One thing we CAN be happy about is that the developer is still engaged and engaged in a big way.   Too many time's I have seen games that are nothing but a pure money grab and receive nothing in the way of support after initial release.  The fact that there has been significant investment in the client server architecture required to even run this product should be encouraging.  

 

Edited December 10, 2020 by wthomas33065

[Maxis 1,832]

1 hour ago, wthomas33065 said:

We only disagree because we have only conjecture to base that opinion as well.  Neither of us have any insider information on what or how current QA and project management is affecting the product.   

We see the same patch issues, and new bugs being created, but I tend to think that Covid-19 and a biweekly update schedule more likely contributed to the low quality of testing that appeared to occur on the patches.  I also point to my own experience in testing.  More and more testing departments rely on scripting and not blackbox testing.  Test Scripts are great, but rarely test a product the way a real user does.  And unfortunately many test managers today emphasize testing against design documentation  (simply testing what development says works and keeping to that script), not testing on real user cases (test what users most likely will do when using the product.).

This doesn't excuse such mess ups as the livery change or the blank controller CTD which affected so many.   But I do believe the latest AP issues can be directly contributed to only testing the actual PID changes via scripting and not seeing how they affect real life performance.

One thing we CAN be happy about is that the developer is still engaged and engaged in a big way.   Too many time's I have seen games that are nothing but a pure money grab and receive nothing in the way of support after initial release.  The fact that there has been significant investment in the client server architecture required to even run this product should be encouraging.  

 

Fair assessment. Especially on the insider information or lack of it. As to your second paragraph i can buy in to what you think is happening. I loath to bring in my IT background and the environment i operate in so i wont. What i will say about my job and its environment its that it is a whole lot less tolerant of failures of this sort for the end user and if our testing methods are ineffective in flushing out bugs then a better method MUST be designed and implemented post haste.

Granted this is a sim/game so the expectations of immediate response are definitely not the same and i know that. However it is reasonable to expect the developer to move toward a more effective means of QA and testing before the sales and marketing dept starts to call for meetings. 

My enthusiasm for the developers has not been great due to this experience. At the moment i am at a state of indifferent acceptance and am no longer angry at Asobo. For me i consider myself to be in a holding pattern, waiting to see how they respond in the coming months in addressing issues that will change my impression of this experience and whether to invest in the ecosystem or move on going forward.

Hopefully the engagement of the developer is genuine and there are improved processes that reflect on the end user experience.

 

Cheers 

[wthomas33065 994]

43 minutes ago, Maxis said:

What i will say about my job and its environment its that it is a whole lot less tolerant of failures of this sort for the end user and if our testing methods are ineffective in flushing out bugs then a better method MUST be designed and implemented post haste.

I've seen both sides.  When I worked under Ed Fries in Microsoft Games (circa Age of Empires, Midtown Madness, Close Combat III) the testing department was great. and we all read and used Cem Kaner's Testing Computer software as if it was some sort of bible.  Wrote boundary cases, test cases for text box overflows, null values, etc.
Really good stuff.  We had our own software bug tracking system called "RAID".  yeah....funny.  But it was great. 

I was the first hired "full time" tester for CC3 the Russian Front, which really means nothing, except that I got to do a lot of the regression testing.  We had thousands of tests already written for CC2 A Bridge to Far, and I had to start running those against what I could in CC3.  As more and more of us got assigned the the game, I was primarily responsible for map data and movement.  I had to verify that the artwork matched each of the map data, which was nothing more than numerical values in a dimensioned variable.  each value matched a terrain type.   WALL, ridge line, mud, pavement.   Etc.   

We would take the variable and put it into an excel spreadsheet and then assign a color value to each number.  So for instance I could assign "black" to all the walls.  and then the output would show a very rough outline of all the walls.  Which I had to hand match with the artwork.   This wall of the Kremlin had 15 openings.....so count on the map data... 15 openings? and in the right place.   Really tedious stuff.  But there was nothing worse than putting your troops into a building and finding out there was no way for them to get out.... 

When I left Microsoft, I joined a company that was primarily into hardware, and testing and design of hardware is much different.   I had to actually ask the Vice President of Engineering for about $2500 to buy a basic "bug tracking" software suite because up to that point they were using email to track bugs in the firmware.  I said that won't fly for the Windows testing they had hired me to do.  

Eventually moved out of test (basement, windowless room....hahah) and into Technical Support, where I've been for the last 20 years.  But we still work very closely with Test and Development.   Unfortunately, my company has never invested into QA the way I think they should have.   Now everything is based on SCRUM, but TS and Testing never seem to be on the early SCRUM meetings or if we are, our voices are drowned out, so by the time we get a shot at it, It's already "too late".  Very frustrating.  So I've seen both sides.  

