• Announcements

    • turbofandude

      Proper Problem Reporting for the TFDi Design 717   07/09/2017

      Folks, As the user count is increasing, it is becoming more of a task for us to stay on top of problem reports. We are also receiving an increasing amount of incomplete reports that we are unable to follow up on (which frustrates everyone, us included). On that note, we are going to be enforcing a stricter formal problem reporting system. There are a few main points that need to be addressed. ALL problem reports must be submitted via the Bugs and Suggestions system. Open discussion on the forums is fine, but our team simply cannot respond to every single comment (especially when the majority of doing so is asking for information the Bugs and Suggestions system already asks for). Incomplete reports will be ignored. This means reports like "my FPS is bad" will not be given weight. This isn't because we don't care. It is because a report like that does absolutely nothing for anyone - we can't fix it based on that. Please include EVERYTHING you can possibly think of regarding the problem and take as many steps as you can to isolate it. Prepar3D v4, despite what you have heard, can still be unstable. Yes, it is definitely possible to fly flights without issues. But it is still relatively new, as are most of the addons available for it. Please refrain from attacking us or other developers because of this. We have already had one scenario where a 717 CTD was resolved by an update to Prepar3D itself. This is likely not the only time that will be the case. You are still free to report the crashes, as it may help whoever the developers of the faulting code are solve the issue. Duplicate reports will be deleted/merged. For the sake of sanity for us trying to prioritize and evaluate issues, if we discover multiple reports of the same thing, we may merge/delete the duplicates. We strongly encourage you to search and see if your problem has been reported before. Thank you all in advance. This may seem overbearing, but this is the only way we are going to be able to stay on top and continue heading in the direction we, and the community, want.
  • 0
cwdefouw

Displays going black during flight

Question

I have had this happen twice on the most recent version as of June 8, (1.0.9.0, I think). During the flight usually at around an hour in, the four leftmost displays suddenly go black, the two rightmost displays stay on with no issues. I can click the black displays and bring up the 2D windows, which display fine, which allows me to continue, but I don't think this should be happening, and I can't imagine it is a problem with my machine as two of the 6 displays do not exhibit the problem. After the first occurrence, I uninstalled the aircraft, downloaded a new copy of the TFDI Addon Manager Installer and reinstalled everything. Had the same problem on the next flight the next day. I did not have this problem before the last update. A screen capture is attached, system specs are below.

Thanks,

Chris DeFouw

System Specs:

Proc: I5 2570K

Graphics: Nvidia GTX970

RAM: 16GB

Conventional HDD

Prepar3d v3.3

 

 

717_DisplayFailure on 6_9_17.JPG

Share this post


Link to post
Share on other sites

98 answers to this question

Recommended Posts

  • 0

It everyone with this issue could do me a favor - if you could please measure your GPU memory usage while it's happening and report it, it may give me some more insight.

Share this post


Link to post
Share on other sites
  • 0
15 minutes ago, turbofandude said:

It everyone with this issue could do me a favor - if you could please measure your GPU memory usage while it's happening and report it, it may give me some more insight.

I'll do a flight tonight and get a reading but I've had it happen on my EVGA 1080 FTW and now my EVGA 1080ti FTW3. I've monitored the memory in the past and never got close to maxing it out. 

Share this post


Link to post
Share on other sites
  • 0
3 hours ago, turbofandude said:

By that logic, why is it not happening to me on the 717? Not arguing, just making the point. I'm not interested in blaming anyone or deflecting at all - that said, here is an image sent to us by a tester: https://gyazo.com/031b67d5e036a91195db9fe12ca6912d

OK. Then how about a link to a LM forum post where its shown to be their issue. Heard that more than once in this forum by people on the TFDi team. I love this aircraft and want to fly it but cant because what was once tolerable is no longer. Put up with all the problems since it's initial release. Worked around or with those issues. Over 1 year later I would like a aircraft with no major bugs. Not too much to ask I think.

Share this post


Link to post
Share on other sites
  • 0
1 minute ago, Irishcurse said:

OK. Then how about a link to a LM forum post where its shown to be their issue. Heard that more than once in this forum by people on the TFDi team. I love this aircraft and want to fly it but cant because what was once tolerable is no longer. Put up with all the problems since it's initial release. Worked around or with those issues. Over 1 year later I would like a aircraft with no major bugs. Not too much to ask I think.

The forum post is in a hidden developer forum on the LM forums. I could link it for you, but you wouldn't have access to that section of the forums.

Share this post


Link to post
Share on other sites
  • 0

Even if it’s just on the LM developers site you would think there would be a public posting of this issue in the LM, PMDG, FSLabs, Aerosoft, etc by many users if this problem was happening to all these other aircraft. I still think something is being missed in the 717 because once again this happens on more than one of P3D and didn’t happen with early releases of the 717.

Been a user of since day one and still have hopes for a fix.

Edited by Dan Moore

Share this post


Link to post
Share on other sites
  • 0
1 minute ago, Dan Moore said:

Even if it’s just on the LM developers site you would think there would be a public posting of this issue in the LM, PMDG, FSLabs, Aerosoft, etc by many users if this problem was happening to all these other aircraft. I still think something is being missed in the 717 because once again this happens on more than one of P3D and didn’t happen with early releases of the 717.

We haven't ruled out anything. But based on our troubleshooting and the reports, it generally sounds like something within the core code is causing this to happen. We're going to keep looking into why this is happening and why it seems so much more common with our aircraft, believe us, we're equally as frustrated by this. We want to get v1.1 out, but we're not going to release it to the general public with this bug in it.

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites
  • 0

Dan,

One of the users experiencing this hadn't modified his CFG at all, so unless another tool did it, I'm doubtful it's those specific tweaks. That said, thank you for the initiative and idea. LM seems to think it has to do possibly with drivers/GPU texture paging, hence my sudden focus on memory.

Share this post


Link to post
Share on other sites
  • 0

In my experience, P3D has always been very picky about GPU drivers.

It took me awhile to find a stable Nvidia driver for P3D3.4. But in P3D4 this very driver led to constant crashes. I've had to change five or seven versions before I found the stable one (and it wasn't the latest version at all).

So I won't be surprised if this bug is indeed caused by P3D, and would require some yet unknown hack or workaround to avoid it. It's not like Lockheed Martin have the best programmers in the world (after seeing their dynamic lights implementation, I think it's quite the opposite).

Share this post


Link to post
Share on other sites
  • 0
6 hours ago, turbofandude said:

Dan,

One of the users experiencing this hadn't modified his CFG at all, so unless another tool did it, I'm doubtful it's those specific tweaks. That said, thank you for the initiative and idea. LM seems to think it has to do possibly with drivers/GPU texture paging, hence my sudden focus on memory.

Ok, thanks for that information. I will attempt a flight this week and keep an eye on the gpu memory. I will also state my drivers and  video card used.

Share this post


Link to post
Share on other sites
  • 0
17 hours ago, turbofandude said:

It everyone with this issue could do me a favor - if you could please measure your GPU memory usage while it's happening and report it, it may give me some more insight.

1 hour 46 minutes in and up comes the problem with the 4 displays I've attached a few files that may help you track this down2018-01-18.png.7d5c30642105bfc6dce1cac434311a52.png5a60c3a05ff5f_2018-01-18(1).thumb.png.bf31c116f5bd8e308bc7cc0ae4a325ad.png

tfdi-display-error.xlsx

  • Thanks 1

Share this post


Link to post
Share on other sites
  • 0
8 hours ago, BrianT said:

1 hour 46 minutes in and up comes the problem with the 4 displays I've attached a few files that may help you track this down2018-01-18.png.7d5c30642105bfc6dce1cac434311a52.png5a60c3a05ff5f_2018-01-18(1).thumb.png.bf31c116f5bd8e308bc7cc0ae4a325ad.png

tfdi-display-error.xlsx

Hmm. Well, the GPU usage is nearly 100%, and the memory usage is over 90%. I wonder if that's what it is. Do you have hardware rendering on or off for the 717?

Share this post


Link to post
Share on other sites
  • 0

Here is mine. Running a EVGA 1080ti. Did a flight from Flightbeam KMSP and Imaginesim KATL (Latest Update). Departing KMSP my GPU load never exceeded 35%, 5700 Memory Used or 50c. Memory averaged 55xx - 56xx I flew over payware KBNA and GPU didn't climb. As I started arriving @ KATL, I was at 7000 ft, north west of the airport for an approach to 26L. As I turned east for a final descent my GPU load jump as the 4 displays went out. This happened at 1hr45min into the flight and I had the 717 loaded for 2 hours. Not sure if the spike was due to KATL loading.  You can see the spike when the displays went out in the photo.