I would think that Xbox Game Studios would have those resources, but I'm not sure they're really involved in the patching.   Something tells me Asobo is more responsible for that and like others have mentioned, they might be a little overwhelmed by not only the "scope" of the product but also the level of accuracy demanded by the user base.  I just think they underestimated how rabid the "sim community" can be over things most other people would "overlook".  That's not an excuse.  But if you have underestimated those things from the start, it's hard to dig your way out while you are in the middle of the blizzard.   Hopefully monthly updates and possible beta branching will allow them to provide the time they need to provide better quality testing.   And hopefully the sim community will provide the patience necessary for that to occur as well.

Edited December 10, 2020 by wthomas33065

[Virtual-Chris 682]

So are most of us in agreement that visuals have changed in a negative way? 

And we’re now just debating the possible causes?

Or are some people still convinced everything looks as good as it did at launch?

For those that agree there has been some degradation... There’s a few possible explanations...

1. They’re changing things to accommodate other platforms

2. They are recoding aspects of the graphics engine for DX12 which is affecting visuals (this is really related to the first choice)

3. They are changing things and think they’re improving but we don’t like it

4. They are just being jerks

5. Something else?

So for those that think the Xbox (and thus choice 1 and 2) are not the cause, what is your theory?

Edited December 11, 2020 by Virtual-Chris

[wim123 758]

3 hours ago, Virtual-Chris said:

So are most of us in agreement that visuals have changed in a negative way? 

And we’re now just debating the possible causes?

Or are some people still convinced everything looks as good as it did at launch?

For those that agree there has been some degradation... There’s a few possible explanations...

1. They’re changing things to accommodate other platforms

2. They are recoding aspects of the graphics engine for DX12 which is affecting visuals (this is really related to the first choice)

3. They are changing things and think they’re improving but we don’t like it

4. They are just being jerks

5. Something else?

So for those that think the Xbox (and thus choice 1 and 2) are not the cause, what is your theory?

what the h is the use of theories?? its all conjecture.

[ADamiani 470]

36 minutes ago, wim123 said:

what the h is the use of theories?? its all conjecture.

Theories are the next best thing to factual truth.

If you are saying that you'd rather hear some clear statement from the programmers than our conjectures, well I am totally with you. I'd love to hear from them.

While we wait for some explanation for the odd things we are observing, we try to understand what's going on from the limited information we have. 

That's more or less how we humans got here: trying to understand what's going on from limited information.

Andrea

[ADamiani 470]

Oh and btw: I love the new water around Honolulu and Bahamas. Hopefully we are on the right way. 

Andrea

[steve 86 287]

the game has been downgraded some much its not the same looking game as it was before the first patch, 200% lod slider works out to be 90% of day one slider, autogen buildings in the uk got changed to low res models in patch 4, the clouds got down graded in patch 4, water got downgraded patch 3, you can see its all for vr, as when this game was released it could never run on VR with good fps. and this is why now they are talking about giving us more sliders to push it back up, the only people who are happy about the downgrades are the people who has a low/med range pc, 

 

[JRBarrett 3,034]

On 12/9/2020 at 3:17 PM, Jeeper said:

Sorry but this is hard to believe when Asobo themselves have acknowledged that they have reduced some visuals to improve performance. Perhaps your performance has improved, but visual quality at similar settings cannot have improved.

Can you provide a quote by Asobo for that? This question has been asked in at least two Developer Q&As and on both occasions they have denied deliberately reducing any visuals. I have never seen any statement by developers that indicate they have knowingly reduced visual quality.

On my relatively modest system, the visuals have improved since release, not degraded, and I have been using the sim since Alpha 4. After patch 2, I was having major problems with autogen popping when flying toward large urban areas at low altitude - that no longer happens. My test to check this is a short flight from Santa Fe to Albuquerque in the C172 at about 4000 feet AGL. In early release versions, as I approached the city from the relatively empty desert to the north, I would see entire neighborhoods suddenly pop into existence out of nothing as I descended to land at KABQ. Now all the buildings and houses are “complete” from 20 miles away.

In the early release, the terrain would shift from highly detailed to the base Bing satellite imagery at about 10,000 feet in a climb, now the detailed terrain remains visible until about 25,000 feet.

I just did a flight over the heavily forested hills of northern Pennsylvania from KELM to KIPT - a flight I have done many times in r/w, and there were visible trees literally as far as the eye could see. It’s not “realistic” for this time of year, because the trees are now devoid of leaves, and MSFS (like X-Plane) currently has “eternal summer”, but certainly no issues with tree LOD. 

The only widely reported graphics anomaly I do see is pixelation in the edges of clouds, but that is only obvious near sunset or sunrise when the clouds are backlit by a low sun angle.

I will agree that the water looked better in the initial release than in a couple of subsequent patches - especially the one that made the water look “mirror like”, but it is better now. I will take people’s word that water masks are missing. I don’t generally fly in areas that have water masks, so I don’t have any first hand experience with whether that is worse than before.

But in general, the sim looks and performs far better today on my system than it did at release. I’m not accusing those who see degradation of imagining things, but it is definitely not the case here.

[Jeeper 74]

32 minutes ago, JRBarrett said:

Can you provide a quote by Asobo for that? This question has been asked in at least two Developer Q&As and on both occasions they have denied deliberately reducing any visuals. I have never seen any statement by developers that indicate they have knowingly reduced visual quality.

On my relatively modest system, the visuals have improved since release, not degraded, and I have been using the sim since Alpha 4. After patch 2, I was having major problems with autogen popping when flying toward large urban areas at low altitude - that no longer happens. My test to check this is a short flight from Santa Fe to Albuquerque in the C172 at about 4000 feet AGL. In early release versions, as I approached the city from the relatively empty desert to the north, I would see entire neighborhoods suddenly pop into existence out of nothing as I descended to land at KABQ. Now all the buildings and houses are “complete” from 20 miles away.

In the early release, the terrain would shift from highly detailed to the base Bing satellite imagery at about 10,000 feet in a climb, now the detailed terrain remains visible until about 25,000 feet.

I just did a flight over the heavily forested hills of northern Pennsylvania from KELM to KIPT - a flight I have done many times in r/w, and there were visible trees literally as far as the eye could see. It’s not “realistic” for this time of year, because the trees are now devoid of leaves, and MSFS (like X-Plane) currently has “eternal summer”, but certainly no issues with tree LOD. 

The only widely reported graphics anomaly I do see is pixelation in the edges of clouds, but that is only obvious near sunset or sunrise when the clouds are backlit by a low sun angle.

I will agree that the water looked better in the initial release than in a couple of subsequent patches - especially the one that made the water look “mirror like”, but it is better now. I will take people’s word that water masks are missing. I don’t generally fly in areas that have water masks, so I don’t have any first hand experience with whether that is worse than before.

But in general, the sim looks and performs far better today on my system than it did at release. I’m not accusing those who see degradation of imagining things, but it is definitely not the case here.

Why would they admit to downgrading? That would be an admission of taking our money on features that have since been stripped down. There are lots and lots of threads in various flight sim forums with screenshot comparisons as well as many videos from respectable YouTube simmers showing the downgrades. I'm not going to link them as Google is sufficient for finding them.

Buildings in the UK have definitely been downgraded, and of course they haven't admitted to that either. They're still trying to sell copies so why would they publicly state that they're reducing the visual quality of the sim when it's basically the only thing FS2020 has over the legacy sims?

Edited December 11, 2020 by Jeeper

[JRBarrett 3,034]

 

27 minutes ago, Jeeper said:

Why would they admit to downgrading? That would be an admission of taking our money on features that have since been stripped down.

But in your initial post you said: “Sorry but this is hard to believe when Asobo themselves have acknowledged that they have reduced some visuals to improve performance.”

That is the statement I am questioning.

YouTube videos and screen shots are all well and good, but all I can go by is my own experience, on my own system, which I have been using with MSFS since June. 

Edited December 11, 2020 by JRBarrett

[cobalt 798]

On 12/9/2020 at 3:17 PM, Jeeper said:

Sorry but this is hard to believe when Asobo themselves have acknowledged that they have reduced some visuals to improve performance. Perhaps your performance has improved, but visual quality at similar settings cannot have improved. 

I just have to comment on this post, which is a great example of -- sorry to have to use a strong word -- arrogance, which . unfortunately is all too common in these forums. "Cannot have improved"? How can you tell anyone that they are not seeing what they say they are seeing? This has happened to me, for example when I posted that the clouds I was seeing in MSFS looked exactly like the real clouds I was seeing out my window, and was told no, they didn't! A famous scientist once said that before he published any new finding he always asked himself, how might I be wrong? We should follow that rule. A little humility would go a long way in these discussions.

 

[eslader 1,774]

11 hours ago, ADamiani said:

Theories are the next best thing to factual truth.

What?