KMSP ZMBRO7 ODI J30 BRIBE ENL PLESS J45 TINGS BNA NEWBB IHAVE MTHEW CHPPR1 KATL / FL330

Capture.thumb.gif.423da79ad163e7fbe6e8fcacf73bc0de.gif

Doing the same flight, same settings and conduction in the CS757 to see if I get the spike at the same location. So far at FL330 I’m only using 3285 Memory Used and 28% GPU load compared to the 717 in the first flight.

 

Edited by Dan Moore

Share this post


Link to post
Share on other sites
  • 0

Update, just finished the same flight from MSP to ATL in the CS757. During my 717 flight I tracked distance and time when I had the 4 displays go out. I just completed the the 757 flight and had no strange spikes in the same location. Memory also did not spike like the 717 when Atlanta came into view. During this flight GPU averaged 28% load and 3096 Memory. Only once Atlanta loaded that the memory slowly climb but it went back down to 3500 after the scenery completed loading. 

Atlanta.thumb.gif.388c5a3438b3edb6beaa31e644b5842e.gif

Share this post


Link to post
Share on other sites
  • 0
14 hours ago, turbofandude said:

Hmm. Well, the GPU usage is nearly 100%, and the memory usage is over 90%. I wonder if that's what it is. Do you have hardware rendering on or off for the 717?

Hardware rendering is ON for the 717. I'm running with the new dll you supplied last night and I'm about an hour into the flight I'll let you know how it goes

Share this post


Link to post
Share on other sites
  • 0

2 cents worth here. I would try monitoring the VAS usage. I don't recall if 3.3 benefits from the himemfix mod but that what it kinda acts like.

Share this post


Link to post
Share on other sites
  • 0
2 hours ago, BrianT said:

Just got to 2 hours and 5 min with the replacement DLL and it happened again I'm afraid

Capture.PNG

Did you get a spike or sudden jump like I did when the displays went out?

Share this post


Link to post
Share on other sites
  • 0
7 hours ago, Torkermax said:

2 cents worth here. I would try monitoring the VAS usage. I don't recall if 3.3 benefits from the himemfix mod but that what it kinda acts like.

I'm running P3d4.1 so I think VAS is out of the equation.

 

7 hours ago, Dan Moore said:

Did you get a spike or sudden jump like I did when the displays went out?

I didn't notice a jump in the when the displays went I should probably find a utility with a graph that might show it better.

Share this post


Link to post
Share on other sites
  • 0
2 hours ago, turbofandude said:

Just saw this on an Fb group -  https://gyazo.com/95fe5a347edd3548f313aea9cc1f9d1d. I suppose it's not just the drawn gauge surfaces that it can happen to. (There is apparently also an FSLabs forum post - this is good, if it starts gaining traction, many minds working on it is better than one.)

I definitely saw a screenshot of a PMDG 737 with the same bug on the panel below the screens, where brightness knobs are located. So I can confirm that this is not just gauge-related issue.

Share this post


Link to post
Share on other sites
  • 0
7 hours ago, dimkzr said:

I definitely saw a screenshot of a PMDG 737 with the same bug on the panel below the screens, where brightness knobs are located. So I can confirm that this is not just gauge-related issue.

We all saw that, TFDI posted it here. How can you confirm anything yourself when it was TFDI that linked the 737 screen shot? Are you even experiencing the problem or just confirming what has already been discussed? 

Share this post


Link to post
Share on other sites
  • 0
1 hour ago, Dan Moore said:

We all saw that, TFDI posted it here. How can you confirm anything yourself when it was TFDI that linked the 737 screen shot? Are you even experiencing the problem or just confirming what has already been discussed? 

You're right, I should have formulated it differently.
I'm simply backing turbofandude's assumption that this bug is not just gauge-related or TFDi-related.

Share this post


Link to post
Share on other sites
  • 0

Not sure if this was posted anywhere else, does disable VC Texture Mipmapping help this issue? (Sorry if this was already suggested, just saw this on another developers forum as a possible bandaid.)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now